Fullscreen HTML5Med Res (??MB)HTML4

The VERY BEST DIY Projects! Model Headboard Destination Display multiple compatible OLED display sizes; serial input or infrared control; 3.3V operation HF, SSB & QRP TRANSCEIVER TEST SET Terahertz Waves how they are used in astronomy, medicine and more Ultrasonic Cleaner ADJUSTABLE AUGUST 2026 ISSN 1030-2662 08 9 771030 266001 $ 00* NZ $1590 15 INC GST INC GST www.jaycar.com.au Contents Vol.39, No.08 August 2026 12 Beware: Fake Energy Savers Dodgy products are becoming increasingly common online, such as energysaving devices that are useless, fake pest repellers, battery scams etc. By Nicholas Vinen Fake and scam products 16 Terahertz Waves Laser imaging using terahertz electromagnetic waves (ranging from 0.1THz up to 30THz) is a new field that has applications in security, medicine and communications. By Dr David Maddison, VK3DSM Laser imaging Terahertz Waves Page 16 Part 2: p30 64 Power Electronics, Part 8 In this series of articles, we explore the principles of power electronics. In this final instalment we look at AC to AC converters, including cycloconverters. By Andrew Levido Electronic design 70 GM805 Barcode Reader This module can read linear (1D) barcodes, QR codes and other types of 2D codes via its integrated camera. By Tim Blythman Low-cost electronic modules 98 Is this the end of the NE5532? The NE5532, released in 1979, is the ‘classic’ audio op amp. Despite still being widely used today, it might be becoming obsolete. By Nicholas Vinen Electronic components 30 Adjustable Ultrasonic Cleaner Rated at up to 40W, our Ultrasonic Cleaner is fully adjustable for frequency, power and duration. In this final part, we cover all the construction details and how to test and tweak it. Part 2 by John Clarke Cleaning project 40 Phenomenal Pinball Machine This series explains how to design and build every part of your own Pinball Machine. This month we look at the deck and cabinet in detail, then start describing some of the 3D-printed parts required. Part 3 by Phil Prosser Gaming project 51 Destination Display This tiny display is small enough to fit onto a HO or N scale model train. It can be controlled with our microDCC Decoder, infrared remote control or by a serial port. It could also be used for station displays. Part 10 By Tim Blythman Model train project 74 Transceiver Test Set With a 9V rechargeable Li-ion battery, this Test Set is an all-in-one portable device for testing high frequency, ‘low power’ (QRP) single sideband (SSB) transceivers. By Andrew Woodfield, ZL2PD Test equipment project Adjustable Ultrasonic Cleaner The end of the venerable NE5532? Page 98 2 Editorial Viewpoint 4 Mailbag 39 Subscriptions 61 Circuit Notebook 82 Serviceman’s Log 88 Vintage Radio 96 Online Shop 101 Ask Silicon Chip 103 Market Centre 104 Advertising Index 1. Clap-operated switch 2. Auto-recharging battery backup 3. Router Watchdog enhancement 4. Safely measuring HV battery packs 5. Plugpack-powered Little Jim radio Baby Beethoven 555 by Dr Hugo Holden SILICON SILIC CHIP www.siliconchip.com.au Publisher/Editor Nicholas Vinen Technical Editor John Clarke – B.E.(Elec.) Technical Staff Bao Smith – B.Sc. Tim Blythman – B.E., B.Sc. Advertising Enquiries (02) 9939 3295 adverts<at>siliconchip.com.au Regular Contributors Allan Linton-Smith Dave Thompson David Maddison – B.App.Sc. (Hons 1), PhD, Grad.Dip.Entr.Innov. Geoff Graham Associate Professor Graham Parslow Dr Hugo Holden – B.H.B, MB.ChB., FRANZCO Ian Batty – M.Ed. Phil Prosser – B.Sc., B.E.(Elec.) Cartoonist Louis Decrevel loueee.com Founding Editor (retired) Leo Simpson – B.Bus., FAICD Silicon Chip is published 12 times a year by Silicon Chip Publications Pty Ltd. ACN 626 922 870. ABN 20 880 526 923. All material is copyright ©. No part of this publication may be reproduced without the written consent of the publisher. Subscription rates (Australia only) 6 issues (6 months): $77.50 12 issues (1 year): $145 24 issues (2 years): $270 Online subscription (Worldwide) 6 issues (6 months): $55 12 issues (1 year): $105 24 issues (2 years): $200 For overseas rates, see our website or email silicon<at>siliconchip.com.au * recommended & maximum price only Postal address: PO Box 194, Matraville, NSW 2036. Phone: (02) 9939 3295. ISSN: 1030-2662 Printing and Distribution: 1 Huntingwood Dr, Huntingwood NSW 2148 54 Park St, Sydney NSW 2000 2 Silicon Chip Editorial Viewpoint Finally, some open standards! For years, we have had the bizarre situation that the government requires workers to adhere to certain standards (such as AS/NZS 3000, the Wiring Rules), yet those standards are not freely accessible. You have to pay to access them, although there is a certain amount of limited access available without paying, as long as you register an account. This is the sort of thing that really should be provided as a service by the government. If they are going to require that you follow those rules, they should let you see the rules. It’s also very helpful for those who are not in the trades to be able to access the standards for a variety of reasons. For example, say you suspect that your wiring is substandard and you may need to bring in an electrician to fix it. How can you actually tell if it’s substandard if you can’t check the standard to compare it against? In most states in Australia, the average person is allowed to do things like replace plugs on mains cables, do work inside the chassis of electrical equipment and even build such devices from scratch. Even if you aren’t allowed to work on fixed wiring yourself, you certainly can work on things that plug into it, and logic says that doing so safely is easier if you understand the rules. So free access to electrical wiring rules can only be a good thing. Consider that New Zealand homeowners are permitted to do certain limited fixed-wiring work, although new or altered wiring generally has to be tested, certified and connected by a licensed electrical inspector. WorkSafe NZ does provide official guidance and a homeowner electrical code of practice, but the broader point remains: once a standard or code becomes part of the legal safety framework, access to the underlying rules should not depend on someone paying for them. I understand that maintaining these standards costs money, but so does having a police force, a fire brigade and many other government services that are paid for by tax revenue rather than being charged directly to users. There is also a rule-of-law issue here. Once a document is incorporated into legislation, it is no longer just a technical publication for specialists. It becomes part of the rules that citizens and businesses are expected to obey. In that case, free access should be the default, not a special concession. I’m pleased to see that the Australian Federal Government has announced that mandatory Australian Standards are to be made freely accessible through government-funded, sponsored access. In theory, this is a very good thing and will resolve most of the above problems. However, at the moment the whole thing is quite vague. They haven’t explicitly listed the standards that this will include, although it seems pretty certain that AS/NZS 3000:2018 will be included. We don’t yet know if related standards AS/NZS 3008 for cable selection, AS/NZS 3012 for construction sites, AS/NZS 3760 for test-and-tag or AS/ NZS 5139 for battery systems will be included in the free list. We also don’t know exactly what free access will be like. For example, it may allow you to read the standards on a web page but not download a PDF for use offline, which would be better than nothing but certainly far from ideal. Still, it’s a step in the right direction. These standards contribute to safety in general, and that shouldn’t be locked behind a paywall. by Nicholas Vinen Cover background image: https://unsplash.com/photos/green-and-yellow-light-digital-wallpaper-26WixHTutxc Australia's electronics magazine siliconchip.com.au MAILBAG your feedback Letters and emails should contain complete name, address and daytime phone number. Letters to the Editor are submitted on the condition that Silicon Chip Publications Pty Ltd has the right to edit, reproduce in electronic form, and communicate these letters. This also applies to submissions to “Ask Silicon Chip”, “Circuit Notebook” and “Serviceman’s Log”. More on analog computers I was interested to read the article in the May 2026 issue of Silicon Chip about Analog Computers (siliconchip.au/ Series/459). It made a brief reference to the Colossus computer (which was digital, and is shown below) that was used at Bletchley Park during World War 2 to determine the initial settings of the German Lorenz machine. It had some 2400 valves & electromechanical devices such as relays & uniselectors as used in electromechanical telephone exchanges. There is a reconstructed version at the Codebreaking Centre and Museum at Bletchley Park. Perhaps you could make it the subject of an article in Silicon Chip. Also at Bletchley, there is a reconstruction of the “Bombe” that was invented by Alan Turing to determine the initial settings of the German Enigma machine. Len Cox, Forest Hill, Vic. And even more on analog computers I have just read Dr David Maddison’s article in the May 2026 edition of Silicon Chip on the subject of Analog Computing and its historical roots (siliconchip.au/Series/459). The first known analog computer (pending any new archaeological discoveries to the contrary) is not the Antikythera Mechanism but the planetarium built by Archimedes and referenced in several ancient texts. In 2006, an astonishing archaeological discovery was made in Olbia, Sardinia (Italy) of a gear fragment that predated the Antikythera Mechanism by as much as 250 years. Its provenance was judged by experts to be consistent with the known history of Archimedes’ planetarium 4 Silicon Chip and its fate. For more information, see the website at siliconchip.au/link/accn The precise and mathematically accurate tooth-form was cycloidal, which is universally employed in technical measuring devices, not least of which are clocks and calculating machines, as opposed to the involute tooth-form, which is mainly used for power transmission. You may be assured that Archimedes, or whoever crafted the discovered fragment, didn’t stop at machining a single gear and certainly not to create an elaborate piece of decorative jewellery! Andre Rousseau, Auckland South, NZ. Channel 7 DTV has switched to MPEG4 I was surprised to see the letter on page 10 of the July issue regarding problems receiving Channel 7. I would have thought by now someone would have explained the reason. Simply put, Channel 7 has switched to MPEG4 and older receivers/TVs can only decode MPEG2. Your correspondent twigged to the solution but for the wrong reason! Most set-top boxes can decode MPEG4, so if connected to the TV, Channel 7 comes back. I discovered this by accident when I found that most of the TVs in the house I bought could receive Channel 7, but not one of the older TVs that I normally use as a security camera monitor. Ross Tester (former Silicon Chipper), Birtinya, Qld. Analog Devices is restricting part sales to Australia The “Ask Silicon Chip” section in your June issue contained a letter entitled “Analog Devices placing restrictions on parts”. I had the same problem trying to order the Australia's electronics magazine siliconchip.com.au LT8316 Micropower No-Opto Isolated Flyback Controller from Mouser because “Analog Devices won’t allow us to ship this product to Australia”! (Analog Devices bought Linear Technology in 2017.) I had to cancel the whole order of parts for my flyback converter project and re-place the order with DigiKey, who had no problems shipping the ICs to Australia. David Hanslip, Sorrento, WA. Comment: this makes us reluctant to use Analog Devices components in future projects, as we don’t want potential constructors to be unable to get the parts to build them! 40MHz oscilloscope giveaway I want to donate my CRO to a local hobbyist/amateur on the NSW Central Coast. It is in good working order; as I am now retired, it is no longer used. Paul Newman, Central Coast, NSW. Note: please email Silicon Chip if you are interested and we’ll forward it to Paul. A photo of the CRO is shown above. Another magazine giveaway I have been guided in my business by your publication but am now retired. I need the space, so the magazines have to go. They are from 1998 to 2024, mostly in binders, in six cartons. They are free for personal pickup in Perth only. Anthony Hopkins, Koongamia, WA. Comment: if you’re interested, email Silicon Chip and we’ll forward your email to Anthony. Difficulty reflow soldering the HCI PCB I recently purchased the Human Comfort Indicator kit (siliconchip.au/Article/20362). I was in a bit of hurry and didn’t notice it largely used SMDs. I only have a little experience with SMD soldering, but this presented an opportunity to use the Miniware mini hot plate (MHP30) I hadn’t yet tried out. It has worked out well and looks like a fairly good job. However, I paused as your article mentions after mounting the SMDs to perform the voltage test on CON1, and there were only a few millivolts, if any, so I investigated. I found the problem was underneath the USB connector, which I had mounted on the ‘short side’ for landscape use. It seems the board design extends the connector pin pads too far into the connector space and they short out on the metal connector casing. I tried re-mounting the connector 6 Silicon Chip (the hotplate is magic for desoldering/resoldering SMDs) so that it was further out but still on the mounting tab pads, but they still shorted out. So I re-mounted it around on the ‘long side’ of the board, and now it all works fine and the test voltage is as expected. Upon inspecting the board, it seems the relative pad dimensions between the USB connector pads and USB tab mounting pads are different on the short side of the board (for landscape use) compared to the long side. I’m looking forward to using this kit and adjusting the software to show barometric pressure and trend on the EPD. Much thanks for producing this interesting design and kit. Greg Newton, Greenwich, NSW. Comment: Tim built three prototypes by hand and none had this problem. The two USB socket footprints in the PCB design are identical as far as we can tell (apart from the shared pad in the corner where they overlap). The PCB design was optimised for manual assembly, meaning we enlarged some pads to make hand-soldering them easier. It seems like you found out that these changes made the situation worse for reflow soldering. We’ll try to avoid that problem in our future designs. We’re glad you got it working in the end. Advice on charging lead-acid batteries I note in the Serviceman’s Log last month (July 2026; siliconchip.au/Article/20469), your correspondent makes some comments on a sealed lead-acid (SLA) battery and his attempts to recharge it. I offer the following comments. Lead-acid batteries can be either flooded (liquid acid) or contained-acid types, with SLA batteries most often being absorbed glass mat (AGM) or gel cell in construction. Deep-cycle lead-acid batteries can have a cycle life of over 400 cycles, while starting, lighting and ignition (SLI) batteries can have a much shorter cycle life if deeply discharged. All lead-acid batteries should not be regularly discharged below 50% if a good cycle life is to be achieved. Older flooded-cell batteries have vents that can be opened and topped up as required with distilled water, but more modern ones have sealed vents so there is no access to the cells. Flooded-cell batteries are considered dangerous cargo and require precautions when being transported, while AGM and gel batteries are generally easier to transport. The charging requirements for lead-acid batteries are broadly similar, with a nominal 14.4V for boost charging and 13.8V for float charging a “12V” battery at 25°C. The first battery charger I built was in the early years of my apprenticeship (1964) and consisted of a transformer that I had to make coupled with a rectifier made of selenium plates. The first version had a single diode, so it was only half-wave rectified; a modification with more plates converted it to a full-wave rectifier. One advantage of the old, simple chargers was that they only required connection to the battery, regardless of the state of charge. Most modern chargers usually have a low-voltage lockout, nominally 6-9V, so if the battery is discharged below the charger threshold, the charger will not commence charging. A simple workaround is to jump-start the charger by making a temporary connection to the flat battery with a charged battery. Once the charger has commenced the charging cycle, the jump-start battery can be disconnected. Australia's electronics magazine siliconchip.com.au ATEM Mini Pro The compact television studio that lets you create presentation videos and live streams! Now you don’t need to use a webcam for important presentations or workshops. ATEM Mini is a tiny video switcher that’s similar to the professional gear broadcasters use to create television shows! Simply plug in multiple cameras and a computer for your slides, then cut between them at the push of a button! It even has a built in streaming engine so you can live stream to popular social media platforms without any extra streaming software required! Live Stream to a Global Audience Easy to Learn and Use Includes Free ATEM Software Control Panel There’s never been a solution that’s professional but also easy to use. Simply press any of the input buttons on the front panel to cut between video sources. You can select from exciting transitions such as dissolve, or more dramatic effects such as dip to color, DVE squeeze and DVE push. You can even add a DVE for picture in picture effects with customized graphics. ATEM Mini is a full broadcast television switcher, so it has hidden power that’s unlocked using the free ATEM Software Control app. This means if you want to go further, you can start using features such as chroma keying for green screens, media players for graphics and the multiview for monitoring all cameras on a single monitor. There’s even a professional audio mixer! Use Any Software that Supports a USB Webcam You can use any video software with ATEM Mini Pro because the USB connection will emulate a webcam! The software thinks the ATEM Mini is a common webcam, but it’s really a live production switcher. That guarantees full compatibility with any video software and in full resolution 1080HD quality. Imagine doing a presentation from a professional broadcast studio to software such as Zoom or Microsoft Teams! www.blackmagicdesign.com/au ATEM Mini Pro has a built in hardware streaming engine for live streaming to a global audience! That means you can live stream training and presentation videos direct to colleagues all over the world in better video quality with smoother motion. Streaming uses the Ethernet connection to the internet, or you can even connect a smartphone to use mobile data! ATEM Mini Pro .................................................................$475 ATEM Mini Pro ISO......................................................$795 ATEM Mini Extreme ISO G2 ........................$3,109 Learn More! It should be noted that the battery capacity reduces with usage and time. While a battery’s voltage can sometimes be recovered after a deep discharge, its capacity cannot. Regardless of the type of lead-acid battery, regular charging is recommended for a good working life. Gordon Dennis, Mill Park, Vic. Building the Calliope Amplifier with Japanese transistors I was interested in the Calliope Amplifier described in Silicon Chip (April 2026; siliconchip.au/Article/20084). After reading that article, I retrospectively went back to the Hummingbird Amplifier (December 2021; siliconchip. au/Article/15126), which somehow I had missed (or forgotten)! My interest was piqued by the reasons for doing a newer version of the amp due to the unavailability of the KSC3053 VAS transistor. Long story short, I just happen to have a large supply of Toshiba transistors that were ordered directly from Japan by a local, and now defunct, custom amplifier builder. Included in this inventory are some 2SC3423s, which would seem to be an ideal replacement, with a Cob figure of just 1.8pF. So I ordered a couple of Calliope PCBs. I thought perhaps I should try replacing all the transistors with the 2SA and 2SC ‘audio specialist’ Toshiba parts. The list below details the changes I made, which included all transistors except the BD139 Vbe multiplier, as the Toshiba equivalent had a different pinout, which didn’t allow it to be correctly mounted on the heatsink. A number of the other Toshiba products had different pinouts, so some bending of the component leads was required. Most prominently, this included the replacement of the BC556 and BC549 pairs which, as you aware and is mentioned in the article, ideally need to be matched and thermally coupled as much as possible, with 2SA1349 and 2SC3381 dual transistors, which seemed ideal substitutes. The unit worked the first go and, as best I can tell without the audio measuring equipment you have at your disposal, worked very well. One key characteristic that I was able to check, and that was discussed in detail in Phil Prosser’s article, was the sticking on the negative rail by the VAS when overdriven into clipping. The result with this amplifier seems to be at least as good as the original Hummingbird with the KSC3503. The amp is looking (and sounding!) good. I may even design a new version of the PCB to avoid the need for tedious component lead bending! The parts changes are: BC556 (pair)  2SA1349 BC549 (pair)  2SC3381 BC546  2SC1815 BC556  2SA1015 MJE350  2SA1837 KSC3503  2SC3423 MJE15032  2SC5171 MJE15033  2SA1930 MJW21194  2SC5200 MJW21193  2SA1942 Graham Bowman, Duncraig, WA. Comment: there’s certainly nothing wrong with Japanese transistors if you can get them. Many have excellent specifications. 8 Silicon Chip Australia's electronics magazine siliconchip.com.au “Now there is an alternative” ALL NEW, RECHARGEABLE CALIPERS! The long-lasting, rechargeable Lithium-Ion battery completely charges in under three hours, so you will never be stuck without a battery again! Available in three sizes 150mm, 200mm and 300mm! AVAILABLE FROM $89 ALL MODELS FEATURE: • Fast, magnetic charging • Metric, Imperial, and in less than 3 hours Fraction measuring modes • Long battery life, • Auto-Off function with low battery alert • 4-Way measuring ALL MODELS INCLUDE: • USB-A Magnetic Charging Cable • Protective Storage Case (dust and water resistant) CODE SIZE PRICE Q170 Q171 Q172 150mm / 6" 200mm / 8" 300mm / 12" $89 $99 $109 INC GST IP-67 (coolant proof) CODE SIZE PRICE Q175 Q176 Q177 150mm / 6" 200mm / 8" 300mm / 12" $129 $139 $149 INC GST SYDNEY MOVING* BRISBANE MELBOURNE (03) 9212 4422 (08) 9373 9999 PERTH ADELAIDE 1/2 Windsor Rd, Northmead 625 Boundary Rd, Coopers Plains 4 Abbotts Rd, Dandenong 11 Valentine St, Kewdale 11/20 Cheltenham Pde, Woodville (02) 9890 9111 (07) 3715 2200 (08) 9373 9969 06_SIC_270726 View and purchase these items online: www.machineryhouse.com.au/SIC2607 Specifications & Prices are subject to change without notification. *OUR BRISBANE BRANCH IS MOVING TO 680 BOUNDARY ROAD, RICHLANDS, 4077 (8.8KMS AWAY) FROM MID-SEPTEMBER High Energy Ignition System works well I successfully built the 2012 High Energy Ignition System from your magazine (siliconchip.au/Series/18). It worked well overall. The only tricky bit was soldering the wires to the PCB stakes; my soldering is a bit questionable there. I built it after buying the voltage regulator, ICs and PCB from your website. All the other bits and pieces I bought in-store from Jaycar. Soldering it was straightforward, but getting the IGBT attached is fiddly. Still, I’ve made a Jacob’s ladder before, so I knew the procedure. I have a waveform generator from Jaycar’s Short Circuit Book 3 that I’ve built in a UB3 jiffy box. I cranked it to 5V peak-to-peak at 100Hz and connected that to the trigger wire/PCB pin, and it works well. Interestingly, the article says a 5V trigger pulse is required, but I found you can dial down the voltage on the waveform generator to 3-4V and it still works. I used the VN Holden Commodore ignition coil from a Jacob’s Ladder kit and, as that is a wasted-spark type ignition coil with physically isolated primary and secondary coils, which I just tied together with a jumper wire. I also put a plastic tool (resistor lead bender) between the two secondary outlets so it didn’t spark between them. I then used those cheap auto parts ignition coil testers with adjustable gap; I reckon if I had a better setup with proper HT cables, it would jump the largest gap on the coil tester, so your kit must be pretty powerful at generating a spark. The IGBT doesn’t really even seem to get hot. Cheers for another great project and selling the key bits on your website. It kept me busy during several Friday nights in the shed. Edward Menzies, Kew, Vic. A possible PCB manufacturing error I had a reader call me who had the same problem with three Calliope amplifier modules he had built. I had isolated the fault to the bias generation – the modules had similar weird problems like a large DC output offset, with the base drive on the positive and negative output devices being +9V and -9V (the voltages should be much closer). He says that at least one board was missing the through-plating on the 100W resistor on the right-hand side, where it connects to the base of Q4. He has only soldered the bottom, so the top track was not connected. He thinks, but has not yet verified, that all the boards have similar problems. 10 Silicon Chip At least in this case it’s possible to solder the component lead on the top and bottom, thus taking the apparently missing through-hole plating out of the picture. Phil Prosser, Prospect, SA. Comment: it’s pretty unusual to have problems with commercially made PCBs, but it can happen occasionally. Luckily, we tend to buy Calliope amplifier boards in batches of 10, so if he has three faulty boards, there are unlikely to be more than seven others out there. They would all have been sold already. Hopefully, any constructors who run into a similar problem can also solve it by soldering the 100W resistor leads on both sides of the board. Ducted Heat Transfer Controller appreciation Firstly, thank you for a fantastic publication to which I have happily subscribed for many years, after having tired of badgering my newsagent since day one, and also discovered the financial benefits of subscription. The purpose of this letter is to thank you for the Ducted Heat Transfer Controller project (August/September 2025; siliconchip.au/Series/446). It has transformed the comfort level in my home for a second time and will be especially appreciated as winter encroaches. The first transformation, four years ago, was the installation of a chimney fan on our combustion stove. The house has a unique design comprising a lounge room with a cathedral ceiling and mezzanine rooms, then a lower-­ pitched dining area with a flat ceiling and a kitchen/amenities lean-to. Heat for the lounge room is supplied by a wood-burning heater, where most of the heat goes up into the cathedral roof space and mezzanine office, and very little manages to escape to the lower ceiling areas. Hot water was augmented by a cantankerous combustion stove named Bertha in the kitchen; she invariably filled the house with smoke until her horizontal outlet pipe and flue warmed sufficiently to draft properly. I always thought some kind of turbo was needed, and eventually came upon an Italian design, purported to reduce the risk of chimney fires. The resultant clean air and warmth transformed the livability of our home. However, after two chimney fires, I came to believe the stove (and our cheap hot water during winter) had to go. Serendipitously, along came the Heat Transfer Controller articles. Like all of my projects, after ordering parts from Silicon Chip (siliconchip.com.au/Shop/?article=18640), I dove in, completing and bench-testing it in an evening. Then came my lesser-preferred task of fitting it into a suitable enclosure. Then the task I have been putting off until the shadows have become longer and the autumn chill was in the air: installing the insulated ducts and fan kit, drilling holes in the ceiling, installing the wall plates and routing the cables. After twenty years of a sauna-like sitting room and a subzero kitchen, the home has been transformed once more. I love the automated feature and delight in watching the wall switch illuminate when the temperature differential meets the setpoint. No more swapping clothes from tropical island style in the office to Antarctic camp in the dining room. And it’s great to use the wood heater now more efficiently, as mentioned in the article. SC Ian Oldman, Budgeree, Vic. Australia's electronics magazine siliconchip.com.au Fake Energy Savers and other dodgy products This short article is a warning about fake and scam products that are becoming increasingly common online. They waste money, time and, in some cases, can even be dangerous. Warn your friends and family. By Nicholas Vinen S cams are unfortunately very common these days, partly because it has become easier to promote dodgy or outright fraudulent products online, and harder for consumers to tell what is genuine. As the saying goes, “If it sounds too good to be true, it probably is”. Keep that in mind when shopping online. Here are some products to watch out for. Fake energy-saving devices It’s very unlikely that you could save electricity by simply plugging a small device into a wall socket, so don’t bother buying these. They are a waste of money and generally do nothing useful. If anything, most of them will increase your power consumption slightly! As with many lies, there is a grain of truth behind the claims. Many of these devices claim to work by “cleaning up dirty power” or improving the power factor of your appliances. Large industrial customers do use power factor correction (PFC), and many consumer devices include PFC components. However, for ordinary domestic electricity users, improving the power factor generally does not reduce the real energy used, which is what you are charged for. So even if these plug-in devices could do the things they claim – and most cannot – they still would not save you money. Some scammers are even using fake celebrity endorsements to sell these products. Dick Smith was recently surprised to find himself supposedly promoting dodgy energy savers online – unwillingly! Don’t fall for it. Dick knows that you cannot save electricity by plugging a small, cheap box into the wall. It just doesn’t work that way. He has never promoted such devices, so if you see an ad saying he has, you know it is a scam. So how can you reduce electricity costs? Unfortunately, there’s no easy way. Real options include: 1. Buy more efficient appliances (but they often cost more). 2. Be more careful to switch things off when you aren’t using them. 3. Unplug “phantom” loads like chargers that aren’t actively in use. 4. Install a solar system. 5. Install a battery system that charges during off-peak/free times and discharges during peak times (even more effective if you have a solar system). 6. Shop around for cheaper electricity plans. 7. Improve your home’s insulation to reduce heating/cooling needs. Fake fuel-saving devices In the same vein as the fake electricity savers, you can find a device called the “Fuel Shark” or similar that plugs into your car’s cigarette lighter socket, claiming to reduce fuel consumption or increase engine power. It does neither. All that buying it will do is lighten your wallet. Fake pest repellers Similar to the fake energy-saving devices, these plug into a mains outlet and claim to repel pests like cockroaches, mice etc. There is some evidence to suggest that real ultrasonic pest repellers do work, but the cheap ones are similar to the fake “electricity saver” boxes – just a plastic box with an LED that lights up (if you’re lucky) and not much else inside. So if it seems too cheap to be true, it probably is. Real ultrasonic pest repellers are not exactly high-tech, but they generally cost at least $20 Dick Smith sent us this product that he bought online after seeing an ad claiming he was promoting it! He isn’t. Again it’s basically just a mainspowered LED. The large grey object that looks like a capacitor appears to be nothing more than an epoxy-filled plastic box. They couldn’t even be bothered to include a real capacitor! A photo of the back of the box is also shown, which tries to describe how it works. 12 Silicon Chip Australia's electronics magazine siliconchip.com.au The fake ultrasonic pest repeller that L. Ralph Barraclough warned us about in the April 2024 issue (page 6). All it does is light the LED; there’s no piezo transducer or speaker, so there’s no way it can produce ultrasonic sound. Similar items but slightly different-looking products are still for sale on eBay. These cells are about as likely to store 9900mAh each as I am to win a gold medal at the next Olympics. each, sometimes a lot more. So if you find one for $5, it’s likely to be a fake. Do your homework and don’t waste your money. Batteries with fake ratings Buying rechargeable batteries online can be a real gamble. Sometimes you can find a good product; other times, the capacity may be well below what’s claimed. Often, the claimed battery capacity is simply impossible. Even the best 18650 cell will have a capacity of maybe 3600mAh. So anything above that is likely fake. If you buy a cell with a claimed 9900mAh, expect to be disappointed! Smaller cells will obviously store less energy. Torches and LED bulbs with fake ratings Many torches and LEDs sold online have ludicrous lumen ratings. A good torch powered by a single 18650 cell might put out 800-1500 lumens sustained or slightly more in short bursts. Any claims above about 2000 lumens are probably fake unless the battery is much larger. Similarly, car headlamp bulbs sold online will often claim many thousand of lumens more than they can actually produce. Sometimes it seems like if the bulb isn’t selling well, they’ll just add another zero to the lumen rating. talking about blast-wave physics on a catastrophic scale. As for 600dB, that is well into “extinction-level event” territory, so we recommend avoiding those sirens! Power banks with misleading ratings advertised as “12V 3000mAh”. Unfortunately, that is not how physics works. Three 3.7V, 1000mAh cells in series give about 11.1V nominal at 1000mAh, or roughly 11Wh. The voltage adds, but the amp-hour capacity does not. You do not magically get 36Wh (12V × 3Ah) from three cells that only store about 11Wh in total. Caveat emptor. Fake antennas We published a letter from Peter Felton on page 10 of the September 2025 issue that includes a photo of a 4G 700MHz “Yagi” antenna he bought. It is just a piece of aluminium in the shape of an antenna. The wire doesn’t even connect to the elements – it’s just terminated in the junction box! Other fake products Many people have bought an 6000mAh power bank and complained that it can’t fully charge the 4000mAh battery in their phone. That’s because the power bank rating is measured at the 3.7V battery, not at the 5V output of the power bank. It doesn’t take into account conversion efficiency, cable loss, charging losses and many other things. If you mentally halved power bank ratings (assuming they’re honest – many are not!) you would be closer to the truth of what sort of capacity they can actually deliver to a battery under charge. Even worse, we’ve seen 12V devices, such as units using three 3.7V Li-ion cells in series, rated in mAh as though the cell capacities add together. For example, if the unit contains three 1000mAh 3.7V cells, it might be There are tons of fake or fraudulently marketed products online, including but not limited to radiation blockers, air purifiers, health analysers, anti-snoring gadgets, battery rejuvenators, solar chargers, WiFi boosters, car performance chips, audio enhancers, ground-loop eliminators and mosquito repellers. Our best advice is to stick to known brands from reputable retailers. AliExpress, eBay, Temu, Taobao and so on can be great places to snag bargains (and many of the products sold there are actually good) but you need to know enough about what you are looking for to spot fakes or deals that are simply too good to be true. Products are expensive these days enough without wasting your money on items that could never work, so be SC careful when shopping online. If these live up to their ratings, you won’t just be able to see the cars ahead – you’ll be able to see through them! Just $10 from AliExpress, and apparently loud enough to annoy people on other continents. Similar to when Krakatoa erupted. Australia's electronics magazine August 2026  13 Sirens with fake ratings Car horns and sirens are often sold on sites like eBay or Amazon claiming to produce sound at 300dB or even 600dB. For comparison, a firework going off next to you might produce a sound level of around 160dB. Around 200dB is more like standing next to a major explosion. At 300dB, you are no longer talking about a car siren; you are siliconchip.com.au GOT A BRIGHT IDEA? LET’S BUILD IT! Your next big build starts here. Jaycar’s wide range of prototyping accessories delivers great value for every idea. PB8815 HP9572 FROM 6 FROM 6 $ 50 $ 95 . Breadboard Layout Prototyping Boards 6 models available. PB8815 - PB8832 PB8820 QUICK AND EASY PROTOTYPING . Breadboard Layout Prototyping Boards 400 Hole HP9570 | 862 Hole HP9572 MAKE YOUR BREADBOARD PROTOTYPE PERMANENT WC6027 Breadboard Jumper Kit 70 Pieces PB8850 $12.95 150mm Jumper Leads MAKE YOUR OWN CIRCUIT BOARDS Blank Fibreglass Copper Sided PCBs • 4 sizes available HP9510 - HP9515 FROM $6.95 SHOP AT JAYCAR FOR: 8 x 25m Hook-Up Wire Rolls WC6024 - WC6028 FROM $8.25 26AWG WH3009 $39.95 MAKE PCBS IN 4 EASY STEPS 1. PRINT/COPY 2. IRON ON 3. PEEL OFF 4. ETCH 20 Piece Micro Drill Set • Sizes: 0.3 - 1.6mm TD2406 $16.50 • Soldering & Accessories • Components, Cables and Connectors • Magnifiers and Inspection Aids • Tools, Service Aids and Chemicals PBC Wash Defluxing Solution • 1 Litre Bottle NA1070 $15.95 $ JUST 4695 . Press 'n' Peel Film 5 sheets of 215 x 280mm transfer film with full instructions. HG9980 Explore our great range of soldering gear, in stock on our website, or at over 140 stores or 130 resellers across Australia and New Zealand. jaycar.com.au - 1800 022 888 jaycar.co.nz - 0800 452 922 All prices shown in $AUD, and correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. HP9570 THIS ISN’T A NORMAL 3D PRINTER. $ JUST 999 CENTAURI CARBON 2 COMBO TL4986 FOUR COLOURS. ONE PRINT. ZERO FUSS. PRINTS WHAT OTHERS WON’T. FAST. QUIET. DIALED IN. CANVAS: AUTOMATIC COLOUR SWITCHING BUILT FOR ADVANCED MATERIALS SMART CALIBRATION. CONSISTENT PRINTS. SEE THE FULL RANGE OF 3D PRINTING ONLINE jaycar.com.au 1800 022 888 jaycar.co.nz 0800 452 922 Prices shown in $AUD, and correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. Imaging via terahertz electromagnetic waves is a fascinating and rapidly emerging field that blends physics and engineering for practical applications in areas like security, medicine and communications. F or many years, progress in this area was limited by the difficulty of efficiently generating and detecting terahertz radiation, but recent advances have dramatically changed that situation. Terahertz waves occupy the ‘terahertz gap’ between mid-infrared and microwave radiation, sitting at the boundary between photonics (the realm of light and photons) and electronics (the realm of electrons and currents). A one-terahertz signal has a frequency of 1000GHz or 1012 hertz, corresponding to a wavelength of 0.3mm (see Fig.1). The exact frequency range considered as the terahertz band is somewhat flexible and depends on the context. Most researchers and applications define it as spanning roughly 0.1THz to 10THz (100GHz to 10,000GHz), although some extend it down to 0.03THz or up to 30THz depending on the field (eg, astronomy, communications or spectroscopy). For 6G research and future spectrum studies, ITU-R Recommendation SM.2352-1 extends the definition to the range of 100GHz to 10THz (100GHz to 10,000GHz; wavelengths of 3mm to 30µm). The ITU has also identified approximately 137GHz of spectrum space for potential terahertz communications, primarily in the range between 275GHz and 450GHz, allocated for land mobile and fixed service applications. The sub-terahertz range, 100300GHz, is also considered critical for future 6G development due to lower atmospheric absorption and better feasibility with current technology compared to the higher THz frequencies. Frequencies above 3THz are not considered radio waves. ITU Band 13, 3-30THz, is considered part of the optical or infrared spectrum and is free of international radio regulations. According to ISO 20473, frequencies of 0.3-6THz are also considered part of the far-infrared band, while 6-100THz is mid-infrared, 100THz385THz near-infrared and 380-780THz is visible light. Still, the exact boundaries vary according to the field of study. Terahertz radiation has historically been difficult to access until recent times, primarily due to the difficulty of efficiently generating and sensing it. This range is very high compared to regular radio frequencies but very low compared to normal optical frequencies. It is therefore one of the least explored and exploited parts of the electromagnetic spectrum. Terahertz radiation has actual or potential applications in security scanners, medical imaging, 6G telecommunications prototypes and industrial quality control. Early production of terahertz radiation Fig.1: the terahertz band and terahertz gap between microwaves and IR light. 3THz is generally considered the upper limit of radio waves. Although not true terahertz radiation, Sir Jagadish Chandra Bose of India was the first to generate and study extremely high frequency (EHF) waves, approaching the terahertz range. In the 1890s, he produced microwaves with frequencies as high as 60GHz (wavelength ~5mm) using a specialised spark-gap oscillator he designed and built himself. This innovative system generated much higher and more focused frequencies than those achieved by his contemporaries, including Hertz and Lodge. Bose demonstrated 60GHz waves during 1895-1897 – see Fig.2. His oscillator operated by applying a high voltage from an induction coil to a spark gap. A 3mm metal ball (sphere) was mounted on a non-conductive post between two smaller spark balls. Sparks jumped across the gaps, exciting standing waves within the central sphere at a resonant frequency determined by its size and placement. The entire setup was enclosed in a shielded box to minimise interference, with the waves directed outward through a short metal tube, an early form of waveguide. As Bose’s apparatus used a resonant structure to generate the radiation, it Australia's electronics magazine siliconchip.com.au Terahertz waves The International Telecommunication Union (ITU) designates the terahertz range as Band 12 (Tremendously High Frequency or THF), spanning 0.3THz to 3THz (300GHz to 3000GHz; wavelengths of 1mm to 0.1mm). 16 Silicon Chip was narrowband, producing frequencies tightly centred around the cavity’s resonant frequency. Bose’s meticulous engineering extended to other components, including horn antennas, dielectric lenses (made of sulfur or glass) to focus the beam like an optical lens, polarisers and crystal detectors. These enabled pioneering quasi-optical experiments on reflection, refraction and polarisation of radio waves in his small Kolkata (Calcutta) laboratory. How did he measure the frequency? Frequency counters did not yet exist. Bose determined the frequency he was generating using a reflecting diffraction grating made of evenly spaced metal strips (or wires) mounted on a frame. By directing his generated waves at the grating and observing the angles of the reflected diffraction maxima (bright spots), he could apply the standard diffraction grating equation to determine the wavelength, from which he could calculate the frequency. The frequency could also be determined from the size of the resonant oscillator sphere, as well as from interference patterns generated by standing waves. The methods Bose used, which would today be described as microwave optics, were forgotten and not rediscovered until the 1950s. In 1977, Nobel laureate Sir Nevill Mott remarked that Bose was “at least 60 years ahead of his time”. Following Bose’s work on 60GHz waves, the field stagnated for decades due to technological limitations. Bose’s record high of 60GHz was held for nearly 60 years. Discussion of terahertz radiation requires recognition of two types of electromagnetic radiation. In incoherent radiation, all parts of the wave are random in phase and spatially. Coherent radiation is when all parts of the wave are synchronised in phase and space, like a laser beam. The radiation Bose produced was coherent. Thermal (incoherent) blackbody terahertz emissions were detected in the 1960s, eg, using bolometers for astronomy. These absorb radiation, causing a temperature rise, which changes the resistance of a sensing element. Still, THz waves were not generated then. Bose’s 60GHz coherent radiation was not significantly surpassed until electronic sources like backward-wave oscillators (BWOs, we will discuss siliconchip.com.au Fig.2: Sir Jagadish Bose’s spark gap oscillator, waveguide and galena point contact crystal rectifier. Bose patented the galena rectifier in 1901. Source: https://w.wiki/7DAY later) were invented in 1951 by Rudolf Kompfner at Bell Labs in the United States (the O-type BWO). The Soviets quickly advanced this technology, with early BWOs (also known as carcinotrons) reaching millimetre waves (up to about 300-500GHz) in the 1950s and 1960s for military and scientific applications. In the 1970s, Soviet submillimeter BWOs extended to about 0.5-1.0THz, supporting spectroscopy and plasma diagnostics. The first phase-locked loop submillimeter BWO was developed in the USSR around 1970. By the late 1970s and 1980s, the highest reported Soviet BWO frequencies approached 1-2THz in laboratory setups, with papers from the Institute of General Physics (USSR Academy of Sciences) describing BWO-based spectrometers covering 0.03-1.5THz (30-1500GHz) by the mid-1980s. Post1990s, Russian BWOs (eg, from Istok) continued to offer reliable commercial sources up to ~1.25THz, but the 1-2THz experimental frontier was largely achieved during the Soviet era. The Soviets were also interested in military applications of THz waves but there are no reliable reports that any eventuated, primarily due to the low powers and high atmospheric absorption of THz waves. Broadband and narrowband THz sources Soviet BWOs produced narrowband, tuneable continuous-wave THz Australia's electronics magazine radiation. This was excellent for spectroscopy in the frequency domain at specific frequencies (see the panel overleaf) but provided limited spectral coverage. For time-domain spectroscopy, imaging and sensing, what was needed was broadband THz generation that could produce a spectrum of frequencies in a single pulse with no single dominant frequency. A major leap into coherent, broadband THz waves came in the mid1980s to early 1990s with laser-based methods. David H. Auston and collaborators at Bell Labs/AT&T (and later at Columbia University) demonstrated the first practical, coherent pulsed THz radiation in 1984-1988, using photoconductive antennas (see later) excited by femtosecond (10-15s) laser pulses. This produced broadband THz pulses (typically 0.1-3THz or higher) with picosecond durations, marking the breakthrough that opened the modern THz era. The pulsed nature of the radiation produced by Auston enabled time-­ domain spectroscopy (THz-TDS), measuring both the amplitude and phase of the THz field in the time domain, which provides rich information about material properties such as refractive index, absorption and complex dielectric function in a single measurement. This technique has applications in materials science, imaging and non-destructive testing. August 2026  17 Apart from generating THz radiation, Auston’s photoconductive antenna could also detect the THz pulse coherently (time-gated sampling), allowing full reconstruction of the waveform, a key enabler for modern THz systems. Auston’s innovation was practical because it used relatively compact femtosecond lasers (then emerging) rather than the large vacuum tubes, magnetic fields and high voltages required by BWOs, making THz waves more accessible to labs worldwide. Today, broadband THz sources (modern photoconductive antennas, optical rectification, or air plasma generation) produce a continuous spectrum of frequencies in a single pulse (over 0.1-10THz or more), with no single dominant frequency. Auston opened the door to coherent broadband pulsed THz, which became the foundation for the THz time-­ domain spectroscopy, imaging and sensing that dominates the field today. Terahertz analysis in the time and frequency domains Terahertz radiation interacts with molecules in materials by exciting lowenergy resonances of large molecular structures, forming the basis of THz spectroscopy. Alternatively, THz waves can pass directly through a material for imaging. Spectroscopy and imaging can be performed in either the time or frequency domain. In time-domain spectroscopy, extremely short broadband pulses of THz radiation are used to record a waveform (electric field intensity as a function of time) that contains all frequencies simultaneously. Applying a Fourier transform to this time-domain data yields the frequency-domain spectrum (see p14 of the January 2026 article for more on Fourier transforms). Time-domain spectroscopy offers moderate spectral resolution and is commonly used for imaging, material characterisation and thickness measurements in quality control. In frequency-domain spectroscopy, a continuous-wave (CW) THz source is tuned or swept across frequencies, producing very high-resolution spectral data. This approach is very useful for gas sensing, chemical analysis and precise identification of ‘molecular fingerprints’. There are several reasons why terahertz radiation has been difficult to generate and therefore was a virtual scientific and technological no-go zone. However, these difficulties are being overcome. Conventional electronic devices like transistors have difficulty operating at terahertz frequencies because electron transit times (drift and diffusion) through the semiconductor become too long. Capacitive effects and gate resistance dominate, limiting the oscillation frequency. Some specialised transistors such as InP (indium phosphide), HBTs (heterojunction bipolar transistors) and HEMTs (high electron mobility transistors) can reach or exceed 1THz, but often with low power due to thermal limits and low efficiency. Typically, they can only handle less than 10mW, although over 100mW has been demonstrated with an InP HEMT power module at 670GHz. Diodes have difficulties at such frequencies due to the RC time constant of the device, the skin effect, ohmic losses at high frequencies, thermal management, uneven field distribution and other effects. Optical methods such as lasers have difficulties in the same range due to material absorption and phase-­ matching within non-linear crystals. A non-linear crystal is a crystal that, when hit with a higher-frequency optical laser, generates lower-frequency terahertz waves as the laser travels through the crystal. As the laser beam travels through the crystal, terahertz waves are further generated at every point along the laser’s path. For these individual sources to add up constructively, to generate maximum power, the peaks and troughs of the waves must be matched through phase-matching by appropriate adjustment of the crystal Fig.3: a photoconductive antenna. “j” is the photocurrent and “V” the bias voltage. Source: Paul C. Gow Fig.4: generating THz radiation via optical rectification in a non-linear crystal (ω is frequency). Original source: Sonal Saxena Generating terahertz radiation 18 Silicon Chip Australia's electronics magazine properties, such as through orientation of the crystal or laser pulse. Examples of such crystals include complicated organic crystals like DAST Z (4 -dimethy lami no- N-­ methyl-4-stillbazolium tosylate), OH1 (2-(3-(4-hydroxystyryl)-5,5-­ dimethylcyclohex-2-enylidene) malononitrile) and inorganic crystals like GaAs (gallium arsenide) and ZnTe (zinc telluride). There are no known natural materials that resonate at these frequencies. For example, microwaves (ITU Band 9, 300MHz to 300GHz) interact with large-scale rotational modes of molecules in materials like water, enabling the cooking of food. Microwave generation is also simple, as electrons in a magnetic field gyrate at microwave frequencies (cavity magnetron). Infrared radiation (ITU Band 13 and 14, 3-300THz) interacts with the vibrational modes of molecular bonds, making infrared a powerful tool for spectroscopy. Terahertz waves sit in an awkward middle ground: too high for efficient electronic oscillators like transistors or siliconchip.com.au Table 1 – key characteristics of terahertz generation methods Method Type Main advantage Main limitation Photoconductive antenna Pulsed (broadband) High signal-to-noise ratio (SNR), table-top Low average power Optical rectification Pulsed (broadband) High peak power Requires complex laser Spintronic emitters Pulsed (broadband) Extremely high bandwidth, inexpensive Emerging technology Air plasma generation Pulsed (broadband) Ultra-broadband and high intensity Low efficiency, complex setup, needs expensive laser Photomixing CW (narrowband) Tuneable, high-precision Low power output Resonant tunneling diodes CW (narrowband) Compact, room temperature Low power, low frequency IMPATT diode Pulsed or CW (narrowband) Up to 10THz; high power up to 100-200GHz High noise, low efficiency, thermal problems, limited tuneability Quantum cascade lasers CW (narrowband) High power, compact Cryogenic cooling Gyratron Pulsed or CW (narrowband) High power and coherent Large, heavy, needs superconducting magnets Backward wave oscillator CW (narrowband but tuneable) Tuneable, high power, low noise Low efficiency, complex fabrication, high voltage Frequency multiplier – schottky diode CW (narrow to moderate bandwidth) Room temperature, reliable, compact, solid state, up to 2THz, suitable for space and handheld Low power at high frequencies, limited power handling Free-electron laser (FEL) Pulsed (broadband), coherent Very high power Large and expensive Synchrotron Pulsed (broadband), less coherent than FEL High power Large and expensive magnetrons, too low for most molecular vibrations, and without natural material-specific resonances to exploit. We will now look at the production of THz radiation by optoelectronics, electronics, frequency multipliers and other methods. Table 1 summarises the key characteristics of the various methods. Optoelectronic methods – laser-based Photoconductive antennas (Fig.3), also known as “Auston switches” after the person who invented them, generate broadband terahertz waves by directing femtosecond laser pulses onto the gap of a biased semiconductor. The gap is the narrow space (typically a few µm wide) between the two metal electrodes on the semiconductor surface where the bias voltage creates a strong electric field. The intense light pulse creates electron-­hole pairs (photogeneration). Under the influence of an applied DC bias voltage, these charge carriers accelerate, producing a rapid transient photocurrent. This short-lived current siliconchip.com.au acts as a Hertzian dipole radiator, a small oscillating dipole that efficiently radiates electromagnetic waves, here in the terahertz range. It emits broadband THz radiation (typically 0.1-3THz, or higher) in a single pulse. The current then decays rapidly as the carriers recombine, ending the emission for that laser pulse. Optical rectification is one of the most common methods for generating broadband terahertz waves. An intense femtosecond laser pulse is passed through a non-linear crystal (such as ZnTe, GaP, LiNbO3, or organic crystals like DAST, OH1 or BNA [N-benzyl-­2methyl-4-nitroaniline]). A non-linear crystal emits light of a different frequency or polarisation from the input. The strong electric field of the pulse induces a non-linear polarisation in the crystal that follows the pulse’s intensity envelope. Because the broadband femtosecond pulse contains a broad range of optical frequencies, different frequency components combine to form new frequencies (beat frequencies) within its spectrum, generating new coherent oscillations at terahertz frequencies, typically Australia's electronics magazine in the range of 0.1-5THz (up to 10THz or more with optimised crystals). These oscillations radiate broadband THz pulses directly from the crystal (see Fig.4). Most laboratory-­ based terahertz time-domain spectroscopy (THz-TDS) systems rely on this technique due to its simplicity, coherence and broad spectral coverage. A spintronic terahertz emitter converts ultra-short flashes of laser light into terahertz waves by harnessing the spin (a magnetic property) of electrons. It works in three main steps (see Figs.5 & 6): 1. A femtosecond laser pulse strikes a thin bilayer structure consisting of a ferromagnetic metal (eg, iron, cobalt, or nickel) and a heavy metal with strong ‘spin-orbit coupling’ (eg, platinum). Spin-orbit coupling is a magnetic interaction where an electron’s internal spin is locked to its physical movement, causing it to turn sideways when it flows through certain metals. The laser excites electrons in the ferromagnetic layer to higher energy levels. 2. In the magnetic layer, electrons with one spin direction are more August 2026  19 Fig.5: the process of THz emission from a spintronic bilayer after a laser pulse. js – spin current in the z direction; jc – charge current in the y direction; M – magnetic field; FM – ferromagnetic; NM – non-magnetic. Source: www. degruyterbrill.com/document/doi/10.1515/nanoph-2020-0563/html Fig.6 a commercial spintronics Terahertz emitter from TeraSpinTec. Source: https://teraspintec.com/en/ products mobile and preferentially diffuse into the heavy-metal layer. This movement creates a flow of spin (a ‘spin current’) that carries over into the heavy metal layer. 3. In the heavy metal, the inverse spin Hall effect makes the moving electrons suddenly deflect sideways, turning the spin current into a regular electric current. Any fast-moving electric charge produces electromagnetic radiation, so this sudden sideways motion emits broadband terahertz waves. The emitted radiation is broadband (typically 0.1-30THz) and its polarisation can be easily controlled by rotating an external magnet. These emitters are simple to fabricate, using only common metals in layers a few nanometres thick, making them inexpensive and scalable for practical applications. Air plasma generation of terahertz waves works by focusing powerful femtosecond laser pulses into (targeting data rates of 100Gbps or higher), high-resolution spectroscopy and terahertz imaging systems. air, ionising the gas to form a plasma filament. The plasma’s free electrons oscillate under the laser’s electric field and radiate broadband terahertz waves as they accelerate, typically spanning 0.1-10THz or more. A continuous-wave (CW) terahertz source can be produced by photomixing (or optical heterodyne mixing). Two single-mode lasers operating at slightly different frequencies are combined to create a difference frequency (beat frequency) in the terahertz range. These two laser beams are directed onto a high-speed semiconductor (typically a photoconductive material like InGaAs or GaAs), where the beat frequency induces a modulated photocurrent oscillating at the THz difference frequency – see Fig.7. An integrated antenna on the semiconductor radiates this oscillating current as a continuous terahertz wave. The main applications of this method include proposed 6G communications Electronic methods – oscillator-based A resonant tunnelling diode (RTD) is a specialised high-speed semiconductor device that uses quantum mechanics to generate and detect terahertz radiation. Technically, it is a ‘double-barrier quantum well’ (DBQW). Imagine a microscopic structure with two thin insulating ‘walls’ (barriers) and a narrow ‘well’ of semiconductor between them. In classical physics, electrons could not cross the barriers. However, thanks to quantum mechanics we know that if an electron’s energy matches one of the specific energy levels in the well, it can tunnel through the barriers with a high probability. This is called resonant tunnelling. Fig.9: the structure of an RTD transmitter chip. Original source: www.mdpi.com/2076-3417/12/8/3822 20 Silicon Chip Australia's electronics magazine siliconchip.com.au When a voltage is applied across the device, the current initially increases as the electron energy aligns with the well’s energy level. As the voltage increases further, the alignment is lost, and the current drops sharply, creating a region of negative differential resistance (NDR), where more voltage causes less current. This negative resistance allows the device to sustain oscillations when connected to a resonant circuit (such as an antenna or cavity), producing THz waves. RTD THz emission relies on the NDR region of the I-V curve, where the device provides gain to sustain oscillations in a resonant circuit – see Fig.8. While weak or induced radiation can occur elsewhere on the curve, practical, high-performance THz generation and oscillation are effectively limited to the NDR region. Tunnelling occurs predominantly at the discrete energy levels E1 and E2. The current peaks are sharp and localised because resonance only happens when there is good energy matching with one of those discrete states. State-of-the-art RTDs can achieve frequencies up to about 2THz in the milliwatt range. The structure of a practical device is shown in Fig.9. An IMPATT diode (IMPact ionisation Avalanche Transit-Time diode) generates terahertz radiation by exploiting two key phenomena (see Figs.10 & 11): a. High-frequency avalanche breakdown (also called impact ionisation), where energetic charge carriers collide with atoms in the semiconductor lattice, creating additional electron-hole pairs in a rapid cascade. b. Transit-time phase delay, where the time it takes these carriers to travel across the diode’s active region is carefully tuned to be approximately half a cycle of the oscillation frequency. Fig.10: the structure of IMPATT diode. P+ is heavily doped P-type material, N is N-type material, I is intrinsic undoped material, and N+ is heavily doped N-type material. siliconchip.com.au Fig.7: the process of generating terahertz waves using photomixing; PCS stands for photoconductive switch. Original source: https://w. wiki/HyHN Fig.8: the structure of a typical RTD. The energy levels during which tunnelling can occur are around E1 and E2. THz radiation is emitted in the blue region of the plot, while the device can act as a receiver in both the blue and tan regions. Original source: www.mdpi.com/2076-3417/12/8/3822 Together, these effects produce negative differential resistance, a region where an increase in voltage causes a decrease in current, which sustains oscillations when the diode is placed in a resonant cavity (a tuned electromagnetic structure that reinforces the desired frequency, similar to how a guitar string resonates at a specific pitch). A quantum cascade laser (QCL) is a semiconductor laser designed to generate light in the terahertz range, typically 1-5THz, through a cascading process of electrons in an engineered multi-layer quantum structure. In a conventional semiconductor laser, an electron transitioning from a higher to a lower energy level emits a single photon. In a QCL, the active region consists of many repeating stages (often 30-100 or more). An electron emits one photon per stage as it drops to a lower energy level, then tunnels through a thin barrier into the next stage to repeat the process – see Fig.12. Thus, a single electron can generate 30-100 or more photons as it cascades through the entire structure. Most THz QCLs require cryogenic cooling (liquid nitrogen temperatures) to operate efficiently, although recent advances have achieved room-­temperature performance at lower frequencies. These devices produce narrowband, highly monochromatic terahertz radiation with excellent spectral purity, making them ideal for spectroscopy, sensing and high-resolution applications. Work on QCLs is being undertaken in Australia at the Uni of Qld. Fig.11: the assembly of an IMPATT diode into a cavity resonator and then into a power supply/heatsink module. Source: https://terasense.com/ terahertz-technology/impatt-diodes Australia's electronics magazine August 2026  21 Fig.14: the atmospheric absorption of terahertz waves as a spectrum. In this case, the ITU extended definition of Band 12 is used: 100GHz to 10THz. Source: www.researchgate.net/figure/fig3_337266839 A gyrotron (see Fig.15) is a highpower vacuum tube device invented in 1964 in the USSR that generates THz radiation by the ‘cyclotron resonance’ of electrons in a strong magnetic field. Output frequencies range up to about 500GHz, with output powers from tens of kilowatts to 1-2MW. Electrons are emitted from an electron gun and accelerated by a high-voltage DC anode. They then enter a resonant cavity immersed in a strong axial magnetic field produced by a superconducting magnet. This field forces the electrons of the beam to spiral helically around the magnetic field lines at the cyclotron frequency. As the electrons gyrate, they emit electromagnetic radiation (cyclotron emission) at frequencies determined primarily by the magnetic field strength (the cyclotron frequency), with relativistic effects introducing additional spectral broadening. When the cyclotron frequency matches a resonant mode of the cavity, standing waves build up coherently, amplifying the radiation. The resulting high-power THz beam (often in the 0.1-1.5THz range, with some devices reaching 2-3THz) is emitted from the cavity, then converted by a mode converter (where the pattern of the wave is converted to one that can be more efficiently used) to a suitable output mode, reflected by a mirror and directed through a vacuum window into a waveguide or free space. The electron beam is absorbed by a collector. Backward-wave oscillators (BWOs) are vacuum-tube devices that generate high-frequency, narrowband, tuneable terahertz (or sub-terahertz) radiation, either continuous-wave or pulsed. They work by firing a high-velocity electron beam through a slow-wave structure (such as a helix, folded waveguide or grating), which reduces the phase velocity of the electromagnetic wave (the speed of its crests) to match the electron beam velocity – see Fig.13. The electrons interact with this backward-propagating wave, losing energy to amplify it through bunching and collective effects. Because the wave travels opposite to the electron flow, it provides internal feedback that sustains self-oscillations at terahertz frequencies (typically up to ~1-2THz in advanced designs). The output is inherently coherent, Australia's electronics magazine siliconchip.com.au Fig.12: an energy level diagram showing emission of four photons from one electron. Original source: www.teamwavelength.com/using-a-laserdiode-or-quantum-cascade-laser-dontforget-the-electronics Fig.13: a BWO; K = cathode, G = grid, A1 = first anode, A2 = second anode. Original source: https://w.wiki/HyHQ 22 Silicon Chip stable and well-suited for applications such as spectroscopy, interferometry and as local oscillators. Frequency multipliers Schottky diodes can act as a frequency multiplier to produce THz waves by taking a lower frequency signal and converting it into a higher one. This is because the schottky diode is a strongly non-linear device, meaning it distorts the input signal, creating many higher harmonics at odd multiples of the input signal frequency. The fundamental and other harmonics can be filtered to isolate the desired THz signal. Efficiency is reduced as higher harmonics are used, so chains of diodes are often cascaded to produce the desired frequency. Schottky diodes are preferred because, as majority-­carrier devices, they exhibit negligible reverse recovery charge, along with very low junction capacitance and strong non-linearity at high frequencies. Other methods Free electron lasers (FELs) can produce terahertz radiation using a high-energy electron beam that passes through a series of magnetic ‘undulators’ or ‘wigglers’. These alternating magnetic fields force the relativistic electrons to travel in a sinusoidal (wiggling) path, causing them to emit electromagnetic radiation at a frequency determined mainly by the beam energy and the magnetic period – see Fig.16. The emitted radiation is coherent and can reach very high power levels (watts to kilowatts average, or megawatts peak in pulsed systems). Synchrotrons can also produce terahertz radiation in a manner somewhat similar to free-electron lasers (FELs). A relativistic electron beam passes through an undulator or wiggler, causing it to oscillate and emit radiation at the desired wavelength. However, in a synchrotron, the electrons are accelerated and stored in a circular storage ring rather than a linear accelerator, and the radiation is generated primarily from the bending magnets or edge effects. This results in broadband, incoherent THz emission, unlike the narrowband, coherent output of FELs. For further information on synchrotrons, see our article in the May 2012 issue (siliconchip.au/ Article/671). Atmospheric absorption of terahertz waves Fig.14 shows atmospheric attenuation of electromagnetic waves between 10GHz (microwaves) and 1000THz (UV light) for a range of atmospheric conditions. It includes the absorption peaks of atmospheric species such as water vapour, oxygen and carbon dioxide. Based on atmospheric absorption characteristics, the best bands for communications applications are 275320GHz and 335-360GHz, both with losses under 10dB/km (see Table 3). A comparison of the relative advantages and disadvantages of various communication frequencies from microwaves up is shown in Table 2. Detecting terahertz radiation Detectors need to be sensitive to the low photon energies of terahertz waves in an environment where there is high noise and atmospheric absorption, especially by water vapour. Direct detection Bolometers (Fig.17) are broadband thermal detectors that sense THz radiation by measuring resistance changes caused by heating in an absorbing element. They are often paired with antennas to improve radiation coupling or use thin absorbing membranes in MEMS (micro-electromechanical systems) structures to reduce thermal mass and increase sensitivity. Fig.16: a free electron laser. Original source: https://w.wiki/HyHR electrically conductive pillar membrane thermo resistor material thermal insulation using narrow beams ("arms") electrically conductive pillar metallic reflector film Fig.15: the structure of a gyrotron. Original source: https://w.wiki/ HyHP siliconchip.com.au substrate Fig.17: one pixel of a microbolometer array imaging system (MAIS) for medical applications to detect 1-5THz signals on the body’s surface. Source: https://w.wiki/HyHS Australia's electronics magazine August 2026  23 Cryogenic cooling is essential for the highest-performance bolometers, enabling low-noise operation and detection of very weak signals. Bolometers can be assembled in arrays for medical imaging. Pyroelectric detectors generate a voltage when electromagnetic radiation, such as THz waves, impinges on them, as the absorbed energy causes a temperature change that alters the material’s polarisation – see Fig.18. They are a common choice for THz detection because they operate at room temperature, are broadband and require no cooling. Golay cells are highly sensitive, uncooled detectors that operate on a pneumatic principle and are capable of detecting terahertz radiation. THz radiation enters through a window transparent to these wavelengths and heats a small quantity of gas inside the cell, causing it to expand. The expanding gas increases the pressure, which deforms a flexible membrane – see Fig.19. This membrane movement is then detected and measured optically (eg, via a reflected light beam). Golay cells can operate across a broad frequency range up to about 20THz. They offer very high sensitivity and require no cooling, but they are highly susceptible to vibration and mechanical noise. They are commonly used in spectroscopy, astronomy and precise power measurements. Rectifying detectors such as schottky barrier diodes and special field-effect transistors (TeraFETs) can be used to rectify and convert THz signals to DC voltages for detection and measurement. Other detectors Extrinsic detectors use impurities (dopants) added to a semiconductor to create extra energy levels, allowing the material to absorb THz photons and produce a measurable signal. Josephson junction detectors are based on superconducting Josephson junctions. These extremely sensitive devices detect THz radiation by measuring tiny changes in the junction’s current-voltage characteristics caused by the incoming THz field. They offer very high sensitivity but require cryogenic cooling. Heterodyne detection involves mixing the incoming THz signal with a local oscillator (another THz source) to produce a lower-frequency difference signal that is easier to measure. It provides precise amplitude and phase information, similar to the photomixing generation method described earlier, and is widely used in high-­ resolution spectroscopy. Terahertz applications Some of the applications for terahertz laser imaging include: Astronomy Due to atmospheric absorption of THz waves, observatories need to be on the highest mountains or in balloons, high-altitude aircraft or spacecraft. Terahertz radiation is emitted by astronomical objects above about 1.7K and the frequency depends on the temperature. At frequencies where the most thermal emission occurs, 0.1THz corresponds to a temperature of about ~1.7K (-271.5°C), 1THz to ~17K (-256°C) and 10THz to ~170K (-103°C). This enables the observation of the ‘cold universe’, regions too cool to emit strong infrared but too warm to be dominated by radio emission. Terahertz waves are able to penetrate interstellar dust and gas, allowing the following observations to be made: 1 The interstellar medium: the cold, dense molecular gas and dust clouds Fig.18: an experimental pyroelectric sensor for THz radiation with circuit diagram. Original source: www.mdpi. com/2076-3417/14/10/3967 24 Silicon Chip Australia's electronics magazine within the Milky Way and beyond 1 Star and planet formation, mapping the dense cores of gas and dust clouds and the detection of protoplanetary disks 1 Astrochemistry, to identify complex organic molecules for understanding the possible origins of life 1 High-redshift galaxies; those moving away from us fast enough that their infrared radiation is shifted down into the terahertz range 1 Direct detection of exotic ions such as helium hydride from the early universe 1 Black hole environments, such as the event horizon, using the Event Horizon Telescope (EHT) (86345GHz), a global network of linked telescopes, including imaging M87 1 Mapping magnetic fields in space by measuring the polarisation of waves caused by dust particles aligned with magnetic fields The following facilities routinely operate in the terahertz/submillimeter regime, making them critical for studying cold astrophysical phenomena that are invisible at other wavelengths: 1 The ALMA (Atacama Large Millimeter/submillimeter Array, 31.3950GHz) in Chile 1 The James Clerk Maxwell Telescope (86-690GHz) in Hawaii 1 The Submillimeter Array (SMA, 180-418GHz) in Hawaii Fig.20 shows polarised images from ALMA, VLBA and EHT images of the M87 black hole environment, revealing the magnetic fields. Communications As wireless data traffic continues to increase, the demand for higher frequencies has driven spectrum use upward for greater bandwidth and capacity. Current systems operate at the highest practical radio frequencies; Fig.19: a Golay cell. Original source: www.mdpi.com/1424-8220/24/21/6784 siliconchip.com.au Table 2 – wireless communications candidates Technology Microwaves Terahertz Near-infrared Visible light Ultraviolet Data rate Up to 10Gbps Up to 100Gbps Up to 10Gbps Up to 10Gbps A few Gbps Range Short Short-medium Short-long Short Short Power consumption Medium Medium Relatively low Relatively low Expected to be low Network topology Point to multipoint Point to multipoint Point to point Point to point Point to multipoint Noise source Thermal noise Thermal noise Sunlight + ambient light Sunlight + ambient light Sunlight + ambient light Weather conditions Robust Resistant to fog, dust, turbulence and drizzle but not rain Sensitive – Sensitive Security Medium High High High To be determined Wi-Fi 6E and Wi-Fi 7 use the 6GHz band (5.925-7.125GHz), while 5G’s Frequency Range 2 (FR2/mm-wave) operates over 24-71GHz, with the most common bands being 26GHz, 28GHz and 39GHz. Satellite communications reach up to the Ka-band (26.5-40GHz), with experimental V-band trials in the 40-75GHz range. Free-space optical communications such as Li-Fi (visible/near-infrared light) and Starlink’s inter-satellite laser links (operating around 1550nm wavelength, or ~193THz) already achieve massive data rates, but remain limited to line-of-sight, short-range or space environments. In the future, 6G is expected to span a wide range of frequencies, from sub6GHz legacy bands through new centimetric bands (7-15GHz), up to sub-­ terahertz and terahertz candidates (90300GHz or higher). The deployment of THz waves is constrained by two major challenges: the relatively low power levels that can be practically generated, and severe atmospheric absorption that limits range and reliability, although specific frequency ranges have been identified that are less prone to atmospheric absorption, as mentioned earlier. Path losses increase with both the square of the frequency and the distance, making long-distance comms at higher THz frequencies even more difficult. Terahertz 6G connectivity is expected to be complementary and used as a data rate booster if close enough to a transmitter site, rather than exclusive use due to the range limitations of THz comms. For siliconchip.com.au longer-­ range comms with a lower data rate, the lower bands will still be used. Beamforming will be necessary to enhance the range by concentrating the power of THz signals, as is already done for other cellular frequencies (see our article on 5G Mobile Networks from September 2020 – siliconchip. au/Article/14572). The high atmospheric absorption of certain frequencies in the THz band can be put to good use with “whisper radio”. This is a concept intended for ultra-short, ultra-secure comms with ranges from centimetres to metres that cannot be easily intercepted. Frequencies around 183GHz, 325GHz, 380GHz and 450GHz have been identified as suitable. Terahertz imaging for security As mentioned above, the major breakthrough for THz imaging came in the late 1980s to early 1990s with David Auston’s development of pulsed broadband THz radiation, enabling time-domain spectroscopy (THz-TDS) and sparking widespread interest in the field. Table 3 – loss vs frequency Frequency Bandwidth Loss 275-320GHz 45GHz <10dB/km 335-360GHz 25GHz <10dB/km 275-370GHz 95GHz <100dB/km 380-445GHz 65GHz <100dB/km 455-525GHz 70GHz <100dB/km 625-725GHz 100GHz <100dB/km 780-910GHz 130GHz <100dB/km Fig.20: images of the environment around the M87 black hole, taken with the Hubble Space Telescope (HST) in visible light, the ALMA telescope at 230GHz, the VLBA (Very Long Baseline Array) at 43GHz and the Event Horizon Telescope (EHT) at 230GHz. Source: www.eso.org/public/ images/eso2105c/ Australia's electronics magazine August 2026  25 Passive THz imaging was first demonstrated in 2002-2004. For example, MIT’s Lincoln Lab showed passive imaging of concealed objects using microbolometer arrays or Golay cells, followed by portable security cameras around 2004-2008 from Thruvision or Microsemi. Since the 2010s, compact THz sources such as quantum cascade lasers (QCLs) have become available. THz imaging is particularly attractive for human applications because it is harmless and non-ionising, unlike X-rays. Fig.21: the Thruvision security screening system. Source: https:// thruvision.com/case-studies/o2 Passive THz imaging Passive THz imaging detects the natural THz radiation emitted by all living things and objects; thus, no THz radiation source is required. A typical application is security screening, such as Thruvision (https://thruvision. com). It operates around a centre frequency of 250GHz – see Fig.21. Such systems are typically used at airports. Fig.22: a simulated explosive belt imaged by the Terasense Body Scanner 2.0. Source: https://terasense.com/applications/security/ Other applications for this technology include astronomy, non-destructive testing and biomedical imaging. Active THz imaging The Terasense (https://terasense. com) Body Scanner 2.0 system is active and illuminates the target with harmless THz radiation. It can illuminate a target from several meters away and can also do it covertly if desired, from behind any object or wall transparent to THz waves. It can detect objects such as a ceramic knife, which won’t show up on a metal detector, and can view an imaging area of 70×70cm from 3m away with a resolution of 3cm, or a 120×120cm area from 6m away with a 6cm resolution. It operates at 100GHz with six or more sources and has a 3×3mm 32×32 pixel sensor array – see Fig.22. Active imaging can also be performed on items as they pass on a conveyor belt (see Fig.23). Tomography THz radiation can be used for tomography, a technique that creates detailed 3D visualisations of an object’s internal structure. By directing THz waves through an object from multiple angles and measuring how much radiation is transmitted, absorbed, or scattered at each position (often using time-of-flight or phase information), appropriate computational analysis (similar to CT scanning) reconstructs a 3D image. This method is particularly valuable for non-destructive inspection of materials that are opaque to visible light but partially transparent to THz, such as plastics, ceramics, pharmaceuticals, certain composites and medical patients. Quality control and nondestructive testing One application of THz imaging in quality control is to test the integrity of tyres; for example, detecting a missing reinforcing cord. THz imaging can also be used to check the thickness and look for bubbles, defects or inclusions etc. Fig.23: security screening of packages using a THz scanner on a conveyor belt. Source: https://youtu.be/iHOt7Quyduk Medical imaging When surgeons remove a suspected skin cancer or other diseased tissue, they have to excise a margin of healthy tissue around the tumour as it might also be cancerous but not visibly so. Obviously, surgeons wish to minimise Australia's electronics magazine siliconchip.com.au 26 Silicon Chip the amount of healthy tissue they remove. THz imaging has the potential to distinguish healthy areas from unhealthy areas by the different spectral response of the tissues. THz imaging is also being evaluated for applications in dentistry, such as to evaluate the thickness of enamel or examine other parts of the tooth without patient exposure to X-rays. THz imaging is also being evaluated for rapid blood analysis, such as the non-invasive measurement of glucose. Art analysis THz imaging can be used to reveal hidden layers of a painting and can see, or see through, layers that may not be visible to other techniques such as infrared or X-rays. An example is a painting that was thought to be by Francisco Goya but did not have a signature. Fig.24 shows a hidden signature revealed by THz imaging of the painting. Terahertz spectroscopy Terahertz spectroscopy is non-­ destructive and non-ionising. It is used to analyse materials by probing their unique molecular fingerprints through various vibrations, rotations and interactions of molecules within the material under examination. Spectroscopy may be performed in either the time domain using pulses (see Fig.25) or frequency domain using continuous waves, swept in frequency. Terahertz waves can penetrate materials that are non-metallic and non-­ polar such as plastics, clothes and cardboard. Applications include: 1 detecting explosives or illicit drugs in packaging 1 analysing pharmaceutical drugs to detect different crystalline forms 1 checking the quality of food (eg, moisture content) or for contaminants 1 biomedical applications for checking molecular structures, hydration, or disease markers 1 checking plant water content, pesticides or seed quality for agriculture 1 analysing polymers, semiconductors and detecting defects Fig.24: THz imaging of a Goya painting revealing the hidden signature. Source: https://arxiv.org/pdf/1305.3101 Fig.25: the configuration of a time-domain terahertz spectrometer. Source: www2.riken.jp/lab/THz-img/ English/annual_gas.htm distance contact at 30THz (10µm wavelength) of 60m. This frequency is outside of standard radio amateur allocations and is not subject to regulations. The transmitter was a ‘blackbody’ emitter (heated plate) that was mechanically modulated. See siliconchip.au/link/acb5 and the YouTube video at https://youtu. be/6gJtzMLR6T0 for more details. Conclusion Amateur radio In November 2020, Australians The terahertz gap has been signifiAndrew VK3CV/WQ1S and Karl cantly narrowed. Commercial sources VK3LN made a world-record long-­ now exist for 0.3-1.0THz that are in siliconchip.com.au Australia's electronics magazine routine use in laboratories and some industrial systems, for applications like spectroscopy, non-destructive testing and imaging. However, the full 0.1-10THz range is still not fully exploited compared to adjacent bands because of ongoing challenges in developing high-power, tuneable, room-temperature, compact sources and detectors, as well as the atmospheric absorption that limits long-range use. Several different kits of sensors that can detect terahertz band radiation are SC also now available. August 2026  27 SERIOUS PRECISION. SERIOUS SCALE. $ 16K JUST 1849 JUPITER 2 RESIN 3D PRINTER TL4988 LIMITED STOCK. ONLINE ONLY Available for special order in-store SEE THE FULL RANGE OF 3D PRINTING ONLINE jaycar.com.au 1800 022 888 jaycar.co.nz 0800 452 922 Prices shown in $AUD, and correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. NEED TO TOP-UP YOUR SERVICE AIDS AND ESSENTIALS? GREAT RANGE. GREAT VALUE. In-stock at your conveniently located stores nationwide. 4 2 1 3 5 BUY IN BULK & SAVE!!! 1 2 Isopropyl Alcohol 99.8% 250ml Spray NA1066 BUY 1+ $14.50 EA. BUY 4+ $13.00 EA. BUY 10+ $11.50 EA. 99.8% 300g Aerosol NA1067 BUY 1+ $16.50 EA. BUY 4+ $14.50 EA. BUY 10+ $13.00 EA. 70% 1 Litre Bottle NA1071 BUY 1+ $22.95 EA. BUY 4+ $20.45 EA. BUY 10+ $17.95 EA. Electronic Parts Cleaning Solution 1 Litre Bottle NA1070 BUY 1+ $15.95 EA. BUY 4+ $13.95 EA. BUY 10+ $12.45 EA. 3 Liquid Electrical Tape 28g Tube, Black NM2836 28g Tube, Red NM2838 BUY 1+ $25.95 EA. BUY 4+ $22.95 EA. BUY 10+ $20.45 EA. 4 5 175g Aerosols Contact Cleaner Lubricant NA1012 Electronic Cleaning Solvent NA1004 BUY 1+ $14.50EA. BUY 4+ $13.00 EA. BUY 10+ $11.50 EA. PTFE Dry Lubricant NA1013 BUY 1+ $15.95 EA. BUY 4+ $13.95 EA. BUY 10+ $12.45 EA. J-B Weld Products J-B Weld Epoxy 28g NA1518 BUY 1+ $24.95 EA. BUY 4+ $22.45 EA. BUY 10+ $19.95 EA. SuperWeld Extreme 15g NA1539 BUY 1+ $16.50 EA. BUY 4+ $14.50 EA. WaterWeld 57g NA1532 BUY 1+ $25.95 EA. BUY 4+ $22.95 EA. BUY 10+ $20.45 EA. Shop at Jaycar for even more service aids & essentials: • Adhesives & Insulation Tapes • Solder & Soldering Aids • Wire & Heatshrink Tubing • Fasteners & Cable Ties • Ultrasonic Cleaners • Tools & Workbench Accessories Explore our great range of 3D Printing gear, in stock on our website, or at over 140 stores or 130 resellers across Australia and New Zealand. jaycar.com.au 1800 022 888 | jaycar.co.nz 0800 452 922 Prices shown in $AUD and correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. Part 2 by John Clarke Background source: https://unsplash.com/photos/a-person-in-yellowgloves-and-blue-gloves-cleaning-a-floor--dc38HdQR1M Ultrasonic Cleaner adjustable Our new Ultrasonic Cleaner has adjustable frequency, transducer voltage and power. That means you can run it at low or high power, depending on what you are cleaning. The transducer can be brought into resonance for efficient power transfer even with a range of liquid levels. e explained how this new Ultrasonic the cleaning power you need, which attach to the lid. The only external wiring is for DC power to the unit and one W Cleaner works last month. Unlike our can be up to 40W. Since it runs from 12-15V DC at up twin-lead that emerges for the translast Cleaner, which automatically swept the frequency searching for the resonant point of maximum power transfer, this one lets you adjust it yourself. You can also vary the transducer voltage to vary the power level, plus it includes an automatic safety feature that reduces the power if the current draw gets too high. While it might seem odd that you need to find resonance yourself, when the previous version did it automatically, this approach has a few advantages. One is that it makes matching the transformer to the transducer less critical; that made building the previous version more difficult. Another is that it can be more tolerant of different fluid levels in the bath, as that affects the resonant frequency. With this design, you can select one of 16 different frequency bands using pushbutton switches, then adjust the frequency within those bands with a knob. Once it has been set up, the next time you go to use it, it usually won’t take long to make a few tweaks to the frequency and power knobs to achieve 30 Silicon Chip to 4A, that means you can also run it from a standard 12V lead-acid or similar battery, as long as it can provide the required current. Construction Besides the transducer, all the electronic components mount on two PCBs that are housed in a diecast aluminium box. The controls and LCD screen ducer. Both leads pass through the box via cable glands. Building the Ultrasonic Cleaner isn’t too difficult. The main steps are soldering the components to the PCBs and winding the transformer. The enclosure will require drilling and a couple of rectangular holes need to be made. PCB construction The transducer is driven with a high voltage, around 150V AC, in operation, which is more than enough to give you a shock. Touching both of the transducer terminals during operation will give you an electric shock; it will be worse if your hands are wet. You must enclose the transducer in the PVC housing described in this article and only run it when so enclosed and attached to a bath filled to the correct level with cleaning fluid. The main 159 × 111mm PCB is coded 04105261, while the smaller (98 × 60mm) front-panel PCB is coded 04105262. Both fit in a 171 × 121 × 55mm diecast aluminium case. The overlay diagrams for these boards are shown in Figs.6 & 7. Start by fitting the resistors on both PCBs where shown. The resistor colour codes were in the parts list last month, but it’s always best to check the values with a DMM set to measure resistance to make sure they’re going in the right places. The two 0.1W SMD resistors mount on the top of the PCB. Solder one end first and check alignment before soldering the other end. Australia's electronics magazine siliconchip.com.au WARNING! Fig.6: the main Cleaner PCB overlay. Leave Mosfets Q1 & Q2 for last, as need to be mounted to but insulated from the bottom of the case, with access holes for tightening their nuts. REG1, REG2 & D2 are attached to the side of the case, also for heatsinking; the two regulators require insulating washers and bushes. Continuing with just the main PCB, fit diodes D1 and D3, ensuring that their cathode stripes face toward the top edge of the PCB as shown. We recommend that IC1 and IC2 are mounted in sockets. Make sure that the notched end faces toward the lower edge of the PCB. Also mount the two M205 fuse clips now, making sure that they have the correct orientations, with the end stops toward the outside. It is a good idea to insert the fuse before soldering the clips to ensure the fuse is aligned in the clips and that the clips are orientated correctly. CON1, CON2 and CON3 can then be installed. These screw terminals should be orientated so the wire entry faces the edge of the PCB. Mount the 14-way IDC box header (CON4) next. The location notch must face as shown and ensure that the header is pushed all the way down before soldering its pins. Fit the capacitors next, noting that the electrolytic capacitors have their longer (positive) leads through the holes marked “+”. The negative lead is marked with negative (–) signs down the capacitor can. Then solder the two small transistors (Q3 and Q4), which are both BC547s. siliconchip.com.au TO-220 package devices REG1, REG2 and diode D2 are mounted vertically with the mounting hole 22mm above the top of the PCB. REG2 will need pins 1, 3 and 5 bent more forward than pins 2 and 4 to fit the offset mounting holes. VR1 can be fitted now, as well as inductor L1. L1’s leads are inserted into the smaller PCB holes, while the inductor is secured in place with a cable tie that passes through the larger PCB holes provided. Insert the tie from the underside, then through the centre of L1 and back down through the second hole and tighten it up. Mosfets Q1 and Q2 are installed on the underside of the PCB. Bend the three leads for each Mosfet upward by 90°, 5mm from the bottom edge of the Mosfet body. Then insert the leads into the PCB from the underside, but do not solder them yet. Now place the PCB into the enclosure, sitting on the internal mounting corners. Mark where the Mosfets sit, including their mounting hole locations, then remove the PCB and place Fig.7: assembly of the front panel is straightforward, but be careful to orientate the switches and CON5 correctly, as described in the text. CON5 mounts on the back of the board, with the other parts on the front. Australia's electronics magazine August 2026  31 the silicone insulating washers under the Mosfets. Fig.8 shows how these Mosfets will be mounted, although we aren’t attaching them to the case just yet. Reinsert the PCB and adjust the Mosfets so they sit flat on the bottom of the case, on the silicone washers. Now solder the Mosfet leads on the top of the PCB, then remove the PCB and solder the leads on the bottom of the PCB as well. Winding the transformer Fig.9 shows the transformer winding details. The primary windings are made from 1mm diameter enamelled copper wire (ECW), while the secondary windings use 0.5mm diameter enamelled copper wire. The two windings for the primary are shown with different enamel insulation colours for clarity. There is no need to use different colours for your windings. Start with the primary windings. First, cut two 400mm lengths of the 1mm diameter ECW and remove the enamel from one end of each wire using fine emery paper or a hobby knife. Tin the wire ends and wrap one wire around pin 7 on the underside of the transformer bobbin and the other onto pin 8. Solder both close to the bobbin. The bobbin provides for the wire to be wrapped around behind the post before winding a couple of turns around the pin. Now close-wind seven turns of both wires Fig.8: this shows how Mosfets Q1 & Q2 attach to the board and the bottom of the case. Don’t forget the insulators. (side-by-side) until the windings reach the opposite end of the former. The winding direction does not matter as long as both wires are wound together. Using a multimeter set to measure resistance (ohms) or in continuity mode, find the wire that’s terminated to pin 7 and terminate its other end to pin 1 in the same way as before, ie, by stripping the insulating coating before wrapping it around the post and pin, then soldering it. The other wire end from pin 8 terminates at pin 6. Cover the windings in a layer of insulation tape. The secondary winding uses 0.5mm diameter ECW. Terminate one end to pin 4 and wind on 30 turns, then wrap a single layer of insulation tape over this winding and continue winding back over the first layer, in the same clockwise or anti-clockwise direction as before to complete 60 turns. Then, after adding another single layer of insulation tape, wind on another 30 turns in the same direction, for a total of 90 turns. Pass this wire back to pin 3 along the axis of the former. Cover it again with a layer of tape. Once wound, slide the cores into the former and secure them with the clips. These clips push onto the core ends and clip into lugs on the side of the bobbin. It is best not to install the transformer directly onto the PCB just yet. It can be temporarily wired up using some short lengths of 0.7mm diameter tinned copper wire or similar, connecting pins 6, 7, 8 & 1 for the primary plus pins 3 & 4 for the secondary to the corresponding PCB pads. This is so that it will be easier to remove and change the secondary windings should the output voltage not be within the desired range. More on this later. Now insert both IC1 and IC2 into their sockets, taking care to orientate them as shown on the overlay diagram. IC1 needs to be programmed with the firmware before use. The hex file is available from our website at siliconchip.au/Shop/6/3634 for those who have the equipment to program it themselves. Otherwise, you can purchase a programmed PIC from the Silicon Chip Online Shop. VR1 shaft extension VR1 requires a shaft extension to reach and pass through the enclosure lid, as shown in Fig.10. The shaft extension is made using 6mm diameter timber dowelling or several tapped nylon hexagonal spac- The Control and Main Boards for the Adjustable Ultrasonic Cleaner. 32 Silicon Chip Australia's electronics magazine siliconchip.com.au ers that are stacked together, plus a washer. These are held together with a 25mm-long M3 machine screw. If using the timber dowel, a flat will need to be filed at the top on one side for the knob to fit. The hexagonal nylon version already has a flat side to act as a key for the knob. We used a 40A dual-screw mains Earth wire connector as the bush to secure the two shafts together. The outer plastic covering can be removed by first removing the two screws. The inner metal part should then drop out. When securing this to the shaft of VR1, the potentiometer should first be set to its mid-position. Then the securing screws can be orientated toward the right edge of the PCB before tightening to the shaft. In this way, you can rotate VR1 fully in both directions without fouling the screws against the 1000μF capacitors. Fig.9: the transformer primary is bifilar wound in one layer, with thicker ECW, while the sole secondary is wound using thinner wire in three layers, with insulating tape in between each layer. Front panel control board assembly There are only a few parts on this PCB (coded 04105262), but be careful to mount them on the correct side. Most parts go on the top side, but the 14-way IDC box header (CON5) goes on the underside. Install the resistors first, then the 100nF capacitors. Insert the shorter side of the 16-way SIL (single in-line) header into the PCB and solder it in place, place the LCD screen over the pin header and mount it on two 6.3mm-long nylon spacers and secure it with two 12mm-long M3 machine screws & nuts. You can then solder the pin header to the LCD on the top side. Potentiometers VR2 and VR3 can be installed now, along with switches S1-S3. These switches need to be orientated correctly, with the anode and cathode for the internal LED of each placed as shown. The anode (A) and cathode (K) locations are marked on the PCB. Each switch will have a coloured marker (matching the colour of the LED) on the lower black part of the switch on the (K) side. Next, fit CON5 on the underside of the PCB, taking care to orientate it with the location notch as shown. Solder its pins from the top side of the PCB. Now the IDC cable needs to be made and plugged into this header. Fig.11 shows how the IDC cable is made using 14-way ribbon cable and two crimp connectors. The connectors can be clamped onto the cable by adding siliconchip.com.au Fig.10: there are two options for extending the shaft of VR1 to reach through the lid, using either a piece of timber dowel or several hexagonal tapped spacers. Either way, the inside of a double-screw wire connector is used to join it to the plastic pot shaft. Fig.11: the IDC cable is a little unusual in that pin 1 is swapped at each end, but the PCB connector orientations swap it back. The shaft of VR1 can be extended using a 35mm long timber dowel, as shown below. Or you can use a series of tapped spacers (see Fig.10). The IDC cable is made using 14way ribbon cable and two crimp connectors, see Fig.11 Australia's electronics magazine August 2026  33 a small piece of soft timber (eg, pine) over each side of the socket and compressing the lot with a G-clamp or bench vice (or use a specialised tool like Altronics Cat T1540). Make sure the socket orientations are correct before you compress them, with the locating tabs facing as shown in Fig.11. The remaining assembly work for this board is done after the enclosure lid has been prepared. The front panel drilling/cutting template (Fig.12) can be copied or downloaded from our website in PDF format (siliconchip. au/Shop/11/3561) and printed out at ‘actual size’. Attach it to the lid, ensuring the paper template is centred correctly. Next, mark and cut the holes. You can use a centre punch or nail and hammer to mark the centres before drilling. The holes for the power switch and LCD can be made by drilling a series of small holes around the perimeter, knocking out the piece and filing to shape until each fits correctly and is held in position firmly. The PCB is held to the lid using four M3-tapped 12mm standoffs/spacers plus a ~1mm-thick nylon washer to keep the PCB some 13mm back from the inside of the lid. These spacers go between the top side of the PCB and the inside of the lid. Front panel label The front panel label (Fig.13) can be made using overhead projector film, printing the label as a mirror image so that the ink will be between the enclosure and film when affixed. Use projector film that is suitable for your printer (either inkjet or laser) and affix it using clear neutral-cure silicone sealant. Roof and gutter silicone is suitable. Squeegee out the lumps and air bubbles before the silicone cures. Once cured, cut out the holes through the film with a hobby or craft knife. For other options and more detail on making robust labels, see the details on our website at siliconchip. com.au/Help/FrontPanels Two holes are required in each end of the box for the DC power cable gland and the ultrasonic transducer cable gland. Along the sides, 3mm holes are also required for attaching REG1, D2 and REG2 to the inside of the case. You can determine these positions by temporarily mounting the main PCB into the enclosure and marking where the holes need to be. Holes are also required in the base of the enclosure for Mosfets Q1 and Q2. You may have marked the positions earlier; if not, do this now. Drill these to 3mm. Lightly countersink these holes inside the enclosure, plus the ones for REG1, D2 and REG2 on the inside, to prevent the insulating washer from being damaged by a rough hole edge. You can use a large, sharp drill bit turned by hand. Attach Mosfets Q1 and Q2 using silicone insulating washers, plastic bushes and M3 machine screws with nuts, as shown in Fig.8. REG1 and REG2 similarly require an insulating bush and washer, but D2 can be directly mounted with a screw, as it has an insulated tab. Check that the metal tabs are isolated from the case using a multimeter on a high ohms setting. A reading above 1MW (ideally ‘open circuit’, often shown as “0L”) means that the isolation is good. Lower readings may be due to a punctured insulator or a short circuit to the case. Now wire switch S4 to the board using 5A-rated hookup wire, with heatshrink tubing over the soldered terminations. Once the other ends of the wires are secure in the screw terminals for CON2, use a cable tie to hold these wires firmly to the PCB using the larger holes in front of CON2. The cable tie passes through the board and then around the wires. Preparing the ultrasonic transducer There are many suitable 50W/60W 40kHz ultrasonic transducers available online. One such part is the Beijing Ultrasonic BJC-4050T- 45HS PZT4. Alternatives are at siliconchip.com. au/link/ab3g and siliconchip.com.au/ link/ab3h For the wire between the board and the transducer terminals, use mainsrated wiring that can handle at least 7.5A. Figure-8 wire or, preferably, a sheathed dual cable is suitable. The wire ends for the transducer can be soldered to solder lug eyelets and covered in heatshrink tubing. These can then be connected to the transducer terminals with M4 × 10mm machine screws, star washers and nuts. Left: Mosfets Q1 & Q2 are mounted with insulated washers behind them and bushes under the nut, as shown in Fig.8. Right: here is a close-up of VR1, which we’ve extended using multiple tapped nylon spacers, a washer and screw. 34 Silicon Chip Australia's electronics magazine siliconchip.com.au These terminals on the transducer are exposed and need to be protected within a housing to prevent accidental contact, as they are a shock hazard. The high-voltage AC can cause a nasty shock, but only if both contacts are touched. Touching one contact or the front face of the transducer will not cause a shock, since the transformer output is floating from the main circuit (don’t try this, though!). A suitable housing can be made using 50mm PVC DWV (drain, waste and vent) fittings. We used an end cap and an adaptor (with the smaller adaptor section cut off) to extend the length of the end cap to an overall outside length of 50mm. You could use the end cap and a short length of 50mm pipe instead of the adaptor. Wire entry is via a cable gland that is secured in the side or end of the end cap. Place the cable gland hole in a position allowing sufficient room for its securing nut inside. The transducer should be mounted within the DWV fittings using neutral-­ cure silicone sealant (such as roof and gutter sealant). Use just enough silicone to secure the transducer to the inside of the housing, around the outside of the lower bell-shaped section. Fully potting it in silicone will dampen the ultrasonic movements. The face of the transducer should be kept clear of the sealant. This is so that the transducer can be secured to the outside of the bath with an epoxy resin. Connect the ultrasonic driver cable to the PCB at CON3. Make sure there are no strands of copper wire emerging from the terminals that could short together. Testing Before testing, insert the M205 fuse into the clips, if you haven’t already done so. When ready, apply power to the circuit and check the main 5V supply between pins 20 and 1 of IC1, and between pins 4 and 8 of IC2. You should get a reading of 4.75-5.25V across these pin pairs. To properly test the board, you need to have the transducer attached to a suitable container that’s filled with a liquid like water. That’s because you need to check that the transformer is supplying the right voltage to achieve full power. Your transducer could differ from the one we have used, either by being a different type or just coming from a different batch. siliconchip.com.au Fig.12: the lid drilling and cutting details. This can be printed or copied at actual size and used as a template. The larger, rectangular holes can be made by drilling a series of smaller holes inside the perimeter, then filing the edges flat. Metal dishes with thin sheet metal will work best; there must be a flat side or base for the transducer to be attached. If you wish to mount the transducer on the side of the container, the height must be at least 75mm. One of the cleaning dishes we used was from Woolworths. It was an approximately 4L baking pan measuring 225 × 225 × 78mm. The base had a lift-out section that had to be glued in place to prevent leaks. We used roof and gutter silicone sealant for that. Alternatively, there are pans available from Nisbets (www.nisbets.com. Australia's electronics magazine au). They have shops in NSW, Vic, Qld and the ACT, but also sell by mail order. We recommend either the 150mm-deep ¼ gastronorm tray (capacity 4L) or the 100mm-deep ¼ gastronorm tray (capacity 3.7L). The transducer will need to be glued to a flat section on the outside of the container. We tested both side mounting and mounting on the base. The advantage of side mounting is that the container does not need to be raised on a stand to allow room for the transducer that’s mounted on the underside. August 2026  35 Our ultrasonic bath and completed Adjustable Ultrasonic Cleaner (before insulating the transducer). The front panel label is overleaf in Fig.13, you can also download it from siliconchip.com.au/ Shop/11/3561 Some containers are covered in a non-stick surface, which will need to be removed where the transducer mounts to allow glue to adhere (also slightly roughen the surface). Use wet-and-dry emery paper to clean up the area, then clean off any residue before gluing. J-B Weld two-part epoxy resin is recommended as the glue. Tape the transducer in position while it sets. The fluid used in the bath can be tap water with a few drops of detergent as a wetting agent. Other fluids that can be used include deionised water, alcohol (methylated spirits, isopropyl alcohol etc), acetone or similar solvents. The cleaning effectiveness is greatly enhanced when the fluid is warm. Filling it with around two litres of liquid is ideal for the power available from the ultrasonic transducer. Transformer tweaking The 90-turn secondary winding provided sufficient voltage for our test transducer to deliver between 32W and 39W of ultrasonic power into a 4L container filled with 2L of fluid. Any variation in the volume of fluid will affect the operation. More fluid will require a lower drive frequency. Additionally, the transducer impedance will be higher, meaning that more secondary transformer turns will be required for more drive voltage to maintain the power level. Less fluid will mean a higher resonance frequency and a lower drive voltage is required due to the transducer impedance being lower. It is recommended that the transformer windings be set up for use with 2L of fluid. Testing To test the Ultrasonic Cleaner Controller, power it up with the wattage pot wound fully anti-clockwise, then wind this up to show a voltage of around 5V. Press the Stop/Start button to initiate the ultrasonic drive, then adjust the frequency to find the resonance, where there is a peak in the displayed power. You may need to change the Span to find the resonance, including if the resonance is too close to the end of the frequency pot range. Ideally, find the span that allows the resonance to be near the middle of 36 Silicon Chip Australia's electronics magazine siliconchip.com.au the frequency pot adjustment range. Setting the precise resonance requires careful adjustment. Once the resonant frequency has been found, increase the power potentiometer setting. Check that it is possible to get over 30W when the voltage pot is wound up. If the power is over 30W at voltages lower than 10V, the transformer secondary should ideally have a reduced number of turns to avoid inefficient operation. If the power is not sufficient even with the voltage pot fully clockwise, more turns on T1’s secondary are needed. How many turns that need to be added or subtracted can be determined on a trial-and-error basis. If you get 36W, or at least over 30W with the pot wound fully clockwise, the existing number of secondary transformer windings are suitable. Note that the Ultrasonic Cleaner limits the input current to 3.3A, so if you wind up the power pot until that amount of current begins to flow, the voltage supplying the transformer will begin to reduce to maintain the 3.3A draw. Another unexpected effect that can occur is due to the current limit set by the LM2576 (IC2). This is guaranteed to be at least 3A, but could be more. If the current draw exceeds the LM2576’s limit, the output voltage will reduce, even without any change in the power/ voltage pot. So you may find that the voltage reading drops as you adjust the frequency to get maximum power at resonance. In that case, reduce the voltage pot setting and continue searching for the resonant peak. Then the pot can be rotated clockwise for more power, re-adjusting the frequency slowly to more precisely find the resonance point. The reason for the current limit is that the transducer impedance is lower on either side of resonance, so more current is drawn compared to at resonance. At resonance, the higher impedance allows for more voltage to be used without causing regulator overload. Current limiting is less likely when finding the resonance by winding it down from a higher frequency rather than winding it up from a lower frequency. As an example, in our prototype, using a 14V supply, 12V is shown on the LCD screen when the ultrasonic drive is on and the transducer is running at resonance. The power is 37W (so the current is 3.1A = 37W ÷ 12V). If the frequency is adjusted off-­ resonance, the voltage drops to 9.5V and the power shows as 31W. This means that 3.26A is flowing (31W/9.5V) and the regulator has reduced the output voltage from 12V to 9.5V to limit the current. Locking the frequency Locking and unlocking the frequency or span is done by first holding down the stop/start This photo shows how the two boards are joined once they are installed in the case. The insulator (shown at left) fits directly on top of the ultrasonic transducer. siliconchip.com.au Australia's electronics magazine August 2026  37 Silicon Chip Binders REAL VALUE AT $21.50* PLUS P&P Are your copies of Silicon Chip getting damaged or dog-eared just lying around in a cupboard or on a shelf? Can you quickly find a particular issue that you need to refer to? Keep your copies safe, secure and always available with these handy binders These binders will protect your copies of S ilicon C hip . They feature heavy-board covers, hold 12 issues & will look great on your bookshelf. H 80mm internal width H Silicon Chip logo printed in goldcoloured lettering on spine & cover Silicon Chip Publications PO Box 194 Matraville NSW 2036 Order online from www. siliconchip.com.au/Shop/4 or call (02) 9939 3295 and quote your credit card number. *see website for delivery prices. 38 Silicon Chip Fig.13: this front panel label can be printed, laminated and attached to make the unit look nicer and identify the controls. button for over one second. The display then shows OPTION * = LOCK. While still holding the Stop/Start switch, press the down button to lock the Frequency setting (or unlock it if it was already locked). Similarly, pressing the up button while holding the Stop/Start switch locks or unlocks the Span setting. Using the timer Switch S3 starts and stops the Ultrasonic Cleaner. When pressed momentarily, it will start the output drive. An hourglass emptying and filling shows that the timer (and ultrasonic drive) is Australia's electronics magazine running. The value of the timeout, set by VR2, is shown on the LCD screen and drops every 7s or so as the time remaining reduces. Once the timer runs out, the transducer drive ceases, and the hourglass disappears. The timer setting from VR2 is shown instead. During the time-out period, the ultrasonic drive can be stopped by momentarily pressing S3 again. You can change the timeout period while the timer is running to a lower value and stop the timeout with the Stop/Start push button at any time SC during the timeout period. siliconchip.com.au Subscribe to Ultrasonic Cleaner cleaning at up to 40W, with adjus table frequency, power and durat n be o? ca de to her si be ot , may te xt here ion normal JULY 2026 ISSN 1030-2662 07 Australia’s top electronics magazine Silicon Chip is one of the best DIY electronics magazines in the world. Each month is filled with a variety of projects that you can build yourself, along with features on a wide range of topics from in-depth electronics articles to general tech overviews. 9 771030 266001 $15 00* NZ $15 90 INC GST INC GST Published in Silicon Chip If you have an active subscription you receive 10% OFF orders from our Online Shop (siliconchip.com.au/Shop/)* Rest of World New Zealand Australia * does not include the cost of postage Length Print Combined Online 6 months $77.50 $87.50 $55 1 year $145 $165 $105 2 years $270 $305 $200 6 months $95 $105 1 year $180 $200 2 years $335 $370 6 months $115 $125 1 year $220 $240 All prices are in Australian dollars (AUD). Combined subscriptions include both the printed magazine and online access. 2 years $410 $445 Prices are valid for the month of issue. Try our Online Subscription – now with PDF downloads! DCC Accessory Decoders; July 2026 Simple USB Power Monitor; June 2026 Human Comfort Indicator; June 2026 Amplifier Clipping Indicator; May 2026 An online issue is perfect for those who don’t want too much clutter around the house and is the same price worldwide. Issues can be viewed online, or downloaded as a PDF. To start your subscription go to siliconchip.com.au/Shop/Subscribe Part 3: deck and 3D printing Phil Prosser’s Phenomenal B efore we get to building the electromechanical parts of the Pinball Machine, like the flippers, bumpers and kickers, we’re going to explain how the deck is organised and constructed. The reason for this is that you will need a flat surface on which to test and install the parts as you make them. You could use a scrap piece of 12mm timber, possibly an off-cut of the piece you plan to use for the actual deck (and that is not a bad idea). Still, it would save time to just make the deck and install the parts directly, especially if you’re copying our design. We found that the final electronic and electromechanical parts worked the first time, so you can be confident in them. If you change the deck layout or come up with a completely new design, you really do need to prototype it. We encourage this but be ready for the work involved. The play deck Pinball Machine We have now described all the electronics in the pinball machine and will soon get onto the 3D-printed electromechanical parts. But those will need a deck to be mounted to, so first, let’s look at the deck and cabinet. Our play surface is 470mm wide and 900mm tall. If you are happy for yours to be the same width, you can keep the reloading section unchanged. There’s nothing stopping you from varying the height and/or creating your own middle and upper play area. The deck arrangement we came up with is shown in the lead photo. We are supplying all the CAD files. Even if you aren’t planning to create an exact replica of our machine, we encourage you to use these as a starting point to make your own deck parts. There are multiple download packages at siliconchip.au/Shop/6/3628 including one with STL files and one with Fusion360 files. All of the plastic parts can be printed at home. You will need the STL files for printing the parts; some instructions are included in text files along with them. We also provide Fusion360 source files for these, but you only need those if you want to modify our designs. We’ll touch on that later. While you are planning and possibly starting to drill your deck, you can get busy 3D printing the pieces you will need to assemble the electromechanical parts later. The details of those pieces, including the file names, are shown in Table 1. Those files are also in the download package. Have a look at the deck photos to plan what colours you want to print each part. siliconchip.com.au We will assume you have a way to print the parts. If you don’t have a 3D printer or a very accommodating friend with one, it might be cheaper to buy a simple printer than to get these commercially printed. Jaycar and Altronics both have plenty of good options starting from a few hundred dollars. Paying to get the parts printed would cost more than getting your own printer and filament. Note that super cheap filament is not always a great idea. One lot we used went very soft if left in the sun. Wait until you’ve read the instructions at the end of this article to print the targets, since we have some specific instructions on how to print those. All the other parts are generally solid colours, but make sure you use the right materials, wall thicknesses, fill percentages etc as per Table 1. We used 12mm marine ply for the deck as it has a flat surface and is thick enough to provide the mechanical stability we require, but not so thick as to make working it a chore. You will note that our deck does not have fancy box-work on the sides and over the lower deck mechanism, so the side walls are a single thickness of plywood. You certainly have the option of constructing a much nicer machine case and enclosing things like the flipper buttons inside your deck, for example. One thing we have learned in developing this article is the need to try different layouts and prototype the overall playfield such that you place targets and features in areas that result in a fun game. In particular, we found that the ball has a tendency to go in some directions more easily than others, so you want features in these areas that the player should hit easily or often. This means you should be willing to try a deck layout, devise improvements and update this part of the machine. In making your machine case, you need the deck sloped at 6-7°; we chose about 6.5°. You definitely need good access to the underside for wiring, and you will need access for servicing, as pinball machines are notoriously ‘needy’ due to all the switching and electromechanical parts. The last thing you need is a theme and a decoration idea. A pinball machine is virtually defined by colour, light and sound. You should develop siliconchip.com.au Table 1 – 3D-printed pieces required all PLA unless noted Section Qty Filename Fill Wall depth Notes Backboard 1 Player Bezel 30% 2.4mm Backboard 1 Score Bezel 30% 2.4mm Backboard 2 Speaker Grille 30% 2.4mm Three style options Led_Holder >40 Led_Holder 100% solid Five styles; use clear PLA Posts >40 PCB Mounting Washer 8mm 100% solid Any colour PLA Reloader 1 Lower Deck Layout L 10% 1.2mm Print on raft Reloader 1 Lower Deck Outer Runway L 10% 1.2mm Reloader 1 Reload Load Coupling Bushing 30% 2.4mm Reloader 1 Reload Load Coupling 30% 2.4mm Reloader 1 Reload Load Flipper 30% 2.4mm Reloader 1 Reload Load Retaining Washer 100% 2.4mm Lower deck 1 Lower Deck Layout M 10% 1.2mm Print on raft Launcher 1 Lower Deck Layout R 10% 1.2mm Print on raft Launcher 1 Ball Reload Positioner 30% 2.4mm Launcher 1 Launcher 6.6 degree Shim 100% solid Launcher 1 Lower Deck Outer Runway Right 10% 1.2mm Launcher 1 Reload Ball Release Drive 30% 2.4mm Launcher 1 Reload Ball Release Lower Washer 30% 2.4mm Launcher 1 Reload Ball Release Slide Coupling 30% 2.4mm Launcher 1 Reload Ball Release Slide 30% 2.4mm Launcher 1 Reload Ball Release Solenoid Coupling 30% 2.4mm Flippers 1 Double Drive – Driver 100% solid Flippers 1 Double Drive – Driver Mirrored 100% solid Flippers 2 Double Drive – Limiter 40% 2.4mm Flippers 1 Double Drive 12mm Deck 30% 2.4mm Print on raft; use ABS if possible Flippers 1 Double Drive 12mm Deck Mirrored 30% 2.4mm Print on raft; use ABS if possible Flippers 2 Drive Arm Washer 100% 100% solid Flippers 2 Flipper 90 plus 4 40% 5mm Flippers 4 Solenoid Coupling 38mm 100% solid Flippers 2 Washer Lower 100% solid Flippers 2 Washer Upper 100% solid Use ABS The rest of the table is continued overleaf... Australia's electronics magazine August 2026  41 your own ideas on these and do not be at all shy. Some of the most common pinball machine themes fall under these categories: Fantasy and medieval Sci-fi and space Horror/spooky Adventure/treasure Crime/action Music (especially rock bands) Movies and TV shows Superheroes/comics Sports The Wild West Circus/carnival Animals/dinosaurs Historical/military Glamour/party/nightlife Abstract/novelty That doesn’t mean you have to work within one of these. Do whatever you want! You will have seen our approach of stencil-painted retro electronics. That works for us but may not be for you. You could also consider stickers/transfers, spray painting, hand painting (if you are artistically inclined) or the use of small models and figurines. 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 📍 Our example deck Fig.19: the locations of all cutouts, holes and key components in our example deck. We will not go into great detail on this, as we expect you will have your own ideas. The screw and drill locations are determined by your upper deck (likely 3D printed) parts, so many of these locations will be determined by your final layout. 42 Silicon Chip Australia's electronics magazine We built our example deck to demonstrate the controller and parts and to check that they all function as we want. The full deck is shown in Fig.19. You can make the deck in any form you want. We used lots of 3D-printed parts to make it easy, although this did move the complexity into the 3D modelling space. This is shown in Fig.20 overleaf. Our 470 × 900mm deck has a lot of parts screwed to the top, and some to the bottom, using 16mm-long 6G wood screws. There are also some 50mm M4 machine screws and nuts used to secure parts. The drilling details for our deck design are shown in Fig.19. This should be useful to get you started, but as you develop your own deck, you will really need to adjust and fettle stuff to your own needs. Fig.21 (overleaf) shows the overall Pinball Machine and the assembly approach we took. This will give you a kick-start if you intend to follow our initial layout example. We made the back box and legs removable, attaching them to the main deck with M6 machine screws. That will be handy if we ever have to move the machine to another house. siliconchip.com.au Table 1 – 3D-printed pieces required all PLA unless noted | x = number of units The deck is made from painted 12mm marine ply. The blue and black guides and sections are all 3D printed and attached with self-tapping screws. The flippers, bumpers, targets and kickers are all 3D printed too. The back box can be any size and shape, provided you can fit the power supply, controller and wiring in it (or on the back) and the score and player display on the front. We were tempted to add more lights than we did; there are a couple of LED outputs you could put to this use with a very small amount of coding. Our back box is separate from the main deck to allow it to be removed for decoration and for transporting the machine. We used four M6 machine screws to secure it to the deck. We used 80mm holes for routing cabling from the back box through the underside of the deck. These are as small as you should go, as there are lots of cables to run, including some thick ones. Consider making larger holes if you have the tools. siliconchip.com.au Section Qty Filename Fill Thickness Bumpers x Bumper Ball Detect Larger 100% solid Bumpers x Bumper Base Lower 30% 2.4mm Bumpers x Bumper Base 30% 2.4mm Bumpers x Bumper Plunger Coupling 100% solid Bumpers x Bumper Plunger 100% solid Bumpers x Bumper Top 100% solid Make this colourful, maybe transparent or translucent; the LEDs light through it Bumpers x Bumper Under Deck Bracket 30% 2.4mm Should not need supports; remove them if used Bumpers x Bumper Shim for Base to Clear LED Holders 30% 2.4mm Kickers x Kicker Arm 12mm deck 100% solid Kickers x Kicker Base 2.4mm wall 100% solid Kickers x Kicker Coupling 2.4mm 100% solid Kickers 2x Kicker Microswitch Bracket 100% solid Posts 3x Square Rope Kicker 100% solid Posts 1 Posts Top L 100% solid Posts 1 Posts Top R 100% solid Targets x Target Bracket 12mm deck 30% 2.4mm Targets x Target Tiles Numbers 30% 2.4mm Targets x Targets_Top 30% 2.4mm We recommend that you do not attach the legs to the back box and deck until your assembly and wiring are complete. Definitely make the actual play surface removable, as you will probably want to change the layout of your machine as you develop the gameplay. Our play deck is secured by two screws at the top of the deck under the printed runways. There is a lot of wiring and looming required, even though we have used ribbon cables. It is much easier to do this with the deck on its side on a benchtop or trestle table. Be willing to prototype! We did sufficient wiring and assembly to get the machine into a playable state prior to stripping everything off, then painting and decorating it. We Australia's electronics magazine Notes See text in this article regarding multiple colours made quite a few changes to the play surface during our trials. Because it was “prototype” in our minds, we had no hesitation in hacking into the play surface and moving stuff around. We encourage you to bring this approach to your build, as the prototype mindset avoided concerns with scratching or damaging finished artwork. To support this, don’t permanently fix the play surface to the machine. As an observation throughout my career, when creating new concepts and developing challenging concepts, it is important to be constructively critical and willing to change your approach if required. Sometimes radically. We had not built a pinball machine before, and this was certainly the case here where redoing parts like August 2026  43 the flippers led to significant improvements. Buying the parts We’re going to present individual part lists for each section of the machine later and over the next couple of months. Part of the reason for doing this is that we expect many constructors will want to customise their machine, and if we just present one big list, the quantities are not going to be correct for alternative configurations. Having said that, there are quite a few parts that you’re going to go through in large quantities, regardless. Rather than having to make repeated trips to the hardware store, we suggest you buy most or all of the General Hardware parts overleaf so that you have plenty of them. You’re going to need them! The quantities are approximate, but even if you end up with some left over, we’re sure you’ll find a use for them. Printing the parts Assuming you’re going to print the parts as we designed, rather than use them as a starting point to make your own, we suggest that you download the ZIP files linked above, start printing the pieces listed in Table 1 and keep them in order. Even if you don’t know how many bumpers, kickers and targets you need, you can start by printing one of each and then make more later. To get you started now, we will now give details of the targets. The assembly of the remaining parts will be spread over the next couple of issues, as there is quite a bit to cover. Assembly order When building our deck, we performed many test fits, screwing the deck sections to the deck in the following order (you don’t necessarily have to follow this, but it might be a good idea). 1. We started with the lower deck sections and launcher and got this aligned and functioning. 2. We added the flippers and alleyways that abut the flippers, as well as the kicker parts. With these in place, we marked the holes for the kickers and drilled holes for the flippers. This allowed us to remove the upper deck parts, cut and drill the deck, then test-install these parts. 3. At this point, we were able to 44 Silicon Chip Fig.20: this screen grab from Fusion360 shows the main deck top parts but does not include the kickers, bumpers and targets, although you can see the cutouts and holes for mounting them. This should line up with Fig.19. Photo 10: when hit by the ball, the targets cause the LED underneath to light and add to the player’s score. Photo 11: by printing the first couple of layers in a different colour than the rest, we get some contrast for the numbers on the faces. ▶ Australia's electronics magazine siliconchip.com.au drill the holes for the LEDs and sensors local to the flippers. 4. We then added the launch alleyway up the right-hand side of the deck, and from there marked out and cut the target on the right-hand side. 5. Then we installed the upper deck parts on the left-hand side. We were then able to mark the left-hand targets and associated LEDs. 6. With the upper parts in place, we were able to make final markings for the bumpers and associated LEDs. These were then drilled and test-fitted. 7. For the red holes shown in Fig.19, the positions should be pretty close to what’s required, but you should determine the final hole positions on your deck by temporarily placing the 3D-printed parts as you test fit them. Targets Fig.21: the download package (siliconchip.au/Shop/6/3628) includes dimensional drawings for our cabinet, but the exact details will depend on your implementation. Our playdeck is 470 × 900mm, the backboard is 200mm deep, 640mm wide and 400mm tall. We made ours from timber offcuts and used a solid pair of hinges on the rear panel. The cabinet height is 247mm at the front and 400mm at the rear. siliconchip.com.au Australia's electronics magazine The targets do exactly what it says on the box – your aim is to hit them with the ball (see Photo 10). We have four targets in a row, which are target ‘tiles’ hung off the lever of a microswitch. When the ball hits the target, it actuates the microswitch from which the tile hangs, which is sensed by the controller. You could spread them around if you wanted. Targets are quite straightforward to build. The main trick is adjusting the microswitch lever such that it has a ‘hair trigger’ to actuate (similar to the kicker, to be described later). Once adjusted, these install fairly easily, though make sure you have wired in the microswitches before installing them as they are fiddly to solder once installed in the deck. There are Score LED outputs associated with each target. If all targets are hit in one round, a bonus score is achieved. Make sure to install the score LEDs in front of the relevant targets. Target construction You’ll probably want to make several groups of targets (our machine has two). Each set contains four targets and has three parts: the bracket, numbered tiles and a cover. The steps are: 1. Print the target tiles. We printed our tiles using yellow filament on the first two layers (Photo 11), then paused the print and swapped to black filament for the remainder of the tiles. This gives a very bright target with text in black. 2. You need four microswitches; we used KW12 types. These are incredibly August 2026  45 Parts List – General Hardware | many quantities are approximate 1 12.8cm yellow ball launch mechanism [AliExpress 1005006766715473] 1 24V DC 5A+ power supply [Altronics M8973B] 1 panel-mount DC barrel socket to suit the power supply [Altronics P0628] 10 12V 1.5A 20N solenoids [TAU-0826, AliExpress 32618031228] 8 Cree C513A series 30mA LEDs (any mix of colours but in pairs) [Altronics Z0876E] 10 LJ12A3-2 2mm inductive sensors (NPN OCO) [AliExpress 1005002931452610] • 1 5m length of 12V LED strip lighting [Altronics X3203A] 2 100mm loudspeaker drivers [Altronics C0616, Jaycar AS3008] Switches 4 KW12-3C-Z 25mm (straight) lever arm microswitches [AliExpress 1005008233825861] 8 KW11-3Z-E 16mm-20mm (straight) lever arm microswitches 5 4-pin SMD tactile switches with short actuators [Altronics S1112A, Jaycar SP0610] 2 arcade-style momentary pushbutton switches (for flippers) [Altronics SA091x, Jaycar SP0662] 3 SPST square momentary red pushbuttons (front panel) [Altronics S1080, Jaycar SP0717] Connectors 2 6-way vertical pluggable terminal blocks [Altronics P2516, Jaycar HM3126] 30 2-way vertical pluggable terminal blocks [Altronics P2512, Jaycar HM3122] 2 3-way polarised header plugs with matching pins [Altronics P5473 + P5470A, Jaycar HM3403] 20 2-way polarised header plugs with matching pins [Altronics P5472 + P5470A, Jaycar HM3402] 34 10-way IDC ribbon cable connectors [Altronics P5310, Jaycar PS0984] Screws and hardware 3 1200 × 600 sheets of 12mm marine plywood OR 1 2400 × 1200 sheet of 12mm marine plywood 1 1m length of 20 × 1.6mm aluminium flat bar [Bunnings I/N 1064257] 30 9mm-long self-tapping box screws (4G self-tappers) 20 6mm-long self-tapping box screws (4G self-tappers) 1 1m length of M3 threaded rod • this must be NPN normally 100 6G × 16mm wood screws [Bunnings I/N 0201371] open (NO) output with 100 6G × 20mm wood screws [Bunnings I/N 0201372] 2mm or 4mm sense range Panhead machine screws: 25 M3 × 25mm 50 M3 × 20mm 25 M3 × 16mm 100 M3 × 6mm 25 M4 × 50mm 10 M2.5 × 20mm 25 M3 flat washers 50 M3 Nyloc hex nuts 10 M3 × 10mm tapped spacers 25 M4 hex nuts 10 M2.5 Nyloc hex nuts Cable and wire 1 20m length of medium-duty (7.5A+) figure-8 speaker wire [Altronics W2126] 1 20m length of 10-way ribbon cable 2 2m lengths of heavy-duty hookup wire (red & black) [Altronics W2283, W2284] 4 reels of light-duty hookup wire (green, blue, red & black) 2 1.2m lengths of 5mm diameter heatshrink tubing [Altronics W0993A] 50 small cable ties Miscellaneous 1 tube of superglue, Loctite, or another screw thread locker plenty of 3D printer filament in various colours (black, blue, yellow etc) cheap. You could modify an Altronics S3265 microswitch to match, but they are comparatively expensive. Screw the microswitches to the bracket using 2.5mm machine screws and nuts, as shown in Photo 12. 3. The target tiles simply slip into the brackets and sit in place. 4. If you push the target, you should feel the microswitch click before the target hits the target frame. 5. Affix the cover over the targets using three 4G × 9mm self-tapping screws. Coming up 4 KW12-3C-Z 25mm (straight) lever arm microswitches 4 2-way polarised header plugs with matching pins 3D-printed parts (all PLA) 1 Target Tiles Numbers (30% fill, 2.4mm wall; see text regarding multiple colours) 1 Target Bracket 12mm deck (30% fill, 2.4mm wall) 1 Targets_Top (30% fill, 2.4mm wall) Hardware & wire 3 9mm-long self-tapping box screws (4G self-tappers) 8 M2.5 × 20mm panhead machine screws Photo 12: the microswitches are attached 8 M2.5 Nyloc hex nuts to the target assembly using M2.5 2 1m lengths of green light-duty hookup wire machine screws and Nyloc hex nuts. Next month, we will have more detailed instructions on assembling some of the remaining sub-units, mounting them to the deck, and getting them working. That will include the bumpers, kickers, posts (for kickers and guides), flippers, ball return and launch system plus the other parts of the deck. In the meantime, you can start printing some of those parts (using Table 1 as a guide) and gathering the hardware given in the parts list here. Make sure you pay attention to the 3D-printing fill and wall details as they vary between files. Note that there are a couple of parts for the flippers that should be made from ABS rather than PLA for robustness, and we recommend that you print some of the parts on rafts so they don’t distort as they cool. You can also get further along in planning your deck and, if you’re aiming to copy ours, start marking out the hole positions. It might also pay to begin planning the cabinet at this stage and gathering supplies to put it together. If you haven’t chosen a theme yet, you can also start researching that, and once you have decided on it, start making or gathering the artwork, stickers etc that will decorate your machine. If you’re customising your Pinball Machine, you should buy enough timber to make two decks: one to experiment with, where you aren’t too fussed about drilling extra holes and such, and another that will take on the final form, once you have finalised the positions of everything. If you haven’t already assembled all the required electronic modules, that’s another task that you could start on while you are waiting for the next SC instalment. Australia's electronics magazine siliconchip.com.au Parts List – Target Assembly 46 Silicon Chip AUGUST Shop 24/7 <at> altronics.com.au NEW! 129 $ T 2582 Three tools in one! NEW! 359 $ C 5065 Gold C 5068 Black Iroda® 130W Swivel Head Torch & Iron Amazing new multifunction design! Versatile swivel head for different tasks with desk stand. Easily switches between soldering iron, heat gun and blow torch. Up to 550°C soldering and 1100°C blow torch. Introducing HiFuture Passion X Speakers built for your listening passion! These portable BT5.4 speakers work great at home, on picnics, outdoor parties and more. Delivers up to 100W output with 2 way speaker design engineered for a premium listening experience. SAVE $36 SAVE 19% Unique design! 20 $ 99 $ Repair faster! X 0232 T 2128C SAVE 17% 82 X 0437 $ No more eye strain! Zoom into flawless clarity instantly! This portable inspection camera offers up to 12x magnification and supports PC connection for large screen display, plus SD card recording. Delivers powerful visual enhancement wherever you need it. 35 in 1 Motorised Screwdriver Set Folding LED Lantern Light Ideal for camping and more! Innovative design includes a central LED lamp with fold-out side panels. USB C rechargeable. This handy set uses both 4mm and 6.3mm bits making it suitable for a wide range of tasks. From small items like watches & phones, to large items like toys. USB C rechargeable. T 2356 X 7011 SAVE 22% T 2555 SAVE 15% 20 30 $ $ A close up view - hands free! Thousands sold! Offers 1.5x, 3x, 8.5x,10x magnification with LED lamp. Requires 2xAAA batteries. Handy Weather Monitor Check indoor temperature & humidity with high/low display. Easy to read backlit screen. Work bench must have! T 4018 SAVE 24% SAVE 20% 18 12 $ $ Handy Magnetic Bowl 4” stainless bowl with magnetic base. Keep screws from straying. Rotating PCB Holder Work on boards up to 200 x 140mm. Metal base provides a sturdy work platform. Your electronics supplier since 1976. Shop in-store at one of our 11 locations around Australia: WA » PERTH » JOONDALUP » CANNINGTON » MIDLAND » MYAREE » BALCATTA VIC » SPRINGVALE » AIRPORT WEST QLD » VIRGINIA NSW » AUBURN SA » PROSPECT Sale ends August 31st 2026. Build It Yourself Electronics Centre® TOOL Digital Vernier Calipers Great for 3D print design! UPGRADES T 1342 T 2247A 30 Range Multi-Function DMM Precision measuring with ease! 150mm length, suitable for measuring internal, external and depth dimensions. 0.01mm, 0.0005” and 1/128th” display. NEW! SAVE 24% 85 $ Superbly built with an easy to read 2.8” reverse backlit screen and 30 manual ranges across voltage, current, resistance, capacitance. Plus added functions such as non-contact AC detector, diode & transistor test, relative mode and more! NEW! 99 $ 44 $ Q 1066A Ultra High Speed Jet Blower This high power rechargeable fan/vacuum is great for servicing computer equipment, cleaning keyboards - even inflating air mattresses for camping. Handy variable speed slider! High speed jet fan with up to an hour use per charge (recharge via USB C). If you use a few cans of air duster a year, it pays for itself! Three tools in one! T 2580 T 1297 NEW! SAVE $20 89 $ 99 $ Bargain 40W Soldering Station The perfect balance of value for money and features for beginners or cash strapped students and enthusiasts. Slim, lightweight handle with tip cleaning sponge and iron safety holder. Full range of spare tips also available. SAVE $15 50 $ Top buy for students & makers! T 2090 Iroda® 100W Torch & Soldering Iron Benchtop Fume Extractor Amazing new multi-function design! One tip converts between soldering iron, heat gun and blow torch. Includes a 11 piece tip kit for a multitude of uses in home repairs, heat shrinking, model making, wood crafting and cutting. Whisk away irritating solder fumes instantly as you work. The replaceable active carbon filter absorbs fumes for a cleaner work environment. Includes 100mm ducting adaptor. Easily screw clamps to your work bench. SAVE 19% 40 T 2749 $ SAVE 22% 62 35 T 2161 $ Self opening carry case. Jakemy® 62 Piece Screwdriver Kit A quality servicing kit for high tech devices. Includes a huge variety of 4mm driver bits, opening tools, spudger, tweezers & flexible extension. 39 $ T 2181 $ SAVE 23% SAVE 20% T 1528A Stay sharp longer! Tungsten Carbide Side Cutters Tough HRC 72° tungsten carbide construction for 5 times the life of standard side cutters. 130mm. SCAN TO FOLLOW US! Stay up to date on latest releases, exclusive specials and news on our socials. 5 In 1 Plier, Cutter & Crimper Interchangeable heads with ultra secure fit for even the toughest of jobs. Includes pliers, cable cutters, wire strippers, scissors and connector crimping pliers. Wire Stripper & Automotive Crimper Combines a ratchet wire stripper, cutting blade & red, yellow, blue crimper. Suits 10-24 AWG cable. Like our service? Review your store on Google. Every review helps us serve you better. SECURITY & LIGHTING. WAS $149 NEW! S 9845C NEW! 125 $ WAS $65 WAS $89 55 75 $ $ S 9854 WAS $89 75 S 9857 S 9858 NEW! NEW! Indoor Dual Pan & Tilt Camera $ Outdoor Pan & Tilt Camera 5MP Outdoor Solar Powered Wi-Fi Camera Offers 4MP high resolution with pan & tilt functionality in an IP65 rated outdoor enclosure. Powered by USB C. 2-way audio with mic and speaker. Cable free design! Just mount it in a sunny spot for instant surveillance without the hassle of running cabling. LED floodlight in-built with motion detect recording. Two-way audio. IP66 rated. A unique dual camera design allowing both pan/tilt adjustment and fixed cameras ensuring no blind spots in camera coverage. 4MP (2K) resolution on both cameras. Indoor Pan & Tilt Camera Makes a great baby or pet monitor, this camera features intelligent tracking of moving objects within the frame. 2-way audio with mic and speaker. 4MP (2K) resolution. All cameras use Tuya Smart Home app for connectivity & remote viewing. What is Tuya® Smart Home? Tuya is a common application for thousands of products from the worlds leading Smart Home suppliers. It provides a single point of control for home security, lighting and appliance power allowing you to control everything you need from a the one smartphone app. The Tuya IoT platform powers over 250,000 automation products across the globe! 4G Equipped Remote Solar Camera S 9451 SAVE $30 225 $ This 4G LTE equipped trail camera can be used to set up wireless surveillance anywhere! Simply add an SD card and SIM card to the device for wireless real time monitoring of the camera from just about anywhere! SAVE $30 199 S 9445A $ A 4K scouting camera with in-built Wi-Fi and bluetooth file transfer. Ideal for setting up an instant surveillance solution in areas that are difficult to reach with traditional surveillance solutions (such as rural properties, sheds etc). Includes solar panel. Requires SD card. Mega bright! SAVE 24% 22 $ 30 $ X 2387 Lrg LED Solar Sensor Lights Add instant security to your place with these weather and rust resistant solar lights! Require no wiring and are IP54 rated for use outdoors. Available in 3W or 5W versions. SAVE $20 79 $ S 9446E Great for monitoring in remote locations, temporary CCTV etc. 4K video and 20MP still shots. Footage review screen on rear. Requires SD card & 4xAA batteries. SAVE 30% 42 $ X 2340C 10W X2386 Sml SAVE 24% 4K video surveillance anywhere! 4K Solar Powered Wi-Fi Camera SAVE 30% SAVE 29% 2 for $ 70 55 $ X 2315C 20W X 2384 Plug In IP65 PIR Security Lights Cable Free Solar Security Light Lights up to 80sq/m with this powerful multi panel LED light which requires no wiring and is powered by the sun! IP65 rated for outdoor use. Great for the garden or driveway. 600 lumens. Get quality motion activated security lighting at your place for less! Fully adjustable sensitivity, on time and dusk settings. Fitted with 3 pin 240V mains plug. Natural white. NO STRESS 30 DAY RETURNS! GOT A QUESTION? Not satisfied or not suitable? No worries! Return it in original condition within 30 days and get a refund. Ask us! Email us any time at: customerservice<at>altronics.com.au Conditions apply - see website. GADGETS Powerful 6K Lumen Torch GALORE! A super bright USB C rechargeable torch designed to light up areas up to 150m away! With multiple light modes, anodised casing and 4000mAh power bank function. SAVE $40 139 $ M 8196 Easy to set up anywhere! Getaway Generator & Light With powerful LED lantern, torch & emergency dynamo! This compact portable charging solution keeps your devices powered up. The internal 83Wh battery (equivalent to 7Ah at 12V) is paired with three USB outputs, including an 18W USB C PD port and two 18W QC3.0 type A ports. A 3615 SAVE $20 99 $ Up to 150m light beam! X 0214 SAVE $10 89 $ Mini Wi-Fi LED Projector Great for movie nights with friends and family! This compact projector offers excellent LED picture quality with 800x480p resolution for screens up to 4m (170”) wide! Very simple to set up with adjustable focus & projection distance (1-5m). HDMI input or Wi-Fi screen mirroring for playback directly from your device. SAVE $20 X 7018* 59 $ C 9021B Tyre Pressure Monitor System This solar powered TPMS unit sits on your dash and provides wireless monitoring of your tyre pressures. Provides high/ low pressure alarms, leak detection and temp monitoring. Signal booster option Q1302 $95. SAVE $19 SAVE $45 130 $ Q 1300 Brand name noise cancelling performance for a fraction of the price! The Futuretour headphones offer a stylish stealth black finish and excellent noise isolation (up to 20dB reduction) for work and travel. With USB charging. M 8070B SAVE $19 50 $ 240V from your vehicle cup holder. Simply plugs into a 12V cigarette lighter socket & provides mains power! 150W rated (300W surge). Great for fitness! C 9044B Black C 9044W White 42 SAVE 20% Get a crisp close up view Adjustable 5x-7x magnifier with LED backlight. Great for reading fine print, maps, examining parts and more! Perfectly cooked every time! Monitor your roast or BBQ with ease using this smart Bluetooth meat thermometer—track temps, set targets, and get real-time updates right from your phone. Perfect for oven or grill. Easy USB recharging. SAVE 24% 30 $ X 0432B SAVE 15% $ 80 $ HiFuture® ANC Headphones 48 $ HiFuture® Over Ear Headphones Latest design over ear wireless (OWS) headphones - drivers sit on your ear rather than in it for all day comfort. A 0329 Battery Free Door Bell Never change batteries again! Kinetic action with wireless chime unit. 58 ring tones. 150m range. Sale Ends August 31st 2026 Shop in-store at one of our 11 locations around Australia: WA » PERTH » JOONDALUP » CANNINGTON » MIDLAND » MYAREE » BALCATTA VIC » SPRINGVALE » AIRPORT WEST QLD » VIRGINIA NSW » AUBURN SA » PROSPECT Or find a local reseller at: altronics.com.au/storelocations/dealers/ Shop online 24/7 <at> altronics.com.au B 0008 © Altronics 2026. E&OE. Prices stated here in are only valid until date shown or until stocks run out. Prices include GST and exclude freight and insurance. See latest catalogue for freight rates. *Devices for illustration pursposes only. By Tim Blythman μDCC Decoder Accessory Decoder I2C Controller Destination Display Model Headboard Destination Display Background source: https://unsplash.com/photos/ train-tracks-with-trees-on-either-side-3TXv9NJZZTk Miniature electronics have come a long way and are now at the point that it is possible to create displays compact enough to fit into small (HO or N) scale models. This tiny display can be controlled by our microDCC (μDCC) Decoder, an infrared remote control, or a serial port. M ovement is a key aspect of giving realism to models; the motion of trains in a model railway is much of what makes them so engaging. Signals and points also add to an authentic nature. Much of this series has been devoted to adding movement to a DCC (digital command control) model railway. Lights and sound also add to engagement and don’t require moving parts. We noted in the microDCC Decoder article (May 2026 issue, siliconchip. au/Article/20239) that some modellers are now adding small displays to their trains. We showed an example of an LCD (liquid crystal display) fitted to one HO-scale model to emulate a headboard. This Model Headboard Destination Display project provides something similar. While this board can be used without the microDCC decoder, once the graphical data has been loaded onto the chip, with it you will be able to send DCC commands to the loco and have it update the display (eg, select a new destination). Otherwise, you siliconchip.com.au could use an IR remote control aimed at a receiver on the loco to perform the same task. Headboards and destination displays “Headboard” is a term originally used to describe a painted timber Features & Specifications 🛤 The 19 × 12mm PCB drives a tiny OLED panel 🛤 Serial input for control (eg, from our microDCC Decoder) 🛤 Infrared remote control 🛤 8kiB+ NFC EEPROM chip for display data 🛤 3.3V operation; 6mA typical draw at full brightness 🛤 Suits a variety of monochrome display panels and sizes 🛤 Display contents (graphics and animations) are configurable Australia's electronics magazine board placed at the front of a train to indicate its line or destination. Later, adjustable mechanisms using handpainted linen scrolls on rollers were used. Photo 1 shows an example of a scroll from an old Ballarat tram and a Geelong tram, showing how this appears in use. The translucent linen allows effective backlighting. Originally adjusted by hand, these mechanisms later received electronic controls and plastic screen-printed scrolls. Other variants include airport-­ style split-flap displays and flip-dot mechanisms. We created our own version of a flip-dot display, published in the April 2019 issue (siliconchip.au/ Article/11520). Many modern displays use LED matrices. This sort of display is what the Destination Display is intended to replicate. You might wonder about our claims that this unit will fit in an N-scale model, but we have found a tiny 0.32-inch (8.1mm) OLED (organic LED) display panel that is one of a few that can be used for this project. Photo 2 shows the Destination August 2026  51 Display fitted to the same N-scale chassis that we used for testing the DCC Decoder project. The OLED and PCB assembly has an outline that fits within the profile of an N-scale unit. The 8.1mm dimension is the active area; the OLED panel itself is about 10mm wide and 9mm tall, with the total assembly being about 20mm long. We expect that readers will find other uses for the Destination Display. Its size would suit other model railway applications, such as departure boards or even advertising signs. The Destination Display offers animations that include the scrolling motion that would be seen in a changing sign, and it has a mode that emulates the alternating screens of modern digital signs. The video at siliconchip.au/Videos/ Destination shows the Destination Display and a 0.32in OLED panel inside a 3D-printed N-scale model tram. The model is 22mm tall. OLED panels Fig.1: these diagrams show the mapping of the pixels in the displays; the mauve numbers 1-8 indicate the codes used to program the different display orientations. So, to arrange the 0.50in display in landscape with the FFC on the left, you would set the orientation to 2 and the x-value to 40 or higher to ensure the bitmaps are in the active area. 52 Silicon Chip We found several small panels that we have tested with nominal display sizes of 0.32in, 0.50in and 0.54in. Since they offer comprehensive data sheets, we sourced all these from www.buydisplay.com These OLED panels all use a similar controller to the SSD1306 that is found on the common 0.96in and 1.3in OLED modules. The controller types on our panels include the CH1115 and the SSD1315. Like the SSD1306, these display controllers feature a 128 × 64 pixel display memory, but for these small units, not all the pixels map to active elements. For example, the 0.32in panel’s matrix only has 60 × 32 pixels. So the Destination Display must be configurable to work with these limitations. Ample EEPROM memory in the microcontroller allows all manner of graphics and settings to be loaded and displayed. To give you an idea of the scale of these displays, their pixel pitch is about 0.12mm. That means a seven-­ pixel font will be displayed less than one millimetre in height. Fig.1 shows the different panels that we have tested and how their matrices are mapped to their internal memories. We will provide some sample display data sets, but understanding Fig.1 will be important if you wish to generate your own display data. Australia's electronics magazine Fig.1 also shows the main dimensions of the panels (to the nearest 0.1mm); there are detailed dimensional diagrams in the panel data sheets. The panel sizes, model numbers, data sheets and sources are: ● 0.32in panel: ER-OLED0.32-1W data sheet: siliconchip.au/link/acbu > siliconchip.au/link/acc0 ● 0.50in panel: ER-OLED0.50-1W data sheet: siliconchip.au/link/acbv > siliconchip.au/link/acc1 ● 0.54in panel: ER-OLED0.54-1W data sheet: siliconchip.au/link/acbw > siliconchip.au/link/acc2 The “W” suffix indicates that these panels have white light-emitting elements; this colour appears to be the only available variant for panels this small. Importantly, they also feature the same narrow 14-way FFC (flat flexible cable) connector, making them electrically equivalent. We tested some other panels with the same connector that were not compatible or did not work. This included some variants of the 14-way connector that use a different pad arrangement. That’s why we have chosen the three displays noted above. Circuit details For this project, we are using a bare OLED panel, which means that our circuit needs to provide the support circuitry that is typically seen on the modules that we have used for other projects. Fig.2 shows the circuit of the Destination Display. IC1 is an 8-bit PIC16F18115 microcontroller from the same family as the PIC16F18126 and PIC16F18146 that we have used in other projects from the DCC series. The ‘15’ suffix indicates that it is an 8-pin part with 1kiB of RAM and 14kiB of flash memory; this is the largest available memory option for 8-pin parts in this series. Pins 1, 4, 6, 7 & 8 connect to the ICSP programming header (CON1-CON5), which is also used as the main power and data interface during operation. IC1 receives power from CON2 and CON3 and expects serial data to be delivered to pin 7 via CON4. The header is simply a row of surface-­mounting pads, much the same as we have used for the Decoders in this series. The 10kW resistor pulls up pin 4 (MCLR) to allow normal operation. The 3.3V supply (pin 1 of IC1) is bypassed by 1μF and 10μF capacitors to ground (pin 8). Pins 2 and 3 of IC1 siliconchip.com.au are used for the I2C serial interface, so they have 4.7kW pullup resistors to 3.3V. That’s already most of IC1’s pins allocated! CON7 connects the remaining I/O pin on IC1 (pin 5) to a three-way pad header that also includes ground and power. An infrared receiver module can be connected here to allow IC1 to receive commands from a remote control. The pinout matches common IR receivers, so they can be soldered directly to the PCB. The I2C bus also connects to pins on 8-pin IC2; this is a chip from the ST25DV NFC family. These parts are effectively an EEPROM that can be read or written over an I2C bus or via NFC (near-field communication, such as with a device like a mobile phone). The Dynamic NFC Tag from July 2023 (siliconchip.au/Article/15860) demonstrates these features. Unlike the Dynamic NFC Tag, the Destination Display does not use NDEF (NFC Data Exchange Format). Since the data format is unique to this application, it is not important to mark the data with its type. Our prototypes used the ST25DV64KC, which has 8kiB of EEPROM, but we also think that the 2kiB ST25DV16KC part has enough storage to be practical. We’ll look at the memory requirements later. A pair of pads, CON9 and CON10, allow the connection of an external antenna for the NFC interface. We have designed a small, flexible antenna PCB that we tested on our prototypes. We also tested a small coil made from Photo 1: these destination scrolls, seen at the Ballarat Tramway Museum, demonstrate how a tall bitmap can be used to implement the Destination Display. Photo 2: when fitted in this fashion, the assembly is 20mm tall & 12mm wide. This is small enough to fit into the profile of an N-scale model and display a line or two of text at roof level. a length of enamelled copper wire (ECW). The I2C interface is used to read from the EEPROM in IC2 and also to communicate with the controller in the OLED panel connected at CON6. CON6 is simply a row of pads on the PCB, and the OLED panels have an FFC (flat flexible cable) connector that is intended to be soldered directly to the PCB. The OLED panel consists of a controller chip and OLED matrix mounted to the panel glass using a so-called COG (chip on glass) construction. The connections are made with conductive traces of transparent indium tin oxide (ITO). The remaining circuitry is to support the controller chip for the display panel. Two of the 1μF capacitors are part of a charge pump circuit used to generate the voltages needed to drive the OLED matrix. Two of the 10μF capacitors store the voltages generated by the charge pump. The remaining capacitors bypass the 3.3V rail for the ICs and OLED panel. The third 4.7kW resistor pulls up the RESET pin of the controller to allow it to operate, while the 560kW resistor connected between the Iref pin and ground sets the OLED matrix drive current. The 3.3V supply and the I2C signals also connect via the FFC. Software Fig.2: the circuit is fairly straightforward, with most of the passives being needed for the OLED panel’s operation. siliconchip.com.au Australia's electronics magazine Apart from the firmware that runs on the microcontroller to manage the Destination Display, we have also written an application in the cross-­ platform Processing language. That means that you can use a Windows, Linux or macOS computer to generate data for the Destination Display. The program delivers data as a single file, which is simply copied to the EEPROM on IC2 using a suitable NFC-equipped device such as a mobile phone. Information about the program August 2026  53 can be found on page 56 (see “Destination Display Configuration for Processing”), while the file upload process is described later. Firmware The firmware on IC1 loads configuration data from IC2; this contains information about the display format and where the display data can be found on the EEPROM. The display data is simply a tall monochrome bitmap that is laid out in the fashion of the hanging scroll seen in Photo 1 and a display pointer sets a window so that only a small part of the bitmap is visible. IC1 then waits for commands from either the serial or IR interfaces. The command sets which specific item is to be displayed. The item has an index between 0 and 255, which corresponds to the byte received over the serial line. Being a single byte means there are no concerns with data framing or the like. The IR interface is programmed to respond to NEC codes addressed to device ID 0 or device ID 1. That means it should be easy to build a controller using basic hardware such as an Arduino board and IR transmitter. A data byte for device ID 0 is treated the same as a serial data byte and simply indicates the item index. Compact IR transmitters such as Jaycar’s XC3718 send compatible codes, although the data bytes do not have any obvious correspondence to the button markings (see Table 1). Signals addressed with device ID 1 are treated as increment, decrement, or zero commands and simply update the index. This command set has been chosen to allow a suitably programmed NFC IR Keyfob (February 2025; siliconchip.au/Article/17730) to control the Display. A data file to suit the Fob (to use device ID 1) is included in the software downloads for this project. The indexed item contains information about the graphics to be displayed and how it is to be updated. For example, one mode causes the display pointer to steadily increment or decrement until the desired item is reached; this emulates the behaviour of older scroll-type displays. There are several scroll rates that can be selected. Another mode causes the pointer to immediately jump to a specific display, while other modes allow animation, with the display alternating between two or three bitmaps, like a modern LED sign. There are three different speeds for the alternating options. A Destination Display can be programmed with any combination of these modes, since each individual item has a mode setting. Whenever the selected item is changed, the new index is saved into the internal EEPROM on IC1 so that the most recent display is shown if the power is cycled. Construction This project is on a tiny PCB (19 × 12mm, coded 09111252) fitted with very small parts, packed pretty tightly. We recommend patience and experience with surface-­m ounting components, as well as all the gear commonly used for manual SMD assembly. The PCB is 0.8mm thick and has numerous vias inside pads. This is not recommended for automated assembly, since the vias can draw away solder from the joint (although there are techniques to cap the vias to prevent issues). We found that this made the PCB quite thermally conductive, so we had to wait longer for the solder to solidify as we worked. We used a fairly wide-tipped (2mm) iron as we normally recommend for most SMD work. But since the PCB holds heat well, a fine tip may be better in this case, especially to get between the components to avoid forming solder bridges. We did create a couple of accidental bridges, even between the passives, so keep an eye out for that. We have managed to place all the SMD components on one side of the PCB (see the Fig.3 overlay diagram). Here’s how we worked through our prototypes. We started by applying flux to all the pads on the component side. Start with the two ICs, which have their pin 1s at opposite ends. We found that the Microchip part had a dimple, but the ST parts have a less-obvious bevelled edge. Refer to our photos to check the part markings to confirm their orientations. These are the only polarised parts (apart from the OLED panel). Parts List – Model Headboard Destination Display Fig.3: we have kept the components on one side of the PCB. Observe this overlay diagram carefully, since there is no room for silkscreen component designators. This diagram is shown at 400% actual size, and the PCB is shown at actual size in the parts list 1 0.8mm-thick, 19 × 12mm double-sided black PCB coded 09111252 1 OLED panel with a 14-way solderable FFC connector [Buy Display ER-OLED0.32-1W, ER-OLED0.50-1W or ER-OLED0.54-1W; see text] 1 PIC16F18115-I/SN micro programmed with 0911125D.HEX, SOIC-8 (IC1) 1 ST25DV16KC or ST25DV64KC NFC tag chip, SOIC-8 (IC2) 1 antenna to suit IC2 (flexible PCB coded 06101233 or made from 1m of enamelled copper wire) 3 10μF 50V X5R SMD M2012/0805-size MLCC capacitors 3 1μF 50V X5R SMD M2012/0805-size MLCC capacitors 1 560kW ±1% ⅛W SMD M2012/0805-size resistor 1 10kW ±1% ⅛W SMD M2012/0805-size resistor 3 4.7kW ±1% ⅛W SMD M2012/0805-size resistors 1 3.3V-compatible infrared receiver module (optional) [Vishay TSOP33436] 54 Australia's electronics magazine Silicon Chip siliconchip.com.au With the OLED panel fitted, flat against the back of the PCB, the unit is only 20mm tall, 12mm wide and about 4mm thick (shown at twice actual size). We preferred the hand-wound coil antenna to easily communicate with the NFC chip. The narrow neck between the PCB and OLED panel is flexible, which should help with trying to fit the assembly into a small model. In comparison, the flexible PCB antenna is simple to use but not as sensitive as the coil antenna. Subjectively, we thought that 0.54in panels were not as bright as the others. Next, fit the three 10μF capacitors, paying close attention to Fig.3, since there is no room for markings on the PCB silkscreen. Follow with the three 1μF capacitors. Despite the resistors being thinner, we found that it was easier to solder these last, since their profile seems to capture the solder better; you can confirm their locations against the photo. Use a solvent to clean away the flux residue and allow the board to dry. Each of the OLED panels can be fitted in one of two orientations, as seen in the photos above, so be aware of this when planning how you will use the Destination Display. Just make sure the pin 1 markers align. We found soldering the panels to be quite easy; the 0.62mm pitch is comparable to that of SSOP (small shrink outline package) IC leads. If possible, set the FFC back from the edge of the PCB so you can visually confirm the alignment of the FFC traces with those on the PCB. Clean up the joints with additional flux if necessary and check again for bridges before proceeding. Loading the firmware If you need to load the firmware onto the microcontroller (which shouldn’t be necessary if you have purchased the chip from the Silicon Chip Online Shop or as part of a kit), we recommend soldering a standard five-way pin header to the five pads in a row. The pads are placed at slightly less than 2.54mm/0.1-inch, but close enough that this is doable. This header should then plug directly into the socket header of a programmer like a Snap or PICkit. Be sure to align the pins marked with a chevron (>) on the Display and programmer. You’ll need to find a way to provide 3.3V power to the chip if the programmer isn’t able to do so. Table 1: Jaycar XC3718 IR Remote Control codes Button Code 100+ 25 (0x19) CH- 69 (0x45) 200+ 13 (0x0D) CH 70 (0x46) 1 12 (0x0C) CH+ 71 (0x47) 2 24 (0x18) PREV 68 (0x44) 3 94 (0x5E) NEXT 64 (0x40) 4 8 (0x08) PLAY/PAUSE 67 (0x43) 5 28 (0x1C) VOL- 7 (0x07) 6 90 (0x5A) VOL+ 21 (0x15) 7 66 (0x42) EQ 9 (0x09) 8 82 (0x52) 0 22 (0x16) 9 74 (0x4A) siliconchip.com.au Australia's electronics magazine You can use the other pads marked “3” and “G” to apply power if this is easier. Program and verify the chip. You should see some activity on the panel; a test pattern of stripes is shown for half a second if the EEPROM is blank or its contents are invalid. Disconnect the programmer when finished. Antenna options The RFID Antenna flat flex PCB (coded 06101233) can be soldered directly to the main PCB. It worked fine during our testing, but we found that the handmade wire loop antenna was more forgiving. Since it would also be easier to form into a specific shape to be fitted inside an item of model rolling stock, we prefer it. We started with just over 1m of 0.25mm diameter enamelled copper wire. The diameter is not critical, but much finer would be finicky to handle. Wind five turns around a former with a 5cm diameter (we used an isopropyl alcohol spray bottle) and gently twist the trailing leads together. Use some tape or glue to secure the turns against each other; we used short pieces of Kapton tape, as you can see from the photos. Remove the coil from the former, trim the trailing ends to the same lengths and tin their ends to remove the enamel coating. Solder to the antenna pads on the main PCB. Wiring Fig.4 shows the wiring needed for a comprehensive installation, including a microDCC Decoder and IR receiver. For clarity, we have not shown the OLED panel. The microDCC Decoder provides the 3.3V supply and a serial signal, while the IR receiver takes its 3.3V supply from the Destination Display board and sends its data back via the pin labelled IR. August 2026  55 Fig.4: you may not need all the parts shown here; as long as you can supply 3.3V power and ground and one of the control signals, the Destination Display will be fully operational. We soldered the receiver directly to the PCB for our testing and had no problems receiving signals. Some IR receiver data sheets recommend extra components for power supply filtering, so you should consider that in the case of longer wiring runs. If you plan to use only the serial input, the IR receiver can be left off. Alternatively, if you only wish to use the IR receiver, the serial connection can be left off, and you can supply 3.3V power into either 3.3V pad and similar for the ground connection. If you wish to use the Destination Display as a fixed sign, only power needs to be applied. The Display will use index zero initially. The antenna is only needed to upload data to the EEPROM and it can be removed after uploading if it would be awkward to leave it in place. We made a few antennas and moved them around to test the different OLED sizes that we had attached to our prototype PCBs. We did most of our testing with a USB-to-serial adaptor. That would be ideal for a fixed application, such as an advertising sign or station departure board. A suitably programmed microcontroller could even be used to coordinate several different signs, such as the 56 Silicon Chip multiple departure screens at a model railway station. We have not tested this, but it should also be possible to connect multiple Displays to the same serial and IR sources, since they are simply inputs to the Display. Remember that the microDCC Decoder has limited capacity on its 3.3V regulator, so check that the load is suitable if connecting multiple Displays to a single Decoder. Naturally, any power supply used must be stable for proper operation. The default data should also result in an image on the 0.50in and 0.54in displays. We don’t think it will be usable (for a model) with these displays, but should give you confidence that the hardware is working as expected. If you want to just load our sample datasets onto the NFC chip then you can skip these next few sections and go straight to the heading labelled “NFC chip” overleaf. Defaults We used the Processing language to create the configuration programming sketch. The Processing IDE (preferably version 4.4.7 or later) can be downloaded from their website, see: https:// processing.org The sketch presents as a windowed application with buttons, text fields and the like. Processing does not provide these graphical user interface (GUI) features, so we have had to create them from scratch. There is no image editing and only very basic text editing capabilities, so we recommend using other programs to do this. You might find edge cases in the sketch that will cause it to crash, but it should be well-­behaved with sensible inputs. The internet is a wonderful thing, and on it we found a list of Australian railway station names that are duplicated (or triplicated) in different states. We have used this list as the basis for our default demo. The list is shown in Screen 1. The flash memory of IC1 is loaded with these default graphics that are used if the EEPROM does not carry valid data. It is designed to be usable on the 0.32in panel, and should respond to commands on the serial line or from the recommended remote controls. The data corresponds to the DEFAULT 0.32in.bin file in the software downloads, so you can view and edit this data. Australia's electronics magazine Destination Display Configuration for Processing siliconchip.com.au Open the sketch file (Destination_ Config.PDE) and click the Play button at the top of the window to run the program; this will open in a new window. The File Menu also has an option to export a standalone application, which will only work on the same operating system on which it is created. Screen 1 shows this window in use. There are three main steps. Firstly, the image data on the left is created or loaded. Secondly, the individual index items are created based on the image. Finally, the data is generated and exported to a file. Image data You can load image data via a file (eg, PNG, GIF or JPEG formats). The image is converted to monochrome using a threshold (“thresh.” text box) between 0 and 1. A lower value will result in more white pixels, and a higher value more black pixels. The image width is cropped at 128 pixels and the “w” and “h” fields are set based on the image. If you need to adjust the threshold, click on the number box and enter a new value, then reload the image using the “Load image” button. If the image is taller than the window, you can scroll up and down using the up and down arrows at top left. Alternatively, you can enter text in the box under “Load text”. This is a simple multi-line field, so you can’t move around within the field using The prototype being controlled by a μDCC decoder on a loco chassis. siliconchip.com.au the arrow keys; the text can only be edited at the end. You can use Backspace to remove the last character and paste from the clipboard with Ctrl-V. Delete all text with the Delete key. Use the “Load text” button to generate an image from the text in this field. This step responds to the threshold field as well as the “w”, “y pitch” and “cond.” fields. The “w” field sets the width (in pixels) of the generated graphics, while the height depends on the number of lines of text and the “y pitch” value. The “cond.” (condensed) field can be used to narrow the text if it is too wide. For example, the data generated in the adjacent screen uses the value of 0.7. You can use a value larger than 1 to expand the text if desired. The “Save img” button can be used to export this image to a PNG file; there is no prompt for a filename – it is simply saved in the sketch folder with a name based on the current timestamp. Index items The “Create fixed” buttons will do most of the work of generating the single-­screen index items. Press “CLEAR ALL” if necessary to clear any existing entries. If you want the indices to start at a specific value, you can enter this in the “index” field. You should also set the “type” field to suit the scroll rate (or FIXED for non-­scrolling items), or click the button below it to view the options. Any animated or multi-screen items need to refer back to a single-screen item, so these items will need to be generated anyway. You should see the sample views at right populate when the “Create fixed” button is pressed, and you can scroll up and down through these with the “UP” and “DOWN” buttons. Dual-screen items can then be generated by setting the index, type and index 1 and index 2 values before pressing “Create #”. The index 1 and index 2 values can be entered manually or loaded by pressing the “<1” or “<2” buttons next to the desired images. For example, to generate an animation showing “ADELAIDE AIRPORT” over two screens, click the “1X MED” button until it shows one of the 2X types, then click “<1” next to “ADELAIDE” and “<2” next to “AIRPORT” at right. Finally, click “Create #31” to generate the item, which can be previewed by scrolling down the list at right. Triple screen items are created in a similar fashion by choosing one of the 3X types and ensuring that all three pointers are loaded correctly. Note that triple items can only use an index up to 31, since there are only 16 bits available for indexing. The index will increment after each item is created, so it is not hard to create several similar items. Screen 1: the Silicon Chip Destination Display Configuration sketch provides an easy way to generate the data file necessary to create custom displays and can mix and match different styles of animation. Australia's electronics magazine August 2026  57 Silicon Chip PDFs on USB ¯ A treasure trove of Silicon Chip magazines on a 32GB custom-made USB. ¯ Each USB is filled with a set of issues as PDFs – fully searchable and with a separate index – you just need a PDF viewer. ¯ Ordering the USB also provides you with download access for the relevant PDFs, once your order has been processed ¯ 10% off your order (not including postage cost) if you are currently subscribed to the magazine. ¯ Receive an extra discount If you already own digital copies of the magazine (in the block you are ordering). Exporting Check that values such as the x, y, w, h, rotation (from Fig.1) and “bright.” (brightness from 0 to 255) values are correct. The scale values are ignored by the current version of the software, while the x bytes and pointer values are updated automatically. It’s easy to change any header data later using the ST25 NFC Tap app. Press “UPDATE DATA” to generate the dataset. The file size will be reported at the bottom of the column, so you can check that it will fit in IC2’s EEPROM. If all is well, press “SAVE DATA”, which will create a BIN file named for the current timestamp in the sketch directory. You can now transfer the file to your NFC device (such as an Android mobile phone with the ST25 NFC Tap app) to program the EEPROM chip on the Destination Display using the instructions in the main article. EEPROM capacity If you do wish to view or modify the exported file on your computer, we suggest using the HxD hex editor, which can be downloaded from https://mh-nexus.de/en/hxd Since the data is a monochrome, uncompressed bitmap, a byte of data can hold eight pixels. Our sample file is 3776 bytes, of which the headers are 192 bytes and the image EACH BLOCK OF ISSUES COSTS $100 NOVEMBER 1987 – DECEMBER 1994 data is 3584 bytes, corresponding to a bitmap of 56 pixels by 512 pixels. This effectively contains 32 unique screens that each measure 56 × 16 pixels. The 0.50in panel has the most pixels, with 4224 pixels, requiring 528 bytes. The 8kiB ST25DV64KC can store 15 full-sized images to suit the 0.50in panel, with room to spare for 64 index items and the header. NFC chip To program the NFC chip, we strongly recommend ensuring the Destination Display is powered off, since the NFC chip cannot handle the EEPROM being accessed via both channels (I2C and NFC) at the same time. This will also ensure that the EEPROM data is properly reloaded after changes have been made. We used the ST25 NFC Tap mobile app on an Android device (siliconchip. au/link/ac38). There is also a version on the Apple App Store (siliconchip. au/link/ac39) but we have not tested it. Presumably, it works identically. We have created several data files so that you can perform this step without using the configuration program. These files are in the software download package (siliconchip.au/ Shop/6/3629). Although they may be suited to a specific display type, the ST25 NFC Tap App allows individual bytes to be edited, so parameters like the display orientation and size can be adjusted easily to suit different panels. One file has small bitmaps of the numbers from 000 to 255 with matching indexes, so this can be used to make a simple test that your control signals are working correctly. This is JANUARY 1995 – DECEMBER 1999 JANUARY 2000 – DECEMBER 2004 JANUARY 2005 – DECEMBER 2009 JANUARY 2010 – DECEMBER 2014 JANUARY 2015 – DECEMBER 2019 OUR NEWEST BLOCK COSTS $150 JANUARY 2020 – DECEMBER 2024 OR PAY $650 FOR THEM ALL (+ POST) WWW.SILICONCHIP.COM. AU/SHOP/DIGITAL_PDFS 58 Silicon Chip Only three wires are required for the Destination Display: GND, 3.3V and the serial line for control. The μDCC decoder can supply enough current to run the Display. You can also use a USB/serial adaptor to control fixed Displays. Australia's electronics magazine siliconchip.com.au EEPROM data format While many of the details of the data format are not critical to use the Destination Display, some are worth knowing so you can tweak the data files to suit different displays or even to suit specific installations. For example, we envision that some uses of the Display will involve placing it behind a window cut out of a model so that only a small part of the Display is visible. Being able to make small adjustments to the location of the graphics within that window is handy and straightforward. The data file starts with a block of 16 bytes. After this, there are several four-byte index items, followed by the bitmap data. Organising the data in blocks of four bytes makes it easier to navigate in the ST25 NFC Tap App, since it displays the data in four-byte rows. The adjacent screen grab shows the start of a simple data file with four index items. The table directly below summarises the data structures. The first four bytes are simply a file type check, used to confirm that the EEPROM has been correctly loaded. The x and y fields determine the position of the top-left corner of the where the graphics are displayed. The w and h fields determine the extent of the area in which the graphics are displayed, while the orientation field is the value noted in Fig.1, where it indicates the corner that would be at upper left. The x-bytes field after that is used to know how many bytes of data to read for each row of pixels to be displayed. The y-pitch field is not used, although it should match the h field. Brightness is simply a raw value (0-255) that is written to the OLED controller’s brightness (or contrast) register. The bitmap pointer field holds the absolute address of the start of the bitmap data. It is a little-endian value (LE, the least significant byte is first), so the bytes “20 00” in the adjacent screen grab are read as the value 0x0020 or decimal 32. The first three index items (starting at addresses 16, 20 and 24) are single-screen types and will be activated with commands 0, 1 and 2 respectively (values in the first byte of each row). Their pointer fields (0x0000, 0x0010 and 0x0020) are offsets from the bitmap pointer field, so their bitmap data will be at decimal addresses 32, 48 and 64 respectively. The fourth index item (at address 28) responds to command 3 and is type 6, meaning there are two displays and they update every second. The pointers refer to item 0 and item 1. So it will alternate between displaying bitmap data from address 32 and address 48. The table at lower right shows the meaning of the other item types. Editing We mostly found ourselves changing the x, y and orientation fields to quickly modify a file to suit different displays. There would be little need to change the other parameters, although it’s easy enough to change the brightness if this is needed. Note from Fig.1 that the 0.32in panel uses different orientation values. This also encodes information to ensure that the data is ordered correctly for display. Values other than 1-8 are not valid and the display will not show any output if this is not observed. Data Format – 16-byte header You don’t need to know about the data format to use the Processing sketch, but it can help to understand how to make changes if things aren’t working as expected. Value Type description 0 (0x00) Null, used to mark an entry as invalid. 1 (0x01) Single display, slow scroll (7 lines/sec) 2 (0x02) Single display, scroll (10 lines/sec) 3 (0x03) Single display, fast scroll (20 lines/sec) 0x44 ‘D’ 0x45 ‘E’ 0x53 ‘S’ 0x54 ‘T’ 8-bit x-field 8-bit y-field 8-bit w-field 8-bit h-field 0x01 (unused) 0x01 (unused) 8-bit orientation 8-bit x-bytes field 4 (0x04) Single display, fixed 8-bit y-pitch field 8-bit brightness 16-bit (LE) bitmap pointer field 5 (0x05) Two displays alternating every 500ms 16-bit (LE) pointer field 6 (0x06) Two displays alternating every 1s 7 (0x07) Two displays alternating every 2s 8 (0x08) Three displays alternating every 500ms 9 (0x09) Three displays alternating every 1s 10 (0x0A) Three displays alternating every 2s One 4-byte single-screen item 8-bit entry 0x01-0x04 type One 4-byte double-screen item 8-bit entry 0x05-0x07 type 8-bit entry index 8-bit entry index One 4-byte triple-screen item 8-bit entry 0x08-0x0A type Three 5-bit entry indices padded with a leading zero. [0cccccbb bbbaaaaa] Bitmap data Each visible row of bitmap data consists of x-bytes count of bytes, with the left-most pixel being the MSB of the first byte. There should be h rows of bitmap data. siliconchip.com.au Australia's electronics magazine August 2026  59 Screen 2: choose the Memory tab from the main page of the ST25 NFC Tap app and use “Fill memory from file” to upload a data file. Screen 3: select the source file, ensure that the destination offset is zero and tap OK. It may take up to 10 seconds for the memory view to appear, confirming that the write has completed. Screen 4: the Read memory option allows the EEPROM on IC2 to be edited directly. Note that the displayed memory contents are in hexadecimal. the “256 Entries.bin” file. The screen seen in Photo 2 is one of these bitmaps. Open the ST25 NFC Tap App and place the device over the antenna. When the tag is detected, switch to the MEMORY tab (Screen 2) and select “Fill memory from file”. Select the source file and ensure that the Destination offset is zero. Press OK to transfer the file. This might take up to ten seconds, so wait until it completes and the memory contents are shown as in Screen 3. At this point, you can power up the Destination Display and see that it shows the screen that would be expected from index item zero. If you need to tweak the parameters, power off the Display and rescan the antenna with the App. From the MEMORY tab, choose the Read memory window. Press OK to perform a read and then hold your finger on a memory location to edit it; a row of four memory locations will pop up as seen in Screen 4. You can edit the values and then press Write bytes. The values are in hexadecimal, although they do not have a leading base marker like 0x or &H. The EEPROM data format panel has more detail about specific values that you might want to change. DCC PROJECT KITS DCC Destination Display (SC7697, $22.50) includes everything in the parts list, except for the screen (see below). The kit includes 1m of enamelled copper wire for the antenna 0.32in OLED Screen (SC7698, $5.00) 0.50in OLED Screen (SC7699, $6.50) 60 Silicon Chip Australia's electronics magazine Other uses While we intended this design to be used in model railways and the like, we think it could be useful anywhere that a small, simple display is needed. It can interface directly with a 3.3V device and only needs one signal wire. With the right bitmaps loaded, it could be used in much the same way as single-digit devices like Nixie tubes, although the control interface is quite different. Any device that needs a small status display could do so with an appropriately programmed Destination Display, providing dozens of different outputs (or more). These could include numbers, text, images, or anything else that can be encoded in SC a small bitmap. siliconchip.com.au CIRCUIT NOTEBOOK Interesting circuit ideas which we have checked but not built and tested. Contributions will be paid for at standard rates. All submissions should include full name, address & phone number. Simple clap-operated switch This simple clap-operated switch circuit lets you switch a relay on or off by the sound generated by clapping your hand. The range of this clap control relay is a metre or two. Anyone who is interested in electronics can build it – all the components used are easily available. The heart of this circuit is a decade counter IC (a CD4017 or similar) with 16 pins. Two NPN transistors are also used. A small current is applied to the condenser microphone via a 10kW resistor from the 6V supply. Its AC output is coupled to the base of BC547 transistor Q1 via an electrolytic capacitor. Q1 gets a base bias current via the 1MW resistor. Its collector is connected to the 6V supply via a 100kW load resistor and produces the output that feeds the clock input terminal (pin 14) of IC1. Q1 amplifies the small voltage from the microphone to send clock pulses to the counter when a loud enough sound is picked up. When a clap sound is picked up, output pin 2 (O1) goes high, supplying current to the base of transistor Q2 and energising the relay coil. The LED indicates when this happens, and the COM/NO contacts of RLY1 can energise a load, like a lamp. Another clap will cause the O2 output (pin 4) to go high, resetting the counter and de-energising the relay. Diode D1 protects Q2 from the coil’s back-EMF voltage. Raj. K. Gorkhali, Hetauda, Nepal. ($40) Auto-recharging battery backup for a clock radio This circuit was developed as I was annoyed at having to replace the 9V backup battery in my bedside clock radio, which always seems to have failed when a blackout occurs. As well as the inconvenience, the environmental impact of continually throwing away too many batteries concerns me. I decided to use a rechargeable 9V battery that would be trickle charged while power was available. I tapped the 8.6V secondary of the existing transformer in the clock, then rectified and filtered it using the circuit shown, mounted on a small piece of veroboard. However, I found that the 9V backup circuit didn’t like sharing the same power supply as the clock. I salvaged a transformer from another junked clock radio to siliconchip.com.au provide another isolated supply. I managed to shoehorn this all into the clock and get it working. The 9V battery is trickle charged at about 2mA due to the 820W resistor while power is available. This modification involves mains wiring (which should not be attempted if you are not experienced with mains voltages), and modification to the clock, which would void any warranties, if applicable. An alternative arrangement is to do away with the mains wiring, the Australia's electronics magazine extra 8.6V transformer, and use a 12V DC plugpack. Feed power to the 9V battery clip on the clock via the 820W resistor and the 1N4004 diode. This can be done using a reverse-connected 9V battery clip to make connection easy. I found the power consumption of the clock jumped from 1W to 2.5W. This increases the running cost of the clock (at 40¢/kWh) from $3.50 per year to $8.76 per year. I think I can live with that. Geoff Coppa, Toormina, NSW. ($40) August 2026  61 Modem/Router Watchdog enhancement I found the Modem/Router Watchdog project in the November 2023 issue (siliconchip.au/Article/16017) an ideal solution, especially during extended travels away from home. Over time, I noted it suffered a number of reboots, which I believe was due to a combination of IP/NBN service irregularities and some WiFi peculiarities in the packaged WebMite 5.07.07 firmware. While diagnosing this, I also became intrigued by the 0.91-inch OLEDs featured in a November 2024 Modules article (siliconchip.au/Article/17027). While these technically worked, they were a bit tiny for my ageing eyesight, so I adapted the same code to the 1.3inch OLED screen. This just required changing the OLED String Length parameters from 128 to 132. Along with upgrading to the later PicoMite v6 firmware, my MMBasic file (siliconchip.au/Shop/6/2781) is a combination of the Modem Watchdog and the OLED code that shows progress status updates, and some extra delays between internet access calls to improve stability. The OLED increases the unit’s current draw by only about 7mA. Another feature I added is monitoring the CPU boot count, to provide some indication of reboot activity over time. As a tip, this can be re-zeroed before putting the Watchdog into service by using the Pico’s Flash_Nuke file (as described in the March 2025 issue, on page 89). However, note that a full code and option reload is then required. The standard ‘Running Normal’ screen is deliberately a little different (to stand out from the other status conditions), and includes the aforementioned boot count. Hardware-wise, I mounted the OLED on a sub-strip board and connecting it to header pins GP4 (SDA), GP5 (SCL) and adjacent ground pin on the Pico. I sourced the 5V supply directly from D1’s cathode on the Watchdog board. The included photo, just below, shows the checking NTP display, but with the acrylic case lid removed for clarity. Glenn Paterson, Cherrybrook, NSW. ($100) Safely measuring HV battery pack voltages This design was made up to measure voltages in large battery banks, nominally 48V DC. The concern was how to do this without running sense wires; battery banks are massive, can deliver thousands of amps, with positive chassis ground (a telecom standard). This can catch people out with really bad results. I came up with these optical solutions, originally using Broadcom BFBR-xxxx optical TX/RX modules, but they are a bit costly. So I redid design with standard, inexpensive components. There are two versions of the circuit, one which uses an opto-­ coupler, with the other using plastic optic fibre cable between LED1 and phototransistor Q4. In both cases, the ‘transmitter’ is a simple reverse breakdown BJT ‘pulser’, which is a kind of relaxation oscillator that produces a frequency 62 Silicon Chip related to the applied voltage. The 18nF capacitor charges up via a 15kW resistor and once it reaches a particular voltage, Q1’s reverse-­ biased base-­emitter junction breaks down and sends a pulse of current through LED1, producing a light pulse. When the pulse is produced, the capacitor is partially discharged, so the cycle repeats. The higher the battery voltage, the less time it takes for the capacitor to charge to the breakdown voltage. In the top version, a cascade amplifier and 74HC74 IC clean up the pulses from Q4 to produce a 50% duty cycle signal that can be sent to just about any microcontroller (eg, a Raspberry Pi Pico). I tested a standard red LED and IR LED. Both worked, but to get the NPN photo-transistor to fully saturate (switch on), a high-intensity Australia's electronics magazine LED works best. I tested it with plastic optical fibre up to 10 meters long. One way to get this is to chop up a cheap TOSLINK optical audio cable. I quite enjoyed the PicoMite BASIC articles, which brought back good memories of learning to code assembly language and BASIC on a 6502. So I’ve supplied BASIC source code that will run on a PicoMite and drive a WaveShare Pico-LCD-1.8 128×160-pixel TFT display, which plugs straight into a Pico (siliconchip.au/Shop/6/2795). This measures the frequency, converts it to a voltage and displays it on the screen. This requires a linearisation equation, which I created using curve fits from data captured with an oscilloscope. The required data will vary based on the components used. James Langdon, Kalgoorlie, WA. ($90) siliconchip.com.au Plugpack-powered Little Jim Valve Radio Inspired by the Electronics Australia article on the Little Jim valve radio receiver (June 2000, page 90), I have built an updated version using modern available parts; it requires no mains wiring. The original Little Jim dates back much further, to Wireless Weekly, May 20th & 27th, 1938. In my attempt, I have favoured the Hartley configuration, which worked first time, with no separate tickler coil to worry about phasing feedback windings correctly. I used a Jaycar L15 (LO1238) material toroidal former to wind L1, using 18 turns of 0.5mm diameter plastic-­ covered copper wire from an old telephone cable. This allows it to cover most of the AM broadcast band. Plain enamel covered copper wire would also suffice, if you don’t mind the task of removing the enamel for the twoturn tap, ground and grid connections. 18 turns through a toroidal former is far easier than making a 100-turn air-wound coil! Feedback from the first triode (V1a) cathode is coupled into L1’s two-turn tap. I used Jaycar’s RV5728 tuning capacitor for VC1, as these are available off-the-shelf in most stores. As I mentioned, this covers most of the AM band, and with a series capacitor and suitable coil, it can even tune into shortwave amateur radio bands. The valve heaters are wired in series to run from the incoming 12V AC (with a 4.7W series resistor to slightly lower the filament power), while the ‘HT’ of just over 40V is produced using a voltage tripler. The highest voltage in the set is around 52V DC, which is still high enough to be a shock hazard, so it needs to be built into an insulated enclosure. The set can be built on an aluminium chassis (like a mess tin), or a PCB could be designed for it. My test design was strung together in haphazard fashion on my workbench, and worked well with minimal hand capacity effects, despite the messy layout! Any of the 12AU7, 12AT7, 12AX7, ECC82 series and equivalents should function. I tested the first three of the JJ brand – currently manufactured types. The feedback setting is best with the receiver just bringing in background noise, before a whistling noise can be heard on tuning a station. Such a heterodyne whistle means the set is oscillating, which should not be allowed to continue in a set like this without antenna isolation. ANT1 is for a short length of wire, probably two meters or less, while ANT2 is for a greater length if you have room. ANT3 is for a long wire antenna, preferably outside. An indoor antenna can work in high signal strength areas, like a major city. Lance Neame ZL3LAD, Christchurch, New Zealand. ($80) Circuit Ideas Wanted Got an interesting original circuit that you have cleverly devised? We will pay good money to feature it in Circuit Notebook. We can pay you by electronic funds transfer, credit or direct to your PayPal account. Or you can use the funds to purchase anything from the Silicon Chip Online Store. Email your circuit and descriptive text to editor<at> siliconchip.com.au siliconchip.com.au Australia's electronics magazine August 2026  63 By Andrew Levido Power Electronics Part 8: AC to AC Converters Having covered DC-DC, DC-AC and AC-DC converters in the Power Electronics series published in the November 2025 through May 2026 issues (siliconchip.au/Series/452), I have finally been convinced to fill in the missing piece of the puzzle: AC-AC converters. I was not originally planning to include this topic in the series because they are not the sort of converters that readers of this magazine are likely to come across in practice, but since I was asked specifically, here we are! You won’t find an off-the-shelf chip or a ready-made development board to experiment with AC-AC converters, nor are you likely to be called on to repair one for a neighbour or family friend. Still, their operation is interesting enough to warrant a closer look. AC to AC converters, as the name suggests, convert power from one AC form to another. The inputs and outputs can be single-phase or threephase, and they can be configured to change the number of phases, the frequency and/or the voltage. They fall into two broad categories: those with an intermediate DC link and those without. The former is really a combination of an AC-DC converter and a DC-AC converter with a DC energy storage element in between. In this sense, they are not true AC-AC converters at all. We have already covered AC-DC and DC-AC converters in some detail, so I won’t provide any further details on such systems here. This article will focus instead on direct AC to AC converters, the oldest and most common of which is the cycloconverter. Naturally commutated cycloconverters (we’ll cover what this means in a moment) are used almost exclusively to drive very large induction or synchronous motors, typically in the 5-50 megawatt range. Their applications include marine propulsion, where turbogenerators provide the power and large electric motors are used to drive conventional propellers or Azipods. You will also find them driving very large industrial machines, such as rotary cement kilns. These are huge inclined cylindrical 64 Silicon Chip furnaces up to 6m in diameter and often over 200m long. Some of these kilns can process 10,000 tonnes of cement per day, so they require really big motors and drives to control their rotation! Another application that comes to mind is semi-autogenous grinding (SAG) mills used in mineral refining. SAG mills use hardened steel balls or cylinders to pulverise the ore, making it ready for further processing. They are huge steel drums, up to 12m in diameter and a similar length. This large diameter lets the falling balls or rods gain plenty of kinetic energy to hammer the ore into dust. A typical 12m diameter SAG mill can hold up to 175 tonnes of material, so it takes quite some torque to turn. Naturally commutated cycloconverters are used in these ultra-highpower applications because they are pretty much the only electronic drives able to deliver the high voltages and currents required. These motors typically require many thousands of amps per phase at a couple of kilovolts! Operating principles The circuit diagrams and waveforms associated with cycloconverters can get quite complicated, but we can understand what’s going on if we take it step by step. Fig.1 shows the simplest direct AC-AC converter I can think of. A set of four switches selects half-cycles of an AC source to put together a lower-frequency AC output waveform (in this case, one third of the input frequency). Switches S1 and S4 are closed during the first positive half-cycle of the source voltage, while switches S2 and S3 are closed in the first negative half-cycle, to provide the first two positive voltage pulses at the output. Switches S1 and S4 are closed again for the next full cycle of the input, to Australia's electronics magazine produce the third positive pulse and the first negative pulse of the output waveform. This process continues, with the switches selecting the appropriate half-cycles in turn. The output is certainly AC, but if we are interested in the fundamental component – and we almost always are – this is not a very effective circuit. The output is a kind of square wave with notches in it where the voltage falls to zero. You can imagine the harmonic content is pretty bad, especially the harmonics at twice the input frequency. We could do a lot better with a three-phase source and six switches because the resulting output would be much closer to a square wave. However, neither circuit allows us to control the amplitude of the AC output if the input voltage is fixed. You could think of this circuit as a kind of rectifier where we can control the polarity of the output at will. What we really need is a rectifier that can produce a positive or negative output at any arbitrary voltage level. You may recall from the article on AC-DC converters in the September 2025 issue that a full-wave, phase-­ controlled rectifier with an inductive load does exactly this. I have reproduced the schematic of such a rectifier in Fig.2. This converter has the special property of operating in two quadrants – the average output voltage ‹vx› can be positive or negative, depending on the thyristor firing angle, although the load current can only ever be positive. If the filter inductance is large enough, we can ignore the ripple current, so the load voltage Vload is equal to ‹vx›. The output voltage is Vload = (3Vll ÷ π)cos(φ), where Vll is the peak lineline voltage and φ is the thyristor firing angle. As we change the firing angle siliconchip.com.au from zero to π radians (0 to 180°), the load voltage changes from +(3Vll ÷ π) to -(3Vll ÷ π). When we are using this circuit as a rectifier, we fire the thyristors at a fixed phase angle to produce a fixed DC output, as shown in the upper chart. However, there is nothing preventing us from continually changing the firing angle to synthesise a rising or falling waveform, as shown in the lower chart. We could even produce a sinusoidal output with an arbitrary amplitude and frequency if we manipulate the firing angle appropriately. This is the principle on which all cycloconverters work – creating a lower-frequency AC output by ‘stealing’ appropriate sections of the input AC waveform. There are some limits on this, of course. The peak amplitude is limited to (3Vll ÷ π), as mentioned above, and the maximum output frequency is limited to some value well below the mains frequency. We will look into the frequency limit later on. Low-­ frequency output is not a problem for typical cycloconverter applications – 200m-long kilns don’t spin fast! If we want a sinusoidal output with an amplitude of Vo and a frequency of ωo, we have to control the firing angle according to the expression φ(t) = cos-1 (πVo ÷ 3Vll)sin(ωot). This looks messy, but the stuff in brackets after the inverse cosine operator (cos-1) is just a constant relating the input voltage to the desired output voltage, so the inverse cosine of this is also a constant (for a given input and output voltage). The firing angle therefore simply varies sinusoidally at the frequency of the output, just like the duty cycle for sinusoidally modulated PWM. That is obviously pretty easy to implement with a microcontroller and a few lines of code. Fig.1: the simplest conceivable AC-AC converter selects half-cycles of the input waveform to construct a crude AC output at a lower frequency. Fig.2: a phase-controlled rectifier with an inductive load can produce a positive or negative average output voltage. A cycloconverter is such a rectifier with the firing angle varied to produce a sinusoidal output. Fig.3: this six-pulse, four-quadrant, singlephase naturally commutated cycloconverter circuit is the building block for all sorts of multiphase naturally commutated cycloconverters. Naturally commutated cycloconverters If the firing angle is fixed, this circuit is a rectifier, but if we vary the firing angle as described above, this circuit is a six-pulse two-quadrant naturally commutated single-phase cycloconverter. Quite a mouthful for sure, but easy to break down: • Six-pulse because it is a threephase full-bridge arrangement that delivers six half-cycle pulses to the output per input cycle. siliconchip.com.au • Two-quadrant because the output voltage can be positive or negative, but the current must always be positive. • Naturally commutated because each thyristor is switched off when another is switched on, and the current commutates from one to another. • Single-phase because the output has a single phase. The applications mentioned above almost always require three-phase, Australia's electronics magazine four-quadrant operation. It is a simple matter to add a second set of ‘reverse’ thyristors, as shown in Fig.3. The original ‘forward’ thyristors are used when the output current is positive, while the ‘reverse’ thyristors are used when the output current is negative. This circuit is the basic building block for all sorts of naturally commutated cycloconverter configurations. To get three-phase output, we have August 2026  65 a couple of choices. The simplest arrangement is shown in Fig.4(a). You can look at this as three sets of the ‘top half’ of the single-phase building block circuit; one for each output phase. This is a relatively simple arrangement with only(!) 18 thyristors, but because we have chopped off the lower half of the six-pulse circuit, we only have three half-cycles available to construct our output. This is therefore 66 Silicon Chip a three-pulse, four-quadrant, threephase cycloconverter. To retain the six-pulse output, we have to connect three single-phase ‘building block’ cycloconverters in a star arrangement, as shown in Fig.4(b). This has 36 thyristors and is known as a six-pulse, four-quadrant, three-phase cycloconverter. Despite the apparent complexity, I am sure you can see that this is really Figs.4(a) & (b): typical three-phase cycloconverter circuits contain a lot of thyristors and can look very complex. Ultimately, you can break them all down to a series of phase-controlled rectifiers arranged in different ways. Australia's electronics magazine just a set of phase-controlled rectifiers that can synthesise a low-frequency sinusoidal output. 12-pulse cycloconverters At the upper end of the power spectrum, it is common to use 12-pulse cycloconverters, as shown in Fig.5(a) & (b). These consist of six of the sixpulse single-phase cycloconverter building blocks. The configuration on the left has two separate three-phase outputs because it is really two six-pulse three-phase cycloconverters (shaded yellow and blue respectively) feeding a motor that has two separate windings for each phase. Note that this is not a six-phase motor, as it has three pairs of windings, each electrically displaced by 120°. A true six-phase motor would have six windings, each displaced by 60°. The key to achieving 12-pulse operation is to shift the relative phase of the two converters by 60° so their pulses are effectively interleaved, with one converter providing the six odd-­ numbered pulses and the other providing the six even ones. This relative phase shift is obtained by wiring one of the transformer secondaries in star configuration and the other in delta. By the way, it is not a huge difficulty to use a special 2×3-phase motor in these circumstance, since the motor and transformers will almost certainly be custom designed and built for the application. You don’t get 20MW motors or transformers off the shelf! The circuit on the right is a 12-pulse cycloconverter with a standard threephase output. In this case, the star-fed and delta-fed six-pulse converters are connected in series, and the stacked pairs are star-connected, just like the six-pulse converter in Fig.4(b). Other arrangements are possible. Why would we go to the complexity of using a 12-pulse cycloconverter when a three-pulse or six-pulse one will do the job? The first reason is that the larger configurations spread the switching out over many more power devices, resulting in lower device stress. Remember that we are dealing with many megawatts; 10,000A per phase is not uncommon. Secondly, the higher the pulse number, the higher we can make the cycloconverter’s maximum output frequency, the lower the torque ripple, and the higher the efficiency. It all comes down to harmonics – a higher siliconchip.com.au Figs.5(a) & (b): these twelve-pulse cycloconverters like these can be used to drive AC motors in the 10-50MW range for applications like ship propulsion, rotary cement kilns and SAG mills. pulse number reduces the harmonics on both the input and the output of the converter. We already know that on the input side, higher harmonic content means a lower power factor, and this impacts the sizing of the switchgear and transformers in the supply network. On the load side, higher harmonics mean higher motor losses, as the motor is the output filter (or a very large part of it) – and only the fundamental current contributes to useful output torque at the motor shaft. The rest is dissipated as heat or unwanted torque ripple. Harmonics It is way beyond the scope of this article (and my abilities) to attempt to calculate the magnitude of the harmonics of a cycloconverter because they are extremely complex and highly dependent on the application. However, we can get a good feel for them by looking at which harmonics will be produced. We can assume that, in general, the amplitudes of the harmonics fall off as the harmonic number increases. The harmonics on the supply side are influenced by the pulse number siliconchip.com.au and the input frequency. For a p-pulse converter, the input harmonics will occur at frequencies (pn ± 1)fi, where n is an integer from 1 to infinity and fi is the input frequency. A three-pulse converter (p = 3) will have input harmonics at 2fi and 4fi for n = 1, at 5fi and 7fi for n = 2 and so on. A 12-pulse converter will have input harmonics at 11, 13, 23, 25... times fi. The lowest input harmonic present for the 12-pulse cycloconverter is the 11th, compared with the fifth for a sixpulse converter and the second for a three-pulse converter, so the difference is significant. The cycloconverter’s output will have harmonics related to the output frequency, plus those related to the input frequency, because these converters have no internal energy storage to decouple the two. The output is effectively modulated by the input, so the output harmonics contain the ‘beat’ (or sideband) frequencies that result from the two interacting. The expression for the output harmonic frequencies is pn fi ± mfo, where and fi is the input frequency, fo is the output frequency, and both m and n are integers with the condition that pn + m is odd. Australia's electronics magazine For a three-pulse cycloconverter, there will be output harmonics at the frequencies 3fi, 3fi ± 2fo, 3fi ± 4fo, 3fi ± 6fo… for n = 1, 6fi, 6fi ±fo, 6fi ± 3fo, 6fi ± 5fo… for n = 2 and so on. A 12-pulse has harmonics at frequencies 12fi, 12fi ± 2fo, 12fi ± 4fo, 12fi ± 6fo… for n = 1, 24fi, 24fi ± 1fo, 24fi ± 3fo, 24fi ± 5fo… for n = 2 and so on. Because of the sidebands, some harmonics may have a frequency below the input frequency, fi. These subharmonics cannot be filtered by the motor inductance and can result in potentially dangerous or damaging subharmonic oscillations in speed and torque. We want to avoid them if we can, or at least be sure their amplitude is very low. It is this phenomenon that dictates the maximum output frequency of cycloconverters in most cases. Isolated high-frequency cycloconverters There is another application of cycloconverters that is perhaps a bit more relatable, one sometimes used in grid-tied solar inverters. These typically use Mosfet switches that we can turn on and off at will, so we drop the ‘naturally commutated’ label. August 2026  67 A simplified example is shown in Fig.6. The circuit is fed with a high-­ frequency AC current source, usually derived from a resonant DC-AC converter. The input current passes through a high-frequency transformer to provide isolation from the power grid. A blocking capacitor, Cb, prevents the mains frequency from reaching the transformer. The load is the grid, shown here as a voltage source with a peak amplitude of Vg and a frequency of ωg. During the positive half-cycles of the grid voltage, Mosfets Q3 and Q4 are switched on as depicted in the top graph. Q1 and Q2 are switched complimentarily with a 50% duty cycle at the rate of the high-frequency source. When Q1 is on and Q2 off, the current sourced from the transformer is connected to the load. This current, io, flows into or out of the load and returns via Q4. When Q2 is on and Q1 off, the transformer secondary current circulates via Q2 and Q3. We can therefore control which parts of the sinusoidal secondary current we pass to the load. If Q1’s on-time coincides with a positive half-cycle of the high-frequency current, a positive half-cycle of current would be pushed towards the load. You can see this happening in the middle of the lower graph in Fig.6. If Q1’s on-time coincided with a negative half-cycle of the source current, a full negative half-cycle of current would be passed to the load. By controlling the phase of Q1 and Q2’s switching with respect to the high-­ frequency current, we can deliver partial cycles with any average value between the two extremes. In this example, we control the phase shift to ensure that the average current ‹io› is sinusoidal and in phase with the grid voltage, as shown in the lower chart in blue. The LC filter smooths this current to produce the grid current, ig, shown in green. During the negative grid voltage half-cycle, the whole process repeats, but this time Q1 and Q2 are always on, while Q3 and Q4 are switched complimentarily to shape the negative half-cycle current. The resulting current is sinusoidal and in-phase with the grid voltage, meaning a unity power factor at the grid interface. If the switching frequency is fairly high, the filter components can be quite small. This is a cycloconverter because it operates on the principle of ‘stealing’ bits of the high-frequency source to synthesise a lower-frequency AC waveform. The difference here is that we are doing it with current rather than voltage, and that we can turn switches off whenever we want. Matrix converters Earlier, we looked at cycloconverters Fig.6: the isolated highfrequency cycloconverter is sometimes used in grid-tied inverters to ensure the power factor at the grid interface is unity. Fig.7: the matrix converter uses bidirectional controllable switches and complex PWM switching strategies to implement direct AC-AC conversion. Practical implementations are rare because it is cheaper and better harmonically to use an AC-DC and DC-AC converter in series. Fig.8: the solid-state transformer (SST) is another AC-AC converter technology that has not really become practical. Mains-frequency transformers are cheaper and more reliable in power distribution applications. as phase-controlled rectifiers because this is the easiest way to understand their operation (at least in my opinion). However, if you look at the three-pulse cycloconverter circuit in Fig.4(a), you might be able to see it in another light. The circuit consists of nine pairs of back-to-back thyristors, one pair connected between each input phase and each output phase. We can think of this as a kind of switch matrix. Building on this idea, if we could use some kind of bidirectional fully controllable switch in place of each thyristor pair, we could connect any input phase to any output phase at any point in time (subject to not shorting the input or output phases together) to synthesise the output of our choosing. This is known as a matrix converter (Fig.7) and was proposed in the 1980s. The switches can be controlled using high-frequency PWM to generate a variable three-phase output without some of the limitations of cycloconverters. For example, we can improve the harmonics by using higher frequency switching and economical filters, although we won’t escape the beat-frequency problems because there is no DC link storage element to decouple the input and output. It is a complex device to implement because we do not have bidirectional fully controllable semiconductor switches. We would have to use nine pairs of Mosfet or IGBT switching elements and associated floating gate drivers to implement the circuit. A power-factor-corrected threephase rectifier and a three-phase bridge can do exactly the same job while being less complex and having better harmonic performance, so you just don’t see practical matrix converters. They are nonetheless beloved by academics, so they appear in all the textbooks and they seem to have inspired plenty of research papers describing novel switching algorithms. It is an interesting rabbit-hole to explore if you are so inclined. Solid-state transformers The solid-state transformer (SST) was first proposed and patented by William McMurray (a truly famous name in power electronics) in the 1960s. This circuit, shown in Fig.8, modulates the incoming AC waveform at a high frequency and demodulates it on the other side of a high-frequency transformer to reconstruct the original waveform. I have drawn the circuit as McMurray described it, with a centre-tapped transformer and two switches on either side, but you could equally use a transformer with single windings and a full bridge on either side. McMurray envisioned SSTs taking over from conventional transformers in power distribution systems due to their reduced size, but unfortunately, it was not to be. They require bidirectional switches, which means eight devices with isolated drive circuits (or 16 if you use a full bridge on either side) so they are just not as cost-effective or reliable as a conventional passive mains-frequency transformer. Low cost and high reliability are close to the top of the wish list for power distribution systems, so the idea never took off. Conclusion That’s it for AC-AC converters. At the very top end of the power range, the cycloconverter is hard to beat, but in most other applications, combining AC-DC and DC-AC converters is a far SC better option. Back Issues The UK ’s Circui t Pr Electractical onics fT sTa r TE r premier electron ics so and Underst Surgery using anding and gyrato rs Make computin g ma ker ma Mrom Finishin h Mic E Th g gazine light con the PicoMite ite E ‘sP La troller T’ aT Audio softwar smart e Designi Out sw iTc h- on discreteng a practic ! al audio op am a Mi p GPS-Sy Analog nchronise ue Cloc d k it wit Ta 6- d Ec WIN M mTo ICRO A CH u Dev ch AR IP elo 10 adE pm 00 LLionPico r Es Kit ent PrEc MisE VaLU ite smis Ta n c Es Tolight E b ox finECont art YoUr TUnErolle dEsiG r ns WIN! Microch ip Inte grat Graphic ed sE s and Touch M CuriTE osity sT – Pa luation bUEva iLd Kit an rT 3 d Us siL ic E oU on ch r ULTiM Ec kE aT E r Jump start Mini LE Driver D Egg Tim er – eg to pe gcellen rfectio t brea n! kf ast, tim Compl PLUS! APRIL 13 Cover.in dd 1 ractica lelec practic ed etin g WideTechno range the Tal inter Cool Bea k – My tru face, Ohmmet th, you ne r truth techn Net Wo ns – Arduin er and AI o tal t work, cir o Boo rk – Rou k, pic ters, pow tcamp: new n’ mix cuit Surg www.e bo er sup Sep 202 ery, re lectron plies, TEM ards update 3 £5.9 adout, publish 9 ! U and ing.co 09 more m 9 7726 <at>p 32 5730 30 alelec APRIL 201 3 £4.40 tronics Practical Electronics is the UK’s premier electronics, computing & maker magazine. Each month has a wide variety of electronics projects suiting beginners and experts alike. It also includes many different features on topics like audio, radio, computers and more. 14/02/2 013 10:33:4 7 24 YEAR COLLECTION OF PRACTICAL ELECTRONICS Every issue of Practical Electronics published from January 2000 to December 2023. Covering 288 magazines & over 20,000 pages of content; see siliconchip.com.au/Shop/3 PDF Download SC7650 ▸ $165 The articles can be downloaded per month, year or the entire block. Some of the excellent columns of Practical Electronics include Audio Out, Circuit Surgery, Techno Talk and more. The download size is approximately 5.4GB. siliconchip.com.au PDFs on USB SC7645 ▸ $180 plus postage cost Supplied on a 32GB Silicon Chip branded USB flash drive. Purchasing this also gives you access to the download version shown opposite. Australia's electronics magazine August 2026  69 Using Electronic Modules with Tim Blythman GM805 Barcode Reader We recently reviewed the Tiny QR Code Reader module and lamented that it did not support linear barcodes. We have since found a module that fills this gap and can do a lot more, including reading QR codes and other types of 2D codes. W e reviewed the Tiny QR Code Reader in the February 2026 issue (siliconchip.au/Article/19663) and it definitely lived up to its name. It was tiny (just 19 × 16mm) and had no trouble reading QR codes. Since linear (1D) barcodes have been around a lot longer, we wondered if that device could read other types of 1D and 2D codes, but it appears that it could only handle QR codes specifically. While QR codes are great at packing in lots of information, simple codes still abound; UPC (universal product code) and EAN (European article number) barcodes are still used on products in stores. There are many other types of linear codes that are used for other purposes, such as tracking numbers on parcels (Australia Post uses both linear and 2D codes extensively). We searched the usual online stores to see what features could be found in a barcode reader, especially one that would be easy to interface with and use, as well as being compact. We settled on a unit dubbed GM805 on AliExpress, mostly because it appeared to be well-documented. You can find it on sites like AliExpress or eBay by searching for “GM805”. It is made by the Hangzhou Grow Technology Company Ltd (www. growscanning.com). They also have several other scanners, including some with more consumer-oriented enclosures. The GM805 manual can be downloaded from siliconchip.au/ link/acbl 70 Silicon Chip The GM805 module The GM805 features UART (asynchronous serial) and USB interfaces, which are quite different from the I2C interface on the Tiny QR Code Reader. There are GM805-S and GM805-L variants that differ only in their specified reading distances. We tried the GM805-L, which is claimed to work between 7cm and 50cm, as it is the closer match of the two for the Tiny QR Code Reader. We paid around AU$45 for two units, including shipping. It measures 28 × 21 × 10mm, supplied with two six-way 1.27mm pitch pin headers and a six-way socket fitted. A matching plug with flying leads is also included. The socket breaks out 5V power connections and two lines each for the USB and UART interfaces. The UART interface uses a 3.3V logic level. One of the six-pin headers has the same pinout as the six-way socket, while the other is marked with signals for the LEDs, BEEP and trigger functions. Our photos show the module. It is supplied as an assembled PCB with two mounting holes in opposite corners. The front has two LEDs and a camera, while the rear has a processor and related components. There is also a piezo sounder at the rear. The shape and layout of the module make it quite easy to fit into custom project designs. Fig.1: this wiring connects the GM805 to one of our USB-C Serial Adaptors (June 2024; siliconchip.au/ Article/16291), which has the same pinout as some CP2102 adaptors. The Scanner takes a 5V supply but works with 3.3V logic levels. Note that the ground wire is white, not black! Australia's electronics magazine siliconchip.com.au Perhaps unsurprisingly, the main processor chip had its identifying features removed. It had a rectangular area laser-etched out of the back of it, with the pin 1 identifier barely visible. The manual is quite detailed and is scattered with QR codes. It turns out that the easiest way to configure the GM805 is by scanning codes! We wired up the UART pins to a USB-­serial converter and scanned the “Serial Output” code; immediately, the scanner was able to send codes over the serial port to a terminal application. Fig.1 shows the wiring for serial operation. It appeared to simply work with a default baud rate of 9600. This and other settings can be changed via the UART interface (or QR codes), although we did not need to use that option. The GM805 uses a white LED for illumination, which is automatically controlled. A blue LED flashes and the piezo sounder beeps when a code is detected. The white LED and sounder can be disabled with other QR codes; it’s also possible to force the white LED to be on at all times. USB interface The UART interface makes it trivial to interface with a microcontroller, but the GM805 also provides a USB interface; in fact, several of them. One of these is to emulate a keyboard, since this is one of the easiest ways to get data into a computer. Fig.2 shows the wiring we used to connect it to a USB-C breakout board for all the USB interfaces. Fig.3: compared to the Tiny QR Code Reader, the GM805-L has a generally wider and longer operating span, although it was not as good up close. If shortrange operation is needed, there is also the GM805-S variant. Some of the earlier instances of barcode scanners being used with computers used a so-called ‘keyboard wedge’ (KBW) arrangement, where a splitter cable ‘wedged’ the scanner into the PS/2 keyboard interface, and the scanned codes appeared as if keys were being typed on the keyboard. One of the GM805’s interfaces is as a USB keyboard; this is the HID-KBW mode noted in the manual. Many smartphones also allow the connection of a USB keyboard, so this could be used for data entry in a mobile system. It could even be used with one of the PicoMite (February 2025; siliconchip.au/Article/17729) variants that support a USB keyboard. Screen 1 shows an Android phone connected to the GM805. While simple and effective, the HIDKBW mode can be troublesome. If, for example, the wrong program has focus, the scanned codes can cause havoc (ask me how I know!). There is also a USB HID-POS mode, which is intended to communicate directly with POS (point of sale) software. HID-POS also allows bidirectional communication, including from the computer to the scanner. We tried some HID-POS test software to try to communicate with the GM805, but were not able to get it to work. The USB vendor ID (VID) for the GM805 is 0x002C, which we could not find in any VID lists. Its product ID (PID) changes depending on the mode; the HID-KBW mode uses PID 0x261A, while HID-POS mode uses 0x0300. The USB interface can also behave as a virtual USB-serial adaptor, which appears as PID 0x0302. This mode is activated by another QR code, and we found that this worked seamlessly with the TeraTerm terminal program. Working range The manual suggests a working range for the GM805-L of 7-50cm. As we found for the Tiny QR Code Reader, the usable range can depend on the printed size of the code. We tried numerous EAN-13 codes from typical supermarket products and could not read these from more than 25cm away. However, we were able to successfully read such codes from as close as 5cm. Fig.2: this shows the wiring we used for a USB-C socket breakout, including the 5.1kW CC resistors (required for a USB-C to USB-C connection at 5V). If you are using a USB-A or USB-B connector, the four wires would be connected as you might expect. The wire colours here match those found on the breakout leads that were supplied with the readers. Screen 1: the USB-KBW interface is even recognised by Android phones. Australia's electronics magazine August 2026  71 For a comparative test, we used the same printed 62mm QR codes that we used for the Tiny QR Code Reader article and also turned off the illumination LED. The results are shown in a similar fashion in Fig.3. For these specific QR codes, it appears that the GM805 has a wider view and longer range, although it does not do so well at short distances; not surprising for the longer-­ range variant. The spans correspond to wider viewing angles as well. As expected, the actual performance depends a lot on the size of the code being scanned and how it fits in the field of view, combined with the limitations of the focus and camera resolution. The camera sensor in the GM805 is quoted as 640×480 pixels. The GM805 (shown at actual size) is well laid out, with the camera and LEDs on one side and the connectors and remaining components on the rear. Note that the GM805 has at least two variants. The GM805-S (shown in the centre) has a slightly smaller reading distance of 5-30cm, versus the one we used with a reading distance of 7-50cm (-L model; right). The GM805 manual states a maximum operating current of 70mA. In practice, we found that was a typical value with the default settings and with neither LED lit. The white LED drew about 20mA when lit, and the blue LED about 10mA. So we suggest that you should budget at least 100mA for the module. stuck with the defaults, which seemed to work on all the common types of barcodes and QR codes we came across. There are also settings to add prefixes and suffixes to scanned codes, which could be handy if you need to automate an existing system. And of course, there is a code that can be used to reload the defaults. The default settings appeared to be capable of about three scans per second (of our test QR codes), but there are various timeouts, delays and sensitivity settings that can change that. Options Summary There are many options that can be selected by means of QR codes; we counted hundreds of configuration codes in the manual. Apart from the main modes noted earlier, it’s possible to set the reader to require triggering before producing a code; presumably, this is provided via the TRIG input pin. There are also settings to determine the fraction of the camera’s view that can be used for scanning. In other words, it’s possible to set a narrower field of view than noted above. The codes that are produced can be selected by symbology, and many code types have other specific settings. Thus, it is possible to only allow specific types, which might be helpful depending on your application. There is a setting to allow all codes, which would be good for experimenting with the GM805’s abilities. There is also a setting to allow no code types, which then allows just configuration codes to be read. If you needed to limit the reader’s output to just one code type, you could scan this ‘forbid all’ setting and then add just the codes you wanted to accept. We just The GM805 is a very versatile and flexible reader that can handle both QR codes and 1D/2D barcodes. It has numerous interfaces that allow it to easily connect to a microcontroller or a computer. Its mechanical design should also make it easy to incorporate into a larger project. Importantly, it was very easy to get working. The GM805’s scanning performance and versatility are easily better than the Tiny QR Code Reader. It has an inbuilt illumination LED and can be programmed in more ways than we could easily test. Still, the Tiny QR Code Reader is smaller, uses less power and defaults to a higher scan rate. With all this in mind, we would probably tend to favour the GM805 in a project unless size or power was a critical factor in the project’s design. Having reviewed the GM805 module, let’s now investigate the properties of barcodes themselves. Power consumption 72 Silicon Chip Linear (1D) barcodes We covered QR codes in the previous article about the Tiny QR Code Australia's electronics magazine Reader module. Here, we’ll explain some of the history and technology of linear barcodes. Barcodes were inspired by the pattern of long and short pulses used in Morse code. The first barcode system was patented in the USA in 1952. During the 1960s, projects and standards were developed to use coloured barcodes to identify railway rolling stock, such as freight cars, but the system was abandoned because it was not reliable, depending on older technology such as valves and photomultiplier tubes. Advances in computer and laser technology set the scene for its widespread use for product identification in the 1970s. It is on supermarket products that barcodes ultimately succeeded. These are known as UPC (Universal Product Code) barcodes. Other places they are widely used include being printed on letters, parcels and tickets. Just about any situation where an item needs to be identified by a machine could potentially use a barcode. Structure The connection between Morse code and a barcode can be easily seen if you take a visual snippet of Morse code and stretch it out vertically, as in Fig.4. In practice, different encodings are used, since Morse code is intended to be interpreted by a human, while barcodes are designed to be machine readable. Fig.4 shows how the ‘stretching’ of the bars and spaces allows for better Fig.4: a simple barcode can be generated from a printed Morse code message, but most modern barcodes have features that make them better for machine reading. siliconchip.com.au tolerance in reading the barcode. The red lines show that the reader only needs to scan along a single line through the code. Even if that path is not exactly at right angles to the bars and spaces, the relative spacings of the various zones remain the same. The so-called quiet zone is an area with no features that allows the hardware to distinguish the start and end of a valid code from other noise. While the characters in Morse code have varying lengths, most barcodes use a fixed width per symbol. Early patents also suggested that a circular arrangement (looking somewhat like a bullseye) might be preferred. In hindsight, the need to accurately scan through the middle of the bullseye meant that this layout did not succeed (see US patent 2612994A). Symbologies As the QR code is just one type of 2D code, there are numerous types of linear barcodes using different symbologies or encodings. A typical encoding has a fixed width per character and also a fixed number of bars and spaces. This means that it is very easy for a reader to keep track of its position within the code; the code is self-clocking. QR codes use a masking step to try to achieve a balance between light and dark areas over their surface. Similarly, most barcodes are designed to have an equal amount of bar and space, since this provides the best chance for code recovery. The lack of a masking step makes barcodes simpler to decode. You can imagine that these factors severely limit the number of combinations that are valid within a given symbology. We’ll look at the EAN-13 (European Article Number) symbology, which is used for the 13-digit barcodes commonly found on products, including this magazine. So-called UPC-A is very similar. EAN-13 The 13 individual digits of an EAN13 barcode are encoded by a sequence of seven units of width, making up two bars and two spaces; this gives a total of 40 symbols. Twenty start with a bar and twenty start with a space, so there are really only 20 usable symbols, since they must all be the same type to be processed consecutively. EAN-13 encodes 12 of its digits using this scheme. It also adds a start marker, a centre marker and an end marker. On one side of the centre, the symbols start with a bar, and on the other, they start with a space. Since the 20 symbols are only used to encode the 10 digits, zero to nine, two groups of codes exist, and they can provide an extra layer of information and redundancy. Firstly, the codes can be used to determine whether the barcode is being read from right-to-left or left-toright. The choice of code groups in one half of the barcode is used to encode the thirteenth digit. You might also see that most barcodes have a human-readable version printed underneath. This allows the code to be manually entered if the scan fails for whatever reason. One of the thirteen digits is designated as a checksum; since this is also one of the printed numbers, both scanned and manually entered codes can be verified. Fig.5 shows a breakdown of a typical EAN-13 barcode. Other types Fig.5: the start, centre and end markers of a typical EAN-13 barcode. The red lines show how the scanner can use the centre marker to decode it even if the scan does not cover the entire barcode in one pass. Source: https://w.wiki/8WbT siliconchip.com.au You can see that a typical example of a modern barcode actually has numerous layers to store, encode and verify the information that is held. EAN-13 is just one example; there are other types that can have a variable length and can encode other data, such as ASCII text. Code 128 is one such type; there are many other types for differSC ent applications. Australia's electronics magazine Ideal Bridge Rectifiers Choose from six Ideal Diode Bridge Rectifier kits to build: siliconchip. com.au/Shop/?article=16043 28mm spade (SC6850, $30) Compatible with KBPC3504 10A continuous (20A peak), 72V Connectors: 6.3mm spade lugs, 18mm tall IC1 package: MSOP-12 (SMD) Mosfets: TK6R9P08QM,RQ (DPAK) 21mm square pin (SC6851, $30) Compatible with PB1004 10A continuous (20A peak), 72V Connectors: solder pins on a 14mm grid (can be bent to a 13mm grid) IC1 package: MSOP-12 Mosfets: TK6R9P08QM,RQ 5mm pitch SIL (SC6852, $30) Compatible with KBL604 10A continuous (20A peak), 72V Connectors: solder pins at 5mm pitch IC1 package: MSOP-12 Mosfets: TK6R9P08QM,RQ mini SOT-23 (SC6853, $25) Width of W02/W04 2A continuous, 40V Connectors: solder pins 5mm apart at either end IC1 package: MSOP-12 Mosfets: SI2318DS-GE3 (SOT-23) D2PAK standalone (SC6854, $35) 20A continuous, 72V Connectors: 5mm screw terminals at each end IC1 package: MSOP-12 Mosfets: IPB057N06NATMA1 (D2PAK) TO-220 standalone (SC6855, $45) 40A continuous, 72V Connectors: 6.3mm spade lugs, 18mm tall IC1 package: DIP-8 Mosfets: TK5R3E08QM,S1X (TO-220) See our article in the December 2023 issue for more details: siliconchip.au/Article/16043 August 2026  73 By Andrew Woodfield, ZL2PD This Test Set is an integrated solution for simple go/no-go testing of HF QRP SSB transceivers. It measures 155 × 85 × 45mm and weighs under 220g, complete with its 9V rechargeable Li-ion battery. HF, SSB & QRP TRANSCEIVER TEST SET I ’ve been restoring and converting a series of decommissioned portable HF QRP SSB search-and-rescue (SAR) transceivers over the past few years. Converting these to amateur bands ensures that dozens of otherwise useful portable and handheld SSB transceivers don’t need to be discarded. QRP is radio parlance for ‘low power’, typically referring to transmitters up to 5W, while SSB stands for single sideband, a modulation scheme. Much of the design work for the initial conversions took place in my workshop. It’s equipped with the typical array of test equipment. When several of the MRS-1 yellow radios initially arrived on my bench after the closure of the MRS service in 2024, they were accompanied by several very large grey plastic boxes, each about the size of a couple of loaves of bread. A few faded panel labels indicated these were used to test the two-­channel MRS-1 and MRS-3 radios. I used one of these test boxes briefly to verify the status of several transceivers prior to their conversion for use on the 80m and 40m amateur bands. A BNC connector on each box provided a connection for a cable to the transceiver antenna connector. Eight AA alkaline batteries were fitted internally to power the test set via a three-minute timer. A large meter mounted on the front panel displayed the transceiver RF output power. An internal crystal oscillator generated the two fixed 3MHz MRS HF frequencies. These were used to monitor the modulation during transmitter testing and provided accurate signal levels of about -90dBm and -105dBm for receiver testing. Other test box functions permitted a twotone ‘selcall’ system to be tested with the help of other external test assemblies and cables. A quick look inside these boxes revealed a rat’s nest of wiring, a multitude of prototype boards and circuit mysteries worthy of Agatha Christie. There was no documentation. So modifying these for other purposes was out of the question, but the idea of a simple tester stuck with me. Designing a new Transceiver Test Set (TTS) Faced with many radios to be tested during the upcoming club’s conversion workshops, a similar integrated Fig.1: the Transceiver Test Set (TTS) provides a basic test system for HF QRP SSB transmitters, receivers and transceivers. 74 Silicon Chip Australia's electronics magazine test system designed to cover a wider range of HF frequencies would be very useful. So, with the functions of the original test boxes in mind, I set out to create an updated design that would be capable of testing any HF QRP SSB transceiver. The features I considered most useful included: • Frequency range: 400kHz to 40MHz • Tuning steps: from 10Hz up to 1MHz • RF power meter: 5-10W full-scale • 50W dummy load: capable of handling 5W continuously or 10W for brief periods • Transmit modulation monitor: via an internal speaker • Receiver sensitivity test: with internal RF signal generator (-85dBm and -100dBm levels) • Transceiver frequency checking: using receiver and/or transmitter tests • Dual-mode operation: USB and LSB (to cover all HF bands) • Power supply: a 9V PP9 battery or a similar capacity rechargeable battery • Internal calibration: allows precise alignment of the PLL oscillator frequency Fig.1 shows the block diagram of the resulting Transceiver Test Set. The TTS comprises a wideband direct conversion receiver (DCR) with a digital PLL VFO, an attenuator/dummy load and an RF power detector. The attenuator/load and RF power detector form an RF power meter, and the DCR is used to monitor transmitter modulation. In addition, the VFO signal can be used as a simple RF signal generator to test the SSB receiver. The VFO can be tuned from 400kHz to 40MHz. The siliconchip.com.au required SSB mode may be selected, either USB or LSB. This selection will result in a 1kHz audio tone being audible in a USB or LSB receiver correctly tuned to that frequency. The test mode, TTS operating frequency and measured results are shown on a two-line alphanumeric LCD screen. The nominal carrier frequency is displayed, accurate to ±20Hz. The three test modes can be selected using the Mode selection pushbutton. Test mode one is used to test transmitters, while the second and third modes are used for testing receivers. The latter two modes deliver a -85dBm and -100dBm signal to the receiver, respectively. The operating frequency is displayed on the LCD screen. This may be tuned in steps of 1MHz, 100kHz, 10kHz, 1kHz, 100Hz or 10Hz. The selected test mode is shown in the upper-left corner of the LCD screen: TX for test mode one, RX H for mode two, and RX L for mode three. When testing transmitters, the RF power is displayed using a bargraph on the lower line of the LCD. The meter FSD can be adjusted internally and set to a convenient level, typically from 5W to 10W. Circuit details Fig.2 shows the circuit diagram of the main RF section of the TTS. The HF SSB QRP transceiver (or transmitter or receiver) being tested is connected to the RF Input connector, CON1(a), in Fig.2. When testing a transmitter, the RF detector stage (D1 and the 10nF capacitor) rectifies the transmitter RF signal. Typically, for a QRP transmitter or transceiver, this may range from 1W (+30dBm) to 5W (+37dBm). This DC signal is passed to the microcontroller via trimpot VR2, a resistor network and pin 5 of CON2 (more on where it goes later). This voltage is measured and displayed as a bar-graph RF power display on the lower line of the LCD screen. The signal at pin 5 of CON2 must not exceed 3.3V DC, as that is the most that the microcontroller used to measure it can handle. The transmitter signal is both terminated and attenuated by the 50W 60dB attenuator shown in the dashed cyan box in Fig.2. The RF power detection circuit described in the previous paragraph is connected in parallel with this 50W attenuator/load. The first section of the attenuator uses three 2W-rated resistors, selected to handle typical QRP transmitter output power, as part of three series-­ connected T-type attenuators. The frequency response of the resulting 60dB fixed RF attenuator (and transmitter load) is flat (within ±2dB) from below 1MHz to about 150MHz, despite its simple construction. When testing a typical 5W (+37dBm) QRP transmitter, the signal at the output of the attenuator is about -23dBm. This passes through a wideband RF buffer (Q1). This stage introduces a signal loss of 3-4dB. More importantly, it also features 30dB of isolation in the reverse RF path direction, ie, from the buffer stage’s output to input. These losses are near-flat for frequencies from below 1MHz to above 50MHz. The buffer output signal generated by the transmitter is then mixed using a Polyakov dual-diode mixer. The diodes in the Polyakov mixer switch twice per VFO cycle, during the positive and negative peaks of the oscillator sinewave, as the oscillator voltage exceeds the forward voltage of each diode. This requires the VFO to be set at half of the transmitter frequency for correct demodulation. This approach avoids the potential problems with direct low transmitter energy coupled from the transmitter and mixing directly in the VFO when used with conventional diode mixers. The mixer is driven by a square wave output from the Si5351A PLL chip. This waveform is ideal for both conventional diode double-balanced mixers and the Polyakov diode mixer. For more details on this, see H.P. Walker, “Sources of intermodulation in diode-ring mixers”, Radio and Electronic Engineer, Volume 46, Issue 5, May 1967, pp247-255. Since all the preceding wideband stages are untuned, the VFO signal Fig.2: the TTS attenuator/load, RF power detector, RF buffer, Polyakov mixer and audio stages are in this section of the circuit. siliconchip.com.au Australia's electronics magazine August 2026  75 The prototype for the Transistor Test Set was relatively easy to build, with pretty much all wiring made using DuPont-style connectors onto regular pin headers for ease of construction. used in a Polyakov mixer can also be readily radiated via the mixer input. If the buffer stage were not present, this VFO signal could measure as much as -55dBm to -65dBm at the RF input connector. However, the 30dB reverse isolation of the RF buffer, combined with the 60dB loss of the attenuator, ensures that this low mixer port isolation causes no problems during transmitter testing. The unwanted emissions at the test set input are less than -85dBm, more than 120dB below the level of the typical 5W QRP transmitter’s RF output. The Polyakov mixer demodulates the SSB transmitter audio when the VFO and transmitter are correctly tuned. This audio is amplified by NPN transistor Q2, filtered by L1 and its three connected capacitors, and further amplified by IC1 to drive the TTS monitor speaker. Digital oscillator & LCD screen Fig.3 shows the other half of the circuit, on a separate board. The two circuits are joined via CON2 in Fig.2. GND and VBAT on CON2 connect to CON7 in Fig.3; the RF signal from CON2 goes to pin 1 of IC2; and the VFO IN signal on CON2 comes from the OUT0 connector (CON8) in Fig.3. An 8-pin ATtiny85 microcontroller (IC2 in Fig.3) controls the TTS operation, including driving the LCD screen and the 10-pin Si5351A PLL chip (IC3). Rotary encoder RE1 is monitored by the ATtiny85 via its PB3 digital input. The three connected resistors provide a 2-bit DAC function, allowing direction of rotation and switch presses to be detected using a single pin. The pin’s voltage is monitored using the ATtiny85’s internal analog-to-digital converter (ADC) to allow it to distinguish the different actions. The I2C LCD is controlled by an on-glass ST7032 chip, which supports the standard alphanumeric LCD commands via I2C rather than the more typical parallel connections. This display is not equivalent to the standard alphanumeric LCD with the ‘added backpack’ I2C-to-parallel sub-board. This LCD is a more compact and better-­ integrated display. We suggest using the JLX1602 2-line, 16-character alphanumeric LCD for this project. To find one, search for “JLX1602” or “ST7032 I2C LCD”. One source is given in the parts list. The Midas MCCOG21605 range of I2C LCDs are also available in the UK. These are more expensive and differ in size and pinout. The PCB and software are designed to accept these LCD screens without any changes to the hardware or software. Only one of the Si5351A’s three outputs is used in this design (OUT0), although a second output (OUT2) is used briefly during calibration. The VFO output frequency at OUT0 depends on the operating mode of the TTS. When testing a transmitter in test mode one, the +7dBm VFO signal Fig.3: the TTS VFO uses an 8-pin ATtiny85 microcontroller to control the Si5351A PLL chip and the I2C LCD. It also monitors the user inputs from the rotary encoder and switches. 76 Silicon Chip Australia's electronics magazine siliconchip.com.au simply drives the Polyakov mixer’s oscillator input at half the transmitter frequency. “TX” is displayed in the top right-hand corner of the LCD during this mode. For testing receivers, in test mode two, the TTS generates an RF input signal for the receiver of about -85dBm (about an “S5” or “S6” moderately strong signal on a typical receiver S-meter), while test mode three gives a lower signal level of about -100dBm (about “S2” or “S3”). The VFO in mode two operates at a frequency 1kHz offset from the receiver frequency, 1kHz above or below the nominal carrier frequency depending on the USB and LSB selection switch state. “RX H” is shown on the LCD during this test. In mode three, the VFO operates at half the nominal carrier frequency, plus or minus the required offset. “RX L” is then displayed on the LCD. This uses an oscillator harmonic with a reduced output level for this test. The 1kHz offset in each case generates an audible 1kHz tone in the SSB receiver being tested when it is on the correct frequency. In these receiver test modes, the VFO oscillator signal passes through the mixer to the mixer input with a loss of around 6dB, then on through the RF buffer with a reverse isolation loss of 30dB. The signal then flows back through the attenuator, adding another 60dB loss, and finally into the receiver. Both modes two and three produce a 1kHz audio tone in a correctly tuned on-frequency receiver. However, if desired, the VFO frequency can be manually tuned to give other demodulated tones between, say, from 300Hz to 2400Hz, to check the SSB receiver audio response. A low-cost 3.3V regulator (REG3) supplies the majority of the digital sections of the TTS, while the 9V battery voltage coming from the RF board directly supplies the RF and audio sections. Construction The TTS is built using two PCBs, one for the VFO and LCD (coded 06104262, 89 × 36mm), the second for the RF and audio sections (06104261, 75 × 48mm). Their component overlay diagrams are shown in Figs.4 & 5. The parts for the individual boards are listed separately in the parts list, but siliconchip.com.au Converting fixed channel SSB transceivers The first radio converted was the handheld Codan/Condor 8332 1W SSB transceiver (Photo a). A very compact phasing SSB radio, the changes included migrating from USB operation on two crystal-­ controlled 3MHz & 5MHz channels to variable frequency oscillator (VFO) operation on the 80m and 40m amateur bands, and lower sideband (LSB) using a digital PLL VFO (see www.zl2pd.com/Condor_SAR_ Transceiver.html). The conversion was made possible with a version of my low-cost compact SugarCube PLL VFO module (www.zl2pd.com/ sugarcube_plus.html). This module uses an 8-pin ATtiny85 microcontroller, a Silicon Labs Si5351A PLL chip and an OLED display on a compact 25 × 25mm PCB. It delivers up to three synthesised PLL oscillator outputs from 5kHz to about 290MHz. Following the successful Codan/Condor transceiver conversion, the much larger AWA TR-105 transceiver (Photo c) was next in line (www.zl2pd.com/TR105.html). And once that was completed, I moved on to the newest conversion, the recently withdrawn Mountain Radio Service (‘MRS’) MRS-1 portable HF SSB transceivers (www.zl2pd. com/MRS-1_Conversion.html). With a growing number of these transceivers now successfully converted by local club members, I’m considering its successor, the very compact MRS-3/SR-3 transceiver (Photo b). All these transceivers, each finished in a distinctive bright yellow colour, are battery-­powered HF SSB QRP transceivers with RF outputs ranging of 1W (Condor), 3-4W (MRS-1) or 5W (TR-105 and MRS-3). These receivers also feature good sensitivity. The MRS-1 (Photo d) and TR-105 are particularly robust, the former featuring a yellow painted aluminium shell and integrated battery holder, the latter having a very heavy duty ABS plastic case and clip-on battery pack. The MRS-3 handheld has a very sturdy clamshell-style diecast aluminium case. off-board components that are wired to that board are part of the general parts list. The various connectors, switches and other user controls are wired to these boards. The PCB layout diagrams show the location of the components. The prototype RF/audio PCB was built using a single-sided PCB (see the photo on page 76), which was Australia's electronics magazine Photos a & b: the Codan/Condor portable QRP HF SSB transceiver (left), and the MRS-3/SR-3 portable QRP HF SSB transceiver with two-tone call option (right). Photo c: the TR-105 portable QRP HF SSB transceiver. Photo d: the converted MRS-1 transceiver. perfectly satisfactory. Provision was also made for the addition of shields around sections of the attenuator, but that was found to be unnecessary. There are three 2W resistors on the RF/Audio PCB. Space these about 1-2mm above the PCB when mounting them. If you can’t find 2W resistors (local retailers stock 1W and 5W types but nothing in between), you could use pairs of 240W 1W resistors soldered in August 2026  77 Fig.4: follow this diagram while installing the components on the RF board. The only slightly tricky part is T1, which has three windings. Compare the 1-6 numbering of its pads to what’s shown in Fig.2. Fig.5: start assembly of the control board by soldering IC3 as it is delicate. Make sure it’s orientated as shown, with its pin 1 marker at lower left. Apply flux paste to the pins before soldering, and if you accidentally bridge them, use more flux and some solder wick to clear them. parallel for each, with a few millimetres between the bodies. Components of particular note include the 100mH choke used for the audio low-pass filter, the FT3761 toroid (T1 in Figs.2 & 4) used in the Polyakov mixer, and the I2C LCD (LCD1), which mounts on the back via one of three possible header locations. None of these are terribly hard to find or expensive; all three can be found from the usual internet suppliers (see the parts list). A toroidal core from an old fluorescent lamp inverter can also be used for T1. T1 is made using three 200mm lengths of thin enamelled copper wire, say 34SWG or 0.2mm diameter. Hold these three wires together and wind 10 turns onto the toroid. Twisting the three wires together a little to hold them together is helpful but not essential. This arrangement forms a ‘trifilar’ winding. Connect the various wires from T1 as shown in Figs.2 & 4 (the points numbered 1-6 in Fig.2 correspond to the similarly numbered points in Fig.4). I mounted T1 flat on the PCB in the prototype, but some may find it easier to mount it vertically. Either approach is satisfactory. The VFO PCB provides for two sizes of Midas I2C LCDs and the lower-cost JLX1602 I2C LCD. It also allows the Si5351A chip to be mounted on a separate MSOP-10 to DIL-10 adaptor board if necessary. Two of these through-holes (pins 7 & 8) are used to allow the 3.3V rail and ground to be carried to the RF/Audio PCB. This can be seen in the wiring diagram, Fig.6. If you purchased a programmed ATtiny85 chip, it can be carefully plugged into the socket on the control board now, with its pin 1 end lined up with the socket notch. If you have a blank chip, you will need to program it first (see the panel opposite). When all the components have been mounted on the VFO PCB, the LCD may then be mounted on the back. The LCD’s backlight and connection pins require careful handling. If using the recommend (JLX) LCD, the two backlight pins go into the pair of holes visible on the right-hand side of Fig.5. The other LCDs use a pair of slots near REG3 (a different pair depending on the LCD size). In either case, the LCD screen should only be fitted after all other parts have been mounted on the board. I made a simple laser-cut box from 1.6mm birch ply with cutouts for the LCD screen, controls and speaker. All sides except the lid the speaker is attached to are glued together. I used a fast-setting PVA glue. I glued four 15mm-long M3 threaded nylon standoffs in each corner, about 4mm below the upper edge of the box, so the top cover could be attached using 12mm-long M3 panhead screws. The files for the laser-cut box can be downloaded along with the software and 3D-printing files from siliconchip. au/Shop/6/3583 I printed the front panel artwork on a sheet of plain paper, carefully trimmed using a sharp scalpel and covered it with self-adhesive transparent film from a stationery supplier. The reverse side was sprayed with artwork spray adhesive and then applied to the birch ply front panel. This process makes a very tidy, inexpensive and hard-wearing panel, but it is a little time-consuming. The Transceiver Test Set measures just 155 × 85 × 45mm and weighs under 220g. It’s powered by an internal 9V rechargeable Li-ion battery. 78 Silicon Chip Australia's electronics magazine siliconchip.com.au With the panel made, the VFO board and display can be mounted on the front panel using four M3 × 15mm panhead machine screws and bolts. Fig.6 shows how the internal wiring is arranged. Three resistors are mounted at the rear of the rotary encoder. If after assembly your rotary encoder tunes the VFO in the opposite direction, swap the 1.8kW & 3.9kW resistors that are soldered directly to it. The 9V battery used in the prototype is a rechargeable LiPo type with an integrated charger. The manufacturer’s claimed 6600mAh capacity is, unsurprisingly, exaggerated. It is actually closer to 1300mAh, but that is still adequate for many hours of testing. Recharging via a phone charger or other USB-C power source is relatively fast and convenient. I made the VFO tuning knob and volume knob on my 3D printer, but commercial equivalents are readily available. If you wish to print these yourself, the relevant STL files can be found in the download above. The monitor audio level is infrequently adjusted, so that control has been relegated to the rear panel, along with the RF connector and power switch. Programming the ATtiny85 Download the HEX and EEP files for the Transceiver Test Set from siliconchip.au/ Shop/6/3583 If you have an in-circuit programmer like the USBasp, you will also need a way to connect the correct lines to the pins on the chip. This is most easily done using an adaptor board. It saves adding a 6-pin programming socket to each PCB. My 8-pin adaptor was published in the September 2020 (on page 47; siliconchip.au/ Article/14563) and the PCB is still available (siliconchip.au/Shop/8/5642). Once you have the chip plugged into an adaptor, connect the programmer to your computer. Download and open a programming application (such as Extreme Burner) and load the HEX and EEP files into this program. Now program your ATtiny85 with the HEX file, then the EEP file. Click on the “Write” tab in Extreme and select the file you are sending to the ATtiny85. Next, program the hardware configuration fuses in the ATtiny85. Table 1 shows the required fuse settings. You need to set these after loading the HEX and EEP files before the TTS will work. These configure the ATtiny85 for operation from the 8MHz RC clock and the internal reset mode to free up pin 1 for RF power measurement. To set the fuses, click on the Fuse Bits/Setting tab, enter the values shown, and click on the Write selection boxes for the Low and High fuses (the others may safely be ignored). When you have done this, write the fuse settings to the ATtiny85 by clicking on the Write button at the lower right of this tab. If necessary, detailed step-by-step programming instructions can be found on my website, www.zl2pd.com Fuse Hexadecimal value Comment Lock byte FF Flash not locked Extended byte FF Self-programming disabled High Byte 5F Defaults except RSTDISBL=0 Low byte E2 Defaults except CKDIV8=0 Table 1 – the required ATtiny85 fuse settings TTS Frequency Calibration The Si5351a VFO must be calibrated to ensure the TTS is accurately tuned to the nominal carrier frequency shown on the display. This is determined by the precise frequency of the 25MHz reference crystal attached to the Si5351A. The ATtiny85 program calculates the settings of the Si5351A using this value to set the correct VFO output frequency. Fig.6: the front panel (lower ▶ PCB) has been artificially folded flat in this sketch to show the internal wiring. siliconchip.com.au Australia's electronics magazine August 2026  79 Since these crystals are inexpensive, their frequency may vary by more than 3kHz from 25MHz. Any error in the value of this crystal’s frequency that is permanently stored in the VFO microcontroller’s EEPROM will directly impact the accuracy of the VFO output. To ensure this value is accurate, the software contains a VFO calibration routine. To calibrate the TTS, switch off the power, then switch it on again while holding down the Mode pushbutton. After the initial power-up screen has been displayed, “Calibration” will be seen on the LCD. Now release the Mode pushbutton. The VFO will now be delivering what it calculates to be a 25.000000MHz square wave of about 3V peak-to-peak via the Si5351A’s OUT2 output, which can be found at CON10 (“Cal”) on the VFO board. Connect an accurate frequency counter to the output on CON10. This should display a value within 3-4kHz of 25MHz. Tune the TTS VFO tuning knob until the frequency counter displays exactly 25.000000MHz. You can use the tuning step pushbutton shaft switch on the tuning control to select the desired tuning step size. When the frequency counter is displaying a frequency as close as possible to 25.000000MHz, press and hold the Mode pushbutton for about half a second. Now switch off the power to the VFO and reconnect the frequency The TTS rear panel can be seen here with the transmit audio monitor volume control, power switch and RF connector while testing an MRS-3 transceiver. counter to the VFO PCB’s OUT0 connector (CON8). Switch on the power again. Confirm that the frequency counter shows the correct frequency. Be careful to note that each test mode results in an output frequency that differs from the displayed frequency: Test Mode 1: TX Test VFO CLK0 = Displayed frequency ÷2 Test Mode 2: RX High Level Test VFO CLK0 = Displayed frequency ±1kHz Test Mode 3: RX Low Level Test VFO CLK0 = (Displayed frequency ±1kHz) ÷ 2 The 1kHz offset will depend on the setting of the USB/LSB switch, ie, LSB = -1kHz, USB = +1kHz. If the output at OUT0 (CON8) agrees with the frequency counter display, the TTS VFO is calibrated. Final adjustment & operation Songbird An easy-to-build project SC6633 ($30 plus postage): Songbird Kit Connect a fresh 9V battery and switch on the power. The initial poweron message should appear. This is shortly replaced by the transmitter test (Mode 1) display with frequency and that is perfect as a gift. Choose from one of four colours for the PCB (purple, green, yellow or red). The kit includes nearly all parts, plus the piezo buzzer, 3D-printed piezo mount and switched battery box (base/stand not included). See the May 2023 issue for details: siliconchip.au/Article/15785 current tuning step size. Turning the Tune/Step knob will change the frequency, and pressing in the encoder knob will change the tuning step size. Pressing the Mode pushbutton should change the displayed mode. Changing the LSB/USB selection switch will not change the display on the LCD, but it does alter the output frequency in (receiver) test modes two and three. Use the Mode pushbutton to set the test mode to (transmitter) test mode 1. Adjust the TTS for the nominal carrier frequency of the transmitter being tested, then set the USB/LSB switch for the required mode. Set VR2 to approximately midrange, connect a QRP SSB transmitter and adjust VR2 to set the required LCD RF power meter maximum level. The meter is reasonably linear from 0.2W to 5W when set for a full-scale of 5W, for example. Modulate the transmitter with voice or an audio tone. The monitor volume can be adjusted to give a suitable level for monitoring the transmitted audio. Avoid transmitting into the TTS for long periods to avoid overheating the internal attenuator/load. It is designed for testing 5-10W SSB transceivers. Select (receiver) test mode two and the required USB or LSB setting, then confirm that a 1kHz tone is clearly audible in the receiver being tested. This test assumes the receiver has a sensitivity of, say, 1μV for a 10dB signal-­to-noise ratio (SNR) or better, and the displayed TTS frequency matches the nominal carrier frequency of the receiver. Now select test mode three. This reduces the signal into the receiver by about 15dB. On a sensitive receiver, the 1kHz tone should be audible above the noise floor with the standard 3kHz SSB speech filter receiver passband. Parts List – Transceiver Test Set 1 laser-cut or moulded instrument case, 154W × 44H × 84D (mm) or larger 1 9V PP9 battery and snap [for snap: Jaycar PH9232, Altronics P0455] 3 2-pin headers, 2.54mm pitch (CON1, CON3, CON5) 1 BNC female panel-mounting socket (CON1a) [Jaycar PS0658, Altronics P0516A] 1 5-pin header, 2.54mm pitch (CON2) 1 3-pin header, 2.54mm pitch (CON4) 1 panel-mount pulse-type rotary encoder with integrated push switch (RE1) [AliExpress 1005005983134515] 2 SPDT panel-mount toggle switches (S1, S3) [Jaycar ST0336, Altronics S1315] 1 panel-mount pushbutton (S4) [Jaycar SP0711] 1 57mm 8W loudspeaker (SPK1) [Jaycar AS3000, Altronics C0610] 1 10kW log panel mount potentiometer (VR1) [Jaycar RP3610, Altronics R2214] 2 knobs, to suit RE1 & VR1 4 M3 × 20mm tapped nylon spacers 8 M3 × 10mm panhead machine screws 4 M3 × 20mm panhead machine screws and hex nuts 4 M3 × 6mm panhead machine screws and hex nuts a selection of wires terminated with DuPont female connectors (cut jumper wires in half) various lengths and colours of light/medium-duty hookup wire RF/Audio board parts 1 single-sided PCB coded 06104261, 75 × 48mm 1 47μH axial RF choke (RFC1) 1 100mH radial RF choke (L1) [AliExpress 4001355154716] 1 FT37-43 toroidal core (T1) [Minikits FT37-43 or AliExpress 1005009245292057] 1 600mm length of 0.15-0.2mm diameter enamelled copper wire (T1) 1 100kW top-adjust trimpot (VR2) 1 8-pin DIL IC socket (for IC1) Semiconductors 1 LM386 audio amplifier IC, DIP-8 (IC1) 1 J310 N-channel VHF/UHF JFET or equivalent (Q1) 1 BC548 30V 100mA NPN transistor (Q2) 3 1N4148 75V 200mA signal diodes (D1-D3) Capacitors (all 50V radial ceramic unless noted) 1 100μF 16V radial electrolytic 2 15nF polyester or MKT 3 10μF 50V radial electrolytic 2 10nF 5 100nF 1 4.7nF polyester or MKT 1 47nF polyester or MKT Resistors (all ¼W ±1% axial unless noted) 1 220kW 1 10kW 3 120W 2W 2 82W 4 10W ½W 1 18kW 1 1kW 2 100W 1 39W 1 0W VFO/Control board This Transceiver Test Set is a simple, easy to build and lightweight portable test system. It has proven to be ideal for the task, and a very worthwhile successor to the original crystal-locked test box. The TTS has been invaluable in testing a wide variety of QRP SSB transceivers in my workshop, and in the series of conversion workshops undertaken locally at our radio club. I’m certain you’ll find it equally useful for testing your QRP transceivSC ers, too. 1 double-sided PCB coded 06104262, 89 × 36mm 5 2-pin headers, 2.54mm pitch (CON6-8, CON10-11) 1 4-pin header, 2.54mm pitch (for connecting RE1) 1 JLX1602 I2C 16×2 alphanumeric LCD (LCD1) [AliExpress 32807890814] OR 1 Midas MCCOG21605-series 16×2 alphanumeric LCD (LCD1) [element14/RS] 1 25MHz HC-49 crystal (X1) [AliExpress 1005002830871853] 1 8-pin DIL IC socket (for IC2) Semiconductors 1 ATtiny85-20PU 8-bit microcontroller programmed with 0610426A.HEX, DIP-8 (IC2) 1 Si5351A-B-GTR 3-output PLL clock generator IC, MSOP-10 (IC3) [AliExpress 1005008517358757] 1 TS2950CT33, 78L33 or equivalent 3.3V 100mA regulator, TO-92 (REG3) [AliExpress 1005006134947908] Capacitors (all 50V radial ceramic unless noted) 1 10μF 50V radial electrolytic 3 1μF 50V radial electrolytic 4 100nF Resistors (all ¼W ±1% axial) 3 10kW 1 3.9kW 1 2.7kW 1 1.8kW 1 470W 1 0W siliconchip.com.au Australia's electronics magazine Conclusion August 2026  81 SERVICEMAN’S LOG Repair and servicing stories from readers LED gym sign repair A friend rang and asked if I could take a look at his son’s gym sign; some letters were not lighting up correctly. He said they were willing to pay up to $300 for the repair. The sign duly arrived and I connected it up to its 12V, 15A power supply. I could see straight away that several letters had sections missing. The letters have a silicone moulding as a cover that could be easily removed to access them. I could then see the arrangement of the LEDs. They were 12V segments, all connected in parallel. Each segment had three LEDs and a series resistor, with copper pads at each end. The segments could be cut at the copper pads to get the required length. There were numerous faulty segments in each letter; the damage looked like it was from water ingress. The sign had been hanging outside the premises, supposedly under cover. There were 120 LEDs per meter, with some sections bright white, and some blue. I found similar LEDs on the internet and ordered a 5m roll of each colour, hoping that they would match the existing colours. Otherwise, I would have to replace the lot. When the LEDs arrived, I powered a section of each colour from my bench supply and saw that they matched the originals perfectly. Now it was just a matter of replacing the faulty segments. I removed the silicone moulding to access each area of LEDs, then cut out the faulty sections and used Kynar insulated wire-wrap wire to connect the new LEDs to the existing strips. It was quite a fiddly process, which took about five hours. I finally had to fit the silicone mouldings back into place and the sign was back to its original condition, with all letters now complete. I rang my friend, who was delighted that the sign could be used again, this time inside the gym. He asked how much he owed for the job. I was happy to get paid for the parts plus a bottle of scotch for my efforts. John Western, Hillarys, WA. A WiFi repeater and leaf blower My electronics workshop is in a shed behind our house, but the WiFi signal from the house is quite weak inside the workshop, so I’ve been using a WiFi repeater to boost the signal. It was powered via an extension lead under the covered walkway next to the workshop for a long time without any problems. Just recently, we had a big storm with torrential rain, and I didn’t go to my electronics workshop for several days 82 Silicon Chip because of the bad weather. When the rain finally eased, I went there to work on a laptop, but I found that the WiFi repeater was no longer working. I turned the power to the extension lead off and removed the device. It was wet; rain must have blown under the covered walkway, resulting in water getting into the repeater. I prised the two case sections apart and found that the power supply board was completely destroyed. There were components with sections of their leads missing and a lot of corrosion on the circuit board. I remembered that I had another one of these WiFi repeaters that no longer worked, but it was slightly different to the one I was using. I wondered whether I might be able to transplant the working WiFi board from this damaged unit into it. I separated the two case sections and found that the power supply board was completely different, but it had the same plug for the WiFi board. I removed the faulty WiFi board and installed the one from the destroyed unit, then I reassembled it with the original top. I plugged it into a power point to test it and the LEDs lit up. I plugged the WiFi repeater back into the extension lead and re-located it further under the covered walkway to better protect it from the weather. It often pays to hang onto non-working electronic equipment in case you need spare parts from it later. Bruce Pierson, Dundathu, Qld. Leaf blower charger repair (again) Following on from this, once again, I’ve had to repair our leaf blower charger. This time, the repair was not easy as the charger would work intermittently. Still, in the end, it’s finally fixed and working correctly again. Australia's electronics magazine siliconchip.com.au A close-up of the dry joints on the leaf blower charger PCB. When a flat battery is connected, the LED turns red, showing that the battery is being charged. The charger contains circuitry that controls the charging and switches the charger to standby when the battery is charged and the LED turns green. Recently, my wife said that the charger was not working; the LED was staying green when a flat battery was connected. I’d made two previous repairs to this charger. The first was when I had to replace the plug, and the second was when the insulation near the case was broken and the wires were shorting. As I’d already cracked the case open for the second repair and glued it back together with superglue, it was a little easier to crack it open again. I suspected bad capacitors, and in testing the two largest capacitors, they both showed a high ESR reading. The two smaller capacitors tested OK, so I thought they were still good. The capacitors were rated at 4.7µF 400V and 10µF 400V. I did not have either of these in stock, either salvaged or new, so I had to order them. These capacitors are much smaller than regular capacitors, and there is very limited room in the charger’s case. I had to check many listings before I could find the right sizes to order. The capacitors arrived, and there was a problem, with the larger capacitor being a couple of millimetres too high for the case lid to be re-fitted. After some trial and error, I managed to fit the capacitor lying down so the lid would fit. I tested the charger, and it was working again. My wife went to use it and said, it’s not working; the LED is staying green. That was annoying, as I thought it was fixed. I had a close look at the underside of the PCB and found several dry joints on the transformer, which I had not noticed earlier, shown in the photo. I re-soldered the dry joints and gave the charger back to my wife after gluing the lid back on again. She put the battery on charge, and it was charging, but it was bedtime, so she unplugged it for the night. The next morning, when she went to finish charging the battery, the LED was green and flashing slowly. This was becoming annoying, so I removed the lid again. I suspected that the 50V 10µF capacitor before the transformer might be the culprit, even though I measured an ESR of 1.8W, which was within limits. siliconchip.com.au I checked my stock of salvaged capacitors and found one with an ESR of 1.2W, then fitted it. This solved the problem, and the charger was definitely working correctly this time. I glued the lid back on again, and that was the end of the drama. Having a spare charger saved the day, and the repair saved us from spending $39.99 on a new one. Bruce Pierson, Dundathu, Qld. Icom IC-271H Repair About two years ago, I bought a faulty Icom IC-271H 2m transceiver. Dating from around 1985, it was one of the better radios of its day. It operates on 144-148MHz with FM, SSB or CW modulation at up to 100W with 0.3µV sensitivity for a 10dB signal-to-noise ratio (SNR). Although it was advertised as faulty, it didn’t appear to have anything seriously wrong with it, and at $100, it seemed a gamble worth taking. As it turned out, the repair process became quite a saga. The radio came without its 13.8V 20A power supply, but I already had suitable supplies on hand. On initial testing, it worked on both transmit and receive and delivered the specified 100W into a dummy load. However, after a short while, the green RECEIVE LED on the front panel went out, indicating that the phase-locked loop (PLL) had lost lock. When this happened, the receiver was also muted. The full service manual was easy to find online. The circuit of the VFO section (overleaf) is extremely complex; the block diagram (also shown) is a much better way to understand what’s going on. All frequencies are ultimately derived Items Covered This Month • LED gym sign repair • Bruce Pierson’s tales (WiFi repeater & leaf blower) • A faulty Icom transceiver from 1985 • A quick fix for an LG washing machine • The long (air) con Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz Cartoonist – Louis Decrevel Website: loueee.com Australia's electronics magazine August 2026  83 from a 10.24MHz crystal oscillator. There are two PLLs: a coarse loop covering 102.30-106.30MHz with 100Hz resolution, and a fine loop covering 115120MHz. The fine loop is divided by 500, producing a range of 230240kHz with a 10Hz resolution. The 10.24MHz reference is tripled to 30.72MHz and mixed with the 230-240kHz signal to produce a narrow range of 30.95-30.96MHz. This is passed through a crystal filter and then mixed with the coarse PLL output to give a range of 133.25-137.25MHz. With the 10.75MHz IF, this results in the required tuning range of 144-148MHz. The exact frequencies differ slightly from those shown in the block diagram, as there are country-specific versions. I carefully followed the tuning procedure in the service manual, with no success. The radio would lock for a while, sometimes for up to an hour, then drop out again. I also changed most of the electrolytic capacitors on the board for new ones, but that did not help. As you can see from the photo, the circuit board is tightly 84 Silicon Chip packed and finding suitable components to look at waveforms was very difficult. There are no test points as such; the service manual just gives component locations to look at. After some hours of getting nowhere, I gave up. As it wasn’t urgent, I put it aside and moved on to other projects. There it remained for about two years! Recently, I finally came back to it. Tracing the frequency chain more closely showed that the coarse PLL stayed locked, while the fine PLL was the culprit. Referring to the circuit, the control voltage from pin 1 of IC6 drives varicap diode D9. When locked, this voltage sat between 2V and 4V. As soon as the lock was lost, it jumped straight to 5V and remained there. Australia's electronics magazine siliconchip.com.au The divided-down signals from IC9 and IC8 looked correct, which pointed suspicion at IC7 (M54466L) and/or IC6 (M54929P). A quick search suggested that both seemed unobtainable. I decided to try replacing the entire board with a much simpler design using an Si5351A frequency generator, which is cheap and readily available. The radio’s processor board produces five bytes of binary-coded decimal (BCD) data, the coarse range on DA1 to DD1 and the fine range on DA2 to DD2. These represent 10210 to 10470 and 23000 to 24000, respectively. From these values, the required output frequency for the Si5351A can be calculated. The top two digits in each range never change and can be ignored for the calculation. I designed a small PCB using an Arduino Nano module and an Si5351A module, both of which I had. The control software, written in BASCOM-AVR, was straightforward. I also added an OLED display and an RS-232 interface for development and diagnostics, which were disabled once everything was working. The completed module is shown in the photo and is essentially a two-chip solution. After a debugging session, the Si5351A was producing the correct frequencies and was installed in place of the original board. It did work, but two problems quickly became apparent. First, when spinning the tuning knob rapidly, the system couldn’t keep up. Decoding the BCD data and sending multiple bytes serially via I2C to the Si5351A simply took too long. This was not a deal-breaker, but annoying nevertheless. The second problem was more serious. As I tuned across the band, numerous ‘birdies’ appeared. The spectrum plot appeared to be fairly clean; only the second harmonic was significant. I am not certain where the spurious frequencies came from; possibly noise from the processor chip. This made the solution unworkable. In a sense, I was back where I’d started, but with one important difference. In the meantime, I had located a supplier for both the M54929P and the M54466L. The former was only available as a used part, but the latter was new. The total cost was around $35, including postage, from China. The diagram on the opposite page is the circuit diagram for the VFO section of the ICOM IC271H, while the adjacent diagram is the block diagram for the PLL (phase-locked loop) section. The module shown above is an Si5351 frequency generator. siliconchip.com.au Australia's electronics magazine August 2026  85 A couple of weeks later, they arrived and I installed them. I fitted a socket for the 16-pin DIP device and soldered the 8-pin SIP chip directly, as before. Once again, I went through the alignment procedure, and this time everything locked and stayed locked. I left it running for a whole day and it stayed on frequency. Almost by accident, however, I discovered another problem. The adjacent logic board contains many 74-series devices, mostly LS types, but one was a plain 7404. When I touched it, I almost burned my finger. Measuring the case temperature, it was about 65°C. Standard 74-series parts are power hungry, but this was excessive for a simple hex inverter. Despite this, it appeared to be functioning. A local replacement would have cost about $4, but after rummaging through boxes containing decades worth of accumulated chips, I eventually found a 74F04. It looked unused but the pins were badly oxidised and needed cleaning. I fitted a socket just in case, but the F-series device worked perfectly and ran cool. I have not included some blind alleys I went along and red herrings I came across on the way. Diagnosis and coming up with a solution to a problem is not always a linear process! With the benefit of hindsight, I should have foreseen some of the problems that would occur with my digital replacement scheme. While I can’t claim to be particularly proud of my troubleshooting performance, I did learn a great deal in the process. Despite its complexity, the original PLL design produces a clean, stable signal and very smooth tuning, something that proved very difficult to replicate with digital techniques. Full marks to the original designers. To make certain all was working, I connected it to a dummy load and tested the transmitter. It worked well and the power output was as per specifications. In the end, I have a transceiver that cost me $100 to buy, and probably another $80 worth in parts during the various repair attempts, most of which I already had in stock. It’s now a classic radio in excellent condition, with a second-­ hand value of around $800. Charles Kosina VK3BAR, Mooroolbark, Vic. 86 Silicon Chip LG Direct Drive front-loading washing machine fix In early December, my 19yo daughter decided she would do some washing – a rare occurrence given she mostly lives in her bedroom. I received a panicked call saying “the machine is not working and displaying LE Error”. I got her to try to rebalance the load and try again, but the error came back. In the meantime, I went to serviceman Google and found it was a pretty common fault with the HAL (Hall effect) sensor. I got home and tried to take the part out to check its resistance but hit a stumbling block trying to remove the 17mm bolt that holds the cover to the drive. It really needed an impact wrench, but I don’t have one. I tried dial-a-friend (also known as my brother-in-law) but he was away. After two hours of thinking and trying my socket set to no avail, I realised I had an air compressor with an attachment. I wheeled it over to the laundry and voilà, one bolt removed. The rest of the drive was held in place by six Phillips-­head screws. With these removed, the drive assembly came straight out. I was then able to check the resistance on my HAL sensor by probing pins 5 and 1. It should give a 10kW reading (likewise pin 4 and 1, from memory). On my unit, one reading was open. The exposed washine machine motor and the replacement sensor. Australia's electronics magazine siliconchip.com.au As I was about to go away for a work trip the following day, I decided to order a replacement from Amazon as they offered same-day delivery so I could get it working. The part arrived by 7pm. I checked it with the multimeter and the new part was giving me 10kW on both pins. The reinstall took five minutes, and it worked, which made my wife happy as she was worried we might be up for a new machine. I still think 10 years is young for a washing machine, and this one appears reasonably well-built. While the part was only $17, I wanted to see what had gone wrong. I pulled out all the potting epoxy and finally had the circuit board clean enough to inspect. Aside from a few other components, it includes two pairs of 680W resistors in parallel. One pair was fine, but the other side was open-circuit. An online order and a few days later a strip of 680W SMD resistors arrived. The hot air got the old ones off easily, and there was still ample solder to reuse to put the new ones in. Sure enough, the fault disappeared and both readings were 10kW as expected. After reapplying some new potting epoxy, I now have a spare part should the recent replacement part fail. I’m usually more of a tinkerer than an electronics repairman, but this was one of the easiest repairs I have made. Roberto Marin, Earlwood, NSW. Conned by the air-con I buy split-system air conditioners and install them myself. It is not hard, just requiring simple gauges and a vacuum pump. There are some lessons to be learnt about flaring copper tubes, especially regarding flare length and lubrication. Anyway, I installed such a unit from a ‘big green shed’ for a guest during winter (it gets cold here!). It worked perfectly, heating the guest room. Come summer, we had more guests who asked me to check the air conditioner. It turns out that the unit would heat but not cool! I knew it had gas and was acting as a heat pump; otherwise, it wouldn’t heat. I removed the cover from the outside unit because it’s the easier one to work on (like how a drunk person searches for their keys at night under the streetlight because it’s easier to see there). There is very little in these units: the compressor, a four-way valve, a fan and a heat exchanger. When set to heat, the four-way valve was powered up and the fan operated – good. When set to cool, the four-way valve was powered up, but the fan was off! The four-way valve is supposed to be off in this case so it doesn’t reverse the working fluid path to provide cooling. The fan certainly needs to run to dissipate the heat being pumped out. After much head-scratching (you may add lots of words here about troubleshooting), I realised that if the functions were swapped, the unit should work correctly. Long story short, swapping the control wiring solved the problem. I thought that I must have made a mistake in the control wiring, but I checked the instructions and I had done everything correctly. I then suspected the manufacturer had made an error with the control wiring, but again, no; checking it all, it looked correct. So the problem must somehow lie in the control system board itself. It was at this point that I was told, “It ain’t broken, so don’t fix it”! Apparently, I have an issue with having to fix things! Even our guest, who was a senior lecturer in electrical engineering, wouldn’t support me in further troubleshooting! SC Garry Woods, Watson, ACT. siliconchip.com.au Australia's electronics magazine August 2026  87 Vintage Radio Baby Beethoven 555 portable regenerative radio The Baby Beethoven model 555 portable radio comes from the UK. It was offered to the public there around 1937-1938. It is a regenerative receiver using the PM2HL valve or its equivalents, such as the HL2K and VT-50. The Reinartz 2 receiver described in the October 2025 issue was also a regenerative set. By Dr Hugo Holden R egeneration results in ‘Q multiplication’ in a resonant circuit. This happens because of energy injection due to positive feedback. The result is a narrower bandwidth, increased gain and improved selectivity. A good regenerative radio can be nearly as good as a superhet radio for gain and selectivity. It needs just the right amount of regeneration, though. This means a regenerative receiver is a ‘Goldilocks circuit’, because everything has to be ‘just right’ for a good result. The designers strive to ensure that the control of the regeneration is smooth, in that the positive feedback is introduced gradually. Otherwise, the user adjusting the control can result in the stage abruptly going into oscillation. If that happens, the highlevel oscillations heterodyne with the received carrier, resulting in a lot of ‘howling’ from the speaker. A ‘super-regenerative’ radio is a little different again as the oscillations are set up to be blocked at a frequency higher than the audio spectrum but much lower than the carrier frequency. 88 Silicon Chip This type of radio can develop astonishingly high gain levels for a single stage, but the zero-signal noise is often very high. Interestingly, this design has persisted much longer that the plain regenerative receiver. It is often used at the receiving end of short-range remote-control systems for home automation. It was used extensively in cheap children’s transistorised ‘walkie-talkies’ because the super-­ regenerative stage readily oscillates and makes for a good transmitter with only a small circuit change needed to switch between receiver and transmitter modes. The ideal amount of regeneration in a regenerative radio is tricky; the required magnitude of positive feedback is difficult to keep uniform on different parts of the tuned band. It can also be affected, to an extent, by the strength of the received station. That is why, in the Reinartz circuit mentioned earlier, the other gang of the tuning capacitor modifies the regenerative energy injection across the tuned band, helping to even that out. Australia's electronics magazine Because of the difficulty of making smooth regeneration controls, manufacturers had some difficulty in marketing regenerative radios to the public. Many circuit variations, such as the one in the Reinartz radio, were devised to help improve the function of the regeneration control. While the regeneration control can be a potentiometer, it was more often a variable capacitor. From the radio user’s perspective, it is a volume control. Thus, most regenerative radios did not have an actual volume control potentiometer in their audio amplifier chains. The regenerative stage can also act as the AM detector as well, as it does in the Reinartz design or the popular Hiker’s 1 radio and in the Beethoven 555. In this case, it is called a regenerative detector or regenerative gridleak detector. Regenerative vs TRF radios Regenerative radios are a subset of the TRF type but behave substantially differently from a standard TRF set. siliconchip.com.au TRF radio receivers were very popular in the 1920s and 1930s era. They were analogous to a crystal set but with active stage amplification; no positive feedback was used. They were simply a chain of tuned bandpass amplifiers, with a detector to recover the audio modulation. The difference with a regenerative radio is that positive feedback has been added at one or more stages. The trick is to add enough positive feedback to improve the gain without sending the set into oscillation. Because the amount of extra gain possible is very large, the number of valves in the set can be dramatically reduced for similar performance (sometimes to just one RF valve!). Multiple regenerative radio construction articles appeared in radio and hobby magazines of the 1930s. A particularly popular one in Australasia was the Hiker’s One. This radio would run from a 6V B+ because it deployed a space-charge valve. That made it a popular choice for children because no high voltages were involved, and their parents did not have to buy them expensive 45V or 90V B+ batteries. Of course, there is a price to pay for the miracle of regeneration; regenerative sets can be difficult to control. As the regenerative stage is pushed further toward oscillation, the bandwidth narrows, and just before oscillations or ‘howling’ begins, the recovered audio modulation becomes muddy and lacking in high-frequency components. This was the challenge for designers and manufacturers of regenerative radios in the 1920s and 1930s: how to make the radio usable for the average member of the public. The user may have limited technical knowledge, making it difficult to manipulate the regeneration control to obtain a good result. The later superhet radio design did not have this problem. Back to Beethoven The Beethoven 555 receiver was moderately advanced for its time, and it addressed the smooth regeneration control concern very well. I found it difficult to acquire the correct circuit diagram, so initially I traced it out by hand in 1987, with the result shown in Fig.1. Later, I found a circuit of a very similar set: the Beethoven P202, shown in Fig.2. The only apparent major siliconchip.com.au The grid leak detector misnomer The ‘leaky grid detector’ terminology is an endless source of confusion, probably because it was not aptly named. The load resistor of the AM detector is responsible for discharging the capacitor between charging peaks of the RF carrier wave; this capacitor is charged via a diode function on RF peaks. The load resistor of this lightly filtered half-wave rectifier system came to be called the ‘grid leak resistor’. This is a function it also served, but was unrelated to the demodulation of the AM signal. In the valve, electrons tend to accumulate at the grid and develop a space charge around it. If there is no DC path for them to ‘leak away’, a negative charge and voltage builds up on the grid, and this can cut off the valve by repelling electrons back to the cathode (or filament if that is the emitter). This can take the anode current to near zero. If very high value grid resistors are used, in the range of 3-10MW or more, the grid current passing via the resistor, in the order of a few hundred nanoamperes or less, can result in a DC bias for many valves in the order of -1V to -3V. This can be helpful in biasing the valve; it is often done when cathode biasing is inconvenient and the cathode is better grounded. For example, when the valve also has an integral anode electrode to act as a diode with the cathode, it is preferable that the cathode is at common/ground potential. Some have wondered why occasional circuits from the 1920s, including in patents, showed valves without grid resistors. There were two reasons. One was that the grid electron current was very low in early valves, and that capacitors and insulating materials of that era were not as perfect as what we have in modern times. Leakage resistances, even in the range of 10-100MW, were enough to dissipate the grid’s electron charge, so the valve never became ‘cut off’ in practice. The second reason was that when these patents were applied for, Mr. Einstein was no longer working at the patent office to detect this oversight; just because something worked in practice did not mean its design was correct or complete. The thing is that this grid-leak biasing function has little to do with its function as an AM detector. This resistance is simply the load resistance that discharges the filter capacitance between charging peaks by the RF carrier. It is a half-wave rectifier circuit with minimal filtering. The circuit of the RF coil, feeding diode, resistor load and filter capacitor is the same for any AM detector. The diode itself, in the grid-leak detector case, is simply the grid-­cathode (or grid-filament) interface of the valve. The discharge time constant of the RC filter sets the upper frequency limit that can be resolved via the detector without distortion. As the modulation frequency gets higher, or the modulation depth gets higher, the capacitor cannot discharge quickly enough to track the modulation envelope present on the RF carrier before the next carrier peak arrives. The highest audio frequency that a diode-RC detector can resolve, when the modulation level is 50%, is 0.275 ÷ RC. In 1920s radio designs, values such as 3MW and 100pF gave a fairly poor result for high-frequency audio recovery. The detector’s distortion and high-­ frequency roll-off began at about 1kHz with 50% modulation; 0.275 ÷ (3MW × 100pF) = 916Hz. By the post-war period, the AM detector situation had improved, and RCA had moved to values such as 100pF and 250kW, allowing audio recovery up to around 11kHz. But in a vintage TRF radio, 250kW would be a somewhat heavy load and would have damped the driving tuned resonant circuit, lowering the gain and selectivity. The trouble was that, in the early days, there was much more of a quest for gain than fidelity, and extra valve stages were expensive. If the detector stage is also regenerative, the losses due to increased loading could be overcome by the additional energy injection from regeneration, meaning the grid resistor could be in the range of about 250kW to 470kW to improve the detector’s high-frequency audio fidelity. The change in the valve’s DC bias conditions would be minimal. In most practical grid-leak detector circuits using directly heated (filament) valves, to encourage the valve to draw grid current on the positive going peaks of the RF carrier, the grid is often made a little positive with respect to the average cathode voltage. The grid-to-filament potential in a directly heated valve is distributed along the length of the filament. The usual configuration is to return the grid resistor or the RF coil to the positive side of the filament connection for that reason. Fig.a shows a common configuration for a ‘grid-leak detector’ using an indirectly heated valve. Fig.a: how a grid-leak detector works. Positive excursions are limited at the grid, and when the RF is filtered out of the inverted and amplified version of this signal at the anode, it leaves behind a reconstruction of the amplitude modulation signal. difference is that it came in a slightly different cabinet and it sported a power lamp. I subsequently added the P202’s component designators to my hand-drawn circuit. In my 555 radio, there were some resistors added in the switching circuit of the MW and LW frame antenna that might not have been original. Capacitors C1, C4 and C12 also had different values, but of course, these may have been altered in the past during servicing. RF amplifier (V1) The valves used in the 555 and P202 are 2V heater types, with the VP2 RF pentode used as an RF amplifier. At this point in history, pentodes were futuristic parts. The design confers properties well in advance of a triode. One major advantage is isolation between its anode and control grid. This means that, unlike a triode with its high Miller (feedback) capacitance, you can connect tuned resonant circuits to its grid and plate operating at the same frequency without them exchanging any significant energy with each other. It will thus not oscillate, as it would if it were an unneutralised triode. It is important to prevent the tuned coil in the VP2’s anode circuit from feeding back to the frame antenna inside the radio’s cabinet, since this coil essentially sits inside the middle of the frame antenna. Thus, the coil is very well-shielded. Grid-leak detector (V2) The job of the regenerative grid-leak detector is allocated to V2, a PM2HL. The grid return voltage of V2 has been set to the centre of the filament voltage with the two 4MW resistors, R2 & R3, and sits at an average of +1V. This is equivalent to a zero-bias condition for V2 because half of the filament’s structure is negative with respect to the grid, and the other half is positive with respect to the grid’s average potential. The grid’s electron current will shift the grid a little in the negative direction, because the Thévenin resistance is 2MW, and a small standing negative grid bias results from that. The PM2HL is a metallised version of the common triode, such as the HL2K or VT-50. The metallisation makes for a helpful shield at the regenerative detector stage. If you see a modulated carrier passed to the grid circuit of a valve and an audio signal is recovered from the anode, there are only two ways that can work. Both require non-linearity in the way the signal is processed or amplified. The common method is the gridleak detector, which is poorly named. In this case, rectification occurs at the grid-filament (or grid-cathode) interface, and the signal at the plate Fig.1: a hand-drawn circuit of my set. The first valve is a pentode that acts as a regenerative RF amplifier. The second valve, a triode, is the ‘grid leak detector’. The second triode is the audio preamplifier while the final valve, another pentode, is the power amplifier that can deliver around 340mW to the speaker. “C2 Gimmick” refers to a low-value capacitor that’s formed by two closely-spaced but separate wires. 90 Silicon Chip Australia's electronics magazine siliconchip.com.au Fig.2: the Beethoven P202 portable battery-powered set circuit from the “Trader” service sheet. It’s a very similar set to the Beethoven 555, with the only definite differences being the cabinet style and the pilot lamp in the P202. Source: Radiomuseum – www.radiomuseum.org/r/beethoven_p202p_20.html siliconchip.com.au Australia's electronics magazine August 2026  91 The rear view of the Baby Beethoven 555 with and without the 90V and 2V batteries that I made. Note that the original battery for this set was 80V. resembles an inverted and amplified version of the negative half of the RF carrier wave. The audio modulation is recovered by filtering the carrier out of this signal. However, in some cases, if the detector valve is correctly biased, the positive-­ going halves of the carrier wave are preferentially amplified by the RF voltage applied to the grid. This happens because of the curve, or bend, in the function of anode current versus grid voltage. The signal at the anode of the valve looks like an amplified, inverted version of mainly the positive half of the carrier wave. That is called an anodebend detector. In the boundary between the two forms of AM detection, grid leak versus anode bend, there can be no detection at all. In this instance, a perfectly symmetrical modulated RF carrier would appear at the anode, and filtering that out would reveal no audio signal at all! So, if you see a valve AM detector circuit, how do you know if it is a grid-leak detector or an anode bend detector? In most anode bend detector circuits, the grid-filament or grid-cathode of the valve is fed directly by the output coil of the previous stage. In the grid leak case, there is always what appears as a coupling capacitor that relies on being charged by grid current and discharged by a resistor. It is not really a coupling capacitor though; it is more like the energy storage capacitor in a half-wave rectifier system. Audio amplifier (V3) Triode V3, another PM2HL, is deployed as an audio amplifier, so does not require a shield. In my radio, I used a VT-50 for V3 because of its narrower profile; there is not a lot of room in the battery compartment. The audio power output valve is the PM22A. This has multiple equivalents and is listed as a KT2 for the P202 version. These are capable output valves for a battery radio and can deliver 340mW. One good place to find out equivalent valve types is in the original Mullard valve manuals. The manuals even suggest what possible substitutes could be used with small modifications, which is very helpful if you cannot find an exact equivalent. The speaker in this set is a good size 92 Silicon Chip Australia's electronics magazine siliconchip.com.au A VT-50 triode (left) was used in my Beethoven 555 for V3 as it is smaller than the alternative PM2HL (right). Fig.3: to determine the filter response in the days before SPICE was available, I simplified the circuit as shown here, then derived the gain formula shown on the right. The plot on the left is the result. Besides having loss overall rather than gain, the response is very similar to that of a tuned IF transformer. at close to 5.5 inches (140mm) in diameter, with an Alnico magnet. tapering off at the upper and lower ends. However, if the balance of factors is not correct, there can be a significantly peaked response. Coupling transformers One notable feature is the audio inter-stage transformer. It is capacitively coupled and used as an autotransformer. This arrangement gives higher voltage gain for fewer total turns (less copper and less iron) because the primary voltage is in series with the secondary. As a result, the autotransformer can be made more compact and lightweight. Note that general transformer theories of impedance matching and power transfer do not apply to typical valve inter-stage transformers driving a valve in Class-A. They do in Class-B, where power transfer is required. This is because the grid of the audio output valve, operating in a Class-A condition, never draws any significant current. The equations that do apply are those of the damped tuned coupled resonant circuits. In the interstage transformer case, the damping is normally provided by the anode (plate) resistance of the driving valve, although in some cases it can be added to the primary or secondary windings. When the proportions of inductance, winding self-capacitance, mutual coupling and damping are correct, the interstage transformer can possess an astonishingly flat response in the audio frequency spectrum, siliconchip.com.au Generally, the output from a regenerative or grid-leak detector passes to the next stage via an inductor or an audio inter-stage transformer to assist in filtering the RF out of the signal and leaving just the audio signal. In this radio, no inductor or transformer is present; there is simply an RC network feeding V3, the first audio amplifier stage. This makes sense; the radio was already heavy enough, so there was certainly a motivation to save weight and to get rid of at least one of the normally heavy iron-cored inter-stage audio transformers. I wondered what the bandpass frequency response of this RC network would look like over the audio frequency spectrum. This sort of thing is dead easy to measure with a signal generator and scope, but I wanted a theoretical proof. This was in the days before SPICE, so I derived an equation for it, which took a lot of work (see Fig.3). I ran this equation through a graphing program on my university’s mainframe computer. The assumption was made that the source impedance driving the filter (the Thévenin resistance) would be in the order of 18kW, being V2’s anode (plate) resistance of around 37.5kW in series with R6 (6kW), both in parallel with R5 (30kW). I don’t have the original plot, but I’ve reproduced it from memory in Fig.3. The band-pass response was The badge for this radio is located on the top of the lid. The top of a Mullard PM1HL triode, which has a gold metallising paint. An interesting feature Australia's electronics magazine August 2026  93 Restoring valves when their metallisation fails Sometimes with very old valves, the conductive paint bubbles off and falls away from the glass. Fortunately, there is a method to restore it. Jaycar sells a highly conductive colloidal silver paint, made by Kemo Electronics GmbH, which has excellent adherence to glass. You can paint it on with an artist’s brush after the remainder of the old paint has been removed. Once finished, you can spray metallic silver or gold paint over it to restore the original appearance. This often results in the loss of the original label, so it is a good idea to re-label the valve on its base. The adjacent photo shows a restored PM2HL valve. The rear and front of the leather cabinet. A close-up of the rotating base (‘Lazy Susan’) is shown at lower right, which uses ball bearings to rotate. remarkably similar to that of an interstage transformer, except without the signal gain that the transformer would have provided. Running the circuit through SPICE gave the same result (shown in Fig.4), except it was about a hundred times easier. The network behaves as a bandpass filter over the audio frequency spectrum with an insertion loss of about 8.45dB. This is made up for by the audio pre-amplifier valve (V3) and the PM22A output valve (V4). Sometimes people repair radios where the interstage audio transformers have gone open circuit by replacing them with an RC coupling network. Clearly, this can work, but there will be a substantial drop in gain, more than the typical 1:3 ratio of the transformer. That is assuming that the RC filter is crafted to have a similar bandpass characteristic to the transformer it’s replacing. Antenna and miscellaneous Radios with ferrite rod or frame antennas are very directional. The manufacturers of the 555 put a ‘lazy Susan’ type spinning base on the radio. This runs very smoothly as it is supported by ball bearings. For convenience, the manufacturers used a removable handle with ‘lift the dot’ fitting. These were used in the automotive industry for attaching softtops to convertible cars. The leather handle had perished. The one shown is a reproduction I had made back in the late 1980s. At a glance, the speaker grille on the radio might look like plastic, but it is made from a high-strength woven string (not cord) that appears to have been varnished. It has stood the test of time very well. The adjacent photo shows the radio with the back fitted. The rectangular steel spring-metal clip slides under two screw heads to help retain the back. In this era, the user had many radio stations to select from, if the dial is anything to go by. Batteries Fig.4: now that SPICE is available, we can easily plot the circuit’s response in just a few minutes. You can see that it matches my version very well, including having almost-identical -3dB points. The insertion loss is 8.45dB. 94 Silicon Chip Australia's electronics magazine I had to make batteries to run this set. The 2V battery is composed of two groups of three parallel C-sized NiCad cells in series. A 1W, 10W ceramic dropping resistor is used inside the same battery enclosure to get the loaded voltage close to 2V. siliconchip.com.au Interestingly, the original 2V cell was described as a 14Ah “celluloid jelly acid”. The 90V battery was created from 72 AA-sized NiCad batteries in series. Because of its current-delivering capability, I incorporated a fuse inside that battery. The original battery for the radio was actually an 80V type, although I did not find that out until years later when I found the P202 circuit. I put an Eveready label on the 90V battery for a bit of fun. Summary The Beethoven 555 is a remarkable MW & LW band regenerative radio from the pre-WW2 era. The manufacturers had largely perfected user-controlled regeneration. The radio is well made and has stood the test of time. Its weight was reduced by eliminating one interstage transformer in favour of an RC filter. The resulting gain reduction was made up with a two-stage audio amplifier and a frontend RF pentode, giving up to 380mW drive for the loudspeaker. From a commercial perspective, regenerative radios were destined to become far less common than superhet types. Over time, the cost of valves came down, and superhet portable valve radios appeared with a converter valve. This combined the function of the mixer and oscillator into one valve. With one IF pentode, one diode detector/triode valve and one audio output pentode, the radio was complete. That was the same number of valves for a superhet (four) as a regenerative radio like the Beethoven 555. However, the ease of use and performance of superhet radios out-classed the regenerative designs. History has shown that regenerative designs remained popular with home constructors. This was because of the large amount of signal gain from a onevalve regenerative detector stage; the economy is hard to ignore. In many cases, one valve was enough to drive a set of headphones directly, and even enough to drive a small speaker with only one or two additional valves. It is fair to say, looking back, that the regenerative radios of the early years demonstrated remarkable innovation by the designers. It is fun to restore these radios, experiment with their operation and study their operating SC principles. siliconchip.com.au RP2350B Computer A Fully-assembled general-use computer The RP2350B Computer runs BASIC and is excellent for creating your own programs, games, tinkering with external circuits and more. And we are selling it pre-assembled, with little to no soldering required to have it up and running. It supports a keyboard, mouse or even a SNES controller. Video output: DVI via an HDMI connector <at> 640 × 480, 720 × 400, 800 × 600, 848 × 480, 1280 × 720 or 1024 × 768 pixels Removable file storage: microSD Card, FAT16/FAT32, up to 32GiB Clock Speed: 252-375MHz Non-volatile program memory: 184kiB General usage RAM: 220kiB (expandable to over 6MiB) Internal File Storage: 14MiB Audio formats: single-frequency tones, stereo WAV, FLAC, MP3 & MOD USB ports: four Type-A for peripherals, one Type-C for power/console and one micro Type-B for firmware loading Clock: battery-backed real-time clock & calendar External console: serial over USB <at> 115,200 baud via the USB Type-C socket External I/O connector: 30 pins with 22 GPIOs, including 7 with analog input ability, plus ground, 3.3V and 5V outputs Power supply: 5V <at> 220mA RP2350B Computer Assembled Module [ SC7531 | $90.00 + post ] fully-assembled PCB, except for the optional components (instrument case, mounting screws, 3-pin header for serial wire debugging and APS6404L PSRAM IC [SC7530 | $5]) Front & Rear Panels [ SC7532 | $7.50 + postage ] pre-cut panels, white silkscreen and black solder mask; not included with the kit above For all the details on how to build it, check out the article in the November 2025 issue of Silicon Chip (siliconchip.au/Article/19220). Australia's electronics magazine August 2026  95 SILICON CHIP .com.au/shop ONLINESHOP HOW TO ORDER INTERNET (24/7) PAYPAL (24/7) eMAIL (24/7) MAIL (24/7) PHONE – (9-5:00 AET, Mon-Fri) siliconchip.com.au/Shop silicon<at>siliconchip.com.au silicon<at>siliconchip.com.au PO Box 194, MATRAVILLE, NSW 2036 (02) 9939 3295, +612 for international You can also pay by cheque/money order (Orders by mail only) or bank transfer. Make cheques payable to Silicon Chip. 08/26 YES! You can also order or renew your Silicon Chip subscription via any of these methods as well! The best benefit, apart from the magazine? Subscribers get a 10% discount on all orders for parts. PRE-PROGRAMMED MICROS For a complete list, go to siliconchip.com.au/Shop/9 $10 MICROS $15 MICROS ATtiny85-20PU Graphing Thermometer (Mar26), Simple LC Meter (May26) Simple USB Power Monitor (Jun26), Transceiver Test Set (Aug26) PIC12F617-I/P Active Mains Soft Starter (Feb23), Model Railway Uncoupler (Jul23) Battery-Powered Model Railway Transmitter (Jan25) PIC16F1455-I/P Battery-Powered Model Railway TH Receiver (Jan25) Dual Train Controller (Transmitter / TH Receiver, Oct25) PIC16F1455-I/SL Battery-Powered Model Railway SMD Receiver (Jan25) USB Programmable Frequency Divider (Feb25) Dual Train Controller (SMD Receiver, Oct25) PIC16LF1455-I/P New GPS-Synchronised Analog Clock (Sep22) PIC16F1459-I/P K-Type Thermostat (Nov23), Secure Remote Switch (RX, Dec23) Mains Power-Up Sequencer (Feb24 | repurposed firmware Jul24) 8CH Learning IR Remote (Oct24), Heat Transfer Controller (Aug25) Vacuum Controller (Oct25), Adjustable Ultrasonic Cleaner (Jul26) PIC16F15214-I/SN Silicon Chirp Cricket (Apr23), Mic The Mouse (Aug25) PIC16F15214-I/P Filament Dryer (Oct24), Tool Safety Timer (May25) PIC16F15224-I/SL Multi-Channel Volume Control (OLED Module; Dec23) NFC IR Keyfob Transmitter (Feb25), Rotating Light (Apr25) PIC16F18115-I/SN Model Railway Destination Display (Aug26) PIC16F18126-I/SL RGB LED Star (Dec25), DCC/DC Stepper Motor Driver (Apr26) μDCC Decoder (May26; bell [G] or whistle [W]) PIC16F18146-I/SO Versatile Battery Checker (May25), RGB LED ‘Analog’ Clock (May25) USB-C Power Monitor (Aug25), DCC Remote Controller (Feb26) DCC Booster & Reverse Loop Controller (Mar26) DCC Accessory Decoder (Snap / Servo-type, Jul26) PIC16LF15323-I/SL Remote Mains Switch (TX, Jul22), Secure Remote Switch (TX, Dec23) STM32G030K6T6 Variable Speed Drive Mk2 (Nov24) PIC16F1847-I/P PIC16F18877-I/PT Digital Capacitance Meter (Jan25) Dual-Channel Breadboard PSU Display Adaptor (Dec22) Wideband Fuel Mixture Display (WFMD; Apr23) PIC16F88-I/P Battery Charge Controller (Jun22), Railway Semaphore (Apr22) PIC24FJ256GA702-I/SS Ohmmeter (Aug22), Advanced SMD Test Tweezers (Feb23) ESR Test Tweezers (Jun24), Human Comfort Indicator (Jun26) PIC32MX170F256D-501P/T 44-pin Micromite Mk2 (Aug14), 4DoF Simulation Seat (Sep19) PIC32MX170F256B-50I/SP Micromite LCD BackPack V1-V3 (Feb16 / May17 / Aug19) Advanced GPS Computer (Jun21), Touchscreen Digital Preamp (Sep21) PIC32MX170F256B-I/SO Battery Multi Logger (Feb21), Battery Manager BackPack (Aug21) PIC32MX270F256B-50I/SP ASCII Video Terminal (Jul14), USB M&K Adaptor (Feb19) STM32L031F6P6 SmartProbe (Jul25) $20 MICROS ATmega32U4 ATmega644PA-AU PIC32MK0128MCA048 PIC32MX270F256D-50I/PT Wii Nunchuk RGB Light Driver (Mar24) AM-FM DDS Signal Generator (May22) Power LCR Meter (Mar25) Digital Preamplifier (Oct25) $25 MICROS PIC32MX170F256B-50I/SO + PIC16F1455-I/SL Micromite Explore-40 (SC5157, Oct24) PIC32MX470F512H-120/PT Micromite Explore 64 (Aug 16), Micromite Plus (Nov16) PIC32MX470F512L-120/PT Micromite Explore 100 (Sep16) $30 MICROS PIC32MX695F512H-80I/PT Touchscreen Audio Recorder (Jun14) PIC32MZ2048EFH064-I/PT DSP Crossover/Equaliser (May19), Low-Distortion DDS (Feb20) DIY Reflow Oven Controller (Apr20), Dual Hybrid Supply (Feb22) KITS, SPECIALISED COMPONENTS ETC MODEL RAILWAY DESTINATION DISPLAY (SC7697) Kit: includes all parts, except for the OLED screen (see p54, Aug26) - 0.32in white OLED screen (SC7698) - 0.50in white OLED screen (SC7699) DCC ACCESSORY DECODERS (AUG 26) (JUL 26) Snap-type (SC7685): includes the PCB and all non-optional onboard parts Servo-type (SC7686): includes the PCB and all non-optional onboard parts I2C CONTROLLER COMPLETE KIT (SC7690) (JUL 26) HUMAN COMFORT INDICATOR (SC7646) (JUN 26) Includes the PCB and all onboard parts (see p83, Jul26) Kit: includes all parts, except the case and battery (see p49, Jun26) - white 3D-printed case: portrait (SC7453) or landscape (SC7684) version - 3.3V GY-BME280 module (SC5482) PINBALL MACHINE KITS (JUN 26) SIMPLE USB POWER MONITOR (SC7683) (JUN 26) $22.50 $5.00 $6.50 $40.00 $40.00 $30.00 $60.00 $12.50 $10.00 siliconchip.com.au/Shop/ STEPPER MOTOR DRIVER KIT (SC7601) (APR 26) CALLIOPE AMPLIFIER PARTS (SC6021) (APR 26) DCC BOOSTER / REVERSE LOOP CONTROLLER KIT (SC7579) (MAR 26) Includes all required parts for DCC or DC mode (see p55, Apr26) Includes some of the harder-to-get transistors, resistors and a capacitor Includes all required parts, except for the Jiffy box, OLED screen (see below), power supply and front panel (see p58, Mar26) - 0.91-inch OLED screen (SC7484) DCC REMOTE CONTROLLER KIT (SC7552) (FEB 26) MAINS HUM NOTCH FILTER (SC7598) (FEB 26) DCC BASE STATION KIT (SC7539) (JAN 26) DCC DECODER KIT (SC7524) (DEC 25) EARTH RADIO KIT (SC7582) (DEC 25) RP2350B COMPUTER (NOV 25) $35.00 $15.00 $45.00 $7.50 Includes all required parts, except for the case and wire/cable (see p63, Feb26) $35.00 Includes everything except for the case and power supply (see p53, Feb26) $50.00 Control Board (SC7659): includes the PCB and all non-optional onboard parts $150.00 Includes everything but the plastic case, power supply and some optional parts. $90.00 Power Supply (SC7680): includes the PCB and all onboard parts $50.00 The Pico 2 is supplied but not programmed (see p39, Jan26) Cable & Connector Set (SC7681): includes 17 10-pin box headers, 34 10-pin IDC RGB LED STAR KIT (SC7535) (DEC 25) connectors, 10m of 10-way ribbon cable, 30 2-way pluggable terminal blocks Includes the mostly-assembled board and all non-optional components and 20 2-way polarised headers $65.00 except the power supply (see p43, Dec25) $80.00 Includes the PCB and all onboard parts (see p63, Jun26) - 0.96in 128x64 cyan OLED screen (USB Power Monitor, Jun26; SC6176) - 0.96in 128x64 white OLED module (USB Power Monitor, Jun26; SC6936) μDCC DECODER KIT (SC7617) (MAY 26) Includes all the parts and the optional piezo (wire not included). Specify if you want a bell or whistle sound for the microcontroller (see p88, May26) SIMPLE LC METER COMPLETE KIT (SC7657) (MAY 26) POWER AMPLIFIER CLIPPING INDICATOR (SC7649) (MAY 26) Includes all the parts and the 3D-printed enclosure (see p67, May26) $50.00 $10.00 $10.00 $25.00 $45.00 Short-form kit: includes the PCB and all onboard parts, the case and power supply are not included (see p35, May26) $95.00 - pair of red & white PCB-mounting RCA sockets (SC2615) $4.00 Includes everything in the parts list (see p73, Dec25) Includes everything to build the radio itself except the case and battery, plus the plug for the antenna (see p65, Dec25) $55.00 Assembled Board: a fully-assembled PCB with all non-optional components, front and rear panels are sold separately below (SC7531; see p28, Nov25) - front & rear panels (SC7532) - 8MiB APS6404L-3SQR-SN PSRAM SOIC-8 IC (SC7530) PICKIT BASIC POWER BREAKOUT KIT (SC7512) Includes all parts except the jumper wire and glue (see p39, Sep25) *Prices valid for month of magazine issue only. All prices in Australian dollars and include GST where applicable. # Overseas? Place an order on our website for a quote. $25.00 (SEP 25) $90.00 $7.50 $5.00 $20.00 PRINTED CIRCUIT BOARDS PRINTED CIRCUIT BOARD TO SUIT PROJECT IDEAL BRIDGE RECTIFIER, 28mm SQUARE SPADE ↳ 21mm SQUARE PIN RASPBERRY PI CLOCK RADIO MAIN PCB ↳ DISPLAY PCB MICROPHONE PREAMPLIFIER ↳ EMBEDDED VERSION RAILWAY POINTS CONTROLLER TRANSMITTER ↳ RECEIVER LASER COMMUNICATOR TRANSMITTER ↳ RECEIVER PICO DIGITAL VIDEO TERMINAL ↳ FRONT PANEL FOR ALTRONICS H0190 (BLACK) ↳ FRONT PANEL FOR ALTRONICS H0191 (BLACK) ARDUINO FOR ARDUINIANS (PACK OF SIX PCBS) ↳ PROJECT 27 PCB WII NUNCHUK RGB LIGHT DRIVER (BLACK) SKILL TESTER 9000 PICO GAMER ESP32-CAM BACKPACK WIFI DDS FUNCTION GENERATOR 10MHz to 1MHz / 1Hz FREQUENCY DIVIDER (BLUE) FAN SPEED CONTROLLER MK2 ESR TEST TWEEZERS (SET OF FOUR, WHITE) DC SUPPLY PROTECTOR (ADJUSTABLE SMD) ↳ ADJUSTABLE THROUGH-HOLE ↳ FIXED THROUGH-HOLE USB-C SERIAL ADAPTOR (BLACK) AUTOMATIC LQ METER MAIN AUTOMATIC LQ METER FRONT PANEL (BLACK) 180-230V DC MOTOR SPEED CONTROLLER STYLOCLONE (CASE VERSION) ↳ STANDALONE VERSION DUAL MINI LED DICE (THROUGH-HOLE LEDs) ↳ SMD LEDs GUITAR PICKGUARD (FENDER JAZZ BASS) ↳ J&D T-STYLE BASS ↳ MUSIC MAN STINGRAY BASS ↳ FENDER TELECASTER COMPACT OLED CLOCK & TIMER USB MIXED-SIGNAL LOGIC ANALYSER (PicoMSA) DISCRETE IDEAL BRIDGE RECTIFIER (TH) ↳ SMD VERSION MICROMITE EXPLORE-40 (BLUE) PICO BACKPACK AUDIO BREAKOUT (with conns.) 8-CHANNEL LEARNING IR REMOTE (BLUE) 3D PRINTER FILAMENT DRYER DUAL-RAIL LOAD PROTECTOR VARIABLE SPEED DRIVE Mk2 (BLACK) FLEXIDICE (RED, PAIR OF PCBs) SURF SOUND SIMULATOR (BLUE) COMPACT HIFI HEADPHONE AMP (BLUE) CAPACITOR DISCHARGER PICO COMPUTER ↳ FRONT PANEL (BLACK) ↳ PWM AUDIO MODULE DIGITAL CAPACITANCE METER 5MHZ 40A CURRENT PROBE (BLACK) BATTERY MODEL RAILWAY TRANSMITTER ↳ THROUGH-HOLE (TH) RECEIVER ↳ SMD RECEIVER ↳ CHARGER USB PROGRAMMABLE FREQUENCY DIVIDER HIGH-BANDWIDTH DIFFERENTIAL PROBE NFC IR KEYFOB TRANSMITTER POWER LCR METER WAVEFORM GENERATOR PICO 2 AUDIO ANALYSER (BLACK) PICO/2/COMPUTER ↳ FRONT & REAR PANELS (BLACK) ROTATING LIGHT (BLACK) 433MHZ TRANSMITTER VERSATILE BATTERY CHECKER ↳ FRONT PANEL (BLACK, 0.8mm) DATE DEC23 DEC23 JAN24 JAN24 FEB24 FEB24 FEB24 FEB24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 APR24 APR24 APR24 MAY24 MAY24 MAY24 JUN24 JUN24 JUN24 JUN24 JUN24 JUL24 JUL24 JUL24 AUG24 AUG24 AUG24 AUG24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 OCT24 OCT24 OCT24 OCT24 OCT24 NOV24 NOV24 NOV24 DEC24 DEC24 DEC24 DEC24 DEC24 JAN25 JAN25 JAN25 JAN25 JAN25 JAN25 FEB25 FEB25 FEB25 MAR25 MAR25 MAR25 APR25 APR25 APR25 APR25 MAY25 MAY25 For a complete list, go to siliconchip.com.au/Shop/8 PCB CODE 18101241 18101242 19101241 19101242 01110231 01110232 09101241 09101242 16102241 16102242 07112231 07112232 07112233 SC6903 SC6904 16103241 08101241 08104241 07102241 04104241 04112231 10104241 SC6963 08106241 08106242 08106243 24106241 CSE240203A CSE240204A 11104241 23106241 23106242 08103241 08103242 23109241 23109242 23109243 23109244 19101231 04109241 18108241 18108242 07106241 07101222 15108241 28110241 18109241 11111241 08107241/2 01111241 01103241 9047-01 07112234 07112235 07112238 04111241 9049-01 09110241 09110242 09110243 09110244 04108241 9015-D 15109231 04103251 04104251 04107231 07104251 07104252/3 09101251 15103251 11104251 11104252 Price $2.00 $2.00 $12.50 $7.50 $7.50 $7.50 $5.00 $2.50 $5.00 $2.50 $5.00 $2.50 $2.50 $20.00 $7.50 $20.00 $15.00 $10.00 $5.00 $10.00 $2.50 $5.00 $10.00 $2.50 $2.50 $2.50 $2.50 $5.00 $5.00 $15.00 $10.00 $12.50 $2.50 $2.50 $10.00 $10.00 $10.00 $5.00 $5.00 $7.50 $5.00 $2.50 $2.50 $2.50 $7.50 $7.50 $5.00 $15.00 $5.00 $10.00 $7.50 $5.00 $5.00 $2.50 $2.50 $5.00 $5.00 $2.50 $2.50 $2.50 $2.50 $5.00 $5.00 $2.50 $10.00 $5.00 $5.00 $5.00 $10.00 $2.50 $2.50 $5.00 $7.50 PRINTED CIRCUIT BOARD TO SUIT PROJECT TOOL SAFETY TIMER RGB LED ANALOG CLOCK (BLACK) USB POWER ADAPTOR (BLACK, 1mm) HWS SOLAR DIVERTER PCB & INSULATING PANELS SSB SHORTWAVE RECEIVER PCB SET ↳ FRONT PANEL (BLACK) 433MHz RECEIVER SMARTPROBE ↳ SWD PROGRAMMING ADAPTOR DUCTED HEAT TRANSFER CONTROLLER ↳ TEMPERATURE SENSOR ADAPTOR ↳ CONTROL PANEL MIC THE MOUSE (PCB SET, WHITE) USB-C POWER MONITOR (PCB SET, INCLUDES FFC) HOME AUTOMATION SATELLITE PICKIT BASIC POWER BREAKOUT DUAL TRAIN CONTROLLER TRANSMITTER DIGITAL PREAMPLIFIER MAIN PCB (4 LAYERS) ↳ FRONT PANEL CONTROL ↳ POWER SUPPLY VACUUM CONTROLLER MAIN PCB ↳ BLAST GATE ADAPTOR POWER RAIL PROBE RGB LED STAR EARTH RADIO DCC DECODER DCC BASE STATION MAIN PCB ↳ FRONT PANEL REMOTE SPEAKER SWITCH ↳ CONTROL PANEL DCC REMOTE CONTROLLER MAINS HUM NOTCH FILTER MAINS LED INDICATOR DCC BOOSTER / REVERSE LOOP CONTROLLER ↳ FRONT PANEL SOLAR PANEL PROTECTOR (WHITE) GRAPHING THERMOMETER PICOSDR CONTROL PCB ↳ RF PCB ↳ FRONT PANEL (BLACK) DCC/DC STEPPER MOTOR DRIVER CALLIOPE AMPLIFIER MICROMITE AUDIO PLAYER ADD-ON ↳ ALL-IN-ONE μDCC DECODER SIMPLE LC METER WIFI ALARM MONITOR POWER AMPLIFIER CLIPPING INDICATOR PINBALL MACHINE CONTROL BOARD ↳ POWER SUPPLY ↳ PLAYER LED BOARD ↳ SCORE LED BOARD ↳ LED OUTPUT BOARD ↳ BUMPER LED BOARD ↳ CASCADE LED BOARD ↳ SWITCH INPUT BOARD ↳ GENERAL INPUT BOARD ↳ HIGH-CURRENT INTERFACE ↳ ROLLOVER INTERFACE ↳ BUMPER DRIVER SSB TRANSMITTER (MikeOne/Two/Three) SIMPLE USB POWER MONITOR HUMAN COMFORT INDICATOR ADJUSTABLE ULTRASONIC CLEANER MAIN PCB ↳ FRONT PANEL CONTROL PCB SNAP-TYPE DCC ACCESSORY DECODER ↳ SERVO-TYPE I2C CONTROLLER DATE MAY25 MAY25 MAY25 JUN25 JUN25 JUN25 JUN25 JUL25 JUL25 AUG25 AUG25 AUG25 AUG25 AUG25 SEP25 SEP25 OCT25 OCT25 OCT25 OCT25 OCT25 OCT25 NOV25 DEC25 DEC25 DEC25 JAN26 JAN26 JAN26 JAN26 FEB26 FEB26 FEB26 MAR26 MAR26 MAR26 MAR26 APR26 APR26 APR26 APR26 APR26 APR26 APR26 MAY26 MAY26 MAY26 MAY26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUN26 JUL26 JUL26 JUL26 JUL26 JUL26 PCB CODE Price 10104251 $5.00 19101251 $15.00 18101251 $2.50 18110241 $20.00 CSE250202-3 $15.00 CSE250204 $7.50 15103252 $2.50 P9054-04 $5.00 P9045-A $2.50 17101251 $10.00 17101252 $2.50 17101253 $2.50 SC7528 $7.50 SC7527 $7.50 15104251 $3.50 18106251 $2.00 09110245 $3.00 01107251 $30.00 01107252 $2.50 01107253 $7.50 10109251 $10.00 10109252 $2.50 P9058-1-C $5.00 16112251 $12.50 06110251 $5.00 09111241 $2.50 09111243 $5.00 09111244 $5.00 01106251 $5.00 01106252 $2.50 09111245 $5.00 01003261 $7.50 10111251 $2.50 09111248 $5.00 09111249 $5.00 17112251 $7.50 04102261 $3.00 CSE251101 $5.00 CSE251102 $5.00 CSE251103 $7.50 09111242 $2.00 01111212 $5.00 01110251 $2.50 01110252 $5.00 09111247 $1.50 04103261 $2.50 01304261 $2.50 01104261 $15.00 08107261 $25.00 08107262 $7.50 08107263 $2.50 08107264 $5.00 08107265 $2.50 08107266 $5.00 08107267 $5.00 08107268 $2.50 08107269 $2.50 08107260 $2.50 08117261 $2.50 08117262 $5.00 06103261 $2.50 04104261 $5.00 21105261 $5.00 04105261 $7.50 04105262 $5.00 09111254 $3.00 09111255 $3.00 09111256 $3.00 TRANSCEIVER TEST SET RF/AUDIO PCB ↳ CONTROL PCB MODEL RAILWAY DESTINATION DISPLAY ↳ FLEX ANTENNA PCB AUG26 AUG26 AUG26 AUG26 06104261 06104262 09111252 06101233 NEW PCBs $5.00 $5.00 $2.00 $2.00 We also sell the Silicon Chip PDFs on USB, RTV&H USB, Vintage Radio USB and more at siliconchip.com.au/Shop/3 The end of the venerable NE5532 could be used, minimising their thermal noise contribution. Of course, there were places where discrete transistors still ruled; for example, in circuits that needed to run from higher supply rails than the generous ±22V limit of the NE553x series. Still, in a lot of cases, the size and component count could be seriously reduced compared to discrete transistors (and possibly the cost as well) with little to no loss in performance. Fast forward to today There are hundreds of different op amps available, but the NE5532 (initially released in 1979) is ‘the classic’ audio op amp. It was such a good design that it is still widely used today, 55+ years later. That may be finally changing; unfortunately, not voluntarily. By Nicholas Vinen Various NE5532 variants and derivatives. From left-to-right, top-to-bottom: an original 1977 TDA1034, Signetics NE5534A, a couple of older TI NE5534As, a Signetics ceramic NE5532, a Signetics plastic DIP part, a couple of older TI NE5532s and a JRC SIL NJM2114L. Source: Jacob Rothman. I t is not surprising that a 1970s IC design may become obsolete in 2026. However, the way it is happening is far from satisfactory, and there are some real pitfalls for users. You may be wondering: why do we care so much about this part? Aren’t there better options? Well, yes... and no. There are good reasons why we’ll be genuinely sad to see it go. When it was released in November 1979, the NE5532 (and related NE5534) was revolutionary. IC op amps had been around for a while by then; the classic μa741 was released in 1968. If you wanted to build a hifi audio circuit in the early-to-mid 1970s, you’d use discrete transistors, as they 98 Silicon Chip could give much better performance. That was especially true for record player preamps that needed a lot of gain (about 100 times, more at low frequencies). That pretty much ended with the introduction of the NE553x series by Signetics. Finally, an op amp had performance that rivalled discrete transistors. They had extremely low noise – around 4-5nV/√Hz – making them good enough for use in phono preamp stages. They also gave extremely low distortion, ≤0.0004% THD+N across most of the audio band at moderate gains. The NE5532’s party trick was its ability to drive 600W loads with virtually no degradation in performance. That meant that low-value resistors Australia's electronics magazine The situation is clearly a lot different now than it was back then. We have many different high-performance op amps to choose from, including JFET and CMOS input types, railto-rail types, wide-bandwidth types, low-voltage types and so on. There are clearly applications now where other op amps, like those from Texas Instruments’ OPAxxxx series, are superior. That might be a battery-­ powered circuit, where the op amp needs to run from a supply voltage where the NE5532 wouldn’t work well, with a rail-to-rail swing to maximise signal handling. Or where even lower noise is required, or high-­ impedance inputs. As with anything, though, there are trade-offs. There are disadvantages to these fancy new op amps. Most of them are not available in DIP; it’s an SMD package (eg, SOIC-8) or nothing. Also, over time, the NE5532 became incredibly cheap; for much of the last decade, it cost under 50¢ per piece, even in smaller quantities. Some of the new, better op amps are not too dear (say $1-2 each), but others can be $5 or more. $2 per dual op amp isn’t so bad when you need one, but if you’re designing a device with many op amps, the cost can add up quickly! So if you wanted a high-­performance DIP op amp, until recently, the NE5532 was usually the best choice. The LM833 is sometimes a good alternative, but it’s best for driving 1kW+ loads and is limited to ±18V supply rails, so often the NE5532 was a better choice. The LM833 also usually costs a bit more. In addition, many of the new ‘whizzbang’ op amps can be a lot more ‘fussy’ in use. They require quieter supplies, better bypassing, and better layouts (keeping magnetic loops small). Otherwise, they could oscillate, leading to high distortion – exactly the sort of siliconchip.com.au thing you’re trying to avoid with an expensive op amp. In contrast, the NE5532 generally performs well with a single 100nF bypass capacitor. As long as the PCB layout isn’t terrible, you usually won’t have any problems with it. In other words, it’s a forgiving part to use in your designs. This is probably one reason these new parts are not available in dual inline packages; lead and trace inductance would make proper supply bypassing very difficult. Besides, these days, most manufacturers (the customers buying these parts in volume) prefer SMD packages for easier, cheaper assembly and better board density. The result of all this is that, for a long time, the NE5532 was a safe choice. Not any more, unfortunately... What happened? Signetics was acquired by Philips in 1975, and the brand was subsequently phased out. The NE5532 continued to be manufactured by other companies such as Texas Instruments (TI) and National Semiconductor (NS). TI bought NS in 2011, leaving them as (almost) the sole manufacturer of the NE553x. The exception was (and is) onsemi, who still makes the SMDonly NE5532D. However, in December 2025, TI released a new “RevK” data sheet for the NE5532 with significantly degraded specifications, as shown in Table 1. This was apparently spotted by diyAudio forum member diyralf in February this year – see siliconchip. au/link/accj We were informed of this change by Practical Electronics contributor Jacob Rothman this May. We had heard rumours of problems with NE5532s made by TI, even before December 2025, but we hadn’t realised what was going on until Jacob wrote about this topic in the July 2026 issue of Practical Electronics magazine, for his Audio Out column. After all, who expects a part that has been in production for 50+ years, with pretty consistent performance during that time, to suddenly change? Surely a reputable manufacturer wouldn’t silently replace it with an inferior version – would they? There’s some good news here – the ‘new version’ of the NE5532 has slightly more bandwidth and slightly lower current consumption. But the siliconchip.com.au Fig.1: total harmonic distortion plus noise (THD+N) versus frequency for the old (RevJ data sheet) and new (RevK) NE5532 ICs by Jacob Rothman. Note that this is in a reasonably ‘challenging’ circuit; the results will be more similar for something like a simple unity-gain buffer. bad news is much worse. The maximum slew rate is almost halved, leading to higher measured distortion in our tests. Critically, with the maximum supply voltage reduced to ±18V, in circuits that ran the NE5532s at (say) ±20V – within the old specification! – these new parts will now go up in a puff of smoke. This has been verified – it isn’t that they might fail at ±20V, they will fail. Bizarrely, the internal circuit structure has changed so much that the input transistors, which were NPN types in the original NE5532 and subsequent versions, have now become PNP. In circuits where the input transistor polarity matters (which is not uncommon), they won’t work correctly either. Frankly, it’s hard to see how it’s possible to sell these parts under the same code. That’s the real problem; those Table 1 – NE5532 RevJ vs RevK Specification RevJ RevK Bandwidth 10MHz 12MHz Slew rate 9V/μs 5V/μs Distortion Very low Higher Max. voltage ±22V ±18V Rec. voltage ±20V ±15V Supply current 8mA 6mA Input transistors NPN PNP Input clamp diodes? Yes No Australia's electronics magazine who are unaware of this change might buy the new parts thinking, quite reasonably, that they’re a drop-in replacement for the old ones with unchanged performance. They aren’t. For example, our Compact Hifi Headphone Amp (December 2024 & January 2025; siliconchip.au/ Series/432) uses NE5532 op amps and we provide distortion plots and other performance specifications. So you’d expect that if you build our circuit using the specified parts, you will get a device with very close to the same performance. However, if you use these new NE5532s, that is not guaranteed. There may not even be a good way to tell which version you receive, although it probably won’t be long before all the old stock is out of the system and the new/worse device is all you’ll get. Another change to the design is that for most of the life, the NE5532 had ESD protection/clamp diodes on its inputs. The new design apparently lacks these, and the data sheet reflects this by halving the ESD protection rating from 2kV to 1kV under standard test conditions. Any circuit design that relied on these clamp diodes for protection or otherwise can’t use the new version. All these changes mean that not only are they not fully compatible, they also don’t result in the same performance – see Fig.1. We suspect that the new, lower-power version struggles to drive the same low load impedances the old one could. What about the NE5534? The NE5534 is, effectively, a single version of the dual NE5532. There are some differences; for example, the NE5534 is not unity-gain stable without an external compensation capacitor, and the NE5534 has a lower noise specification, but otherwise, they have a similar design and performance. There’s no evidence currently that the NE5534 will face the same fate as the NE5532. Still, given how the NE5532 change came as a surprise, we would not assume it won’t happen. Therefore, we plan to stock up on those parts as well, just in case, for those applications where they are worth using (or for boards already designed for the single part). A silver lining New Japan Radio (NJR) is another source of the NE5532, but they call August 2026  99 Table 2 – selection of alternative high-performance dual op amps Device Supply Input type Noise Bandwidth Distortion Package Rail-to-rail? Cost NE5532 (old) ±5-22V NPN NJM5532D ±5-22V NPN 5nV/√Hz 10MHz ~0.0003% DIP/SOIC No 40¢ 5nV/√Hz 10MHz ~0.0003% DIP/SOIC No $2+ NE5532 (onsemi) ±5-22V* NPN 5nV/√Hz 10MHz ~0.0003% SOIC No $1.50 NE5534 (single) ±5-22V NPN 3.5nV/√Hz 10MHz ~0.0003% DIP/SOIC No $1.30 NE5532 (new) ±5-18V PNP 5nV/√Hz 12MHz ~0.001% DIP/SOIC No 40¢ LM833 ±5-18V PNP 4.5nV/√Hz 15MHz ~0.0003% DIP/SOIC No 40¢ NJM4580D ±2-18V PNP 3nV/√Hz 15MHz 0.0005% DIP/SOIC No $1.10 OPA1602 ±2.25-18V NPN 2.5nV/√Hz 35MHz 0.00003% SOIC Output only $3.50 OPA1612 ±2.25-18V NPN 1.1nV/√Hz 40MHz 0.000015% SOIC Output only $8.30 OPA1642 ±2.25-18V JFET 5.1nV/√Hz 11MHz 0.00005% SOIC Output only $2.50 OPA1656 ±2.25-18V JFET 2.9nV/√Hz 53MHz 0.00003% SOIC Output only $1.40 OPA1679 ±2.25-18V JFET 4.5nV/√Hz 16MHz 0.0001% SOIC Output only $3.25 OPA1692 ±1.75-18V NPN 4.2nV/√Hz 16MHz 0.000045% SOIC Output only $2.20 OPA2210 ±2.25-18V NPN 2.2nV/√Hz 18MHz 0.000025% SOIC Output only $5.50 * recommended limit ±20V for thermal reasons their version the NJM5532. NJR was acquired by Nisshinbo in 2021, so its parts are now sold under that brand. According to the data sheet, its performance is an exact match to the chip we have become accustomed to. The good news is that they are still in production, including in dual in-line packages (DIP). The bad news is that they don’t plan to keep making them forever; at least, not in DIP. But they say they will continue for a few more years. At the time of writing, DigiKey has 47,167 NJM5532Ds in stock, while Mouser has 2268. Other vendors worldwide will also have some of these. For the convenience of Silicon Chip readers, and in case those are all sold out quickly, we will also have a reasonable stock of NJM5532Ds, the lower-­ noise version (NJM5532DD) and original NE5534s (the single channel version; the new versions may eventually have the same problems as the new NE5532s). In the UK, Jacob Rothman’s AOShop will also have numerous NE5532Ds and NJM5532Ds available for Practical Electronics readers. Most shops will still sell NE5532s and NE5534s, but you may not know which type you’re going to get when you order them, so we think using the NJM versions will be safer if you need guaranteed performance. Another op amp available in DIP that might be a good choice for circuits designed for the NE5532 or NJM5532D 100 Silicon Chip is the NJM4580D. This part has lower noise (3nV/√Hz compared to 5nV/√Hz) but is only rated for absolute maximum supply rails of ±18V, so it’s better suited to applications using lower supply voltages like ±15V. Still, it is available and its price is reasonable. Newer op amps As we mentioned earlier, there are better op amps than the NE5532 available, but almost all of them are SMDonly. Table 2 summarises some of the better options. Conclusion There are plenty of high-performance op amps available at the moment, but the NE5532/NJM5532D remains an excellent choice for mains-powered equipment, with its low cost, low noise, low distortion, high supply voltage capability and ease of use. It’s also one of the last good choices that’s available in through-hole (DIP) packages. It would be a pity if we couldn’t use this part anymore because it has been ‘dumbed down’. The disappointing part is not so much that these nearly 50-year-old parts are being discontinued; it’s that, if we hadn’t found out they were being replaced with an inferior version, we might have kept using them and wondering why our circuits didn’t work as well any more! If you have one of these parts and aren’t sure which version it is, apply a suitable supply voltage and check the current draw. If it’s less than 7mA, it’s likely the newer version; if it’s more than 8mA, it’s probably the older type. If you’re happy to use an SOIC part, the onsemi NE5532 remains a fine choice, but it’s much more expensive than the TI parts. At that price, you might consider something like the OPA1656 instead, unless you need the high supply rail voltage support. Or, if the output loading is not too severe, you could use the LM833, at a similar cost to the NE5532. If you must have the same performance as the good old NE5532 in DIP, the NJM5532D (or even better, DD) is a fine choice, but at a higher cost. SC Fake NE5532s from AliExpress. They’re some other kind of dual op amp with the markings ground off and NE5532 etched instead. Australia's electronics magazine siliconchip.com.au ASK SILICON CHIP Got a technical problem? Can’t understand a piece of jargon or some technical principle? Drop us a line and we’ll answer your question. Send your email to silicon<at>siliconchip.com.au Using Pinball Machine controller as a PLC I thought the control board Phil Prosser designed for his Pinball Machine (June-October 2026 issues; siliconchip.au/Series/460) might make a good universal control system. There are lots of systems that could do with a controller that has inputs, high-­current outputs and logic. Still, I imagine making universal software to do that might be difficult. Would it be possible? (J. E., Dalton, NSW) ● Phil Prosser responds: You are right, the board and state machine is more or less a PLC, but customised to the Pinball Machine application. To make the design generic, we would need to: • Make the core logic board generic (easy) • Make more generic I/O interfaces (easy) • Change the software to be easily programmable and with more structure (tedious) To turn this into a proper PLC, you would almost certainly need to have access to a ladder logic compiler (the language used to program a typical PLC) and support for it. So I think it would be a lot of work for a relatively low-­demand project. My feeling is that if someone needs a PLC, they would be better off simply buying one. Still, if they wanted to modify the Pico software for the Pinball Machine controller for another application, the source code is available, and an experienced programmer could do that pretty easily. ● We have specified a lithium-ion cell for this project in the Parts List, and that is what should be used. The circuit can be powered by a single LiFePO4 cell, as shown in the photo, but it would need to be charged offboard since the onboard charger wouldn’t terminate charging at the correct voltage (3.65V). Questions on ideal rectifiers I have a follow-up to the question/ answer you published in the May 2026 issue, on page 110, under the heading “An ideal bridge rectifier for power amplifiers”. When you mention “a transformer with two separate secondaries, not one centre-tapped secondary”, I am slightly confused. Can you show me an example of such a transformer? I have a 40-0-40V transformer and am not sure what type it is. On the description of the Ideal Rectifier in the magazine, it mentions that the benefits of the Ideal Rectifier is that of a lower loss in voltage and power. So does the calculation of the DC rails being 1.41× the AC secondary voltages still apply? (W. L., Singapore) ● Many high-power toroidal transformers have separate secondaries since it adds flexibility and doesn’t cost the manufacturer anything. One example is the Altronics M5530C. If you look that up on their web page, you will see that the diagram shows two separate 30V secondaries that can be wired in series to give a 30-030V output. That means you can also connect the secondaries in parallel to get a single higher-current 30V winding, or use them separately to get two separate lower-current, isolated 30V windings. If your transformer only has a single secondary with a centre tap, you will either need to use a standard bridge rectifier or wait until we publish a higher-voltage version of the active rectifier from the September 2024 issue (siliconchip.au/Article/16580) that can work with centre-tapped transformers. Phil Prosser is working on such a design, but it isn’t finished yet. Using Remote Speaker Switch to switch input signals Taking a closer look at this project in the June issue (siliconchip.au/ Article/20362), I noticed that while all the documentation has/lists a 3.7V Li-ion cell, the photo on page 48 clearly shows a 3.2V LiFePO4 cell being used. Which type is correct? (B. P., Toowoomba, Qld) I suspect the Remote Speaker Switch project (January 2026; siliconchip.au/ Article/19561) would be just as effective switching the line audio signal instead of switching the higher-voltage speaker connection, and thus having a dedicated amplifier in or close to the actual speaker and powered locally. Therefore, taking advantage of the ton of small, cheap and powerful amplifiers online, switching 1V line-level audio signals using shielded audio cabling shouldn’t be a problem, I would imagine. (B. A., Dee Why, NSW) ● The speaker switch was designed for switching the power amplifier outputs to speakers. In the particular application the prototype was used (see the June 2026 issue, starting on page 54) there was no room to mount the amplifiers near the speakers. Instead, they were installed together in a loft with wires running to the indoor and outdoor speakers. If you switch the line-level signals instead and use long runs of single-core shielded cable, you are likely to pick up mains hum. To prevent this, you need to use balanced cabling, such as twin-cored shielded microphone cable. You would need an unbalanced-to-balanced converter at one end and a balanced-to-unbalanced converter at the other end. A suitable project for balance/unbalanced conversion was published in the June 2008 issue (Balanced/Unbalanced Converter For Audio Signals; siliconchip.au/ Article/1857). The Remote Speaker Switch could switch balanced audio signals using its stereo support, with the left and right channels instead switching the ‘hot’ and ‘cold’ balanced conductors. siliconchip.com.au Australia's electronics magazine Correct cell type to use for Comfort Indicator August 2026  101 GPS altimeter wanted Has Silicon Chip ever published a project to build a GPS altimeter? I have searched and found an altimeter using a barometer module but nothing using a GPS module (have I missed it?). I happen to have a spare GPS module that I bought from you previously. I just drove back from Narrogin to Bunbury (WA), approximately 200km from inland to the coast, and was staggered to see the height difference between Narrogin and Bunbury right on the coast. So now I am fascinated to find out how much height difference there is. I have always wanted a simple barometer; the ones on eBay all have clocks and other features. I just want the basic barometer, if possible. How does GPS compare to the barometer sensor for accuracy? PS, my GPS sweeping-hand clock has been going beautifully for five years now without a hitch. (G. S., Eaton, WA) ● The June/July 2021 Advanced GPS Computer (siliconchip.au/Series/366) can display altitude, among other things. It should be very accurate. Barometerbased altimeters will have their accuracy affected by atmospheric changes, but GPS has no such problems and should be within a metre or two. The rough formula Vdc = Vac × 1.414 doesn’t take into account rectifier losses because they are basically cancelled out by the fact that the transformer output voltage will be slightly higher than nominal under light loads. The output with an active rectifier will be around 1-2V higher than with a standard bridge rectifier under the same conditions. Notch Filter oscillates at high Q settings I have constructed the kit for John Clarke’s Notch Filter (February 2026; siliconchip.au/Article/19660). While it works, the filter circuit is oscillating at 1.99MHz with a 0.3V peak-to-peak signal at the pin 14 outputs of IC1d & IC2d. I have tuned the channel filters independently, one to 50Hz and the other to 150Hz, to eliminate strong power supply harmonics. In the Fliege filter, I used the VR3 & 4 configuration rather than the matched fixed resistors. As discussed in the text, it was fun trying to match the caps. The values I ended up with are: Left Right Cx 46.7nF 15.4nF Cy 47.3nF 15.3nF VR3/7 64.5kW 69.0kW VR4/8 70.4kW 69.8kW VR1/5 170kW 195kW With or without any input or output connections, both channels are oscillating at this frequency. Both power supply rails are clean. I seek a suggested resolution with a bypass or stabiliser on IC1d (IC2d) without 102 Silicon Chip impacting the operation of the notch filter at the low end. Thanks; I love your work and the magazine. (M. B., Elanora, Qld) ● Oscillation can occur if the Q is set too high. Check if the oscillation stops if you reduce the Q settings (VR1 & VR5). If not, a low-value capacitance across pins 13 & 14 of IC1d/IC2d could shut the oscillation down. Start with 10pF and increase to higher values, up to 100pF if required. Note: the reader responded to say the oscillation was due to faulty op amps; new ones from Jaycar fixed it. Colour selection on the RP2350B Computer Regarding the 4-bit colour Mode 3 (640×480 pixels) with 16 colours on the RP2350B Computer (November 2025; siliconchip.au/Article/19220), what are the 16 colours that are available? (R. M., Melville, WA) ● This colour mode is explained on page 142 of the PicoMite manual that’s part of the RP2350B Computer firmware download package. You can select which colour is used for each of the 16 possible values, although there is a default mapping that should suit many users. It reads as follows: MAP The MAP commands allow the programmer to select the colours used in 4-bit colour modes. Each value in the 4-bit colour palette can be set to one of the 16 available colours. See the MAP function for more information. MAP(n) = rgb% This will assign the 24-bit colour ‘rgb% to all pixels with the 4-bit colour Australia's electronics magazine value of ‘n’. The RGB value is converted to one of the available 16 VGA RGB121 colours as set by the resistor network. The change is activated after the MAP SET command. MAP RESET This will reset the colour map to the default colours which in 4-bit mode are: ‘n’ Colour Value 15 WHITE RGB(255, 255, 255) 14 YELLOW RGB(255, 255, 0) 13 LILAC RGB(255, 128, 255) 12 BROWN RGB(255, 128, 0) 11 FUCHSIA RGB(255, 64, 255) 10 RUST RGB(255, 64, 0) 9 MAGENTA RGB(255, 0, 255) 8 RED RGB(255, 0, 0) 7 CYAN RGB(0, 255, 255) 6 GREEN RGB(0, 255, 0) 5 CERULEAN RGB(0, 128, 255) 4 MIDGREEN RGB(0, 128, 0) 3 COBALT RGB(0, 64, 255) 2 MYRTLE RGB(0, 64, 0) 1 BLUE RGB(0, 0, 255) 0 BLACK RGB(0, 0, 0) Programmable DC Load regulator confusion I’ve been slowly assembling the WiFi Programmable DC Load designed by Richard Palmer (September & October 2022; siliconchip.au/Series/388). Last weekend, I finally powered up the project for the first time. Unfortunately, I ran into a self-inflicted problem and destruction of at least the ESP32 dev module. I had installed the buck 5V regulator on the power board in reverse. l followed the construction image, Fig.14, in the October article but failed to cross-check the circuit diagram and photos. As a consequence, the 5V rail was receiving 12V. It was probably only a few seconds of exposure, and I did have current limiting on my bench PSU when I was doing the test, but the damage was done. I expect the 5V chips on the power board may have been damaged, but I won’t know for sure until I replace the ESP32 and test again fully assembled. I have re-tested the power board statically and checked for problems with a thermal camera and multimeter. So far, nothing gets hot, and there are no odd signal voltages showing on the 20-pin header. What I discovered and want your feedback on is that the outputs of the continued on page 104 siliconchip.com.au MARKET CENTRE Advertise your product or services here in Silicon Chip KIT ASSEMBLY & REPAIR FOR SALE DAVE THOMPSON (the Serviceman from Silicon Chip) is available to help you with kit assembly, project troubleshooting, general electronics and custom design work. No job too small. Based in Christchurch, New Zealand, but service available Australia/NZ wide. Email dave<at>davethompson.co.nz LEDsales KEITH RIPPON KIT ASSEMBLY & REPAIR: * Australia & New Zealand; * Small production runs. Phone Keith: 0409 662 794 keith.rippon<at>gmail.com WIDE TECHNICAL: * Adelaide area. * Kit and small production run, assembly and repair. Embedded controller support. * * Network/firewall support. * Industry certified operators. widetech.au<at>gmail.com LEDS, BRAND NAME AND GENERIC LEDs, filament LEDs, LED drivers, heatsinks, power supplies, kits and modules, components, breadboards, hardware, magnets. Please visit www. ledsales.com.au Micromite Explore-40 October 2024 Complete Kit SC6991: $35 siliconchip.au/Article/16677 PMD WAY offers (almost) everything for the electronics enthusiast – with full warranty, technical support and free delivery worldwide. Visit pmdway.com to get started. Includes the PCB and all onboard parts. Audio Breakout board and Pico BackPack are sold separately. WANTED PCB PRODUCTION TELEX MACHINE OR TELEPRINTER: I am looking to purchase a Siemens Telex Machine or Creed teleprinter (any models). Also, any teleprinter parts. Phone Pete: 0419 036 213 PCB MANUFACTURE: single to multilayer. Bare board tested. One-offs to any quantity. 48 hour service. Artwork design. Excellent prices. Check out our specials: www.ldelectronics.com.au ADVERTISING IN MARKET CENTRE Classified Ad Rates: $32.00 for up to 20 words (punctuation not charged) plus $1.20 for each additional word. Display ads in Market Centre (minimum 2cm deep, maximum 10cm deep): $82.50 per column centimetre per insertion. All prices include GST. Closing date: 5 weeks prior to month of sale. To book, email the text to silicon<at>siliconchip.com.au and include your name, address & credit card details, or phone (02) 9939 3295. WARNING! Silicon Chip magazine regularly describes projects which employ a mains power supply or produce high voltage. All such projects should be considered dangerous or even lethal if not used safely. Readers are warned that high voltage wiring should be carried out according to the instructions in the articles. When working on these projects use extreme care to ensure that you do not accidentally come into contact with mains AC voltages or high voltage DC. If you are not confident about working with projects employing mains voltages or other high voltages, you are advised not to attempt work on them. Silicon Chip Publications Pty Ltd disclaims any liability for damages should anyone be killed or injured while working on a project or circuit described in any issue of Silicon Chip magazine. Devices or circuits described in Silicon Chip may be covered by patents. Silicon Chip disclaims any liability for the infringement of such patents by the manufacturing or selling of any such equipment. Silicon Chip also disclaims any liability for projects which are used in such a way as to infringe relevant government regulations and by-laws. Advertisers are warned that they are responsible for the content of all advertisements and that they must conform to the Competition & Consumer Act 2010 or as subsequently amended and to any governmental regulations which are applicable. siliconchip.com.au Australia's electronics magazine August 2026  103 VXO7805 on the power board and the linear 7805 on the ESP32/display board are bridged via the 20-pin ribbon cable on pins 18 & 22. This seems a bit odd and perhaps an oversight. If these hadn’t been bridged, the over-voltage would have been limited to just the power boards. I’ve cut the link between pins 18 & 20 on the ESP display board now. I have also compiled the source myself on the more recent Arduino IDE 2.3.9 under Linux. This required a few fixes that I have made and published as a pull request to Richard’s GitHub Advertising Index Altronics.................................47-50 Blackmagic Design....................... 7 Dave Thompson........................ 103 DigiKey Electronics..................OBC Emona Instruments.................. IBC Hare & Forbes............................... 9 Jaycar.................. IFC, 14-15, 28-29 Keith Rippon Kit Assembly....... 103 LD Electronics........................... 103 LEDsales................................... 103 Microchip Technology.................. 5 Mouser Electronics....................... 3 PCBWay....................................... 11 PMD Way................................... 103 Practical Electronics PDFs......... 69 SC Micromite Explore-40......... 103 SC Ideal Bridge Rectifiers........... 73 SC RP2350B Computer.............. 95 Silicon Chip Binders.................. 38 Silicon Chip PDFs on USB......... 58 Silicon Chip Shop.................96-97 Silicon Chip Songbird................ 80 Silicon Chip Subscriptions........ 39 The Loudspeaker Kit.com............ 8 Wide Technical.......................... 103 Wagner Electronics..................... 87 Next issue on-sale date Next Issue: the September 2026 issue is due on sale in newsagents by Monday, August 24th. Expect postal delivery of subscription copies in Australia between August 24th and September 11th. 104 Silicon Chip repo. These fixes include removing Richard’s home WiFi credentials from the WiFi secrets header file. I hope this is useful to others. I note, and have commented in Richard’s bug report, that there is a design bug with Kelvin mode that can cause damage to the INA232. I’d like to know more about this and whether I can help resolve the issue. (R. E., Dover Gardens, SA) ● Richard Palmer responds: there are two problems with the PCB layout that were noted in Silicon Chip’s Notes & Errata in the December 2022 issue: a via that is too close to a track and the voltage regulator pin reversal. From your letter, you have discovered the second problem the hard way. I have added an illustrated version of the errata to the GitHub repository for this project (https://github.com/ palmerr23/ESP32-DCLOAD/tree/ main/errata). Regarding Kelvin sensing, when the negative Kelvin sensing terminal is connected to the DUT, any resistance in the negative current lead will induce a negative voltage at the Kelvin terminal, which is connected directly to the ADC’s negative input. It is not a problem if the main +/− terminals are used to measure voltage as well as current. In that case, the ADC’s negative input will be connected to local ground via the 390W resistor. The updated document in the repository explains the concern in more detail and provides a solution, as well as a few minor circuit and software improvements suggested by two readers. 2007 issues (siliconchip.au/Series/56) and have a question regarding the operation of the software “ignprgm.asm”. I intend to use a MAP sensor for load determination and would like to understand how you normalise or average the MAP sensor voltage. The MAP sensor response will be relatively fast and will contain considerable peaks and troughs, particularly at idle. Lowering the ADC gain would probably not give me the resolution I would need for an engine with low vacuum (aggressive cam). I have gone through the source code and can find the ADC but cannot find where the MAP sensor voltage is ‘treated’, if indeed it is. I prefer not to use a TPS signal for load, and external hardware filtering (R/C) will introduce delays. Any insights you have would be appreciated. (G. M., Vermont, Vic) ● The load value as measured from the MAP sensor isn’t averaged. Due to the quick response required, the load is read in real time without adding any delay by software averaging or hardware averaging. The air pressure measured by the MAP sensor at that time is the one used to adjust timing. There isn’t any way to introduce averaging without affecting response time. The beginning of the code for the MAP sensor read is at line 1693: MEAS_LOAD ; measure LOAD input ; set A/D to channel 2 movlb D’1’ movlw B’00001001 movwf ADCON0 call ACQUIRE_AD movf A_DRESH,w movwf LOAD_D_A Alternative LCD for 8-digit Frequency Meter Modifying Flexitimer I want to build the Compact 8-digit for 0-60 minutes Frequency Meter from August 2016 (siliconchip.au/Article/10037) but Jaycar has discontinued supply of the LCD screen. Their replacement is completely different in all respects. Where can I get a compatible device? (J. A., Townsville, Qld) ● The Altronics Z7013 screen specified as the alternative in the parts list is still available. MAP sensing for programmable ignition I am building your Programmable Ignition System from the March-May Australia's electronics magazine Can the Jaycar Flexitimer kit (KA1732), based on the article from Electronics Australia, March 1991, be modified to produce a timer adjustable from 0 to 60 minutes? (R. M., Melville, WA) ● The Flexitimer circuit can produce a timeout of one hour using the Q13 output with RV2 set at mid-travel, but a timeout close to zero is not possible with this timer unless you choose a lower counter output from the 4020 counter IC, and even then, it will not be that close to zero. A digital timer would be a better choice for such a SC requirement. siliconchip.com.au “Rigol Offer Australia’s Best Value Test Instruments” New 2026 Products Oscilloscopes New 12Bit Scopes New Up To 13GHz RIGOL DS-1000Z/E - FREE OPTIONS RIGOL DHO/MHO Series RIGOL DSO-8000/A Series 450MHz to 200MHz, 2/4 Ch 41GS/s Real Time Sampling 424Mpts Standard Memory Depth 470MHz to 800MHz, 2/4 Ch 412Bit Vertical Resolution 4Ultra Low Noise Floor 4600MHz to 13GHz, 4Ch 410GS/s to 40GS/s Real Time Sampling 4Up to 4Gpts Memory Depth FROM $ 649 FROM $ ex GST 684 FROM $ ex GST 13,191 Multimeters Function/Arbitrary Function Generators RIGOL DG-800/900 Pro Series RIGOL DG-1000Z Series RIGOL DM-858/E 425MHz to 200MHz, 1/2 Ch 416Bit, Up to 1.25GS/s 47” Colour Touch Screen 425MHz, 30MHz & 60MHz 42 Output Channels 4160 In-Built Waveforms 45 1/2 Digits 47” Colour Touch Screen 4USB & LAN FROM $ 713 FROM $ ex GST Power Supplies ex GST 604 FROM $ ex GST Spectrum Analysers 632 ex GST Real-Time Analysers New Up To 26.5GHz RIGOL DP-932E RIGOL DSA Series RIGOL RSA Series 4Triple Output 2 x 32V/3A & 6V/3A 43 Electrically Isolated Channels 4Internal Series/Parallel Operation 4500MHz to 7.5GHz 4RBW settable down to 10 Hz 4Optional Tracking Generator 41.5GHz to 26.5Hz 4Modes: Real Time, Swept, VSA, EMI, VNA 4Optional Tracking Generator ONLY $ 799 FROM $ ex GST 1,321 FROM $ ex GST 3,210 ex GST Buy on-line at www.emona.com.au/rigol Sydney Tel 02 9519 3933 Fax 02 9550 1378 Melbourne Tel 03 9889 0427 Fax 03 9889 0715 email testinst<at>emona.com.au Brisbane Tel 07 3392 7170 Fax 07 3848 9046 Adelaide Tel 08 8363 5733 Fax 08 83635799 Perth Tel 08 9361 4200 Fax 08 9361 4300 web www.emona.com.au EMONA