Fullscreen HTML5Med Res (??MB)HTML4

The VERY BEST DIY Projects! Cleaning at up to 40W, with adjustable frequency, power and duration Ultrasonic Cleaner ADJUSTABLE DCC Accessory Decoders and an I2C Controller JULY 2026 ISSN 1030-2662 07 9 771030 266001 $ 00* NZ $1590 15 INC GST INC GST www.jaycar.com.au www.jaycar.co.nz Contents Vol.39, No.07 July 2026 12 Soft Robots Part 1: p28 Soft robots are flexible and adaptable machines that are useful in many different roles, like performing delicate surgery, squeezing through rubble, or even just working in crowded environments. By Dr David Maddison, VK3DSM Robotics feature 38 T50 Robot Mop & Vacuum The Ecovacs DEEBOT T50 Pro Omni is one of the more advanced cleaning robots available today. It has some advanced mapping and navigation features. Review by Nicholas Vinen Automated cleaning robot 66 Making Simple Enclosures Off-the-shelf enclosures for projects can be expensive; they’re also usually not the exact size you’re looking for. So I came up with a way to make my own low-cost enclosures in any size. By Andrew Woodfield, ZL2PD Enclosures for projects Adjustable Ultrasonic Cleaner DEEBOT T50 Pro Omni Robot Mop & Vacuum 84 Altium Designer 2026 Each year brings a new major version of Altium Designer, the electronic design automation software that we use to produce our circuits and PCBs. This version improves wiring harness support, ActiveBoM & file importing. Review by Tim Blythman Software feature 28 Adjustable Ultrasonic Cleaner Rated at up to 40W, our Ultrasonic Cleaner is fully adjustable for frequency, power and duration. It can use a variety of baths ranging from 2.5L to 4L in volume, and it is powered from a 12-15V DC supply. Part 1 by John Clarke Cleaning project 50 Phenomenal Pinball Machine This series explains how to design and build every part of your own Pinball Machine. This month we cover the multiple PCBs and their respective circuit diagrams that control the Pinball Machine. Part 2 by Phil Prosser Gaming project 70 DCC Accessory Decoders These two accessory decoders are used to control fixed devices such as points and signals in a model railway layout. One design is for snap-type & slow-motion point motors, while the other works with servo-type motors. Part 8 By Tim Blythman Model train project 80 I2C Controller This I2C Controller makes programming our Accessory Decoders above a breeze. However, this is not its only use; you could connect it to other devices that need a simple user interface. Part 9 By Tim Blythman Model train project Page 38 2 Editorial Viewpoint 4 Mailbag 25 Subscriptions 47 Circuit Notebook 69 Online Shop 88 Serviceman’s Log 94 Vintage Radio 101 Ask Silicon Chip 103 Market Centre 104 Advertising Index 104 Notes & Errata 1. A 15V split-supply from 5V USB 2. 3-digit mini LCD module 3. Randomly-timed model traffic lights 4. Capacitive proximity sensor National R-72 “Toot-a-Loop” by Ian Batty 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 Looming smartphone obsolescence We will soon reach the point where hundreds of millions of perfectly usable smartphones are made obsolete through software rather than hardware failure. Flagship and midrange phones released around 2019 still have hardware that is perfectly adequate today, with good cameras, good screens and processors fast enough for everyday use. Yet many of those phones will be abandoned. Some 2019 Android phones are stuck on Android 10 or 11, with no further updates from the manufacturer. Increasingly, apps are dropping support for those older versions. That means people with otherwise usable phones may soon be unable to run important apps, including banking, authentication, payment and other essential services. That is not because the hardware has suddenly become useless. There is nothing about a five- or six-year-old smartphone that makes it inherently incapable of doing its fundamental job. The problem is that the software support chain has been cut off. Whether or not this is deliberate in every case, the result is indistinguishable from planned obsolescence. The industry has created a system where software support is tied to hardware replacement cycles. When updates stop, app support gradually disappears, and consumers are pushed toward replacing devices that may still work perfectly well. Drivers and firmware may depend on the chip vendor. Testing and certification also take time and money. But those explanations do not change the end result: usable hardware is being discarded because the software ecosystem has been designed that way. Compare this with a laptop or desktop computer, where you can usually install a newer operating system yourself. That is true even though there is a much wider variety of hardware in PCs than in smartphones. The operating systems are designed to cope with that variety. Smartphones are more locked down than PCs, and that is part of the problem. It prevents otherwise serviceable hardware from having a longer life. If a phone is powerful enough to run a newer version of Android, users should not be entirely dependent on the original manufacturer choosing to provide it. We do not rely on Dell, Hewlett Packard, Asus or other PC makers to keep our desktop and laptop computers up to date forever. The operating system vendor provides updates, and in the case of Linux, you can still install a current operating system on very old hardware. I know because I have done it. It may be slower or limited in some ways, but you can keep using the computer as long as it remains practical. Five-to-six-year-old hardware should not be considered obsolete, especially when the advances in smartphone hardware over that period have been fairly modest for normal use. Windows 11 has attracted criticism for artificially excluding older PCs that are still capable of useful work. We should be just as concerned about smartphones. In fact, the smartphone problem may be worse, because phones are replaced more often, sold in much larger numbers, and contain batteries, rare metals and other materials that are costly to produce and recycle. This is going to create a giant pile of unnecessary e-waste. This is not just an Android problem either. iPhones have the same basic problem, although they are generally supported for longer. An iPhone can be physically fine, but eventually it stops receiving major iOS updates. Once that happens, app support gradually drops away as developers raise their minimum supported iOS version. At the time of writing, Apple’s current iOS compatibility list starts at the iPhone 11 generation, so anything older than that is already outside the current major iOS line. A device that still works and has a usable battery should not become useless because the software update chain has been cut off. That is wasteful, unnecessary and one of the more indefensible aspects of the modern smartphone ecosystem. Cover robot image: UC San Diego Jacobs School of Engineering www.flickr.com/photos/jsoe/46570014664 (CC-BY-2.0) Australia's electronics magazine by Nicholas Vinen 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”. Comments on analog computers I was interested to read the article in the May issue about Analog Computers (May & June 2026 issues; siliconchip.au/Series/459). In 1982, I designed an analog computer to provide simulated inertia for model trains. The voltage at the output of the analog computer controlled the train speed via a train speed controller. I’ve attached a scan of the papers I wrote at the time (shown opposite). Len Cox, Forest Hill, Vic. More on vintage analog computers Regarding the Analog Computers article in the May 2026 issue, many years ago as a young Army Telecommunication Mechanic, I was able to examine an obsolete British WWII No.10 Predictor. I was intrigued by a major component of the Predictor: a large wire-wound logarithmic potentiometer. It was about 500mm in diameter, with a Bakelite former about 200mm wide. The wiper arm was connected to the optical mechanism. The Predictor was used by the operators to optically track aircraft using a telescope. It indicated bearing, computed range and height. This information was sent to a searchlight or anti-­aircraft gun by a servo-type system, thus predicting the aiming point for the searchlights and AA guns. I guess this is an example of an early analog computer. Peter Johnston, Merimbula, NSW. The skin effect can be a major efficiency penalty As a design engineer, I recently had to develop a boost converter for use in a product. The requirements weren’t particularly onerous, with a 29V/1.2A output. My first port of call was the TI Webench design tool. I’ve used this quite a few times in the past and have had good results from it, although I mostly use it for buck converters. I didn’t have a lot of space available on my PCB, so I selected the “Small Footprint” option with my design parameters, and after a bit of deliberation, I selected the LM5155 IC, designed the circuit to match what Webench 4 Silicon Chip provided and integrated the parts into my board following the layout recommendations in the data sheet. When the PCB came back, I ran it up and was pleased with the results – between Webench giving an accurate design and KiCad’s netlist checking, it’s fairly hard to go wrong, at least with the basics. However, load testing revealed it was running way too hot – on my bench, in the open air, I measured the inductor at around 100°C! I tried a bigger inductor with a higher current rating and found similar results. Obviously, this was not acceptable and needed to be sorted out before going into production. Because it was the inductor that was getting so hot, the first thing I did was look into its specifications. Most importantly, I looked at the current rating; boost converters can be pretty demanding on inductors, with peak currents way higher than the output current. In my case, the current rating of the inductor I was using was about what Webench had asked for, and it was a good quality part from Würth. While I was wondering how I was going to figure this out, I remembered something I read in a recent issue of Silicon Chip – it was in the January instalment of the “Power Electronics” series (pages 28 & 29; siliconchip.au/Series/452) where Andrew Levido mentioned skin depth. I thought the article had stated not to use switching frequencies above 100kHz unless you have special inductors, but on re-reading, that’s not what it says. It just shows the skin depth at 1MHz as being way smaller than at 100kHz. Australia's electronics magazine siliconchip.com.au For what I was doing, however, with an amp of load or so, it turns out that a 100kHz maximum switching frequency is not a bad rule of thumb. Because I had selected the “Small Footprint” option, Webench had chosen to run at the highest frequency available for the LM5155 to get the smallest inductor. The resulting operating frequency was around 2.2MHz – no wonder my inductor was cooking. At this frequency, the current was travelling in the outer 45μm of the copper, so the effective resistance was much higher. I looked at the specifications of the inductor I was using, and it made no mention of an upper frequency limit or any frequency derating; it didn’t really have any high-­frequency rating at all. I went back to Webench and looked at some of its other part recommendations for similar 2MHz converters. One it recommended was a Coilcraft part that had a nice inductance/frequency graph; however, this graph topped out at 100kHz. I came to the conclusion that Webench doesn’t make any allowance for high-frequency effects at all. To confirm I was on the right track, I tried reducing the frequency of the converter by changing the requisite resistor. It was an 0402 package SMD resistor, and I managed to graft on a 100kW 0603 size part I had on hand in place. This took the frequency down to more like 220kHz. A quick test of this configuration showed the temperature stabilising at more like 75°C, so I was definitely on the right track. Currently, I’m planning to redesign it to run at around 100kHz. There are a few takeaways from this: • Webench doesn’t take into account frequency effects in inductors. I realise the difficulty they would have in doing so, but I still feel this is a bit of a flaw in the tool, especially when they calculate an efficiency to tenths of a percent; I doubt mine comes anywhere near the ~95% it predicted! It selected parts that were completely inappropriate for the job. • Unless a manufacturer states an inductor is rated for a particular frequency, assume its maximum is around 100kHz. • Try to keep the switching frequency somewhere around 100kHz, perhaps a bit higher for lower-current designs and vice versa. • It’s good to read magazine articles; it’s surprising what you pick up even if it’s not the prime focus of the article. I contacted TI about this, but they didn’t respond. I still think Webench is a good tool, but watch out regarding the frequency it chooses for you. D.T., Sylvania, NSW. Comment: you make some good points. We have had success with switchmode converters running well above 100kHz, although in retrospect this may have depended heavily on choosing suitable inductors. For example, the April-June 2014 40V Switchmode/ Linear Bench Supply (siliconchip.au/Series/241) used an LM5118 buck/boost controller running at around 500kHz, and we found it to be reasonably efficient up to 5A with the Signal Transformer SCIHP1367 inductor we chose. Its data sheet specifies performance at 200kHz and mentions operation up to 5MHz, suggesting that it was intended for relatively high-frequency switchmode use. That may be due to its winding construction, core material, or other design features that reduce high-frequency losses. 6 Silicon Chip So while 100kHz is probably too conservative as a hard limit, your broader warning is sound: at higher switching frequencies, the inductor’s DC current rating alone is not enough. Designers need to check frequency-related losses and temperature rise, and should be wary of design tools that optimise mainly for size without fully accounting for those effects. Suggestion for an updated ‘comfort indicator’ Once again, many thanks for the June 2026 edition of Silicon Chip. I really enjoyed Nicholas Vinen’s editorial on device “presets” that seem to lead to corrupted reality (in this case, sound). One wonders how often good engineering design is diminished by marketing and other groupthink ideas. Regarding the Comfort Indicator project in that same issue (siliconchip.au/Article/20362), for many years I worked in a professional engineering office. The building was a transformed factory brought up to a very good set of thermal designs, including quiet and diffuse air circulation (these looked like large fabric ducts that ‘leaked’ air according to the diameter and weave). Unfortunately, my workstation seemed to get above-­ average airflow and velocity, so I was always cold, no matter the season. Facilities management was dutiful in using several instruments to determine if it was me or the HVAC system at fault. To me, the fundamental missing component of the test instruments used was that there was no sensor or measurement that measured the effect of air velocity on skin. Despite volunteering to attach an array of temperature sensors to my hands, ears, etc to get a true picture, nothing was achieved. My final thought on ‘my’ problem was to design a comfort instrument that uses multiple sets of temperature sensors: a dry bulb, a wet bulb, plus a second wet bulb sensor in its own draft-free enclosure, in multiple locations. I didn’t attempt to construct one before management decided to move me to another workstation with less air movement. The daily solution was to use a fan heater under my desk to keep me warm in the height of Melbourne summers! Nonetheless, this idea remains fertile in my mind, as I wonder if there is an opportunity to design and build a ‘comfort sensor’ that is able to include the airflow effect on skin and predict true skin temperature. Mark Schijf, Doncaster East, Vic. Comment: wind speed is certainly taken into account for “feels like” temperatures given on weather reports. We may be able to update the Human Comfort Indicator one day with some kind of draft-sensing facility. An automotive ‘hot wire’ style airflow meter would likely do the job, but it would probably be too power-hungry for a small battery-­ powered device! Comments on the Airzone 6552A vintage radio The 6U7 is a ‘triple-grid’ valve, with a separate external connection for the suppressor (pin 5) that is not shown on the Airzone 6552A circuit (May 2026, p107; siliconchip. au/Article/20247). Separate suppressor connections were common in this generation of octal pentodes, and in their predecessor UX-based types. The Airzone circuit’s symbol for the 6A8 is also wrong as it shows only one of the two internally connected screens; Australia's electronics magazine siliconchip.com.au Introducing ATEM Mini The compact television studio that lets you create presentation videos and live streams! Blackmagic Design is a leader in video for the television industry, and now you can create your own streaming videos with ATEM Mini. Simply connect HDMI cameras, computers or even microphones. Then push the buttons on the panel to switch video sources just like a professional broadcaster! You can even add titles, picture in picture overlays and mix audio! Then live stream to Zoom, Teams or YouTube! Live Stream Training and Conferences Create Training and Educational Videos Monitor all Video Inputs! ATEM Mini’s includes everything you need. All the buttons are positioned on With so many cameras, computers and effects, things can get busy fast! The All models have built in hardware streaming engine for live streaming via its ethernet connection. This means you can live stream to YouTube, Facebook and Teams in much better quality and with perfectly smooth motion. You can even connect a hard disk or flash storage to the USB connection and record your stream for upload later! the front panel so it’s very easy to learn. There are 4 HDMI video inputs for ATEM Mini features a “multi-view” that lets you see all cameras, titles and program, connecting cameras and computers, plus a USB output that looks like a webcam plus streaming and recording status all on a single TV or monitor. There are even so you can connect to Zoom or Skype. ATEM Software Control for Mac and PC tally indicators to show when a camera is on air! Only ATEM Mini is a true is also included, which allows access to more advanced “broadcast” features! professional television studio in a small compact design! Use Professional Video Effects ATEM Mini is really a professional broadcast switcher used by television stations. This means it has professional effects such as a DVE for picture in picture effects commonly used for commentating over a computer slide show. There are titles for presenter names, wipe effects for transitioning between sources and a green screen keyer for replacing backgrounds with graphics. www.blackmagicdesign.com/au ATEM Mini Pro..........$475 ATEM Software Control..........FREE Learn More! the screen between the second grid/oscillator anode is missing. It’s not our job to correct manufacturer’s mistakes, but it’s worth noting that until at least the 1950s, some manufacturers used their own confusing symbology, with AWA’s inexplicable insistence on drawing their valve symbols upside-down and with the connecting pins in order clockwise, and confusing internal routing inside the symbol’s envelope as the worst, most common examples. It is much better to draw the electrodes in a logical order, then distribute the pin numbers to match. Ian Batty, Malvern East, Vic. Power Electronics series appreciated I want to thank Andrew Levido for his article series on Power Electronics (siliconchip.au/Series/452). He has a way of making it all seem easier. I always understood electronics through its formulas and their application. Norm Boundy, Melbourne, Vic. Amplifiers should have a VAS clamp diode I really cannot agree with Douglas Self’s opinion not to use a Baker clamp diode on the voltage amplification stage (VAS) in his Blameless amplifier design. It makes no logical sense. I think he said something like he was building linear amplifiers, not fuzz boxes, and dismissed them. But of course, when a linear amplifier is driven into clipping, its behaviour in this mode is important. Most of the time, until approaching clipping, the diode is significantly reverse-biased and really only represents a small, slightly non-linear capacitance of a few picofarads, if that. It is little different, in fact, from the reverse-biased collector-to-base junction capacitance of the VAS transistor itself. I doubt that the distortion from that would be anywhere near as detectable as that caused by the VAS transistor’s storage time, which is easily visible in oscilloscope recordings. As clipping is approaching, the diode will start to conduct in the forward direction, and one could argue there would be a trace of non-linearity there. But the thing is, the collector voltage of the VAS is already falling below its base voltage and approaching the saturation voltage; the system is right on clipping then anyway, and the idea of a perfectly reproduced waveform at that point is far from logical. It pays to recall a transistor amplifier’s behaviour at clipping; it is the higher-frequency harmonics that make it sound bad. Music has a very wide dynamic range, and even with the amplifier on half of the maximum listening volume, the peaks will be getting clipped. Multiple high-frequency harmonics are then introduced into the sound because of the VAS transistor’s storage time. It is not just about a distortion test at a level before clipping; it is about what happens with music as the source. With the diode added, the VAS transistor would never enter hard saturation, and the storage time concern would not be there to cause the ‘rail sticking’. It is the storage time that matters; it is not really about the cited transistor’s output capacitance, although transistors with low base-collector feedback capacitances and low-range collector output capacitances, like video output transistors, do tend to have shorter storage times. There is another trick to avoid the actually not-so-sinister 8 Silicon Chip Baker clamp, perhaps not commonly used, and that is if the single BJT stage is replaced by a Darlington pair. The Darlington arrangement prevents the output transistor of the pair from going into hard saturation on overdrive. This also drastically increases the open-loop gain. However, with the loop closed as it is, the AC gain would remain about the same. It would be a matter of trying that one out to see if instability or other issues occurred, but the simple solution is a schottky diode from the base to the collector of the VAS. Dr Hugo Holden, Buddina, Qld. Comment: we agree that there is no real disadvantage in adding the clamp diode. Note that our Ultra-LD Mk.3 & Mk.4 amplifiers, as well as the SC200, use the Darlington VAS approach and we found it worked very well, although we had to go to some effort to ensure stability. The SC200 also includes a BAV21 low-capacitance silicon clamp diode for good measure. Nano Pong trouble with HDMI adaptor I saw the discussion about difficulties running your Nano Pong project (August 2021; siliconchip.au/Article/14988) through a composite to HDMI converter in the June 26 issue (Ask Silicon Chip, page 104). From the research I have done so far in converting VGA to HDMI, there are strict synchronisation and pixel clock requirements. The minimum VGA resolution is 640×480 pixels and a 25.175MHz pixel clock. I have also found out that the converter has the same bandwidth limits as a normal TV. That means that a 7MHz pixel clock is OK, but the double-resolution from an Apple ][ using a 14MHz clock won’t work. The other requirement for any composite-to-HDMI converter is a clean sync signal. If the sync signal does not conform to the standard or is too weak, it cannot be locked onto by the converter and will result in a few possible display problems: • a rolling or jittering picture • tearing or skewed lines • no-signal errors • a blank screen From what the reader wrote, I believe the problem may be related to the sync signal. From reading the Nano Pong article, I cannot determine whether it complies with the PAL 625-line, 50Hz standard, nor the voltage levels for sync and video. The HDMI converter may have stricter tolerances compared to an old TV. The missing colour burst (as the Nano Pong has a monochrome output) should not affect video beyond colour display. The converter should simply revert to monochrome mode. It also does not care whether it is NTSC or PAL, as the converters are universal. Example converters are the Simplecom CM401 for analog composite video and the CM201 for VGA. In the Television Engineering Handbook by K. Blair Benson, Chapter 21 on page 21.75, it states the basic voltage percentages are: sync level 25±2.5%, video 75±2.5%, blanking level to black level 7.5±2.5%. The black level is higher than the blanking level. Both NTSC and PAL use the same voltage levels. I believe that the Nano Pong sync voltage level probably does not meet the standard. It is about 30.4%, which Australia's electronics magazine siliconchip.com.au SERIOUS STORAGE SOLUTIONS Unlock the full potential of your workspace with HAFCO’s range of storage solution... built to save space, boost productivity, and stand the test of time. PA C K A G E D E A L INDUSTRIAL STORAGE & TOOLING CABINET INDUSTRIAL MOBILE TOOLING CABINET WORKSTATION PROFESSIONAL SERIES MOBILE WORK STATION IBLE INCRED ITY! IL T A S VER LAST BUILT TO X TOOLBO MOBILE TION U SOL 1,199 (T773) $ 1,155 (T6890) $ 1,309 (K033) $ SAVE $88 OFF RRP PA C K A G E D E A L MOBILE STORAGE BUCKET SYSTEMS INDUSTRIAL MODULAR STORAGE RACK PARTS STORAGE BIN CABINET LE IN AVAILAB IZES LS SEVERA – 60 BOX FROM 20 NS OPTIO 649 (K210) $ 385 (S018) 616 (T7958) $ SYDNEY BRISBANE MELBOURNE PERTH ADELAIDE (02) 9890 9111 (07) 3715 2200 (03) 9212 4422 (08) 9373 9999 (08) 9373 9969 1/2 Windsor Rd, Northmead 625 Boundary Rd, Coopers Plains 4 Abbotts Rd, Dandenong 11 Valentine St, Kewdale 11/20 Cheltenham Pde, Woodville Specifications & prices are subject to change without notification and valid until 29-07-26 05_SIC290626 $ SAVE $136.40 OFF RRP FOR MORE PRODUCTS VISIT MACHINERYHOUSE.COM.AU/ SIC2606 exceeds the 25±2.5% standard. To get it within specification, increase the resistance of the 1kW sync resistor to 1.21.3kW. This should bring it into voltage specifications and may fix the reader’s problem. However, this will not fix any frequency-dependent problems, like sync pulse widths and position etc. They too need to be within the tolerances of the standard for the HDMI converter to lock on and convert each frame. I have found out that some converters do not support non-interlaced video, expecting the true interlaced composite video signal, including the serrated vertical sync pulse as per the standard. It will try to de-interlace the video, but it is not there and fails there. Of the Simplecom series, the ordinary CM401 apparently needs it, while the CM461 is specifically intended for old PCs like the Commodore 64, which produce progressive scan non-interlaced video. However, the CM461 needs a special cable and connector to separate power and signal from the combined connector. I cannot confirm my conclusions because I do not have a Nano Pong on hand, but hopefully, some of this information helps. Wolf-Dieter Kuenne, Bayswater, Vic. UHF transmitter supplier seems to have changed I am in the process of assembling the Secure Remote Mains Switch described in the July & August 2022 issues (siliconchip.au/Series/383). You mention using a Jaycar ZW3100 UHF ASK 433.92MHz transmitter. The pinout of this particular transmitter is not suitable; the GND and Vcc pins don’t go to the right pads on your PCB. Additionally, its overall thickness is 6.5mm, compared to the Altronics Z6900 which, according to the data sheet, is 5.7mm. Wayne Favier, Atherton, Qld. Comment: thanks for your information. Jaycar must have changed its supplier since the publication of the Secure Remote Mains Switch articles in 2022. The UHF transmitter module used in the project was a ZW3100 that we purchased from Jaycar, but it looks different from the photos you sent. Recent problems with TV reception Regarding Bruce Pierson’s comments on degraded freeto-air TV reception (March 2026 issue, Mailbag, page 10), I have similar problems with accessing Channel 7 here in Melbourne. However, the problems are usually only with CH7. In our case, the TV fires up with CH7 “not available” but after switching on the set-top box (a Panasonic), it works again. I had an “expert” look into it, and the conclusion was that the high-quality cabling I had installed had water damage. The cabling was replaced, but the problem returned and remains, such that the only way to restore on-air reception of CH7 is to reset my set-top box. Given that this does not happen with my Humax set-top box, it is somewhat intriguing. Having been involved in the installation of all major city transmitters back in the early 1970s and ongoing involvement in the supply of various interface and monitoring systems, all I can think of is that some organisations are trying to stuff more into the available bandwidth than is possible. SC Robert Forbes, Forest Hill, Vic. 10 Silicon Chip Australia's electronics magazine siliconchip.com.au SOFT ROBOTS by Dr David Maddison, VK3DSM Prof. Cecilia Laschi’s robotic octopus from the National University of Singapore. Source: Jennie Hills, The Science Museum, London Imagine a robot navigating through a disaster zone, squeezing through rubble like a worm to find trapped survivors, or a medical soft gripper handling human tissues or even squeezing through veins and arteries. This is beyond the capabilities of traditional robots, but within the emerging field of soft robotics. S oft robotics is still mostly confined to laboratory research, but there are emerging areas of commercial devices. Soft robotics is a blend of biology, materials science, engineering and AI to create flexible, adaptable machines that can mimic living organisms. Unlike conventional robots with fixed joints and stiff links, soft robots are built primarily from compliant, deformable materials like elastomers or elastomer-like materials (eg, silicones and hydrogels) that can stretch, bend, twist and squash. This flexibility enables safer interaction with humans, adaptability to unpredictable environments and the imitation of natural movement, from 12 Silicon Chip the muscular hydrostats of an octopus tentacle to the crawling of a caterpillar. Soft robotics might be used for just part of a traditional robot, such as a gripper at the end of a conventional robot arm (an ‘end effector’), to pick up delicate items like fruits and vegetables, or perhaps for legs, as with the robot turtle we will discuss. The emergence of soft robotics is driven partly by the demand for new and more versatile robots for applications not suitable for existing robots, such as delicate surgery; operating in restricted, unstructured environments like disaster zones; or crowded environments like factories. Soft robotics is being made possible with technologies like: Australia's electronics magazine • advanced 3D printing techniques that allow elastomers and other specialised materials to be printed • artificial intelligence and microfluidic ‘brains’ for control • novel materials, such as those with self-healing and stimuli-­ responsive properties • diverse actuation mechanisms (pneumatic, hydraulic, electrostatic and more) to provide powerful yet gentle motion • flexible (bendable) sensors and electronics This article explores the materials, mechanisms and biological inspirations behind soft robots, plus examines their growing applications across medicine, manufacturing, exploration siliconchip.com.au and more. We will look at the role of microfluidics in some designs, and investigate real-world commercial products, before addressing remaining challenges and glimpsing at possible future applications. Fluidics is a concept that will come up throughout this article. We published a detailed article on its principles in August 2019 (siliconchip.au/ Article/11762). Characteristics of soft robots Soft robots are constructed primarily from soft, highly deformable and compliant materials that enable them to bend, twist, stretch, and conform to complex shapes. This inherent softness and conformability give their movements a fluidity far more reminiscent of biological organisms than traditional rigid robots. Rather than relying solely on conventional electric motors and geared joints, as with rigid robots, soft robots frequently employ pneumatic or hydraulic actuators (inflating or pressurising fluid-filled chambers) to generate motion. Control can come from traditional embedded electronics and/ or AI, or in some designs, from microfluidic logic circuits for certain tasks like locomotion. While microfluidic logic circuits can provide basic rhythmic or sequential locomotion capabilities (eg, alternating leg motion like an insect or simple oscillating gaits), they cannot perform the complex, high-speed processing or adaptive decision-making possible with traditional CPUs, GPUs, NPUs and AI algorithms. Historical evolution and biological inspiration The first entirely soft and autonomous robot is generally regarded as the Octobot (2016), which will be discussed later. It is made of soft silicone gel and uses pneumatic actuation to control its limbs from a microfluidic logic circuit. As implied by its name, it is inspired by the octopus. Prior to the Octobot, there were various foundational developments, including smart and flexible materials, advanced fabrication techniques like 3D printing of soft materials, and new methods of mathematical modelling and control for compliant systems. Pneumatic actuators, as used in modern soft robots, trace their origins back to 1957 with the McKibben siliconchip.com.au artificial muscle (also known as the pneumatic artificial muscle or “air muscle”). These compliant actuators featured a rubber bladder inside a braided sleeve that contracted when pressurised, mimicking biological muscle contraction. That was one of the earliest uses of soft, deformable materials for actuation. The artificial muscle was initially developed for artificial limbs and orthotic devices to help paralysed patients grasp objects. Although industrial robots existed at the time (eg, Unimate, described in our May 2017 article; siliconchip.au/ Article/10641), McKibben’s innovation laid the foundation for the compliant pneumatic actuators that power many present-day soft robots. In the early 1990s, researchers including S. Shimachi and M. Matsumoto pioneered the use of silicone micro-actuators and soft compliant fingers for robotic manipulation. Their work modelled deformation, friction and grasping stability of deformable silicone fingertips. This marked one of the earliest systematic investigations into compliant, soft end-effectors, demonstrating improved adaptability, reduced object damage and enhanced force control compared to rigid fingers. This research paved the way for the modern soft robotics era, influencing later pneumatic soft grippers and compliant actuators. During the 1990s, robotics research also saw growing interest in bio-­ inspired designs, with projects such as Joseph Ayers’ lobster-like robots, which drew on animal locomotion for inspiration. However, these early systems remained rigid, using hard exoskeletons and conventional motors rather than compliant materials. Together, pneumatic artificial muscles, the systematic modelling of soft silicone actuators and the increasing emphasis on biological inspiration laid the conceptual and technical groundwork that would enable the emergence of modern soft robotics in the following decade. Materials Many different materials are used in the fabrication of soft robots. Elastomers and silicones are used for their high levels of flexibility, stretchability and formability. Shape memory materials are also used, which change shape Australia's electronics magazine in response to heat, light and electric fields. They may be polymers (plastics) or metals. Electroactive polymers deform when an electric field is applied. Hydrogels can be engineered with many desired properties, such as responsiveness to environmental parameters like pH, humidity, light and magnetic fields. A hydrogel is a polymer material that can retain large amounts of water, giving them a soft, flexible consistency similar to living tissue or jelly. Elastomers such as silicone (eg, PDMS & Ecoflex) and thermoplastic polyurethane (TPU) are commonly used in soft robotics, as are hydrogels. Shape memory polymers and metal alloys are ‘smart’ materials that can be deformed into a temporary shape and then, by the action of some stimulus, return to their original shape. For polymers, the stimulus may be heat, light or electricity; for alloys, it can be heat. Piezoelectric polymers can convert motion to electricity and vice versa. Self-healing materials are being researched for soft robotics, to enable robots to autonomously repair damage and extend their operational lifespan in challenging environments. These materials, typically based on polymers like silicones or hydrogels, incorporate reversible chemical bonds, embedded microcapsules or vascular networks containing healing agents. They activate upon crack formation and release monomers to polymerise or flow to seal breaches, a bit like self-sealing tanks on military aircraft or the platelets in our blood. When damage occurs, such as cuts, punctures or fatigue, the material can restore its mechanical properties. In soft robotics, self-healing enhances their durability for applications like medical devices (eg, catheters that can repair themselves inside the body), disaster-response robots operating in harsh conditions, or wearable exosuits subject to wear. Fabrication Soft robotic components are typically either moulded or 3D printed. Mould casting is a technique where liquid elastomers are poured into moulds, then cured and removed. Channels or chambers can be cast within the part for pneumatic or hydraulic actuation. Multiple parts July 2026  13 Fig.1: two flexible parts can be moulded, then dipped in adhesive, glued together and cured. Fig.2: a dielectric elastomer actuator changes its dimensions in response to an electric field. Original source: www.digikey.com/en/maker/projects/ diy-soft-robotics-dielectric-elastomerdot-actuator/5b77674365634d86b1f97 87fa4501c9b Fig.3: a dielectric elastomer actuator configured so the tip bends as the electric field is cycled. Source: www. digikey.com/en/maker/projects/diysoft-robotics-dielectric-elastomer-dotactuator/5b77674365634d86b1f9787f a4501c9b can be adhered together – see Fig.1. This technique was used for the Octobot and is commonly used today for robot grippers and many other components. Mould casting is used for PneuNets (pneumatic networks), a very common type of soft robotic pneumatic actuator, described later. 3D printing includes a variety of techniques such as FDM/FFF (fused deposition modelling and fused filament fabrication) and DIW (direct ink writing) to extrude materials like TPU, silicones and hydrogels. Vat photopolymerisation (including SLA [stereolithography] and DLP [digital light processing]) is also used to cure liquid resins with light. Material jetting can be used to deposit multiple materials simultaneously, including soft and rigid materials, possibly even embedding electronics. “4D printing” is an emerging technique in which a material changes shape after fabrication. Such a device could be fabricated using the moulding or 3D printing techniques described above. Shape change is brought about under the influence of heat, light, moisture etc. This technique enables self-folding or self-­ assembling robots, often biomedical or miniature types. Microfluidic controls for soft robots are fabricated using a variety of techniques. These include subtractive manufacturing (where materials are removed to create microchannels), moulding, micromachining and 3D printing. Actuation mechanisms As typical motors are usually incompatible with soft robots, actuators to generate movement typically rely on methods involving electrical or fluidic activation and sometimes magnetic activation. Electroactive polymer actuators use various combinations of electrodes, insulating polymers, conducting polymers and piezoelectric polymers to achieve movement. Electroactive polymer actuators include: Dielectric elastomer actuators consist of a thin elastomer film sandwiched between two compliant (stretchable) electrodes. When a high voltage is applied across the electrodes, the resulting electric field compresses the elastomer and causes it to expand in area, producing shape change and mechanical work (see Fig.2). The elastomer film is typically tens to hundreds of micrometres (µm) thick, and operating voltages range from tens of volts to several kilovolts. In practical devices, multiple layers are stacked to achieve greater force and stroke. A common configuration uses two elastomer layers with a shared ground electrode in the middle and separate high-voltage electrodes on the outer sides, as in Fig.3. This arrangement mimics antagonistic muscle pairs: applying a voltage to one layer causes it to expand (contracting the opposing layer), enabling bidirectional bending or linear motion. Examples of dielectric elastomer grippers are shown in Fig.4. An article on how to make your own device can be found at siliconchip.au/ link/acao Liquid crystal elastomer actuators are advanced stimuli-responsive materials used in soft robotics for their ability to undergo large, reversible shape changes, often changing dimension by 50% or more when triggered by external stimuli such as heat, light or electric fields. These materials combine the ordered molecular alignment of liquid crystals with the elasticity of polymer networks, allowing programmed molecular orientation during fabrication Ground Vchuck Vactuator P Electrode terminals DEA units Vchuck = 0 Fig.4: examples of dielectric elastomer grippers and structure. Source: www.mdpi.com/2076-3417/10/2/640 14 Silicon Chip Australia's electronics magazine siliconchip.com.au (eg, via 3D printing or alignment techniques) to dictate precise deformation patterns like bending, twisting, or contracting. Challenges remain in response speed, force output and durability. Fig.5 shows a variety of liquid crystal polymers and the shape transition of liquid crystal elastomers. The liquid crystal main chain polymers (LCP) are shown, then a liquid crystal polymer network (LCN), then a liquid crystal elastomer (LCE). Only LCEs can perform a shape change. The difference between LCNs and LCEs is that LCNs have many more cross-links between the polymer chains (too many to allow a shape change). Ionic polymer actuators bend or deform when a voltage is applied, mimicking muscles by moving ions within a polymer membrane, causing swelling in one area and shrinkage in another, resulting in motion. Key types are ionic polymer metal composites and ionic polymer gels. They work through the application of a low voltage, causing ions to migrate to the oppositely charged electrode, as shown in Fig.6. Piezoelectric polymers are just like ceramic piezoelectric crystals, such as quartz. Motion is converted into electrical energy or vice versa. The main difference is that a polymer is used rather than a ceramic. PVDF (polyvinylidene fluoride) is a common piezoelectric polymer. Not only can such polymers be used for actuators, they can be used for sensors as well. A single sheet of polymer will not generate enough motion, so typically they are assembled in two layers. When an electric field is applied, one layer shrinks and the other expands, causing motion, as shown in Fig.7. Conducting polymers (see our November 2015 article on the topic) are polymers such as polypyrrole that are intrinsically electrically conducting and don’t rely on metal or carbon fillers to render them conductive. Ions can be moved into and out of them in an appropriate solution, causing them to change shape, as in Fig.8. Hydraulic actuators use water or oil to inflate a bladder or similar structure. Magnetic actuators generate motion via an external magnetic field interacting with magnetic materials inside the robot. Photoresponsive actuators react to light by changing shape, stiffness siliconchip.com.au Liquid crystal mainchain polymers (LCPs) (A) Liquid crystal polymer networks (LCNs) Liquid crystal phase (B) Liquid crystal elastomers (LCEs) Isotropic phase Cooling Heating Fig.5: liquid crystal polymers. LCEs can change shape upon heating, cooling or some other stimulus. Source: https://encyclopedia.pub/entry/history/ show/60582 Fig.6: the operation of ionic polymer actuators. Original Source: www.mdpi.com/20734360/17/6/746#polymers-17-00746-f002 Fig.7: an activated PVDF bimorph showing motion from the vertical position. Source: https://physics. montana.edu/eam/polymers/ bimorphs.html Fig.8: a conducting polymer (polypyrrole) actuator with motion as the voltage is switched. Australia's electronics magazine July 2026  15 or volume. These include materials such as: Liquid-crystal elastomers, which can change length by up to 50% when exposed to light (mentioned above) Polymers containing photosensitive compounds that bend or twist on light exposure Photothermal composites, which contain layers of materials with different properties in which the structure bends when exposed to light Hydrogels with photosensitive compounds that change stiffness when exposed to light Pneumatic actuators are probably the most common type of actuators for soft robots. They use compressed air, another gas or the decomposition of a fuel like hydrogen peroxide to generate gas. A PneuNet is an example of such an actuator. It typically has two layers or areas; one that is extensible with an internal air chamber, and another inextensible layer or area. When the extensible chamber is inflated, the assembly bends, constrained by the inextensible layer or area – see Fig.9. They can be fabricated by casting in 3D-printed moulds. Thermally responsive actuators use shape memory alloys, polymers or thermally responsive hydrogels that change shape in response to heat. Sensors Sensors for soft robots are chosen for their ability to maintain compliance and stretchability so they can be integrated into soft, deformable bodies. Examples include: Stretch/strain sensors are thin, stretchable films or fibres that change electrical resistance or capacitance when deformed. Materials include carbon black in elastomers, or lowmelting-point liquid metal alloys like EGaIn (liquid eutectic gallium-­ indium) in microchannels. Applications may include force estimation with grippers and soft exosuit strain monitoring. Soft pressure/tactile sensors measure contact pressure or distributed force via changes in resistance, capacitance or optical properties. Techniques include piezoresistive (conductive foam/rubber), capacitive (elastomer layers with flexible electrodes) or optical (light intensity change through deformable waveguides). Applications include gentle grasping of fragile objects (eg, fruit, eggs), human-robot safe interaction and texture discrimination. A commercial example is the grippers from Soft Robotics Inc, which have embedded soft tactile sensors for slip detection. Embedded optical fibre sensors detect strain, curvature or temperature by changes in light wavelength or intensity. Applications include surgical catheters or soft robot arms/ tentacles, and many other snake-like soft manipulators for minimally invasive surgery. Soft magnetic sensors use Hall-­effect Fig.9: a variety of configurations of PneuNet type actuators to give different shapes or motions. Source: https://elveflow. com/microfluidicreviews/soft-robot/ sensors or magnetometers to detect changes in magnetic fields from embedded soft magnets or ferrofluids. They can be used for curvature/angle sensing in pneumatic actuators or soft exosuits for joint angle measurement. Ionic/electroactive polymer sensors use ionic polymer-metal composites (IPMC) or dielectric elastomers to generate a voltage/current when bent or stretched (self-sensing). Applications include self-sensing actuators (one material acts as both the actuator and sensor). Emerging/bio-inspired sensors such as hydrogel-based chemo-­sensors to detect pH, temperature or specific chemicals (eg, for environmental monitoring); bio-hybrid sensors such as living cells (eg, muscle cells) integrated with soft robots for chemical or biological sensing; stretchable cameras (miniature soft cameras for visual feedback, eg, in medical or underwater soft robots). These sensors are often integrated directly into the soft elastomer body during fabrication (3D printing, moulding or embedding), making soft robots sensor-filled from the inside out. Mathematical modelling Because of the ability to continually deform with infinite degrees of freedom, soft robots need different control and modelling strategies compared to traditional rigid-body robots. New mathematical models have been developed based on “Cosserat rod theory” to predict the complex non-linear behaviour of such robots (see Fig.10). Traditional and finite element analysis are also used. Cosserot rod theory is a mathematical framework that applies to soft, deformable slender structures like rubbery tubes to accurately model behaviours like bending, extension, shear and twisting. Normal structural models cannot account for these Fig.10: a variety of soft robot elements that are modelled with Cosserat rod theory. They would be difficult or impossible to model using other methods. Source: www. researchgate.net/publication/383153694 16 Silicon Chip Australia's electronics magazine siliconchip.com.au Soft controller Fuel reservoirs Reaction chambers Actuators Vent orifices Fuel inlets Upstream check valves Pinch Downstream valves check valves Outlets Figs.12 & 13: Octobot, the first fully soft, electronics-free, autonomous robot (left) and the microfluidic device that controls it (right). Sources: https://wyss.harvard.edu/news/the-first-autonomous-entirely-soft-robot/ | https://newatlas.com/ chemical-power-soft-robot-autnomous-harvard/45073/ properties, have difficulty with it or requiring excessive computational resources. Soft robot control Soft robots can be controlled by various means, such as embedded traditional control using a CPU, artificial intelligence (AI) and machine learning (ML). AI/ML can also be used to process data from sensors to manage the control of soft robots, enabling autonomous and adaptive behaviours. Integrated sensing and feedback loops using soft, stretchable sensors enable soft robots to perceive their own shape by detecting pressure and touch. Microfluidics is also used to produce simple, repetitive motions, but not complex decision making and control as with a CPU and AI. Traditional control systems are well known, so we will just discuss microfluidic controls in detail. Here are some examples: Conventional computing platforms, such as Arduinos, can be converted into stretchable, compliant controllers by embedding them in flexible carriers with highly stretchable conductors, as demonstrated by researchers at the Yale University Faboratory (see Fig.11). The team created stretchable versions of Arduino Pro Mini boards that function at over 300% strain, embedding them directly into soft robots for locomotion control and wearables for motion sensing. The conductors are made from biphasic gallium-indium alloys, particularly oxidised gallium-indium (OGaIn), a foam of amorphous gallium oxide particles mixed with EGaIn. It was patterned on or within silicone substrates for high conductivity, extreme stretchability and reliable interfaces with rigid components. For more on this, see the video at https://youtu.be/VgNwUPpOY9A We also looked at flexible electronics in our November 2015 issue. Microfluidics; the main purpose of microfluidics in soft robotics is to enable autonomous, electronics-free (or minimal-electronics) control of soft robotic systems, particularly for untethered and lightweight designs. Microfluidics involves routing small volumes of fluids (gases or liquids) through tiny embedded channels and valves within the soft robot’s body. These channels form fluidic logic circuits, analogous to electronic circuits that can perform basic computation (eg, AND/OR/NOT gates, oscillators, timers) using pressure differences instead of electricity. Fig.11: an Arduino microcontroller module with the components mounted on a stretchable substrate with flexible conductors. Source: https://engineering.yale. edu/news-and-events/news/flexible-electronics-stretching-possibilities-softrobots siliconchip.com.au Australia's electronics magazine Microfluidic controllers contain components like pumps, fluid logic gates analogous to transistors, oscillators, shift registers, multiplexers and fluidic amplifiers. Microfluidics is good for generating rhythmic or sequential motion, reducing weight and complexity. It also enables some degree of untethered autonomy because the fluidic controllers can be powered by onboard chemical reactions or stored pressurised gas, allowing operation in environments where electronics would fail (eg, underwater, in MRI machines or explosive areas). Also, distributed control is possible as pressure signals propagate through the body like nerves, enabling coordinated multi-limb movement without a central processor. Microfluidic logic excels at simple, repetitive tasks (eg, walking gaits, pulsing, or basic sensing feedback) but not complex tasks. It is therefore most valuable in minimalist untethered robots or as a low-level controller complemented by higher-level electronics in hybrid designs. It serves as the brain and nervous system for the simplest fully soft, autonomous robots, trading computational power for lightness, robustness and independence from external power/control tethers. An example of a microfluidic controller is the one used in the Octobot (Fig.12; described in more detail later). Its controller (Fig.13) is basically an oscillator circuit. It was created using soft lithography, in which PDMS (polydimethylsiloxane) is poured into moulds with etched channels. The polymer was cured and solidified, with multiple layers being precisely aligned and bonded to form July 2026  17 Fig.14: in a Quake valve, the control air can deform the flexible membrane and block the flow of the process fluid. Fig.15: a pneumatic ring oscillator. Source: https://is.mpg.de/ publications/preston19-scir-oscillator Fig.16: the Squishy Robotics mobile robot. Source: https://squishy-robotics.com/ research-and-development/ the fluidic valves, channels and oscillator network. The Quake valve, invented by Stephen Quake and collaborators in 2000, is a widely used pneumatic microfluidic valve in soft robotics and lab-ona-chip devices. It typically consists of multi-layer soft elastomer structures with two crossing channels: a flow channel (for the main fluid/gas) and a perpendicular control channel (for pressurised air) – see Fig.14. When pressure is applied to the control channel, a thin elastomeric membrane between the channels deflects and pinches off the flow channel, closing the valve (acting like a transistor for fluid flow). Releasing pressure opens it again. This design can be used to make fluidic logic gates (AND, OR, NOT, oscillators etc) by combining multiple valves, enabling complex control without electronics, a key for untethered soft robots It is scalable (thousands can be integrated on one chip), has a fast response time and is compatible with pneumatic actuators. A disadvantage is that it is limited to low flow rates, so it is often better for microscale or control signals rather than high-force actuation. Microfluidic pumps are used to pump fluids around the microfluidic chip and through devices like the Quake valve shown in Fig.14. Each tube is activated in sequence to push fluid along in a peristaltic fashion. Microfluidic ring oscillators are pneumatic (or hydraulic) devices that offer a clever way to generate periodic motion in soft robots without any electronic control – see Fig.15. A ring oscillator consists of an odd number (typically three or five) of pneumatic inverters, which are fluidic valves made from deformable elastomeric membranes that function like inverting logic gates with hysteresis, similar to a Schmitt trigger inverter. When connected in a closed loop, pressure builds sequentially in each stage, causing the inverters to switch between on and off states. This creates a self-sustaining oscillation that can drive rhythmic movements, such as the coordinated leg motion in multi-legged crawling robots. These fully soft oscillators enable truly untethered operation, as demonstrated in some autonomous soft walkers such as the UC San Diego 18 Silicon Chip Australia's electronics magazine soft-legged walking robot described later and shown in Fig.19. Advantages of soft robots The advantages include: • Superior safety in human-robot interaction due to compliant, deformable bodies, making them inherently safe for close collaboration with people (eg, in factories or assistive wearables). • Adaptability to unstructured environments, being able to squeeze through narrow gaps, navigate irregular terrain or handle objects of varying shapes and fragility without precise programming or complex sensors. • Biomimicry and natural movement; some soft robots replicate biological systems (octopus tentacles, elephant trunks, worms), enabling smooth, energy-efficient, and fluid motion that rigid robots struggle to achieve. • Robustness to impacts and overload; compliant materials distribute forces, allowing them to survive drops, collisions or compression (eg, Squishy Robotics’ airdropped platforms). • They are lightweight and potentially low-cost because many use inexpensive elastomers and simple siliconchip.com.au fabrication methods (3D printing, moulding), reducing overall system weight and enabling untethered operation in some cases. issue (siliconchip.au/Article/17782). It produced the ReStore Exo-Suit, designed to promote the redevelopment of correct gait patterns in people who have suffered a stroke or similar Disadvantages of soft robots neurological injury. The disadvantages include: It is a lightweight, cable-driven soft • Limited force and precision; soft robotic device that provides timed materials generally produce lower assistive forces to a partially paralysed forces and less precise positioning ankle and is commercially used in compared to rigid actuators and metal rehabilitation clinics. It uses Bowden linkages, making them unsuitable for cables (like bicycle brake cables) heavy lifting or high-accuracy tasks. pulled by motors in a waist-mounted • Durability and fatigue; repeated unit. For more details, see the video at deformation causes material fatigue, https://youtu.be/1pC3fUGOdFw wear, tearing or degradation over time, Somnox (https://somnox.com) is especially under high strains or cyclic a pillow-like robot that simulates loading. breathing patterns to help with sleep • Complex modelling and control; disorders. It is pneumatically opertheir nonlinear, infinite-degree-of-­ ated. According to the manufacturer, freedom deformation makes accurate “Somnox detects and matches your modelling, simulation and precise breathing rate and rhythm; then, the control difficult. rhythm gradually slows to a tranquil, • Power and actuation challenges; sleep-inducing pace”. many rely on bulky external compressors (for pneumatics/hydraulics) or Rescue robots high voltages (for dielectric elastoSquishy Robotics (https:// mers), limiting true untethered perfor- squishy-robotics.com) produces both mance. Onboard chemical or battery a commercial stationary soft robot and solutions are still emerging. a developmental mobile one, shown • Manufacturing and scalabil- in Fig.16. The stationary robots can be ity; while prototyping is relatively deployed from aircraft to be dropped easy, producing durable, repeatable, into disaster areas from 300m and can high-performance soft robots at scale carry a variety of payloads such as gas remains challenging and expensive sensors for CO, H2S, lower explosive compared to conventional robotics. limit of gas and O2, cameras for 360° video, GPS and mesh networking. Commercial applications The structural concept is tensegrity (tension and integrity), which uses a network of rigid struts in compression and elastic cables in tension to create compliant, lightweight robots that are deformable and can absorb impacts such as being dropped from an aircraft. The developmental mobile robot moves via the concept of paired cable actuators, where one cable pulls and the other pushes like pairs of muscles work in opposition. In these robots, there are multiple sets of paired actuators acting in a coordinated way to propel the robot, as shown in the video at https://youtu.be/oDaLb63iCPI Soft grippers and food handling Schmalz Group (www.schmalz. com) produces the mGrip range of pneumatically powered soft robotic grippers for picking up fragile items like produce, baked goods and poultry without damage (see Fig.17). The grippers are activated by air channels inside the ‘fingers’. They are an example of a soft robotic component attached to a traditional rigid robot. The Festo HPSX (https://press.festo. com/en/node/5135) silicone gripper is a soft robot component for high-speed picking up of delicate products such as various foods – see Fig.18. It can work safely in human-robot collaborative environments with low risk to humans. It is pneumatically operated. Soft robots excel where safety, adaptability and gentle interaction are priorities, such as in medical devices, food handling or exploration. Still, they lag behind rigid robots in strength, precision and long-term reliability. Most practical applications today use hybrid approaches, combining soft elements for interaction with rigid frameworks for power and control. Here are some examples. Healthcare and emotional support Moflin from Casio (www.casio.com/ us/moflin) is an AI fluffy companion for emotional support that develops personality traits. It is regarded as a soft robot by some, but while its exterior is soft and fluffy, its interior is standard mechatronics. ReWalk Robotics (now Lifeward) produced the walking assistance exoskeleton described in our article on prosthetic limbs in the March 2025 siliconchip.com.au Fig.17: the mGrip soft robotic gripper. Source: Schmalz – siliconchip.au/ link/acaw Australia's electronics magazine Fig.18: the Festo HPSX silicone soft robot gripper. July 2026  19 iCobots (https://icobots.com) is an Israeli company providing plug-andplay soft robotic grippers that integrate seamlessly with existing industrial robots and cobots. A cobot is a collaborative robot designed to work alongside people rather than replace them. These grippers combine the speed of automation with the gentle, adaptive touch of human hands, making them ideal for handling delicate items such as eggs, fruit, chocolate and other fragile produce without damage or the need for complex vision systems. Wearable and assistive devices NEO from 1X Technologies (www. 1x.tech/neo) is a domestic humanoid robot with a soft-bodied design, launched in late 2025 and now available for pre-order. It is intended for everyday household tasks and uses biology-inspired tendon-driven actuation (with high-torque-density motors) for smooth, compliant and safe interactions with people. Conceptual soft robots (experimental) Apart from commercial soft robots, many have been designed or are being researched as in the following: • In 2016, the Wyss Institute at Harvard University unveiled the first entirely soft autonomous robot, called Octobot. It was electronics-free, mostly 3D printed and used microfluidic logic oscillator circuits for control. It used hydrogen peroxide as fuel, which generated gas to drive the robot – see Fig.12 and the video at https://vimeo. com/179510230 • Stanford University has developed an “isoperimetric” soft robot in the form of an inflatable tube truss with relocatable joints which enable it to change shape due to changes in the length of individual truss members and move or perform other tasks (siliconchip.au/link/acap). The joints are moved by roller modules that create new joints by pinching the tube at different locations. See the video at https://youtu.be/XqgbLb8m77U • Researchers at UC San Diego have developed an electronics-free softlegged walking robot (Fig.19). It is powered by pressurised air and has no electronics. Its movement is controlled by pneumatic ring oscillator fluidic control circuits to generate rhythmic movement, similar to animals. The robot was also equipped with simple sensors in the form of bubbles of fluid at the end of the legs which, when depressed, flip a valve and cause the robot to change direction in response to environmental interactions. The biological inspiration for this machine comes from the African sideneck turtle. For more, see the video at https://youtu.be/bnT6BBkDYlc • Researchers at the University of California, San Diego, have developed an electronics-free autonomous walking robot with an embedded pneumatic oscillating control circuit. After 3D-printing the six-legged robot, shown in Figs.20 & 21, is ready to operate as soon as a gas supply (CO2 cylinder, tube and pressure regulator) is added. It uses a 3D-printable four-phase bistable oscillating valve, capable of generating coordinated motion of multiple limbs from a steady source of gas. Each of the six legs has four chambers, each of which generates one of up, down, forward or backward motion. See the videos at https://youtu.be/ f8hTK7AabM8 and https://youtu.be/ PDoiguTdLXs • SoFi is a soft robotic fish developed by MIT’s CSAIL in 2018. It is a remote-controlled underwater robot equipped with a camera to observe marine life without disturbing it. A diver directs it from a console (using acoustic signals) while hydraulic fluidic actuators in the tail mimic natural fish swimming. See Fig.22 and the videos at https://youtu.be/BSA_zb1ajes and https://youtu.be/Dy5ZETdaC9k • In recent years, Chinese researchers have pioneered soft robots for exploring the Mariana Trench, the ocean’s deepest point. Unlike traditional submersibles made of expensive Fig.20: a six-legged walking robot that needs no electronics; just a CO2 canister. Source: UC San Diego – siliconchip.au/link/acax Fig.19: the soft-legged walking robot from UC San Diego. Source: https:// newatlas.com/robotics/air-powered-robot-no-electronics-turtle/ 20 Silicon Chip Australia's electronics magazine Fig.21: the embedded pneumatic oscillating control circuit of the robot shown in Fig.20. siliconchip.com.au Fig.22: a SoFi robotic fish. Source: MIT News – siliconchip.au/link/acay Fig.23: Zhejiang University’s 2021 robot compared to a snailfish. Source: www. zju.edu.cn/english/2021/0317/c65148a2268191/page.psp hard metal shells to resist the extreme pressures at depth, these robots are soft, and external pressure is distributed evenly throughout them, just as in the fish they mimic. A landmark 2021 design from Zhejiang University mimicked the hadal snailfish, using a silicone body (22cm long with a 28cm fin span) and dielectric elastomer actuators for a flapping motion. It reached a depth of 10,900m, a world record for a soft robot – see Fig.23. Building on this, a 2025 robot from Beihang University (inspired by batfish locomotion) reached 10,666m, enduring 1100 bar. This larger version (50cm long and weighing 2.7 kg) uses shape-memory alloy actuators that oscillate with periodic heating for multimodal movement of swimming, gliding and crawling across the seafloor. • The DARPA ChemBot (Chemical Robots) program was a research initiative launched around 2007-2008 to develop soft, flexible, shape-shifting robots capable of squeezing through tiny openings (smaller than their normal size), reconstituting their shape and regaining function on the other side, to perform tasks like reconnaissance or payload delivery in denied/ hostile environments. The research program finished in 2011-2012 with no further announcements. • The Amphibious Robotic Turtle (ART), developed by Yale University researchers (Figs.24 & 25) is a bio-­ inspired soft robotic platform with a solid body but soft robotic legs that employs ‘adaptive morphogenesis’ to dynamically adapt its limbs for multi-environment locomotion. Its cylindrical legs can morph into flattened flippers for efficient swimming in water, mimicking sea turtles while reverting to load-bearing legs for land travel like tortoises. This transformation takes 1-2 minutes and is achieved using a thermally responsive polymer composite that softens when heated (via embedded heaters) and stiffens when cooled to hold the new shape. An internal soft pneumatic ‘muscle’ (balloon-like structure) inflates or deflates to drive the shape change during the malleable phase, enabling seamless transitions between terrestrial gaits (creep/crawl) and aquatic propulsion (flapping/paddling). • A survivable amputation of a body part to escape danger is a survival strategy used by certain lower animals like lizards, starfish and crabs. A soft robot has been developed at Yale University to do the same thing. For example, if the leg of a search and rescue robot gets trapped by falling debris, a built-in heating element can melt it away. See the video at https://youtu. be/qPd9x9-bALo • Harvard University’s Whitesides Research Group has developed the “arthrobot”, with an exoskeleton constructed from thin polymeric tubes. It also has pneumatic joints modelled after the hydrostatic joints of spiders to provide actuation and mechanical compliance to external forces. An inflatable elastomeric tube extends a Figs.24 & 25: an amphibious robotic turtle. Source: https://yaledailynews.com/blog/2022/10/25/yale-led-team-developsshape-shifting-turtle-robot/ siliconchip.com.au Australia's electronics magazine July 2026  21 Fig.26: Harvard University’s Whitesides Research Group arthrobot. Source: www.gmwgroup.harvard. edu/soft-robotics limb while an opposing elastic tendon retracts it – see Fig.26. Experimental grippers • The Festo Bionic Learning Network Octopus Gripper uses pneumatic tentacles and vacuum suction to hold objects of any shape. It is currently not a commercial product, but aspects of its technology have been incorporated into other Festo products. See Fig.27 and the video at https://youtu. be/w1zU7FNKm_w • The University of California, San Diego (UCSD) has developed a 3D-printed (in one print) gripper that has an embedded microfluidic controller and needs no electronics to operate. When the gripper is moved horizontally, it drops the object. See Fig.28: the UCSD no-electronics gripper. Source: Iguana Robot – siliconchip.au/link/acaz 22 Silicon Chip Fig.27: the Festo Octopus Gripper, also known as the TentacleGripper. Source: www.festo.com/gb/en/e/about-festo/research-and-development/bioniclearning-network/bionic-grippers-and-soft-robots/tentaclegripper-id_33321/ Fig.28 and the video at https://youtu. be/A5mpy3X1dcc • A granular jammer is a soft robotics gripper concept where grains or grain-like materials are placed inside a membrane, such as a balloon. It is placed around an irregularly shaped object and then a vacuum is applied to tighten the grip. The object can then be moved – see Fig.29. • The Jamming Donut is a universal soft-robotics gripper designed by Australia’s CSIRO. This doughnut-shaped gripper can grab round objects like doorknobs. • A team at the University of North Carolina at Chapel Hill has developed soft robots made of two layers, one simulating skin and the other muscle, that can autonomously detect and respond to different physiological stimuli. The robot’s base layer is made from a thermally responsive hydrogel that can contract and relax like muscle, allowing the robot to bend. The other layer is an electronic ‘skin’ made of another soft polymer, which can host a variety of sensors or stimulators. Such sensors can detect acidity, electrical activity, mechanical strain and temperature; mini electrodes could stimulate tissue, while electrical heaters could trigger the robot’s hydrogel ‘muscle’ layer to contract – see Fig.30. Onboard electronics allow for wireless power and data transmission. Fig.29: a granular jammer or “jamming gripper”. Source: www. creativemachineslab.com/jamminggripper.html Fig.30: sensors from an experimental soft robot. Source: www.nibib.nih.gov/ news-events/newsroom/taking-cuesnature-medical-soft-robots-get-smart Experimental medical applications Australia's electronics magazine siliconchip.com.au (A) (B) (C) Fig.33: the concept of Vine Robot’s movement. It lengthens and doesn’t slide, thus it has no problem getting through small openings. Fig.34: the principle of steering and extension of a vine robot. (a) Air pressure is applied to the core. (b) the pressure causes extension. (c) differential pressure is used for steering. Source: www.researchgate.net/publication/368389948 • Harvard University’s Wyss Institute has developed a soft-robotic sleeve that uses pneumatic actuators and wraps around a human heart to assist its beating. As there is no direct contact with blood, the patient does not have to take blood thinners, as with conventional artificial hearts or ventricular assist devices. The device is currently in advanced preclinical testing stages – see Fig.31. • Fig.32 shows a soft prosthetic robotic hand for amputees or service robots. It has the advantage of feeling more like a normal human hand. It was developed by Rob Scharff from the Delft University of Technology. • Researchers at AMOLF, a leading Dutch institute for fundamental physics and soft matter, have developed a remarkable soft robotic artificial heart prototype that demonstrates autonomous beating with minimal electronics. Driven entirely by pneumatic pressure from an external pump, the device uses advanced soft biomaterials and clever fluidic logic to produce a repetitive heartbeat. A key innovation is a passive soft valve inspired by the sputtering effect when squeezing an almost-empty plastic sauce bottle. As pressure builds, a soft tube buckles and rapidly opens/ closes, creating self-oscillating flow that drives rhythmic contraction of the heart without electronic controllers. • The DARPA Soft Exosuit is a wearable soft robot developed at the Wyss Institute at Harvard University. It works with the body’s muscles to reduce fatigue and assist movement. Cables and motors are used to apply forces to hips and ankles, mimicking tendons and muscles. It can be used in applications such as assisting stroke patients or soldiers. See the video at https://youtu.be/aeDm5yFYt10 Experimental soft robots that mimic plant tendrils These robots grow and navigate by extruding material or inflating tubes. Examples include Vine Robots and the FiloRobot. Vine Robots from Stanford University and UCSB, Dr Elliot Hawkes (www.vinerobots.org/about/) use soft pressurised polyethylene tubes that evert (turn inside out) to form a body that can navigate numerous types of obstacles, even a field of pointed nails or glued-together boards – see Figs.33, 34, 35 & 36. It does not slide; the body extends and lengthens from the tip. That means there is no friction between Fig.31: the Harvard cardiac assistance device. It’s a sleeve that wraps around the heart. Source: https://seas. harvard.edu/news/2017/01/soft-robot-helpsheart-beat Fig.32: a soft robotic prosthetic had for amputees or service robots. Source: https://elveflow.com/ microfluidic-reviews/soft-robot/ siliconchip.com.au Australia's electronics magazine July 2026  23 – see Fig.38 and the video at https:// youtu.be/e1mOac3wRsw Experimental photoresponsive soft robots Fig.35. Vine Robot navigates through a hole. Source: ExtremeTech – siliconchip.au/link/acb0 Fig.36: Vine Robot navigates a maze. Source: www.science.org/doi/10.1126/ scirobotics.aan3028 the external body of the robot and the surfaces it contacts. It can reach up to 72m from its base. Fig.34 shows the principle of extending and steering a Vine Robot. In (a), air pressure is applied to the core. In (b), the air pressure causes extension due to eversion of the tube. The outside of the tube doesn’t move with respect to the surface it contacts. In (c), the end is steered by varying the air pressure by applying air pressure to one or two of the serial pneumatic actuator muscles (sPAM) mounted around the robot’s circumference at the tip of the robot. Guidance is through the use of an inertial measurement unit (IMU) at the tip and a shaft encoder at the base to sense the amount of extension. There is also a camera at the tip. For more information, see the video at https:// youtu.be/q2Q-taHAo7Q The Vine Robot is not commercially available, but has inspired spinoffs using the eversion technology and a robotic gripper for industrial and elder care used to assist in lifting objects or people from beds; see Fig.37. Systems are also being tested for the inspection of commercial pipe networks. There are instructions to build your own Vine Robot at www.vinerobots. org/build-one Miniaturised 1.8mm-diameter versions are also being developed for non-invasive surgeries and intubation – see https://pmc.ncbi.nlm.nih. gov/articles/PMC12370164/ Filobot from the Istituto Italiano di Tecnologia (IIT) is an experimental robot and is not strictly a soft robot, but it bears some resemblance to the Vine Robot. It prints its own stem by 3D-printing its own body from the tip as it ‘grows’ Fig.37: a Vine Robot inspired gripper from MIT. Source: https://news.mit. edu/2025/vine-inspired-robotic-gripper-gently-lifts-heavy-and-fragileobjects-1210 24 Silicon Chip Australia's electronics magazine The Max Planck Institute in Germany has developed untethered soft micro-robots that walk or roll toward/ away from light sources (phototaxis). The Chinese Academy of Sciences has developed near-infrared-driven soft grippers for remote manipulation in confined spaces. Conclusion Soft robots represent a shift toward more organic, resilient machines, with the potential to revolutionise fields from healthcare to exploration, although challenges remain. More videos If you’re interested in seeing more about soft robots, then check out these videos below: “DIY a Food-Grade Soft Gripper for Your Delta X S Robot for Just... $10”: https://youtu.be/Eo3y-UqJ100 “DIY Soft Robotic Tentacle”: https://youtu.be/gPYjo-W2ctU “Can You 3D Print a Robot’s Brain Out of Air?”: https://youtu.be/Cn7jC6YamGE “I Printed a Microchip That Runs on Air — A Nervous System for Squishy Robots!’: https://youtu.be/QJdBp5dGrww SC Fig.38: FiloBot climbs a tree. Credit: Del Dottore et al., Sci. Robot. 9, eadi5908 (2024) siliconchip.com.au Subscribe to Pinball M The VERY BEST DIY Projects ! achine Customisable and built from scratch Comfort Indicator Measures temperature, humidi Australia’s top electronics magazine Analog Computers Part 2: modern-day examples and 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. If you have an active subscription you receive 10% OFF orders from our Online Shop (siliconchip.com.au/Shop/)* New Zealand Australia * does not include the cost of postage Rest of World JUNE 2026 ty and dew point 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! how they’re used in neural networ ks ISSN 1030-2662 06 9 771030 266001 $15 00* NZ $15 90 INC GST INC GST Published in Silicon Chip Simple USB Power Monitor; June 2026 Human Comfort Indicator; June 2026 Simple LC Meter; May 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 LOOKING TO REFRESH YOUR WORKBENCH? We’ve got the essentials to power up your next project, at the best value. PERFECT FOR COMPACT WORKSPACES $ ADJUSTABLE ARM IDEAL FOR SOLDERING, PLASTIC CUTTING, HEAT SHRINKING, ETC. ONLY 5995 . QM3552 FOR A BETTER VIEW Illuminated Desktop Magnifier • 100mm 3-dioptre glass lens • 30 bright LEDs • Mains powered QM3552 SOLDER ANYTHING, ANYWHERE! $ SuperPro Gas Soldering Tool Kit • Up to 580°C temp range • Up to 120 min run time • Durable case with extra tip storage TS1328 ONLY 209 TS1328 VOLTAGE AND CURRENT DISPLAY $ ONLY 39 95 CONSTANT CURRENT & VOLTAGE IN A SLIMLINE FORM FACTOR . TH1827 CABLE MAKING ONE HANDED! Heavy Duty Wire Stripper • Cutter, crimper & wire guide • Strips 10-24 AWG/0.13-6.0mm • Single handed operation TH1827 $ SLIM IN SIZE, BIG IN OUTPUT Slimline Lab Power Supply • 0-16VDC <at> 0-5A (max.) 0-27VDC <at> 0-3A (max.) 0-36VDC <at> 0-2.2A (max.) • Up to 80W max. • Just 300L x 138H x 53Wmm MP3842 ONLY 5995 . QM1323 MEASUREMENTS MADE EASY! Autoranging Digital Multimeter with Temperature • Cat III 600V • 10A AC or DC current • 40MΩ resistance • 100µF capacitance • 760°C temperature • K-type probe & case incl. QM1323 PROVIDES THE PERFECT SOLUTION FOR YOUR SOLDERING & ASSEMBLY NEEDS MEET YOUR NEW RIGHT-HAND Third Hand PCB Holder • Four adjustable arms • Sturdy metal base • Anti-slip ONLY TH1981 $ 1495 . TH1981 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. $ ONLY 219 MP3842 SERIOUS POWER, SERIOUSLY GOOD VALUE! $ INTEGRATED BLUETOOTH® FOR SMARTPHONE CONNECTIVITY ONLY 999 INTEGRATED CARRY HANDLES MB4102 IP21 RATED 12V CIGARETTE SOCKET & 2 X 12VDC OUTLETS USB-A WITH QUICK CHARGE 3.0 & USB-C WITH POWER DELIVERY PURE SINE WAVE OUTPUT FOR SENSITIVE ELECTRONICS CHARGE VIA 12V, MAINS POWER OR VIA SOLAR PANEL BUILT-IN HIGH CAPACITY LIFEP04 BATTERY WITH ADVANCED BATTERY MANAGEMENT SYSTEM INTEGRATED MPPT SOLAR CHARGER UPS FUNCTION TO KEEP DEVICES RUNNING DURING POWER BLACKOUTS MB4102 HIGH POWER MODELS INCLUDE 12V 30A HIGH CURRENT OUTPUT 5 YEAR WARRANTY WHEELS & LUGGAGE STYLE HANDLE SCAN QR CODE TO LEARN MORE MB4102 (S1) MB4104 (S2) MB4106 (S3) MB4108 (S5) Output Power (Total) 2000W Continuous 4500W Peak 2500W Continuous 5400W Peak 3600W Continuous 7000W Peak 5000W Continuous 7000W Peak Capacity 1024Wh (12.8V 40Ah) 2048Wh (51.2V 40Ah) 3072Wh (51.2V 60Ah) 5040Wh (48V 105Ah) 6 Mains Outputs 4 5 6 Pure Sine Wave Output • • • • USB-C Output 2 (100W max.) 2 (100W max.) 2 (100W max.) 2 (60W max.) USB-A Output 4 x QC3.0 (18W max.) 4 x QC3.0 (18W max.) 3 x QC3.0 (18W max.) & 1 x 10W 3 x QC3.0 (18W max.) & 1 x 10W 12V Outputs 1 x Cig. Lighter Socket & 2 x DC Sockets (10A max.) 1 x Cig. Lighter Socket & 2 x DC Sockets (10A max.) 1 x Cig. Lighter Socket & 2 x DC Sockets (10A max.) 1 x Cig. Lighter Socket & 2 x DC Sockets (10A max.) 12V 30A Output - • • • PV Solar Input 14A / 800W max. 15A / 2100W max. 15A / 2100W max. 15A / 2100W max. Dimensions (W x D x H) 384 x 232 x 295mm 460 x 270 x 305mm 641.5 x 304.5 x 437.5mm 641.5 x 304.5 x 437.5mm Warranty 5 Years 5 Years 5 Years 5 Years Availability In-store and online In-store and online Order online or in-store Order online or in-store RRP RRP $999 RRP $1999 RRP $2999 RRP $3999 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 1 by John Clarke Background source: https://unsplash.com/photos/a-person-in-yellowgloves-and-blue-gloves-cleaning-a-floor--dc38HdQR1M Ultrasonic Cleaner adjustable This 40W Ultrasonic Cleaner is fully adjustable for frequency, power and duration. You can also select the shape and size of the cleaning container you use. It’s powered from a 12-15V DC supply. Ultrasonic Cleaner Controller is ideal reach the small apertures that are usuFor more delicate parts, the power Tlery,hisforornaments, cleaning items such as jewel- ally the most important areas to clean. can be reduced to prevent damage to mechanical parts and An ultrasonic cleaner makes this the items being cleaned. small areas of delicate fabrics. Cleaning fuel injectors, a carburettor, or any other intricate parts is a messy and time-consuming task, requiring soaking them in harsh solvents such as petrol, kerosene, or degreaser and then scrubbing them with various brushes. It is a difficult and tedious task and often does not task so much easier. Just place the components in a solvent bath, press a button, then come back later to remove the parts in sparkling clean condition. It will even clean internal areas! It uses a high-power piezoelectric transducer and an ultrasonic driver to release the dirt and grime with ultrasonic energy. Fig.1: in the ultrasonic transducer we’re using, two piezoelectric (ceramic) discs are sandwiched between the two halves of the body, with electrodes to allow a voltage to be applied across the piezo elements. The compression of the piezoceramics due to the tension from the bolt holding the whole thing together is critical to prevent early failure from the ultrasonic vibrations. 28 Silicon Chip Our previous High Power Ultrasonic Cleaner in September and October 2020 (siliconchip.au/Series/350) was an automatic unit that found the transducer resonance itself. Manual operation was possible, but it wasn’t as easy as this latest offering. Because this one has adjustable power and doesn’t rely on automatically Fig.2: the frequency vs power curve for the transducer. Most transducers with a nominal 40kHz resonance should be similar, but the exact frequency of the peak will vary, as will the steepness of the slopes. Hence, our Cleaner allows you to adjust the frequency to find the peak, from 33.683kHz to 46.859kHz. Australia's electronics magazine siliconchip.com.au Features » » » » » » » » » Background source: https://unsplash.com/photos/frostedwater-with-bubbles-_LHf-WzBYpo Ultrasonic cleaning at up to 40W Screen shows frequency, span, timer, voltage and wattage Manual frequency control Timer from seven seconds to 30 minutes Operates from 12-15V DC at up to 4A Reverse supply polarity protection Over current protection Ultrasonic standing wave minimisation Can use a variety of cleaning bowl sizes and shapes from 2.5L to 4L Specifications » » » » » » » » » Frequency reading: 1Hz resolution, ±3% at 25°C Frequency adjustment: 16 spans from 33.683kHz to 46.859kHz (see Table 1) Fine frequency adjustment in 128 steps of about 37Hz (for Span 0) to 44.5Hz (for Span F) Voltage supplied to T1’s primary: from 1.23V to 1.4V below the input supply, displayed with a 100mV resolution Power readings: 1W resolution Current power limiting: 3.3A (40W with 12V at the transformer primary) Power delivery to 2L of water: 32W with a 12V supply, 39W with a 13.8V supply Timer: seven seconds to 30 minutes in approximately seven second steps Standing wave reduction: ultrasonic drive is switched off every 14s for about 1ms with variation to ensure a near 180° phase change each time. finding the resonant frequency, it’s less fiddly to get up and running. As a bonus, this latest Ultrasonic Cleaner Controller provides much more information than the previous version by having a two-line, 16-­ c olumn liquid crystal display (LCD) screen to convey useful readings, allowing for an easy setup. How does it work? A metal container is filled with a solvent, de-ionised water, or normal hot water with a detergent or wetting agent. The ultrasonic transducer agitates the contents of the bath. At higher power levels, the ultrasonic wavefront causes cavitation, creating bubbles which then collapse, as shown in Fig.3. As the wavefront passes, normal pressure is restored and the bubble collapses to produce a shockwave. This shockwave helps to loosen particles from the item being cleaned (Fig.4). The size of the bubbles depends on the ultrasonic frequency; they are smaller with higher frequencies. We are using the commonly available bolt-clamped Langevin ultrasonic transducer, depicted in Fig.1. It comprises piezoelectric discs sandwiched between metal electrodes. siliconchip.com.au The central bolt not only holds the assembly together, but is critical in ensuring the piezo elements are not damaged when being driven. The bolt is torqued to a predetermined tension and locked (thread glued) in place to prevent it loosening. The bolt tension ensures the piezo discs always remain in compression even while they are operating, preventing the discs from breaking apart. When a voltage is applied to the piezoelectric discs, forces are generated by the piezo elements that move the two metal ends closer together and then further apart at the ultrasonic drive rate. Our Ultrasonic Cleaner drives the piezo transducer at close to its nominal 40kHz resonant frequency. Fig.2 shows the power applied versus frequency for the particular ultrasonic transducer we are using. It claims to have a resonant frequency of 40kHz ±1kHz. When under load, resonance is lower; we found that resonance dropped by a couple of kilohertz. The transducer drive frequency needs to be adjusted to produce the required power level. A small change in frequency from the resonant point will reduce the power quite markedly. Australia's electronics magazine Figs.3 & 4: the sound waves produced by the Ultrasonic Cleaner rapidly create and destroy bubbles in the liquid. When they collapse, they generate localised shockwaves. This ‘cavitation’ stirs up the solvent layer that’s in contact with the dirt, grease and grime, helping to break it up and more rapidly dissolve it away. You can do this by hand – it’s called scrubbing – but it’s a tedious job, and it’s hard to get into nooks, crannies and internal spaces in the parts being cleaned! July 2026  29 The Adjustable Ultrasonic Cleaner is built using two PCBs; the Main Board shown at left, and the Control Panel Board below. Switches S1-S3 have a coloured marker near their cathode pin. Image Source: Jaycar Additionally, the transducer impedance varies depending on the load. So when operating in free air, the impedance is much lower compared to when the transducer is driving a bath full of cleaning fluid. Another factor affecting the power delivered is the voltage applied to the ultrasonic transducer’s driver transformer. Higher voltages produce a higher power output. Presentation The Ultrasonic Cleaner controller fits in a diecast aluminium enclosure with three knobs, three pushbutton switches, a power switch and the LCD screen. Two knobs are for the timer setting and the frequency adjustment. Pushbutton switches are for changing the frequency span selection up and down to select between 16 options, and the start/stop of ultrasonic drive. The 16 spans allow the frequency to be adjusted between 36.140kHz to 46.859kHz. The frequency knob allows for finer frequency adjustment within the range of each span. Below one minute, the timeout is shown in seconds, while above one minute, the timeout is shown in minutes and decimal minutes in 0.1m steps. The third adjustment knob is for the voltage applied to the transformer that drives the ultrasonic transducer. It can be adjusted from 1.23V to around 11-12V depending on the input voltage. This allows the ultrasonic power delivery to be adjusted. This control is labelled as ‘Power’ since that’s what it affects. The transformer voltage and delivered power are shown on the LCD 30 Silicon Chip screen, along with the frequency, span and timeout. Circuit details The Ultrasonic Cleaner circuit is shown in Fig.5. It is based around a PIC16F1459 microcontroller (IC1) that controls the two Mosfets (Q1 & Q2) driving the primary windings of transformer T1 in an alternating fashion. T1 produces a stepped-up voltage of around 150V AC (RMS) to drive the ultrasonic transducer. IC1 also drives the LCD screen, monitors the Timer potentiometer (VR2), Frequency potentiometer (VR3) and switches S1-S3. At the same time, it measures the current flowing through Mosfets Q1 and Q2 at its AN3 analog input (pin 3) via amplifier IC2b and the voltage applied to the transformer (T1) at its pin 8 analog input (AN8) via a voltage divider. IC1 is powered from REG1, a 5V regulator that is supplied input voltage via diode D1, which provides reverse polarity protection. Adjustable transformer supply REG2 is an LM2576 adjustable regulator. It is supplied with 12-15V from CON1 via power switch S4 and 4A fuse F1. Diode D3 provides reverse polarity protection by conducting if the supply voltage goes negative. The fuse then blows, preventing damage to REG2. Australia's electronics magazine The LM2576 is a switch-mode stepdown regulator. It has an internal transistor that switches on to charge inductor L1 via the load and output capacitors. When it switches off, diode D2 provides a path for the inductor current to continue to flow to the load. The duty cycle of the internal transistor being on compared to being off determines the output voltage. Feedback is applied to pin 4 of REG2, and the duty cycle is adjusted by the regulator to maintain 1.23V at this pin. The output voltage can thus be adjusted by varying the resistance of the top divider resistance, which includes 100kW potentiometer VR1. Ideally, a 50kW potentiometer should be used, but 100kW potentiometers are more common, so we shunt it with a 100kW fixed resistor. siliconchip.com.au Fig.5: the complete Cleaner circuit diagram. Microcontroller IC1 drives Mosfets Q1 & Q2 alternately, causing an AC current to flow in T1’s primary. T1 steps up the voltage in the primary to around 150V AC in the secondary for driving the transducer at 40W. REG2 allows the primary voltage to be adjusted, controlling the output power, while op amp IC2b helps to provide current monitoring feedback and IC2a allows IC1 to reduce REG2’s output to prevent overload. That fixed resistor should be omitted if a 50kW potentiometer is used. A 22kW resistor connects to the divider from IC2a’s output; this op amp buffers the analog output from pin 7 of IC1. This allows IC1 to control the output voltage to some extent, limiting power to the ultrasonic transducer. More on this later. REG2’s output provides voltage to transformer T1’s primary winding. Two 1000μF 25V low-ESR capacitors are used to provide storage of voltage from the switch-mode supply and maintain a low source output for the siliconchip.com.au transformer. These capacitors also smooth the supply ripple from REG2. REG2’s output can’t go as high as its input; there is a voltage drop of about 1.4V. So a 12V output cannot be produced if the input voltage is 12V. Typically, the maximum output voltage with a 12V input is 10.6V at 3A. Similarly, with a 13.8V input, a maximum of 12.4V can be produced at 3A. Transformer driving A complementary waveform generator within IC1 is used to drive Mosfets Q1 & Q2 in push-pull mode. The Australia's electronics magazine transformer (T1) is centre-tapped to allow this type of drive, with the supply from REG2 applied to the centre tap. IC1’s pulse-width modulation (PWM) generator includes an adjustable dead time, allowing time for one Mosfet to switch off before the other Mosfet switches on. IC1’s RC5 and RC4 digital outputs provide the complementary gate drive signals for Mosfets Q1 & Q2. Since these outputs only swing from 0V to 5V, we are using logic-level Mosfets. Standard Mosfets require gate signals of at least 10V for full conduction, but July 2026  31 logic-level Mosfets will typically conduct fully at 4.5V, or sometimes at even lower voltages. With the IPP80N06S4L-07 Mosfets we are using, the typical on-resistance (between drain and source) is 7.9mW at 40A with a 4.5V gate voltage. They are rated at 80A continuous and include over-voltage transient protection that clamps the drain-to-source voltage at 60V. Mosfets Q1 & Q2 are driven alternately and these drive the separate halves of the transformer primary of T1, which has its centre tap connected to the adjustable supply. When Mosfet Q1 is switched on, its drain goes low (to 0V and current flows in its section of the transformer primary winding. Q1 remains on for less than 12.5μs (assuming a 40kHz operating frequency) and is then switched off. Both Mosfets are off for two microseconds before Q2 is switched on. Q2 then draws current through its section of the T1 primary winding and remains on for the same duration as for Q1. Both Mosfets remain off again for 2μs before Q1 is switched on again. The gap when both Mosfets are off is the dead time, which allows for the fact that they don’t switch off immediately when their gates reach 0V (discharging the gate capacitance also takes time). Scope 1 shows the gate drives to Q1 (top yellow trace) and Q2 at the lower cyan trace when running at 40kHz. The two Mosfets are each off during the 2μs dead time period and switched on for around 10.2μs. The vertical cursors indicate the dead time. Without dead time, the two Mosfets would both be on together for a short duration. This would cause massive short-circuit current spikes, overheating the Mosfets and also drawing large current spikes from the supply filter capacitor and DC power supply. The inductance and resistance of the transformer primary would limit this to some extent, but it’s still best to avoid it. The alternate switching action of the Mosfets generates an AC square wave in the secondary winding of transformer T1. With a turns ratio of 12.8:1 (assuming a 90-turn secondary and 7-turn primary) and 12V DC at the primary, the secondary winding delivers about 150V to the ultrasonic transducer. The waveform applied to the ultrasonic transducer is shown in Scope 2, with 12V at the transformer primary and 35W delivered to the transducer, both values shown on the LCD screen. The voltage applied to the ultrasonic transducer shown in the yellow trace is around 150V peak (on average; it varies a bit). The cyan trace is the measured current scaled by 1.4V/A. So the 4.07V current reading value equates to 2.9A. Table 1: Typical frequency range adjustment within each span Span # Centre frequency Minimum Maximum 0 36.140kHz 33.683kHz 38.409kHz 1 36.580kHz 34.123kHz 38.665kHz 2 37.040kHz 34.520kHz 39.177kHz 3 37.500kHz 34.980kHz 39.689kHz 4 37.970kHz 35.387kHz 40.458kHz 5 38.460kHz 35.877kHz 40.970kHz 6 38.960kHz 36.314kHz 41.482kHz 7 39.470kHz 36.761kHz 41.994kHz 8 40.000kHz 37.291kHz 42.506kHz 9 40.500kHz 37.728kHz 43.018kHz A 41.138kHz 38.366kHz 43.530kHz B 41.660kHz 38.825kHz 44.299kHz C 42.250kHz 39.415kHz 45.067kHz D 42.860kHz 39.962kHz 45.579kHz E 43.418kHz 40.520kHz 46.091kHz F 44.117kHz 41.156kHz 46.859kHz 32 Silicon Chip Australia's electronics magazine With 12V at the primary of the transformer, the power is 34.9W (2.9A × 12V). Standing waves Running the Ultrasonic Cleaner at a constant frequency near resonance is efficient, since the impedance of the transducer is almost purely resistive under those conditions. However, this is not ideal for minimising standing waves within the cleaning bath. Standing waves can build in strength while the frequency remains constant. These waves are caused by reflections from the parts being cleaned and the tank walls being in phase. This can damage delicate parts. To avoid standing waves, the drive is stopped every 14s for about 1ms with variation to ensure a near-180° phase change each time. This out-of-phase change attempts to calm the standing waves. Additionally, our Ultrasonic Cleaner Controller can reduce the power so it can be used with delicate parts and parts that have delicate sections within them, especially thin-walled cavities. The power is reduced by lowering the voltage applied to the driver transformer. Over-current protection Overcurrent protection for the Mosfets is provided in two ways. Both rely on current detection via the voltage across the 0.1W resistors between the sources of Q1 and Q2 and ground. The first method uses NPN transistors Q3 and Q4. These have their base-emitter junctions connected across those 0.1W current-­sense resistors. The protection starts when the voltage across the 0.1W resistor exceeds about 0.5V, ie, more than 5A through either Q1 or Q2. The associated transistor Q3 or Q4 then begins to conduct. The current flowing from its collector to its emitter reduces the gate voltage of the associated Mosfet, effectively increasing its on-resistance, which then reduces the current. This protection is a fast-acting, cycle-by-cycle measure. At the same time, the voltages across the two 0.1W current-sense resistors are averaged by a pair of 10kW resistors and filtered by a 100nF capacitor. This averaged voltage is then applied to the non-inverting pin 5 input of op amp IC2, which amplifies the signal 28 times (27kW ÷ 1kW + 1). siliconchip.com.au The averaging effectively halves the sensed voltage, so this results in an overall amplification of 14 times, meaning that pin 7 of IC2b produces 1.4V per amp. This is measured by the AN3 analog input of IC1 (pin 3) and is converted to a digital value and processed by IC1. Should this voltage reach 4.9V or more, the drive to the transducer is switched off. 4.9V represents a 3.5A average current flow (4.9V ÷ 1.4V/A). This voltage can also be measured at the TP CURRENT test point. An overcurrent error is indicated as “OVR” on the LCD screen. When this happens, OVR will momentarily be displayed and the voltage reading will drop to reduce the current. With reduced current, the overload indication will cease as the voltage returns to its original setting. However, if the overload still exists, OVR will show again and the drop in voltage will be repeated. The OVR display will occur around once per second. To prevent this, the voltage/ power pot will need to be rotated anti-clockwise, and the frequency will then need to be adjusted to be closer to resonance. There is also a warning displayed if there is no voltage supply to transformer T1. This could be due to a blown fuse (F1). The display shows “FUSE NO V”, although there could be other reasons for the lack of voltage, such as an incorrectly wound transformer, a short circuit, or a supply break. Power limit control The current measured at the AN3 input is also used for controlling the maximum power applied to the ultrasonic transducer. The maximum power rating of the transducer is 50W, but this is not a continuous rating; the recommended continuous power is 43W. We limit power by reducing the voltage applied to T1 when the current reaches 3.3A. This equates to almost 40W (39.6W) when there is 12V applied to the transformer. The analog DAC output from pin 7 of IC1 is normally set to the same 1.23V as is at the pin 4 feedback input of REG2. With that voltage, the 22kW resistor from IC2a’s output has no effect on the regulator voltage as it has the same voltage at each end of the resistor, so no current flows. siliconchip.com.au Scope 1: the yellow trace shows the gate drives to Q1, while the cyan trace shows Q2, both being driven at 40kHz. Scope 2: the yellow trace shows the voltage applied to the ultrasonic transducer, while the cyan trace is the measured current scaled at 1.4V/A. However, if IC1 detects that the transducer current rises above 3.3A, IC1 increases the analog output from pin 7 of IC1, causing current to flow through the 22kW resistor, raising the voltage across the 5.1kW resistor. The regulator compensates for this extra voltage at the 5.1kW resistor by reducing its output voltage to maintain the 1.23V at its pin 4 feedback input. Frequency adjustment VR3 is used for fine frequency adjustment, while S1 and S2 move the span down or up, respectively. There are 16 spans labelled from Span 0 through to 9, then A to F. For the fine frequency adjustment, the voltage at VR3’s wiper is converted to a digital value in IC1 via its AN4 input pin. Since the voltage across the potentiometer is the same as the microcontroller’s supply voltage, this maps to the full ADC range. A 100nF capacitor from that pin to ground lowers the pin source impedance during the analog-to-digital conversion process. Australia's electronics magazine The internal oscillator for IC1 runs at 48MHz and can be adjusted in small steps using the OSCTUNE register. This can vary the internal oscillator frequency over about a 15% range in 128 steps. For Span 8, with a 40kHz centre frequency in driving the ultrasonic transducer, this allows a 5.2kHz control range in 37.5Hz steps. The cleaning timer also depends on the oscillator for accuracy. We compensate for any variance from the nominal 48MHz due to this fine frequency adjustment to maintain timer accuracy. The 37.5Hz step resolution in frequency change is sufficiently small to drive the ultrasonic transducer at its resonant point. However, the OSCTUNE register does not have sufficient range to ensure we can drive an ultrasonic transducer that is resonant outside the range of 37.291kHz to 42.506kHz that can be obtained by simply changing OSCTUNE. Thus, a coarser adjustment is used to widen the operating range. July 2026  33 Parts List – Adjustable Ultrasonic Cleaner 1 111 × 159mm double-sided plated-through PCB, 04105261 1 98 × 60mm double-sided plated-through PCB, 04105262 1 110 × 159mm front panel label 1 50W 40kHz ultrasonic transducer 1 compact 16×2 character alphanumeric LCD screen [Altronics Z7013] 1 M205 4A fuse (F1) 1 100μH 5A toroidal inductor (L1) [Altronics L6622, Jaycar LF1270] 1 ETD29 transformer assembly: 1 former, 2 N87 ferrite cores & 2 clips (T1) [Silicon Chip SC3888] Switches/potentiometers 3 tactile illuminated pushbutton momentary switches (blue, green or red LEDs) [Altronics S1174/5/7, Jaycar SP0612-4] 1 SPST 250V 6A rocker switch (S4) [Altronics S3210, Jaycar SK0984] 1 100kW linear 9mm vertical PCB-mount potentiometer with 6mm, 7mm-long spline shaft (VR1) [Altronics R1978] 2 10kW linear 9mm vertical PCB-mount potentiometer with 6mm, 17.4mm-long spline shaft (VR2, VR3) [Altronics R1946] 1 10kW miniature top-adjust trimpot, 3386F style or similar (VR4) 1 push-on D-shape knob for ¼-inch shafts [Altronics H6024, Jaycar HK7709] 2 18t spline 6mm knobs [Altronics H6109, Jaycar HK7733] Connectors 1 16-pin header, 2.54mm pitch (for LCD screen) 1 20-pin DIL IC socket 1 8-pin DIL IC socket 2 M205 PCB mount fuse clips 2 2-way 20A 5/5.08mm-pitch screw terminals (CON1, CON2) 1 3-way 20A 5/5.08mm-pitch screw terminal (CON3) 2 14-way IDC crimp connectors [Altronics P5314] 2 14-pin keyed box headers (CON4, CON5) [Altronics P5014] Hardware 1 171 × 121 × 55mm IP66 diecast aluminium enclosure [Jaycar HB5046] 1 2-4L (stainless) steel, aluminium round or square cross-section baking tray, 75mm tall or higher 1 65mm diameter DWV (drain, waste and vent) end cap [eg, Holman DWVF0194] 1 35mm-long 65mm DWV pipe or 65mm to 45mm pipe reducer [eg, Holman DWVF0382] 2 MG12 or PG7 cable glands (for the transducer cable) 1 mains Earth connector for attaching VR1’s shaft extension (6mm ID wire entry) [Altronics P2125A] 4 TO-220 insulating kits [Altronics H7210, Jaycar HP1140] 1 MG16 or PG11 cable gland (for the power supply cable) 1 200mm length of electrical insulation tape 4 100mm-long cable ties Screws etc 2 solder lug eyelets, M4 × 6mm screws, nuts and star washers (for transducer connection) 2 M3.5 × 6mm screws for mounting PCB to enclosure (in addition to the two supplied with the enclosure) 9 M3 × 12mm panhead machine screws 8 M3 × 6mm panhead machine screws 4 M3 × 12mm tapped spacers 34 Silicon Chip 2 M3 × 6.3mm tapped nylon spacers 4 3mm inner diameter nylon washers 7 M3 hex nuts 1 M3 × 25mm panhead machine screw + M3 × (15mm, 12mm, 6.3mm) tapped nylon spacers + 3mm ID nylon washer OR 1 35mm length of 6mm timber dowel (for VR1 shaft extension) Wire & cable 1 800mm length of 1mm diameter enamelled copper wire (for T1’s primary) 1 6.2m length of 0.5mm diameter enamelled copper wire (for T1’s secondary) 1 1m length of 7.5A sheathed figure-8 mains rated wire (for connecting the transducer) 1 400mm length of 10A hookup wire (for S4) 1 120mm length of 14-way 1.27mm pitch ribbon cable 1 50mm length of 5mm heatshrink tubing (for transducer terminals) Semiconductors 1 PIC16F1459-I/P microcontroller programmed with 0410526A, DIP-20 (IC1) 1 MCP6272E/P or LMC6482AIN dual rail-to-rail CMOS op amp, DIP-8 (IC2) 1 7805 5V 1A regulator, TO-220 (REG1) 1 LM2576T-ADJ 3A adjustable regulator, TO-220-5 (REG2) 2 IPP80N06S4L-07 or equivalent high-current N-channel Mosfets, TO-220 (Q1,Q2) [Silicon Chip SC6184] 2 BC547 NPN transistors, TO-92 (Q3,Q4) 1 1N4004 400V 1A diode (D1) 1 STPS1545F 45V 15A schottky diode (D2) [Altronics Z0065] 1 1N5404 400V 3A diode (D3) Capacitors 3 1000μF 25V low-ESR electrolytic 3 100μF 16V PC electrolytic 1 470nF MKT polyester 6 100nF MKT polyester 1 1.5nF MKT polyester Resistors (all ¼W, ±1% axial unless noted) 2 100kW 1 27kW 1 22kW 1 20kW 9 10kW 1 5.1kW 1 1kW 3 560W 1 68W 2 47W 2 100mW 1W SMD M6331/2512 resistors [ERJM1W, RS Cat 566989 or similar] Miscellaneous amounts of: Solder, JB Weld epoxy resin, neutral-cure silicone sealant and electrical tape Australia's electronics magazine siliconchip.com.au Fine-tuning is then done via OSCTUNE. The wider frequency range allows a variety of different transducers to be used, as the resonance range can be adjusted to suit. This coarser calibration is performed using the PR2 register within IC1. This sets the period and thus the frequency of the PWM drive waveform for the ultrasonic transducer. For our circuit, the PR2 adjustment provides steps of approximately 530Hz. We restrict this coarse adjustment to the range 33.683kHz to 46.859kHz. This caters to transducers that have a nominal 40kHz resonance. The value of the PR2 register is stored in flash memory, so it is recalled when power is applied. The PR2 value sets the Span setting (0-F) displayed on the LCD. The OSCTUNE value is effectively ‘stored’ in the position of VR3. There is the option to lock the Frequency setting and the Span so that they remain fixed at their last settings even when the power is switched off. Each setting can be independently locked or unlocked. When locked, the related control has no effect. Switches S1, S2 and S3 connect to the RA1, RA3 and RA0 inputs of IC1, respectively. The inputs are each held high (at 5V) by 10kW pull-up resistors. A closed switch is detected when it is pressed, as the input is pulled to 0V. Power supply 12-15V DC power for the circuit is fed in via CON1. The supply needs to be rated to deliver 4A or more. If using a 12V battery, it should have a capacity of 10Ah or more. More power can be produced with a higher voltage supply, such as a 13.8V 4A supply or a universal laptop power supply like Jaycar’s MP-3476. This supplies 12V at 6A or 14V/15V at 5A. Do not use a 16V or higher supply since the input capacitor for REG1 is only rated at 16V. If your supply has a power plug, remove it and strip the wires to connect to the screw connector at CON1. Power is switched by S4, which is wired back to the PCB via the CON2 screw terminal. LCD screen driving The LCD is driven in 4-bit mode, with the most-significant data bits D4-D7 of the LCD connected to IC1’s RB4-RB7 outputs. D0-D3, the least-­ significant data input bits of the LCD, are tied to ground. The enable and register select (EN and RS) also connect to IC1 pins RC2 and RA5. The contrast potentiometer (VR4) provides a voltage to the contrast input of the LCD and is adjusted for best display clarity. Display backlighting is via the BLA (backlight anode) connection to 5V and the BLK (backlight cathode) of the LED backlight connected to ground via a 68W current-limiting resistor. Switches S1-S3 are also lit while the unit is powered with their internal LEDs via 560W current-limiting resistors across the 5V supply. Next month The second half of this article next month will mainly cover the construction, setup and usage of the Adjustable Ultrasonic Cleaner. Assembling it is mostly straightforward, just requiring a bit of trickery so that all the controls (including those on the main and control boards) project through the front panel neatly. We’ll have all the details in the secSC ond part. For the enclosure (left), we have used a IP66-rated aluminium case measuring 171 × 121 × 55mm, this is the smallest size that will fit the Main Board. Next to it is the underside of the baking tray which we’ve fitted a 50W 40kHz ultrasonic transducer to, functioning as our ultrasonic bath. siliconchip.com.au Australia's electronics magazine July 2026  35 BUILDING BEYOND THE BENCH? Big ideas don't stay put. Power your Arduino® -compatible projects with our range of modules, batteries, and accessories. Made to go wherever innovation takes you. USB OUTPUT $ LED VOLTAGE DISPLAY ONLY 495 $ . ONLY 2195 . POWER YOUR 5V PROJECT FROM BATTERIES POWER YOUR PROJECT FROM A LOWER VOLTAGE Boost Module Converts 2.5-5VDC from a single Li-Po or two Alkaline cells up to 5VDC. 500mA max. XC4512 DC-DC Boost Module with Display Converts 3-35VDC up to 4-35VDC. 2A max. XC4609 USB OR SOLDER TAB INPUTS EASILY ADJUSTABLE BY MULTI-TURN POTENTIOMETER ONLY 5 $ 95 $ . ONLY 1025 . RUN ARDUINO BOARDS OFF HIGHER VOLTAGE POWER MAKE YOUR PROJECT BATTERY POWERED Lithium Battery Charger Module Charges a single Lithium cell from 5VDC. XC4502 DC Voltage Regulator Accepts any voltage from 4.5-35VDC, and outputs any lower voltage from 3-34V. XC4514 PROJECT POWER Single 18650 Battery Holder PH9205 $4.40 SHOP AT JAYCAR FOR: Switched 4xAA Battery Enclosure with USB Port MP3083 $6.50 • Step Up and Step Down DC-DC Converters • Huge range of Batteries and Battery Holders • Great selection of USB and DC Connectors and Leads • Regulated DC Plugpacks and Lab Power Supplies Switched 4xAA Battery Enclosure with DC Plug PH9283 $6.75 3.7V 18650 2600mAh Li-ion Battery SB2308 $16.95 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. OUTGROWN YOUR FIRST 3D PRINTER? NEW TO JAYCAR FULLY ENCLOSED PRINT CHAMBER More stable prints. Less warping. SMART MULTI-MATERIAL WORKFLOW Less Untangling. Less Failed Prints. 350°C CARBON-READY HOTEND Ready for carbon fibre and advanced materials. RFID FILAMENT RECOGNITION Faster setup with automatic filament detection. ADVANCED HEPA + CARBON FILTRATION Cleaner air and reduced odours while printing. $ ONLY 799 TL5104 5" HIGH-RESOLUTION TOUCHSCREEN Bigger, clearer and easier to navigate. FAST PRINTS. TOUGH MATERIALS. LESS FUSS. The new ELEGOO Centauri Carbon V2 combines speed, strength and smarter multi-material printing in one enclosed desktop machine. Designed for makers, engineers and serious hobbyists who want cleaner results with less trial and error. Explore our great range of 3D Printers, accessories, and consumables, 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. Ecovacs DEEBOT T50 Pro Omni Robot Mop & Vacuum It wasn’t that long ago that the original iRobot Roomba robot vacuum bumbled around picking up a few odd hairs but mostly just got in the way. Today there are many new robots from Chinese manufacturers that can also mop, with advanced navigation and mapping, at increasingly reasonable prices. Review by Nicholas Vinen S ome of the more prominent robot vacuum/mop brands now include Dreame, Ecovacs, Roborock, Dyson, Samsung, LG and Xiaomi. Most of those are Korean or Chinese companies. Recently, we were in the market for a new vacuum cleaner because we were sick of our Dyson vacuum, which was very noisy and had a relatively short battery life. We also don’t have a lot of time for manual vacuuming, and even less if you consider that we mostly have hard floors that really could use frequent mopping. Those factors, and the fact that an increasing number of our acquaintances use robot vacuums/mops, led us to consider going down that route. are so efficient at dirtying the floor with food, grass, sand, confetti and so on. Having a robot would mean that, as long as we could keep the floor clear of larger items, we could clean it as often as needed. Installing and setting up the robot was easier than I expected. Unpacking the box, I quickly found the large base station, the smaller robot, plus a few accessories like a spinning brush and power cord. I had to partly disassemble the base station to remove all the bits of tape they put on it for transport, but it only took a couple of minutes and the parts snapped back in place easily. After that, all I had to do to set up the base station was find a location for it, install the ramp (it also clicks into place), fill up the water reservoir and then put the detergent in the internal bottle. Then I plugged it into mains power. It came with a vacuum bag pre-installed. Annoyingly, the box (which costs $1500 at full retail price!) didn’t come with any detergent, and you have to use a special one – you can’t just use whatever you have on hand. So I had to run out and buy some before I could finish setting it up. The robot itself came almost fully pre-assembled. I just had to remove a bit of tape and snap the spinning brush into place on the bottom of the unit. I My robot After looking at numerous brands and models, we ended up purchasing the Ecovacs T50 Pro Omni “Deebot” robot vacuum because it had all the features we wanted at a price we could afford. It can vacuum and mop, frequently cleaning the mops with hot water and detergent back at its base station. That’s ideal for keeping our timber and tile floors clean. After last Christmas, it was on sale at half price, for $750. I’ve since discovered that these robot vacuums are frequently on sale (it went on sale again in late January). So we took the plunge, especially since our children 38 Silicon Chip Photo 1: the mops are attached magnetically so they’re easy to replace and one can extend outward to clean edges, as shown here (there’s also an extended brush at the front to pick up fluff and dust). Australia's electronics magazine siliconchip.com.au You can see the camera/lidar system on the front of the T50. The hatch at the top gives access to the power slide switch, pairing button and QR code to connect the app. On the underside, you can see the cliff edge sensor & drive wheels (which have a suspension system). The mop pads and brush are held on by magnets, so if they get caught, they’ll fall off and you can pop them back on. then set it on the floor and switched it on. The next step was to connect it to my WiFi network and set up the app. Not another app I’ve previously written about how I don’t like devices that rely on apps and I intend to avoid them as much as possible (eg, in the February 2022 Editorial Viewpoint; siliconchip.au/ Article/15192). However, I have to be realistic in this case and accept that a robot vacuum is going to need an app to control it. I’ve accepted that it may well be that this app stops functioning before the robot does. That will be very frustrating, especially given what it cost, but I think it would be received very poorly if Ecovacs or any of the other manufacturers pulled support prematurely, so hopefully that keeps them from doing anything too stupid in the near future. Connecting to WiFi was pretty easy. First, the app connects to the robot’s WiFi network. Then you select the network for it to connect to and enter your password. It then connects to your network and you’re ready to control it via the app. The robot automatically returns to its base station and begins charging. The battery came more than half charged, so I could have immediately initiated cleaning, but I let it charge first. siliconchip.com.au Options While the battery charged, I went through the options in the app. Some of them are shown in Screen 1 (there are more). You can choose whether it just vacuums, just mops, vacuums and mops at the same time, or vacuums first and then mops. You can also control things like the vacuum power (higher power will lift dust better but make more noise and discharge the battery quicker), how much water it uses for mopping, how quickly it goes about its business, and whether it makes one pass or two. Having set those, and with the battery, water and detergent all full, I told it to go ahead and clean. The first thing it does is drive around your home to build a map using its camera and lidar. That only takes a few minutes. Once it has built a map, you have the opportunity to name rooms. You can also divide large rooms into smaller ones (eg, in case it accidentally considered two separate rooms as one), merge rooms and make other changes to the map – see Screen 2. I then set it to work. By default, it picks a room, then drives around its perimeter while vacuuming and mopping. It mops using two microfibre disc mops at the back, one of which can extend outwards to reach edges and corners (Photo 1). The vacuum is in the middle and the mops at the back of the robot (see above), so it will Australia's electronics magazine Screen 1: these are the standard options that you would be most likely to change. The Cleaning Modes are vacuum only, mop only, vacuum+mop in one pass, or vacuum then mop. Screen 2: it only takes a few minutes for the robot to build a lidar map of your home. You can name the rooms, split them, join them, specify which areas have carpet or hard floor etc. The white lines and lighter shaded areas show its cleaning progress, the base station is shown in dark grey and obstacles are shown as darker areas. July 2026  39 normally pick up dust, dirt and hair before mopping that area. After it has driven all around the perimeter of the room, it starts making linear passes, going back and forth until it has cleaned all the areas it can reach. It automatically drives around obstacles like furniture, pets or people, even if they’re in a different location each time. You can track its progress in the app, as shown in Screen 2. Because it’s low, it can get under couches, beds, tables and some doors (Photo 3). Once it has finished one whole room, it will move onto the next and repeat until all rooms are clean. I found it surprisingly quiet, especially if you set it to the lowest vacuum power, which seems to be adequate for hard floors. You can hear it moving around and doing its thing, but it isn’t that annoying – much less bothersome than a person vacuuming. Thoroughness Its mops certainly do a good job of cleaning our timber floors. The two rotating microfibre mops constantly have water added from an internal tank as it drives around. By default, after every 15 minutes of cleaning, it returns to the base station to empty its dustbin into the main bag and wash its mops with hot water and detergent. Then it goes back to where it was and carries on. You can change that interval (eg, to 10 or 20 minutes), or ask it to go back to the station after cleaning each room. Since this model uses hot water and detergent to clean the mops in the base station, they stay relatively fresh throughout the process. You keep the clean water tank on the base station full (it holds several litres) and it dumps dirty water in another tank. That water certainly comes out black (see Photo 4)! Even after several passes, the water still came out black, even though we regularly mopped the floors before getting this robot. Part of that is because it cleans areas we can’t easily reach – under couches, cabinets and beds – and the rotating mops do a good job of working grime out of cracks in the timber. After we ran it once or twice a week for a few weeks (probably 6-8 total passes), the water stopped coming out so black and turns grey instead. That suggests it has picked up most of the remaining grime. After running the robot the first time, the floors looked, felt and smelled cleaner than they had done for years. Our whole home smells better now. I can tell from all the dust, hair and detritus that has accumulated in the bag in the base station that the vacuum function works too. And similarly, because it goes under things we can’t easily vacuum under, it picks up dust and dirt that has been sitting there a long time. Living with it While the robot is very convenient, it isn’t completely hands-off. If you want it to do a good job, you need to go around picking things up off the floor before you start it (otherwise it’ll clean around them). For example, if you have a table surrounded by chairs, you’ll want to move at least a couple to let it get under the table. Most of the time, the only maintenance you need to do is refill the clean water tank if it’s getting low and periodically empty the dirty water tank. Every couple of months you’ll also need to empty or replace the vacuum bag (maybe more frequently if you have pets that drop a lot of fur). The mop pads and vacuum filter will need to be replaced eventually (perhaps every 6-12 months). A year’s worth of ‘consumables’ comes in at roughly $100, depending on whether you buy genuine or thirdparty parts, and how often you use it. It’s actually quite clever the way the robot announces what it is doing: “Starting cleaning!” “Washing the mop with hot water!” “Drying the mop!” You can track its progress on the map; it shows where the robot is, which way it’s facing, what areas have already been cleaned, and in what order it will clean the rooms. With the T50 Pro Omni model, you can even watch its view from its onboard video camera, although I haven’t tested that feature. Apparently there are comprehensive security features to prevent others from accessing that camera, but I’m not sure how much I trust them. Its coverage is pretty good; with the central vacuum, roller brush, spinning side brush and two mops (one of which can extend to clean edges, or retract so Photo 2: you can see the sheen from the wet areas of the floor it has already mopped. It’s cleaning the middle of the room in a racetrack pattern. Photo 3: it can fit under furniture to clean where you can’t easily, including under this Majestic loudspeaker! 40 Silicon Chip Australia's electronics magazine siliconchip.com.au it doesn’t get caught), it cleans pretty much 100% of the floor area it can reach. The only places it won’t clean are where it can’t fit through gaps. It can detect the difference between carpet/rugs and hard floors, deploying the mops for hard floors or retracting them, and increasing the vacuum suction power, for carpeted areas or rugs. You can also specify which areas have which flooring types, but it seems to get it right. While it’s very good at avoiding obstacles, if it bumps one, it does so gently with a spring-loaded bumper. A switch in the bumper lets it know when that happens and it will make its way around the obstacle. Once it knows where it is, it will avoid it for the rest of the cleaning run. By the way, you don’t have to use the app once it’s set up. You can just press the button on the top of the unit and it will start a standard cleaning pass (it also supports voice commands). The same button can also be used to pause it (or you can pause it via the app). It will just sit there until you ask it to resume. That’s quite handy if it’s getting in your way. Cleaning multi-level houses You can carry the robot up and down stairs to clean another level. I think if you have carpet and just need to vacuum, that would be OK. You’d ensure the battery was 100% charged, carry it up/down and let it clean that level. Then you’d bring it back and let it dock and charge when it was finished. When you pause it, bring it to a new level and unpause it, it realises it is not in an area it knows and generates a new map. However, since we’re mopping hard floors, it needs to return to the dock roughly every 15 minutes to refill with water and clean the mop. That means, for a full clean upstairs, I had to carry it up and down at least four times to do the whole job. While it was easier than doing the cleaning myself, it wasn’t exactly “set and forget”. Annoyingly, you can’t buy a new base station to put on another level for your existing robot. If you want two base stations, you have to buy two robots. That does at least mean you don’t need to carry them up and down the stairs. I therefore considered buying a second unit. The one we bought for $750 was back up to its full $1500 price, siliconchip.com.au Photo 4: you can tell the mops do their job by how dirty the wastewater is (the water used to clean the mops periodically). Even after multiple passes, it’s still picking up grime. Photo 5: the T50 Omni base station is the same size as the T50 Pro Omni’s. It takes up room but not too much. It’s slightly deeper than the nightstand it’s next to but only about 2/3 as wide. which I was not going to pay on top of what we had already paid. However, Ecovacs gives existing customers discounts, and the slightly cheaper T50 Omni model was on sale for $650 (no “Pro” in the model name; full price is around $1300). I got an additional $60 discount for already owning one of their robots. So I decided that for $590, it was worthwhile to get a second robot for the convenience. Luckily there was a space upstairs that was unused and perfect for the base station, right next to a power point. We ended up getting a white one this time (Photo 5), making it obvious which robot we’re dealing with. The main differences I’ve noticed are: • The non-Pro model doesn’t have a detergent dispenser in the station, so it seems you can’t use detergent with it. • The non-Pro model has a smaller battery capacity. • The non-Pro model apparently has less advanced mapping, although I haven’t noticed any differences. They seem to work equally well in terms of navigation. • The non-Pro model doesn’t let you look at the camera feed. I think the processor on the non-Pro model is overall less powerful, so it lacks some AI-type features (detecting stains etc) but in our usage, I haven’t found that it makes any difference. One advantage of having a robot for each floor is that you can run them at the same time if you want. And the upstairs robot seems to detect our stairs just fine and keeps itself from falling down. I’ve heard the ‘cliff edge’ sensors can get dirty and fail, though; I hope we don’t hear an expensive robot tumbling down the stairs one day! Australia's electronics magazine Conclusion It’s nice to finally see a truly useful application of fairly advanced robotics in the home! I think these robots are worth getting at their sale price. July 2026  41 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). EACH BLOCK OF ISSUES COSTS $100 NOVEMBER 1987 – DECEMBER 1994 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 42 Silicon Chip If you’re wealthy, perhaps you could consider paying full price, but it’s better to wait for a sale if you can. The bottom line is that they work very well, if not quite perfectly. If they’re both on sale, I suggest you pay the extra $100 and get the T50 Pro Omni model. The detergent reservoir (Photo 6) is nice to have, although I think the non-Pro model does a fine job of mopping without it (the hot water cleaning makes a bigger difference). The extra battery capacity is nice to have, mainly because it will offset some of the ageing effect of Li-ion batteries. The non-Pro model does not support detergent use at all; there is no detergent dispenser in the base station (just a space where the T50 Pro Omni has it), and Ecovacs specifies wateronly operation for both the robot and base station. Oddly, the detergent they sell is listed as being compatible with the T50 Omni, but it’s unclear how you’re supposed to add it. Mentioning it could be a mistake. It appears that the battery packs for the T50 Omni (5200mAh) and T50 Pro Omni (6700mAh) share the same voltage, physical dimensions and connector. If that proves to be the case, it may be possible to replace the smaller pack with the higher-capacity one when the battery eventually fails, although that would be unofficial and unsupported. Pros ☑ Thorough cleaning of hard floors (likely good at vacuuming carpet too, but we don’t have any) Pushbutton convenience Relatively quiet Easy setup, both hardware-wise and software-wise Many cleaning options Self-cleaning mops Minimal day-to-day maintenance Avoids obstacles even if they move run-to-run (toys, chairs etc) Can be run while people are around or when you’re out (avoids people and pets) Can clean under furniture and right up to edges Announces progress audibly and can be tracked via the app, including an end-of-cleaning report Won’t fall down stairs (unless it malfunctions...) ☑ ☑ ☑ ☑ ☑ ☑ ☑ ☑ ☑ ☑ ☑ Cons ❎ Relatively expensive Australia's electronics magazine Photo 6: the T50 Pro Omni base station with the door opened, revealing the vacuum bag on the right and the detergent dispenser on the left. In the T50 Omni base station, there’s just an empty space on the left. ❎ App & camera privacy concerns ❎ Premature obsolescence concerns ❎ The moderately large base sta- tion permanently takes up space in your home Finite rechargeable battery life (although the battery is user-replaceable) Presence of people and pets can reduce cleaning effectiveness Can sometimes get stuck and require manual intervention, making it less convenient to run unattended Limited support for multi-level dwellings without multiple robots You can’t buy an extra base station for an existing robot ❎ ❎ ❎ ❎ ❎ Videos You can see a few short videos of the T50 Pro Omni in action at these links. Leaving the dock: siliconchip.au/ Videos/Deebot+start Cleaning the floor: siliconchip.au/ Videos/Deebot+clean Avoiding a table: siliconchip.au/ Videos/Deebot+avoid Returning to the dock: siliconchip. SC au/Videos/Deebot+dock siliconchip.com.au Winter DEALS across the range. Great deals in electronics Shop 24/7 <at> altronics.com.au Iroda® 100W Torch & Soldering Iron Amazing new multi-function design! One tip converts between soldering iron, heat gun and blow torch. Includes a 11 piece tip kit allowing a multitude of uses in home repairs, cable heat shrinking, model making, wood crafting and cutting. T 2580 NEW! 89 $ Sale prices end July 31st. SAVE $40 119 $ Iroda® 130W Swivel Head Torch & Iron A0323 65W SAVE $30 69 $ A0319A 20W Recharge anywhere you go! Available in either 20W or 65W power delivery with in-built mag safe compatible wireless charging they come with travel adaptors (US inbuilt + 3 adaptors) for travel all over the world. All supplied in a handy zip up case. Handy Battery Maintainer Hassle free maintenance charging for vehicle, lead acid & LiFePO4 lithium batteries. 6/12V 1.5A output. SAVE 16% 49 $ M 8521B T 2582 X 4300B NEW! SAVE $20 129 $ 99 $ Portable Digital Microscope Camera Completely portable perfect for schools! A compact design with a 4.3” colour screen, ideal for classrooms, hobbyists, and PCB inspection. Offers adjustable focus, zoom, LED lighting, USB connectivity, photo storage, and portable rechargeable operation. All-Rounder Multimeter The perfect beginner, student or enthusiast multimeter. 12 auto ranging test modes with good accuracy and an easy to read jumbo digit screen. SAVE 22% Includes test leads. $ 30 Q 1129 Amazing new multi-function 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. A light for camping, fishing, anything! Compact and rechargeable! Standard torch PLUS 5W LED flood light in one convenient folding design. Magnetic folding bracket in-built. Magnetic Sensor Lights Light up your work space! 30s on time or manual on/off. Easy USB C recharging. Also great for inside cabinets. SAVE 10% 2 for $ 50 X 2394 X 0226 SAVE 10% 2 for $ 40 SAVE $9.95 40 $ X 0210B s Work er w as a potoo! bank 1200 Lumen Torch Durable aluminium body with double sided adjustable beam and double sided lantern lights. USB C rechargeable. 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 July 31st 2026. Build It Yourself Electronics Centre® Adventure DEALS N 1117A 200W N 1114A 100W SAVE $110 SAVE $50 339 $199 $ T GREA ! E U L VA Dynalink® 14” Go-Anywhere Digital TV Perfect for the car or caravan! Powered off internal rechargeable battery, your vehicle battery or mains plugpack. Also fitted with USB connection for recording TV. SAVE $80 Compact 20A Solar Regulator Heavy Duty Solar Blankets Portable solar charging made for life off the beaten track. Provide portable charging power for your campsite set up. Double stitched panels, durable webbing straps and metal hanging loops and zippered cable pocket. 200W version has folding legs which allow the panel to be used freestanding. Folds up and secures with velcro for a fast getaway! 5m Anderson cable connection. See right for solar regulator to suit (N2018). SAVE 29% N 2018 5W 80 Channel UHF Walkie Talkie 24.95 Portable design, use it anywhere! Superb build quality for a great price! Our new Powerhouse walkie talkie is great for recreational and commercial comms use with IP67 protection and charging dock $ A 1102B NEW! 29.95 $ SAVE $40 Folding LED Lantern Light Innovative folding design includes a central LED lamp with fold-out side panels. USB C rechargeable. 159 $ X 0664 Versatile design transmits or receives audio. Up to 10 hours portable use per charge. Bluetooth 5.4 for superior range and sync. Works as a hands free system for calls plus plays audio from your phone. USB C 20W PD & QC3.0 chargers in-built. 2 for $ 40 34.95 28 SAVE 15% 75 $ $ $ A 2798A D 2201 Magnetic Phone Charger Magsafe compatible. Provides 15W wireless charging to keep your phone juiced up on the road. Best paired with a QC3.0 charger. AM/FM Pocket Radio Features easy-to-use digital tuning, a bright LCD display, and a built-in speaker with headphone jack. Requires 2xAAA batteries. SCAN TO FOLLOW US! Stay up to date on latest releases, exclusive specials and news on our socials. 30 $ Bluetooth Car Audio Adapter SAVE 33% SAVE 12% SAVE 15% X 0604D Mini Bluetooth Audio Adaptor D 0514 NEW! 40 $ Works with 12V or 24V lead acid batteries. Easy on screen monitoring. Huge range of protection features. *350A shunt pictured. X 0232 NEW! 209 $ S 8864A Q 0594 50A Handy Magnetic Wireless Power Bank Magnetic 15W wireless charging plus USB C power delivery. Just 120 grams! 5000mAh capacity. SAVE $30 129 $ Q 0596 350A* Ultimate Battery Fuel Gauge. Accurately measures battery voltage, current, power, real capacity and remaining run time of your connected battery (suits any chemistry, 8-120V). Includes shunt. 1% accuracy Like our service? Review your store on Google. Every review helps us serve you better. Tool SAVERS BEST SELLER! SAVE $26 99 SAVE $124 Super fast desoldering for quick repairs or recycling parts T 2052 $ 375 $ T 2040 Micron Soldering & Vacuum Desoldering Station Micron® 68W Compact Soldering Station Save space on your bench with this top performing 60W soldering iron and 90W vacuum desoldering station. Removes a 16 pin through hole IC in 30 seconds! Sucks molten solder away from components & pads in no time and is easily cleaned. 160° to 480°C adjustable. Includes 0.2mm soldering tip and three desoldering tips. Offers convenience and plenty of power with precise dial control and temperature lock. Sleeper stand shuts down the unit when not in use saving on power costs. Includes fine 1.2mm chisel tip, solder reel holder and tip sponge. ® Buy 2 & save 15% SAVE $50 T 1289 119 $ Hot Air Re-Work Desoldering Iron SAVE 21% 55 Micron® Lithium Rotary Tool Set Drills, cuts, sharpens, cleans, polishes and engraves most surfaces, this rotary tool is ideal for enthusiasts, makers, or just odd jobs! 5 speeds (5000 to 25000RPM). USB C recharge. 42 accessories included. T 1460A W/magnifier SAVE 23% 15 $ T 1450A SAVE 20% 10 $ Hobby Helping Hands. The hobbyists dream - just like having an extra hand to get things done! Great for gluing, painting or soldering. T 2825B 39 $ 2 For $ 22 Wire Stripper & Kwik Crimper Combines a ratchet wire stripper, cutting blade & kwik crimper. Suits 10-24 AWG cable. 130mm Nipper Side Cutters Ideal for cutting component legs, small core wiring, cleaning up 3D prints etc. 15 15 $ Durable all metal design. Suits rolls up to 1kg. Dual roll version available (T1302A) T 2127 1000V Screwdriver Set $ SAVE 10% Fibreglass Eraser Made from fine fibreglass threads for cleaning and polishing surfaces. 20 $ T 2188 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. 65 $ Repair faster with a lithium screwdriver. T 1926 SAVE 24% SAVE 24% T 1300A SAVE 19% A USB rechargeable screwdriver has an adjustable torque drive for accurate driving of precision screws. Suits 4mm driver bits. 2 hrs use per charge. Two way control. Great for contact cleaning Solder Holder Provides 300W of hot air for quick and easy desolder and re-work of surface mount boards. 200-500°C adjustable. Also makes a great hot air gun. $ T 1528A *solder not included. T 2125 SAVE 20% Smaller sizes than most 1000V driver sets. 3 flat blade (2.0, 2.5 & 3mm) & 3 phillips (#000, #00, #0). Tech &Gadgets. NEW! 85 $ X 7060A With outdoor sensors! NEW! X 7063A Live and local weather monitoring. 299 $ Our new fully updated weather station displays your local weather data - great for boaties & gardeners. Bright & clear base station provides readings for indoor/ outdoor temperature, humidity, air pressure, rainfall, wind speed and direction. Plus handy weather trends. You can even connect it to wi-fi for monitoring readings & upload data to the internet for review. 100m sensor range. Indoor / Outdoor Weather Station. Track your local conditions with this sleek, 9.5” slimline weather station. Featuring a vibrant, backlit LCD, this monitor provides real-time data via a high-precision wireless sensor (range 30m). Tracks temperature, humidity, air pressure & moon phase. Each system can connect up to 3 external sensors (X 7061A $19.95). C 9040B Black C 9040G Green C 9040R Red C 9040W Latte One hub for all your AV gear SAVE $40 SAVE $25 $ A 3089A 79 $ C 5060A 4K 5 Way HDMI Switcher HiFuture® FlyBuds ANC Ear Buds Top Value Microlab® Bookshelf Speakers Ultra compact deisgn supporting 4K <at> 60Hz switching with IR remote control for easy source selection from your couch. Offering great performance at an amazing price point, the Microlab B77BT offer rich full sound & wireless connectivity. Perfect for gaming, music and movies. Featuring walnut finish cabinets and cloth grilles. Winter boredom buster! SAVE $60 K 8673 SAVE $20 179 50 $ 199 $ Available in 4 colours! These true wireless earbuds offer high performance noise cancelling and up to 30 hours of listening time. Makerzoid® 200 in 1 STEM Set. Makerzoid® Smart Robot Set Build up to 200 different projects or create your own! Great for inspiring kids to invent and design with staged learning from the basics up to advanced Scratch programming. Each set contains hundreds of blocks, plus multifunction sensors, programmable motor and host controller. Compatible with big brand name blocks. For ages 6 and up. This 300pc set contains the parts for 72 different project designs which can be built using Makerzoids 3D app tutorials. It includes an intelligent motor controller which can be programmed with Scratch, plus distance and light sensors. Includes 4 wheels for mobile projects. Blocks are compatible with other major brands. Includes storage box. SAVE $20 K 8671 79 $ Sale Ends July 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 0007 © 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. 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. Generating ±15V from USB 5V with a common-mode choke The circuit generates clean, regulated ±15V rails at up to 50mA (to run circuitry like op amps) from a USB power bank, phone charger etc. It uses the sort of parts that are frequently recovered from junk circuit boards and is pretty handy to have in the toolbox. It’s based on the venerable MC­ 34063A IC (REG1) in boost configuration to generate +24V from the 5V supply. Instead of a regular inductor, it uses a coupled inductor (sometimes called a common-mode choke [CMC]), so the second winding can be used to generate the negative rail. When REG1 pulls its pin 1 low using its internal transistor, current builds up in the lower winding of L1. When the drive to that internal transistor is cut, the energy stored in L1’s magnetic field causes pin 1’s voltage to shoot up, forward-biasing diode D2 and charging the output capacitor up to 24V (regulated using feedback to REG1’s pin 5). The same magnetic field also drives current through the upper winding of L1, with D1 conducting at the same time as D2. As D1’s cathode ultimately connects to ground, its anode goes negative, charging the other capacitor to around -18V. This is not directly regulated, but it remains close to -18V as a consequence of the positive output being regulated to +24V. 7815 & 7915 linear post-regulators are used to clean up the outputs of the switch-mode chip. You can download the PCB design and simulation from https://github.com/lightbox-eng/pm15 The PCB is designed to fit in a small Hammond 1551L plastic case. The 1W resistor in series with the input shown is not on the PCB, but it is a good idea to add it if you ever plan to plug this device into a computer’s USB port. Otherwise, the high surge current drawn by the 100μF input capacitor charging could cause the USB port to be disabled or possibly even damage it. Matthew Curlis, Lightbox Solar, Leonards Hill, Vic. ($100) 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 July 2026  47 Driving a mini 3-digit low-power LCD module Digital displays are useful for showing voltage, current, power, duty cycle ratio etc. This minimalist design uses a very small and lowcost 3-digit LCD screen, which lacks an onboard controller, available from AliExpress: 1005003745628043 It is driven by an STM32C011F6P6, an entry-level 20-pin microcontroller based on a 32-bit ARM CortexM0+ core. Since the supply voltage of the LCD screen has a maximum limit of 3.0V, a micropower low-dropout (LDO) voltage regulator (MCP17023002) is used to power the microcontroller, which drives the LCD screen directly from its outputs. The 200kW resistors bias the LCD common pins 7-10 at 1.5V (3.0V ÷ 2) when the microcontroller GPIO pins PA0 to PA3 are configured in input mode, while the other LCD pins are directly driven by the GPIO outputs PA5 to PA12. The table opposite shows which pin pairs need to be driven to darken a particular segment, where A-G correspond to the standard seven segments and 1-3 are the digits. The wires between the microcontroller and LCD screen should be kept as short as possible to minimise DC bias artefacts and prevent ghosting or image retention on the LCD screen. Higher-value resistors can be used, since the LCD draws negligible current. The microcontroller consumes only about 3.4mA at 48MHz and only 85µA in sleep mode. For testing purposes, I used a mini development core board programmed using the Arduino IDE. The development board only costs less than $2! (AliExpress 1005009677670000). The development board circuit can be downloaded from siliconchip.au/link/ acan The microcontroller code is written in C++. You can download it from siliconchip.au/Shop/6/3610 The test consists only of driving the LCD screen as a decimal counter, incrementing once per second. You can see the prototype counting up in the video at siliconchip.au/Videos/ minLCD+demo Once you get that working, you can modify the code to perform other jobs. Djouad Saada, Oran, Algeria. ($75) Randomly timed model traffic lights A friend of mine bought a set of model traffic lights, which he expected to operate randomly. However, he found that they followed an all-too-predictable pattern. Could I change them to something slightly more erratic? This circuit is the result. IC1 provides a random 3-bit input to IC2, a BCD-to-decimal converter. With a 3-bit input, eight out of ten decimal outputs are used. These 48 Silicon Chip are fed via diodes D2-D9 to three coloured LEDs. As shown, these LEDs are weighted green/amber/red with a 3:1:4 ratio. IC2’s A3 input pin is tied low to prevent any uncertain logic state. In practice, the circuit illuminates the LEDs for a few seconds each on average. There is some room for experimentation. For instance, the values of the capacitors may be altered, or the number of diodes Australia's electronics magazine feeding each LED can be rearranged. As a joke, I included a “New South Wales” switch S2, which introduces random power outages. A supply voltage of 9-12V is recommended (3V minimum and 15V maximum). The circuit draws roughly 20mA, so a plugpack power supply is recommended for longterm use. Thomas O. Scarborough, Cape Town, South Africa. ($70) siliconchip.com.au LCD segment to pin mapping Pin 1 2 COM1 – 3D – 3 COM2 3C 3E COM3 3B COM4 3A 3F 4 5 6 2D – 1D 2C 2E 3G 2B 1C 1E 2G 1B 2A 2F 1G 1A 1F Diode-biased capacitive proximity sensor This is a proximity sensor which, with a large square tin foil sensor, will reliably detect two hands at a distance of one metre. It has exceptional stability and low current consumption. Capacitive proximity sensors pick up a minuscule charge on the human body, equivalent to about 100pF. Thus, a human hand will usually be detected on contact with a sensor, or within a few centimetres. Proximity sensors commonly have two types of front end: sensitive transistor preamplifiers or RC oscillators with a very small capacitance value. However, there is another way – arguably, a better way – based on a diodic divider, a concept I introduced in the Circuit Notebook section of the February 2024 issue (page 47). siliconchip.com.au A CMOS TL072 dual op-amp has input impedances of around 1TW. These inputs may be regarded as floating. In this circuit, the non-inverting input of IC1a is biased with a fixed but adjustable voltage using 1MW potentiometer VR1 and, for fine adjustment, a 100kW potentiometer (VR2). The inverting input uses two reverse-biased 1N4148 diodes to provide a roughly mid-rail bias voltage with an extremely high source impedance of a few GW (gigohms). It is then possible to present a human charge directly to this input. Power consumption is just 2mA on standby, while drift is virtually non-existent. This means that sensitivity can be greatly increased. However, on switch-on, allow a few minutes for everything to settle before adjusting the trimpots for maximum sensitivity without false triggering. Various sensors may be used, and these need not only be metal. A timber tabletop served well in experiments. Note that ambient EMF has some influence on the charge on a human body, so the performance of the circuit may be influenced by nearby mains wiring. The supply voltage for the circuit is 7V to 36V, although 9-12V is ideal. A small 9V PP3 zinc-carbon battery should last about a week. A friend suggested that, by inserting a meter at the output of IC1a, one could turn this circuit into a joke ‘lie detector’. The sensor could be a small metal pad. Thomas O. Scarborough, Cape Town, South Africa. ($75) July 2026  49 Part 2: electronics Phil Prosser’s Phenomenal Pinball L ast month, we showed the overall configuration of the pinball machine and introduced pretty much all the modules that make it up. Besides the cabinet, controller, score display and deck, pretty much everything else is modular. Those modules fall into two broad categories: electronic and electromechanical. It helps to have the electronic parts working as you build the electromechanical parts, so that you can test and actuate them properly. Therefore, we will present the electronics first, starting with the circuitry and then the PCBs and assembly instructions (we published the parts list last month). Some parts are required, like the Control Board and Power Supply. Most of the others are optional, although you’ll almost certainly want to build most of them. In some cases, like the bumpers, targets and kickers, most good machines will have several. With pinball, more is more! Our Control Board has been designed to have enough inputs and outputs for what most constructors will need. Later, we will eventually present an expansion board, in case someone wants to build a monster pinball game! While not strictly necessary, it’s very helpful to have a computer with a USB port for testing. You also need a serial terminal program such as PuTTY to access the debugging information. Now let’s get stuck into the electronic side of the game. Circuit details Machine The full circuit of the Control Board is shown in Fig.4 overleaf. The sections in dashed boxes are repeated multiple times, as described in the notes at the top of those boxes. We use 74HC595 serial-to-parallel shift registers to drive all the outputs and 74HC165 parallel-to-serial shift registers for monitoring all inputs. This allows us to have hundreds of I/Os with just a few pins used on the Pico 2. These chips are relatively inexpensive, so the board doesn’t cost a huge amount to build despite its size. All this I/O could have been handled in a single very-high-pin-count FPGA or microcontroller, but this would probably cost about the same as the discrete solution and would definitely need to be a surface-mount device. We thought it was best to make this easy to work on. There is also a level of nostalgia in using old-school devices. Australia's electronics magazine siliconchip.com.au This customisable pinball machine has everything you’d expect: a ball launcher, flippers, bumpers, ramps, targets, rollovers, sound effects, flashing lights – the works. You can build it just like ours or design your own using the electronic and mechanical modules we’ve designed and tested. 50 Silicon Chip Adding to the number of parts on the Control Board, there is substantial input and output protection. The inputs can be expected to be subject to some pretty serious EMI. Given that this is a large device and very mechanical, it is also likely that during construction and servicing, the inputs will be subject to abuse. All inputs have 1kW series resistors and clamp diodes limiting the 74HC165 input voltages to safe levels. We also have 1kW pullup resistors to 3.3V on all inputs, making it easy to connect switches between these pins and GND. This means that if you choose not to use a particular input or sensor, it defaults to an inactive state. The 1kW pull-ups provide a relatively low impedance, which makes coupled noise less likely to be a problem. We run a ground line along with each input group from the controller to the pinball deck, which should minimise ground-related noise problems that can occur when switching high currents. There are four 74HC165 devices on the board, providing 32 discrete inputs. This is just enough to make a decent pinball machine. We considered using resistor arrays on the inputs and outputs, but the cost from reputable suppliers was far more than individual parts. We felt the trade-off between parts count and cost to constructors fell well on the side of individual resistors. LED outputs We need a lot of lights to make the machine pretty, meaning we also need a lot of controllable outputs. We chose to use LEDs for most outputs, though you could connect incandescent bulbs with some modifications. The supply current draw could get out of hand unless you are careful, though. There are thirteen 74HC595 chips on the board, providing a total of 104 outputs. 40 of these are dedicated to the score and player number displays. The remainder is buffered with open-collector transistors. This allows us to drive LEDs at much higher currents than the 74HC595 supports, siliconchip.com.au which is 50mA total per chip. We run the white LEDs at 25mA, although the transistors will happily sink much more current than that with lower-value current-limiting resistors. The LED outputs use 64 transistors, but these are cheap and any pin-compatible NPN device will do (BC337/8, BC546/7/8/9 etc). If you need to handle more current, that is quite possible; just watch the 5V supply total limit. All the 74HC595 shift register inputs are driven in parallel with the SER (serial data) and SRCLK (serial clock) signals, so data is clocked into all shift registers simultaneously. However, this has no effect on the outputs of any chips until one of the individually driven RCLK pins is driven, allowing us to clock eight bits of data to any of these chips at any time. Every time eight bits (one byte) of data is clocked to a register and latched, the states of the eight connected outputs are updated simultaneously. We are driving these devices way below their maximum rate and can still update all the outputs in a couple of hundred microseconds. From a modern data communications perspective, this is terrible. But since we are interacting with human beings, this allows a solid 1kHz update rate for everything on the pinball table, which is more than fast enough. One concern we had was driving SRCLK and SER to so many devices across such a large board. We have included a 100W series resistor at the driving end. The signal measured across the board is quite clean, showing very safe setup and hold times without undue ringing. Power outputs Pinball machines need to be able to drive solenoids at relatively high voltages and current, as well as things like bells and lamps. This is a job for a power Mosfet. Having only 3.3V to drive the Mosfet gates from the Pico 2 demands the use of logic-level devices. In our machine, we use 12V 1.5/2A solenoids but drive them at 24V. The flippers use two in parallel, which amounts to a short-term demand of 6A, although only for 100ms or so. This is why we have 10,000μF of supply bulk storage on the 24V rail (split between the Control Board and the Power Supply Board). For these Mosfets, we have specified IRLZ44NPBF devices, which are about $1 each if you buy 12. These are rated at 55V and 47A with a maximum 1.8V Vgs (gate-source voltage) threshold. This means we can drive them straight from the 74HC595 outputs, given that the chip is running from a 3.3V supply, for compatibility with the Pico 2. Make sure you use logic-level devices (ideally the ones we’ve specified) or they won’t work properly. Be careful as not all ‘logic-level’ Mosfets are equal; for example, we cheekily used some MTP3055VL devices in development, but these are only “on” enough to allow testing, and should not be used in a permanent installation. Each power output has a normally reverse-biased 1N4004 diode across it to absorb the significant inductive spikes from the solenoids. These are included on the controller as ‘belts and braces’; you will see that we also specify diodes across the solenoids themselves under the pinball deck. Once you see the machine in operation, you will understand our conservative approach here. ...continued on page 54 Photo 6: the Control Board for the Pinball Machine will take a while to build. But since it’s split up into sections, you can tackle it one bit at a time. July 2026  51 Fig.4: the Control Board is large but it’s made of lots of repeated sections, so we’re only showing each one once. There are four instances of the input section inside the dashed red box, giving 32 total inputs. The tables show their default functions. There are five low-current output sections (green box) to drive the Player and Score 7-segment displays and eight medium-current outputs (cyan box) to drive up to 64 LEDs (note the use of different resistor values in some sections). The 12 high-current outputs are inside the purple dashed box, with the default functions of each listed. 52 Silicon Chip Australia's electronics magazine siliconchip.com.au siliconchip.com.au Australia's electronics magazine July 2026  53 Sound interface We used a ‘1-bit PWM DAC’ library for sound, which uses the onboard PWM modulator and an interrupt service routine (ISR) to generate an analog 8-bit output at a sampling rate of 11kHz. This is not hifi, but it does the job. The output is from a single digital I/O line at pin 7 (digital output GP5). It goes through a low-pass RC filter before being amplified by an LM384 power amplifier, producing sound from the small connected loudspeaker. Future expansion We’ve considered what might happen if someone more ambitious than us runs out of inputs or outputs, so we’ve provided a four-pin header (CON20) and three two-pin headers (CON35CON37) for future expansion. In a pinch, CON35-CON37 can be used as three extra switch inputs. We plan to describe a board in a future issue that can be connected to these four headers to provide even more inputs and outputs. In theory, we (or you) could add any number of both, but most likely we will add one bank of 8 inputs, two to four banks of 8 LED outputs and one bank of 4-8 high-current Mosfet-based outputs. That should be enough for a very complex Pinball Machine indeed! Keep in mind that the existing Photo 7: the Power Supply Board provides 3.3V and 5V DC rails to power the Pico 2, LEDs and so on from the 24V supply. It also passes the 24V supply through to power the solenoids and audio amplifier. design already has several spare inputs and outputs, so it’s possible to build a somewhat more complex machine than ours without needing any expansion. It depends on how ambitious you are! The software will need to be modified to handle the extra I/Os but that should not be difficult. Power Supply Board Old-school pinball machines ran their solenoids at quite high voltages, in many cases exceeding what are currently considered ‘safe’ levels. We want to make our pinball machine something that anyone can fiddle with, without fear of a significant shock. So the whole thing operates from a 24V DC 5A plugpack or power brick. A rating higher than 5A won’t hurt. We have used several different supplies while working on this project, including a 20V 6A laptop docking station supply. This is a touch short of our target of 24V, but works well enough and it was free. We succeeded in achieving excellent performance from the flippers with a 24V rail, but had to use dual solenoids per flipper and overdrive the 12V DC rated solenoids at 24V. This gives us the oomph we need without the use of hazardous voltages. To ensure the power supply rail handles the high current pulses, we have 6600μF of storage on the Power Supply Board and another 4400μF on the Control Board. Our lighting in the game is all LED-based, so the power supply has Fig.5: the power supply is mercifully simple. The incoming 24V DC (or thereabouts) is fed straight through with some capacitors to help handle current spikes. That supply is also converted efficiently to 3.3V and 5V rails to power digital logic and LEDs by a pair of integrated buck (step-down) regulators, REG1 & REG2. 54 Silicon Chip Australia's electronics magazine siliconchip.com.au high-efficiency buck (step-down) conversion of the 24V DC to 3.3V DC and 5V DC rails for logic and lighting. The current draw on these rails can exceed 1A, so linear regulators are not a sensible option. The Power Supply circuit is shown in Fig.5. It’s intended to be mounted reasonably close to the Control Board, as there are some quite high current spikes that will be drawn when solenoids are actuated. We won’t linger on the power supply design, as it is quite conventional, with the two stepdown converter sections basically being lifted straight from the LM2576 data sheet. The only difference between the two buck regulator sections is in the feedback divider resistor ratios. The LM2576T-ADJ uses negative feedback to regulate its feedback pin to 1.23V. So with a feedback ratio of 2.6 (1 + 1kW ÷ 1.6kW), that results in an output of 3.198V (1.23V × 2.6; close enough to 3.3V). Similarly, 1 + 3kW ÷ 1kW = 4 and 1.23V × 4 = 4.92V. All rails are fused, as we have a creeping suspicion that there will be quite some ‘poking around under the deck’ for a machine that is well used. The 3.3V and 5V converters are pretty efficient (typically about 80%), so their normal draw from the 24V rail will be a couple of hundred milliamperes in the worst case. For example, a 1A draw from the 5V rail is a load power of 5W. At 80% efficiency, that’s 6.25W drawn from the input, which is just over 250mA for a 24V supply. We have used rather chunky pluggable terminal connectors for the outputs on this board, and in many other places in this project, such as for solenoids. The current will see 3A pulses when each flipper is operated, with brief pulses to 6A. So we cannot use lightweight plugs and wiring. These connectors are rated at 10A and allow you to unplug parts of the machine during construction and service. Photo 8: the finished Pinball Machine (legs not shown). Note that the backboard has a strip of white LEDs run around the inside of the bezel that flash when certain events are triggered. Remaining circuits The Control Board and Power Supply contain about 95% of the electronics in the Pinball Machine, but there are another 10 simple circuits/boards used, mostly to keep the wiring manageable: 1. Player Number Board: this is a simple 7-segment display on a small carrier board wired to a 10-pin IDC siliconchip.com.au Fig.6: these helper circuits (starting with the Player Number Board) mostly serve to simplify wiring the various LEDs, switches, sensors and solenoids up to the Control Board without the wiring becoming a mess. They all connect back to the Control Board with some combination of 10-way ribbon cables and figure-8 cables for the solenoids. Australia's electronics magazine July 2026  55 Fig.7: the circuit diagram and PCB overlay for the Score Board, which uses six 7-segment LED displays. These displays must have commonanode wiring. Don’t trick yourself and accidentally install common-cathode parts, they look identical but don’t work. header (Fig.6). Connecting it to one of the low-current output headers on the Control Board allows the current player number to be displayed. 2. Score Board: this board has six 7-segment displays, four 10-pin IDC headers and a small amount of drive circuitry (Fig.7). It’s driven by four low-current output sets to show the score as four digits plus two zeroonly digits (so the score is always a multiple of 100). That means you can 56 Silicon Chip get a score approaching one million – much more impressive than mere thousands! The onboard resistors and transistors allow the zero digits to be switched on or off using pin 8 of CON101, meaning there’s no connection to the decimal point segment of DISP1. 3. General LED Board: this connects up to eight separate LEDs to a 10-way ribbon cable (Fig.8). The LEDs connect to this board via two-pin polarised Australia's electronics magazine headers. It’s driven from one of the medium-power outputs on the Control Board and is used for general lighting and effects. 4. Bumper LED Board: this has eight LEDs in a circle and fits around the outside of bumpers (Fig.9). Like the General LED Board, it connects to a medium-power output set on the Control Board. The Bumper LEDs use one 8-bit output port each from the Controller and siliconchip.com.au Fig.8: the General LED Board, which connects up to eight separate LEDs. generate patterns triggered by time and when the ball hits the bumper hard enough. We used the brightest reasonably-priced LEDs we could find. The resistors on the Control PCB set their drive current to 20mA. Our deck had drilled holes into which we inserted 3D-printed clear LED bezels, to be described in a future article. The PCB is mounted with the LEDs pushed into the bezels, and we glue a couple of the LEDs to the bezels to secure the assembly. As shown in Photo 4 last month (and reproduced on page 62), this board is sized to fit around the bumper mechanisms. 5. Cascade LED Board: this has 15 LEDs in a triangle pattern (Fig.10). The extra LEDs let us flash some interesting patterns. It’s typically placed in the middle of the deck and is driven by two medium-­ power output sets on the Control Board. Note that there is no LED16 due to the triangular layout. 6. Switch Input Board: this connects up to six regular switches and two inductive sensors to an 8-way input port on the Control Board (Fig.14). The inductive sensors differ by needing a 24V supply voltage, hence the 3-way connectors for them. These should connect to CON2-CON4 on the Control Board as those are the input headers with a connection to 24V (CON1 supplies 3.3V). 7. General Input Board: this connects up to eight regular switches to an 8-way input port on the Control Board (Fig.15). 8. High-Current Interface: this adds four back-EMF clamp diodes across the wires to up to four solenoids (Fig.11). It’s important that these are close to the solenoids. It can also simplify the wiring by keeping the four figure-8 cables from the Control Board siliconchip.com.au Fig.9: the circuit diagram and PCB overlay for the Bumper LED Board is shown above and to the left. Depending on how you plan your Pinball Machine, you might need several of these. Fig.10(a): the PCB overlay for the Cascade LED Board. The circuit diagram for this board is shown overleaf. Australia's electronics magazine July 2026  57 Fig.11: the HighCurrent Interface Board. This board should be kept close to the solenoids. Fig.10(b): the circuit overlay for the Cascade LED Board. Note that it only has 15 LEDs instead of 16 due to the layout. together up to this board. This board is used for the flippers and reload mechanisms or other things you want to control. 9. Rollover Board: this connects up to eight inductive sensors to an 8-way input port on the Control Board (Fig.12). 10. Bumper Driver Board: this is like a combination of the General Input Board and High-Current Interface (Fig.13). Having them together means one less board to mount, as they are both required for bumpers and kickers. It also provides one extra high-­ current channel, allowing for the three bumpers and two kickers to be connected via a single board, plus three extra headers and current-­ limiting resistors for high-power LEDs (mounted on top of the bumpers) to be wired in parallel with the bumper solenoids. Control Board assembly First, make sure you have a good couple of hours spare and are armed with a cup of your favourite beverage. Construction is not hard, but it is a proper task. Get all the parts together and ready. Follow the plan, and if you stop, make sure you stop somewhere sensible so you can pick up in an orderly manner. The Control Board PCB overlay is shown in Fig.16 (at actual size, but split across two pages). Start by fitting all the resistors. There are only a few values, which is a mercy, but there are a lot of the 1kW, 220W, 150W and 82W parts (plus just a few of the 100W, 2.2W and 2.7W). We mounted them all in the marked sections on the PCB, one value at a time. An old trick when loading a lot of parts is to get a sheet of packing foam or similar. Once you have bent the leads of the parts and inserted them through the pads, put this on top of the board, then holding the foam to the PCB, you can flip it over knowing all your parts won’t fall out. Another cheeky trick if you find yourself stuck is to solder some parts from the top of the board. Next, install the diodes. Start with the 1N4148s, which are the most numerous (64), and watch their polarities as they don’t all face the same way. After that, solder the larger 1N4004s, again paying attention to their orientation (don’t forget D1, all by itself next to the Pico mounting position). Fig.14: the Switch Input Board. This connects up to six regular switches and two inductive sensors. 58 Silicon Chip Australia's electronics magazine siliconchip.com.au Fig.12: the Rollover Board (left and below) connects up to eight inductive sensors. Fig.13: the Bumper Driver Board PCB overlay is shown directly above, and its respective circuit diagram directly below. It provides an extra high-current channel allowing for three bumpers and two kickers to be connected via a single board. Now is a good time to solder all the ICs to the board. There are a few things to watch out for here. First, don’t get the 74HC165 and 74HC595 chips mixed up, as they come in the same package but have different functions. Second, double-check the orientation of each part before soldering it! It’s really annoying to fix a rotated chip and usually involves destroying it to get it off the board safely. Third, we suggest you solder chips directly to the board rather than use sockets, as sockets can oxidise and become a source of unreliability. Also, since these are fairly sturdy logic chips with lots of protection, they are unlikely to fail, and sockets are an additional cost. Still, if you want to use sockets, you certainly can. Fig.15: the General Input Board connects up to eight switches to the Control Board. siliconchip.com.au Australia's electronics magazine July 2026  59 Next, load all the 100nF capacitors. We used (and suggest you use) ceramic capacitors as they are better for bypassing digital ICs. However, if you have a box full of 100nF greencaps or MKTs, you certainly could use them. Now mount all the terminal block sockets and headers, including the box headers. All connectors are keyed; be careful to get them in the right way around, as we rely on the ribbon cables to simplify a lot of wiring and you don’t want them back to front. The notch is marked on the silkscreen. We have specified boxed connectors for the 10-way cables so that as long as you mount the headers the right way around, it should be almost impossible to break anything no matter where you plug things in. Loading the transistors is easiest if you mount them close to the board and solder all the outside legs from the top of the board. This gives you plenty of room to get your soldering iron in, then flip the board, solder the outside pins on the bottom, then snip off the outside legs before soldering the middle pin. Now you can install the pushbutton switch and LED next to it, followed by all remaining capacitors. Mount all the Mosfets, making sure you do not short any of the tabs together; heatsinks are not required. Follow with the volume potentiometer. Now fit two SIL header strips to the Raspberry Pi Pico 2. These can be cut or snapped from a longer header (eg, snap a 40-pin header in half to get two 20-pin headers). While you could solder the Pico 2 directly to the board for reliability (making sure you get it the right way around!), we recommend that you instead solder header sockets to the board, allowing it to be unplugged if necessary. Phew, you are finished. Sit back and behold that Control Board, straight from the 1970s, except for the Raspberry Pi board, of course! You will need a power supply before you can test it, though... Power Supply The locations of all the components on the Power Supply Board are shown in Fig.17 overleaf. Fit the resistors, then the diodes (watch the orientations), followed by the fuse clips and fuses, then the capacitors. Next, 60 Silicon Chip Australia's electronics magazine siliconchip.com.au solder the output socket in place, double checking that you have it the right way around. The regulators do not need heatsinks the way we use them. If yours came with five pins side-by-side, use fine-nose pliers to crank them out to match the PCB pad pattern, then solder them in place with the tabs orientated as shown in the overlay diagram. Power supply testing Check that all the components are on the Power Supply Board, including the fuses; then you are ready for testing. Your power supply doesn’t need to be rated at exactly 24V DC, but lower voltages will result in less ‘oomph’ for the electromechanical parts. It does need to be able to deliver at least 5A. We have not tried voltages above 24V, but it should work up to about 30V, with 19V being about the lowest we would bother with. We encourage you to look in your drawer of outdated laptop and docking station power supplies, as these are usually 19-21V at a very high current. To test the board, apply 24V DC or thereabouts to either of the input connectors. There is no reverse polarity protection, so take care with your wiring! Check the 3.3V output; it should be between 3.1V and 3.4V. Similarly, check the 5V DC output; it should be between 4.7V and 5.1V. If either voltage is too high, our prototypes regulated fine with no load, but it’s possible yours needs a load on it, so add a 100W resistor across that output and check again. If it’s still too high (or too low), there’s something wrong with the board, so switch off the power and check it carefully for dry joints, short circuits, incorrectly orientated or wrong-value components. Check the orientations of the diodes and verify that the LM2576s are -ADJ versions. Also check that your power supply is working properly and that the input voltage is as expected with the board powered up. Fig.16: the Control Board is substantial, in part because it completely avoids the use of surfacemount parts. A lot of the work is simply in fitting the many resistors and diodes, but once you’ve done that, the rest is not too bad. siliconchip.com.au July 2026  61 At this point, you should have a functional power supply and are almost ready to test the Control Board. However, it will be much easier to test it once you’ve built some of the various LED and breakout boards. Remaining board construction Photo 4: this photo (from last month) gives an idea of what the wiring is like on the underside of the Pinball Machine. 62 Silicon Chip Australia's electronics magazine These boards are all pretty simple to build, so we will just give some brief notes for each. Use the overlay and circuit diagrams, Figs.6 to 15, as a guide to assemble them. Importantly, it will also help to have an idea at this stage of how many of each board you will need, which depends on your intended deck layout (if it’s similar to ours, you can stick to our suggestions). Even if you don’t know, it’s pretty safe to build one of each for now. You will almost certainly need more than one Bumper LED Board, as bumpers are a staple of a good pinball game. Most of the smaller boards have all the components on one side. The exceptions are the Bumper and Cascade LED Boards, which have the connectors on the back, and the Player and Score LED Boards, which have the displays on the front and everything else on the back. When building the High-Current Interface Board, watch the orientations of the pluggable headers. These need to accommodate the screw terminals, as shown on the overlay, and be in the same orientation as on the Control Board, or the diode will short out the controller output and most likely blow a fuse. For the Player and Score Boards, make sure you use common-anode 7-segment displays; common-­cathode types will not work. Also, get the header the right way around. For those boards with LEDs or transistors, be careful to orientate them as shown in the diagrams. That also goes for all the connectors, including the pluggable terminal blocks. For those boards with LEDs, you may need to solder them on extended leads to fit the deck; if you’re unsure, solder them with maximum lead length. Once these are installed under the deck, some extra lead length is not a problem, and makes installation easier. Keep the heights consistent regardless. Also consider that you may want to use different colour LEDs in some places. We used a mix of red and white; there’s nothing stopping you siliconchip.com.au Fig.17: the power supply layout. You will note that the output connectors are in a different position than shown in the photos, so that a 6-way connector can be used to match the Control Board. from using other colours if it suits your build, just make sure they are high-brightness LEDs. The parts lists last month includes everything you need to build the boards, but nothing to mount them. We’ll get to that later when we start assembling the Machine. You’ll probably need big bags of M3 machine screws and tapped spacers, although we will also describe the 3D-printed mounts we used in our build. Controller testing The first step is to load the software onto the Pico 2; you can download it from siliconchip.au/Shop/6/3628 Once you have the ZIP, locate the UF2 file within and extract it. With the Pico 2 unplugged from the Control Board, hold the BOOTSEL button on it while plugging the cable into your computer using a USB data cable. BOOTSEL is the small-surface mount button on the Pico 2. It will appear as a removable drive on your computer. Drag-and-drop the UF2 file onto that drive. It will copy and, after a few seconds, the drive should disappear and the Pico 2 will reboot. Before plugging the Pico 2 into the Control Board, make sure the Power Supply Board is producing the right voltages. Wire up the two 6-way pluggable terminal blocks to each other, ensuring the correct voltage is applied to each input on the Control Board siliconchip.com.au with the right polarity. Double-check this before applying power! Apply power and check that all three supply rails are still delivering the correct voltages. If not, check that the electrolytic capacitors on the Control Board are the right way around, and poke around for anything getting hot. Without the Pico 2 plugged in, not much should be happening. Now remove the power, wait a few seconds, plug in the programmed Raspberry Pi (the right way around!) and apply power again. Check that the heartbeat LED next to the Pico 2 is blinking. If it is not, the Pico 2 is either not programmed or is very unhappy. Check the 3.3V and 5V rails; if they are OK and the LED is not blinking, unplug the Pico 2 from the controller board and power it using a USB cable. The LED on the Pico 2 should blink. If not, the Pico 2 is not programmed. If it does not blink, there is a fault on the Control Board (could the LED be reversed?). Assuming it’s ‘alive’, now it’s time to make a few 10-pin ribbon cables with IDC plugs on each end. Make them long enough to be reused in the machine as you build it later. Check that the triangle on each connector indicating pin 1 aligns with the redstriped part of the cable, or at least that it points to the same side of the cable at each end – see Fig.18. It’s important when crimping the IDC connectors that you use enough force to compress the connector so the blades slice fully through the insulation and contact the wires within, but not so much that you break the plastic. It’s a tricky balancing act, but it helps to apply force evenly across the top of the connector during crimping. Now, using five such cables, join the Score and Player displays to the Control Board. Connect CON107 on the Player display to CON5 and CON101104 on the Score Board to CON6CON9, respectively. Apply power and you should see “SC Pin Ball” scroll across the four display segments on the left a couple of times. If that does not happen at all, check the data and control lines, especially SER and SRCLK. These lines run to most of the ICs on the Control Board, so a failure anywhere (like a short circuit) could stop the whole thing from working. Look for solder bridges on the Pico 2 connections. Also check your cables, which are easily overlooked culprits. Check pins 9 and 10 especially carefully if no LEDs work. If only some displays or segments are working, look for the control lines having a problem, including those that go to pin 12 on IC5 through IC9 (L_PLAYER & L_DIGITS0-3). The most likely cause would be in the soldering or an improperly crimped ribbon cable. We use the score display in the SelfTest mode, so you will need to get it working before proceeding. Fig.18: here’s how to crimp the ribbon cables. Most IDC connectors come with strain reliefs like this and are in three pieces. If yours lack that, only having two pieces, the cable can just pass straight through. Pin 1 (red stripe) is usually also marked with a moulded triangle on each connector Australia's electronics magazine July 2026  63 Screen 1: while we used PuTTY, you could use any serial terminal program to monitor the Control Board’s output in SelfTest mode. You will see a scrolling status list in the window. Your COM port will differ; check in Device Manager to see which port you should use. Screen 2: any input that is active (pulled low) is reported by name in this mode. Normally, you make a single input active at a time and check that it was correctly detected. Here, we have intentionally pulled Bumper 1 and Kicker 1 to ground. Assuming the display works, hold the Self-Test button (S1) on the Control Board while powering the system up. This puts the controller into test mode. It provides serial data via a USB serial port regarding its status and relevant data for a series of tests for the inputs, LED outputs and power outputs. You really should use this for all testing and debugging. In Self-Test mode, data is also written to the score display, but considering we have four display digits to work with, the output is pretty brief. The controller emulates a serial connection, so all you need to do is plug it into a computer and run a terminal like PuTTY or Tera Term Pro to display what the controller is sending. On a Windows PC, you can go into Device Manager and look at which serial port your computer has assigned to the Pico 2. This varies – search for and open Device Manager, then look at Ports (COM & LPT), which will show you the serial port number. Tera Term also shows the available ports and their names when you launch it. We ran PuTTY on our computer, clicked on serial connection and then “Open”. This opens a window that prints out data on the serial port – see Screen 1. In test mode, you will get relevant data sent out about once per second, depending on what test you are running. You will know that you are in Self-Test mode as the heartbeat LED does not blink. Instead, data is sent on the serial line and to the Score display. Input tests Screen 3: testing the LED ports one at a time. Screen 4: testing the high-current (‘power’) output ports, one at a time. 64 Silicon Chip Australia's electronics magazine The first test is reading the inputs. This will present data to you on the serial console as shown in Screen 2. Simultaneously, the Score display will scroll the value of the inputs in two four-character HEX numbers. These are arranged as MSB first, LSB last. There is a space in between to let you see the two words. We have inverted the logic in the display, so 1 means the input is active, ie, pulled low. You can stimulate each input to the controller by shorting that input to ground and you should see the corresponding input value change. We used the General Input PCB for this, as it breaks out all eight inputs run from the ribbon cable to eight two-pin headers. We shorted each in turn with siliconchip.com.au a screwdriver. You could also use a jumper shunt. A quick-and-dirty test is to short each input in turn and look for the reported input state changing. Even if you don’t bother decoding the input, if the value changes each time you short a different input, it’s likely that all the inputs are working. If an input does not work, check the associated input soldering, resistor, cabling and diodes. If a whole input bank does not work, check that IC’s soldering. Before desoldering chips, check the soldering and the control lines carefully, as getting those ICs off the board realistically requires a hot air gun or snipping every lead off and individually desoldering them. Output tests Once you are satisfied that all the inputs are working, press and hold the Self-Test button for a second or so to progress to the output tests. A quick press might not work because of how the software works in test mode. A message will be presented on the serial output and also the score display. Press the Self-Test button again; this makes the Player LED port blink on and off at 1Hz. A second press runs each LED in series for that port. The serial report on your computer should show something like Screen 3. Repeat this test for the four Score board headers, the three Bumper headers, the two Pattern LED headers; then the Rollover LEDs, Target LEDs and General LEDs (those three are on one header each). If a whole port does not work, debug the control lines and check your soldering, especially around the associated IC. If individual LEDs do not work, look from the output IC through the transistor to the output connector. Power Output tests Next come the power outputs. These tests make the output active for 100ms, then off for 600ms. We run the test this way, as if you have a solenoid connected, it will be driven very hard, and we do not want to leave power applied for an indefinite period. If you have the serial port connected still, you will see these states reported on the terminal, as shown in Screen 4. The next three button presses will start bumpers 1-3 pulsing on and off. Connecting a bumper or LED with a 1kW series resistor to see these operate. siliconchip.com.au Photo 9: this photo gives you an idea of the size of the Pinball Machine. It stands 153cm tall (71cm for the legs alone), 60cm wide and 112cm deep. After that are the two kickers, then the left and right flippers. The eighth press activates the second kicker, although I didn’t use that in my machine. A ninth press triggers the ball load solenoid, then after that the ball release solenoid. The bell comes next; this operates at 50% duty cycle since we use a 12V bell, and if you run this test with the bell plugged in, we don’t want to melt it. If all of these fail, then we need to look at the SER and SRCLK lines (although if you got this far, surely they are OK). If the first eight or second four fail but not all, look at IC22-IC23 plus the L_Power0 and L_Power_1 lines. If an individual output fails, check Australia's electronics magazine for soldering problems and wiring problems, especially around the Mosfets. These Mosfets are pretty chunky, so it is very unlikely that they will fail if you have the diodes installed correctly. Now plug the speaker into CON10 and turn up the volume to a moderate level. Apply power and listen for a tune at start-up. If this fails, look around the power amplifier and volume pot. Next month You should now have the controller up and running. Next month we will start describing how to 3D-print and assemble parts like the bumpers, kickSC ers, targets and flippers. July 2026  65 Feature By Andrew Woodfield Making simple & good-loºking boxes If you are anything like me, you make all sorts of electronic stuff. When I show off my latest creation, an attractive-looking device with a few pretty lights always gets a more positive response than my rough, lashed-up designs on unetched PCBs. T he final step of creating a new electronic device typically involves finding a box to suit it as it’s nearing completion. It’s a Goldilocks moment, except that I usually find absolutely nothing is quite the right size. I might buy something, only to change the design and then discover that it no longer fits in that expensive box! Frequently, I just need a simple box. One well-known approach is to lasercut something from thin birch ply (see Photo 1 above). You can also use this approach with acrylic sheets. Designing and fabricating these boxes is easy, particularly with help from the many free online design tools. These types of boxes are instantly recognisable; it’s the dark and light crenulated edges that are the giveaway. Frankly, that appearance is sometimes seen as less than ‘professional’, even if the electronics within are a marvel of design. An alternative is 3D printing. This approach is great in many cases, especially for unusually shaped prototypes. However, 3D prints can take an appreciable amount of time to design, and often even more time to print. I recently needed to make several boxes, the first for a version of an HF receiver, and the second for a small, short-range, low-power SSB transmitter. In both cases, I was looking for a quick solution while also delivering an attractive appearance. The receiver I was testing was built by Charles Kosina (April 2026 issue; siliconchip.au/Article/20079). The electronics are all mounted on a PCBtype front panel. It is designed to allow the receiver to be mounted in an offthe-shelf plastic box. However, try as I might, I couldn’t find that box or a suitable equivalent locally. Faced with this problem, and wanting to get the receiver into a box quickly to allow testing, I looked for other options. Solution one The first solution I came up with, Fig.1: the cross-section of the 3D-printed rails used in the first of these simple enclosures. 66 Silicon Chip Australia's electronics magazine shown in Photo 2, was surprisingly simple. In this approach, four corner rails are 3D-printed in lengths (and colours) of your choice. For this first enclosure, I made my rails 50mm long. In hindsight, they were probably about 10mm too short, and things got a bit tight inside. I slotted in a set of small rectangular ply panels cut from 2mm basswood ply sheets. That shape, along with the selected material, permits very fast and easy cutting. Being simple rectangles, they can be cut by hand with a metal ruler and a sharp craft knife or, as I did, with a laser cutter. One feature of this type of enclosure is that the burnt laser-cut edges are completely hidden. While the ply thickness was quoted as 2mm according to the label, the material actually measured 1.6mm, so that was the thickness I used in the corner rail design. I double-­checked this again on several other sheets from another retailer, and I found the same result. A third retailer, however, had 2mm sheets that were 2.1mm thick. Clearly, some variation exists in the industry. The cross-section of these rails along with a 3D view is shown in Fig.1. The holes were dimensioned to suit M3 hardware. With the 6.5mm thickness from the rails on each panel edge plus 0.5mm for the materials, a total allowance of 14mm is used when designing the panels. However, the back panel of the case requires no such allowance, and thus siliconchip.com.au ▶ Photo 2 (left): Charles Kosina’s HF receiver needed an alternative enclosure solution prior to bench testing when the specified enclosure could not be located locally. Photo 1 (lead): while easy to design and make, the appearance of many lasercut boxes does not reflect the excellence of the electronics inside. has identical dimensions to that of the front panel. The HF receiver’s front panel measured 160 × 65mm. I cut the following panels directly from the “2mm” ply using a laser cutter: ∎ Top & bottom panels: 2 <at> 146 × 50mm ∎ Left & right panels: 2 <at> 51 × 50mm ∎ Back panel: 1 <at> 160 × 65mm I used PC-based drawing software to create the outline of the back panel. This allowed me to precisely dimension and locate the holes for the DC power supply and the RCA phono-type antenna connector that I used with the receiver. The next version will add a 3.5mm speaker connector and an accessory connector for the optional colour LCD screen. I used CorelDraw to design this panel and to convert it to a suitable format for my laser cutter. However, there are many possible design packages, including free ones (eg, Inkscape). At a pinch, the drawing functions in standard office productivity software could also be used. During assembly, I gently sanded the edges of the panels that were inserted into the corner rails. While the drawing (Fig.1) shows sharp right-angle corners in the base of the slots, when 3D printed, there was a very slight rounding on the corners at the bottom of the slots. Sanding the edges with 180-grit sandpaper very lightly allowed the panels to fit precisely into place. I initially planned to use 6mm-long self-tapping cheesehead screws to hold everything in place. However, siliconchip.com.au Photo 3 (above): a smaller enclosure with curved edges. my elderly 3D printer ended up producing slightly undersized holes in the rails. I found some M3 hex bolts in my parts bin instead and tapped the holes in the rails to match. Incidentally, I designed and 3D-printed all the front panel knobs. This provided a neat overall appearance but, more importantly, it allowed me to use a set of knobs that fitted nicely into the available panel space. A second option I then designed a second smaller enclosure for a very small low-power SSB transmitter, this time an enclosure using rails with curved edges. The box, shown in Photo 3, measured 120 × 25 × 55mm. The small transmitter PCB and a 9V battery fit inside this little box. The new rails I designed for this enclosure were slightly smaller and subsequently faster to print (see Fig.2). The panels were equally fast to cut, again from 2mm ply, and the assembly was completed in just a few minutes. This time, I used 10mm-long self-­ tapping screws. While smaller overall, a total allowance of 10mm is used with the panels and these rails. For this enclosure with its 120 × 25mm front and back panels, the dimensions of the remaining ply panels were: ∎ Top & bottom panels: 2 <at> 110 × 55mm ∎ Left & right panels: 2 <at> 15 × 55mm Available files The industry standard STL files for the two rail styles described here are available for download from the Silicon Chip website at siliconchip.au/ Shop/6/3577 These can be lengthened or shortened using settings in most 3D ‘slicers’. These slicers drive the 3D printer based on the content of the STL files. Just prior to printing the file, the user can adjust several features of the final output, including scaling each axis. Fig.2: the rail cross-section used in Photo 3. Australia's electronics magazine July 2026  67 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 68 Silicon Chip Alternative panel materials The files corresponding to Figs.1 & 2 can also be used with panels cut from 1mm-thick (18 gauge) aluminium sheets available from several electronics retailers. I’ve also included two additional STL rails files for use with 0.5mm-thick (24 gauge) aluminium panels. These are more commonly found at building supply retailers. The 0.5mm sheets can be cut with heavy-duty domestic scissors in situations where access to a metal guillotine is not available. The front and back panels in either of the examples here can also be replaced by aluminium panels without any changes to other parts of the enclosures. The panels may also be made from thick card in cases where designers are uncertain about the final size or placement of parts on the enclosure (and we’ve all been there!). This is also an option for those on a limited budget. By applying inkjet-printed paper labels covered with self-­ a dhesive transparent film to the front and rear panels, a fairly robust and quite professional finish can be achieved. If panels are made from PCB material, these are usually 1.5-1.6mm thick and therefore nicely match the first set of rails. Adhesives Once all the panels are slotted and screwed together, they will all stay locked in place without the need for any adhesive. During prototyping, when you may wish to leave some of the panels off, such as the top cover and back panel, it may be necessary to use a drop of hot glue here and there to keep everything in place. If the panel thickness varies significantly from the slot width, a little adhesive may be required to prevent the panels from vibrating, particularly if that panel is used for mounting a speaker. Different enclosure sizes The enclosure dimensions can be altered to suit individual requirements. Just keep in mind the allowance required for the rails when dimensioning the panels. Let’s use the curved rails to demonstrate this. As designed in the STL files, these are 55mm long. You may, for example, want an enclosure that measures 60 × 35 × 50mm, as shown in Fig.3. The rails can be printed at a Australia's electronics magazine Fig.3: an example of a differentlysized enclosure. Z-scale of 50/55 or 91% to produce rails with the same cross-section but lengths of 50mm. Remember that these rails add 5mm to the left, right, top and bottom edges of the ply panels. So, for this new enclosure with its 60 × 35mm front and back panels, the dimensions of the remaining panels would be: ∎ Top & bottom panels: 2 <at> 50 × 50mm ∎ Left & right panels: 2 <at> 25 × 50mm 3D printing & laser-cutting services 3D printer and laser cutting services are now available for those who do not own such equipment. The first place to check is local public libraries. Over the past decade, many have begun hosting technology services, and facilities often include this type of equipment. Some also provide classes, and many provide a maker service, offering printing and cutting for a very modest fee. In the case of our city libraries, the charges for 3D printing are usually based on the item’s weight in addition to a small setup fee, and availability of the finished item is often very prompt. Failing this, some companies also provide a fee-based printing and cutting service, although these are likely to be slightly more expensive. Final comments This simple, fast, and low-cost approach to making enclosures has allowed me to dramatically improve the final appearance of some of my designs. It delivers a modern professional finish and, importantly, avoids the often distracting appearance of dark and light laser-cut edges on these boxes. It’s also very easy to modify the basic components presented here to cater for a wide range of requirements. Armed with this approach, I feel this solution is certain to also find a use around your SC workshop, too. siliconchip.com.au 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. 07/26 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 ATtiny45-20PU ATtiny85-20PU 2m VHF CW/FM Test Generator (Oct23) Graphing Thermometer (Mar26), Simple LC Meter (May26) Simple USB Power Monitor (Jun26) 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) 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 DCC ACCESSORY DECODERS (JUL 26) I2C CONTROLLER COMPLETE KIT (SC7690) (JUL 26) HUMAN COMFORT INDICATOR (SC7646) (JUN 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 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) $40.00 $40.00 siliconchip.com.au/Shop/ DCC BOOSTER / REVERSE LOOP CONTROLLER KIT (SC7579) (MAR 26) 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) $30.00 DCC REMOTE CONTROLLER KIT (SC7552) (FEB 26) $60.00 $12.50 $10.00 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) PINBALL MACHINE KITS $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 (JUN 26) Includes everything but the plastic case, power supply and some optional parts. Control Board (SC7659): includes the PCB and all non-optional onboard parts $150.00 The Pico 2 is supplied but not programmed (see p39, Jan26) $90.00 Power Supply (SC7680): includes the PCB and all onboard parts $50.00 RGB LED STAR KIT (SC7535) (DEC 25) Cable & Connector Set (SC7681): includes 17 10-pin box headers, 34 10-pin IDC Includes the mostly-assembled board and all non-optional components connectors, 10m of 10-way ribbon cable, 30 2-way pluggable terminal blocks $80.00 and 20 2-way polarised headers $65.00 except the power supply (see p43, Dec25) SIMPLE USB POWER MONITOR (SC7683) (JUN 26) 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 STEPPER MOTOR DRIVER KIT (SC7601) (APR 26) CALLIOPE AMPLIFIER PARTS (SC6021) (APR 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 $35.00 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) (SEP 25) RP2350B DEVELOPMENT BOARD (AUG 25) Includes all parts except the jumper wire and glue (see p39, Sep25) Assembled Board: a pre-assembled PCB with all mandatory parts fitted, optional components are sold separately below (SC7514; see p49, Aug25) - 40-pin header (two are required, SC3189) $15.00 *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 $90.00 $7.50 $5.00 $20.00 $30.00 $1.00ea By Tim Blythman μDCC Decoder Accessory Decoder I2C Controller DCC Accessory Decoders Destination Display Our previous DCC (Digital Command Control) projects can control multiple trains in a model railway, but what about fixed devices such as points (turnouts or switches) and signals? You need Image source: https://unsplash.com/photos/ an accessory decoder; we describe two suitable circuits. a-model-train-set-with-a-red-caboose-iP9kBOECD2U W e have previously published points motor controllers and signal controllers, including a design that interfaces with servo motors – see Circuit Notebook, December 2020 (siliconchip.au/Article/14682). Les Kerr’s past designs include a semaphore signal operated by a servo motor from April 2022 (siliconchip. au/Article/15273) and a points motor controller for snap-type motors in the February 2024 issue (siliconchip.au/ Article/16132). A DCC decoder that can interface with points motors and servo motors allows these devices to be integrated into a DCC system. Let’s have a quick look at the types of accessory devices that might be found on a model railway and how our Accessory Decoders can work with them. We’ll assume you have some experience with model railways. Points motors The most common points motors fall into two main categories. The first is a solenoid or snap-type motor, where the mechanism is actuated by one or more coils. Peco-brand motors, as used by Les, are common and have two coils: a pulse on one coil sets the points for the straight, while a pulse on the other sets the points for the curved track. Other designs have one coil and depend on reversing its drive polarity to change the points. It is possible to convert between different arrangements with cleverly connected diodes. The snap-type motors may come with two, three or four wiring connections. DCC PROJECT KITS Snap-type Accessory Decoder (SC7685, $40) includes the PCB and all onboard parts, including the electrolytic capacitor Servo-type Accessory Decoder (SC7686, $40) includes the PCB and all required onboard parts I2C Controller (SC7690, $30) includes the PCB and all other parts 70 Silicon Chip Australia's electronics magazine Another type is known as a slow-­ motion or stall motor. These are simply brushed DC motors driving a gearbox to slowly move a linkage, and are thus more like their full-size counterparts. As the name suggests, the motors simply stall at their endpoints if they are continuously driven. Better units are designed to handle continuous stalling and may have extra dry switch contacts to provide feedback or control other devices, such as signals. As you might expect, the current requirements for these motors are quite different, with stall motors needing perhaps tens of milliamperes, while the snap-type motors might draw a few amps for a fraction of a second. Our Snap-type Accessory Decoder will work with both of these motor types. Each output consists of a DRV8231 full-bridge motor driver IC, like we used in the DCC Locomotive Decoder earlier in this series. Driving the DC motors in a slow-motion point motor or the coils of a snap-type motor is trivial with this chip. The DRV8231 can be operated as two open-drain outputs, so we also provide a connector to the supply voltage, which becomes the common siliconchip.com.au Fig.1: the Snap-type Decoder uses a simple linear regulator to power its low-voltage circuitry. The 4700μF capacitor provides a reservoir for bursts of current to drive solenoidbased motors. connection when used with three-wire motors. Two-wire point motors simply use the two motor outputs. Signal lights We have also provided a mode that configures the Decoder as four opendrain outputs (with independent controls), so it could be used to operate simple on-off devices like signals or layout lighting. A basic application could wire a red and green bi-colour LED (with an appropriate series resistor) across the motor output to show a different colour depending on the polarity of the output. Independent LEDs (wired with a common anode) or lamps could also use the open-drain outputs. Servo motors Servo motors simply require power and a digital pulse signal. They will move to a position determined by the pulse width. Les’ project shows how a servo motor can be used to operate a semaphore signal arm. It could also be applied to things like level-­crossing boom gates. Commercial suppliers such as Peco are now selling servo motors and brackets that allow their points to be siliconchip.com.au driven by a servo motor. Our Servo-­ type Accessory Decoder simply has four servo motor outputs that are suitable for small hobby servos. It can set each output to be one of two adjustable pulse widths to toggle between two different positions. Circuit details Our original plan combined features of both Decoders into a single circuit, but we figured it would be simpler to create two distinct variants. Thus, there are two boards with much in common; we will start by explaining the common features. Fig.1 shows the circuit for the Snap-type Decoder, while Fig.2 is the Servo-­type Decoder. CON5 is the input for the DCC track voltage, while diodes D1-D4 form a bridge rectifier to create a DC rail, which we’ve labelled as a nominal Features & Specifications 🛤 Separate DC input to allow higher voltage for motor operation (up to 24V) 🛤 Pushbuttons to allow use without DCC 🛤 Headers to allow remote mounting of pushbutton controls 🛤 Programmable running time set by DCC CVs 🛤 Can interface with the I2C Controller (described in this issue) 🛤 Simple address programming Snap-type Accessory Decoder Two bipolar high-current outputs suitable for driving snap-type and slow-motion point motors Can be configured with four open-drain outputs instead Servo-type Accessory Decoder Four servo outputs 5.3V 1A switchmode power supply for servos from DCC input Two independent programmable servo positions per output, set by DCC CVs 🛤 🛤 🛤 🛤 🛤 Australia's electronics magazine July 2026  71 12V here (it could be lower or, more likely, higher). CON6 leads to a single diode D5 that can be used to connect a different source of DC power; effectively, it is diode-ORed with the supply from CON5. Our locomotive decoders can work up to about 17V, but both Accessory Decoders can operate with inputs up to at least 24V. 12V is a common voltage for HO and N scale operation, so the CON6 input allows a higher voltage for accessories operation without having to run the booster at a different voltage. Many snap-type point motors suggest a 16V minimum operating voltage. IC1 is a 20-pin 8-bit microcontroller in both cases, with a nominal 5V supply bypassed at its pins 1 and 20. These, along with pins 4, 18 & 19, connect to the ICSP (in-circuit serial programming) header, CON8. The 10kW resistor pulls up pin 4 for normal operation. CON7 is the connection for the I2C Controller that can be used to easily program the Accessory Decoders. The two 10kW resistors are pullups for the I2C bus that the I2C Controller uses. The 3.3V pullup is from an I/O pin on IC1; this pin can be directed to an internal DAC that can source or sink up to 20mA at an internally set voltage, so it is an easy way to get a suitable voltage at adequate current without needing external components. The two LEDs are provided with a series resistor, with LED2 driven by one of IC1’s digital output pins. LED1 is driven from a different power rail on each Decoder; from the motor supply on the Snap-type Decoder and from the 5V rail on the Servo-type Decoder. The 5V rail here is derived from the 5.3V rail used to power the servo motors. Thus, they show the health of the respective power supplies. The other common items in both circuits are the pushbutton switches (S1-S4 or S1-S2) and jumper shunt (JP2 or JP3). These are simply connected to digital input pins on IC1. The pins are configured with internal pull-up currents to allow detection of the switch or jumper state. The respective CON9s simply break out the switch connections so that the switches can be remote if preferred. These inputs are used in different ways for manually controlling the Accessory Decoder or programming its CVs (configuration variables) to customise its operation. Snap-type Decoder specifics In this Decoder, the 5V power rail for IC1 is provided from a simple 78L05 linear regulator (REG1), with 100μF capacitors on its input and output. The current requirements for the 5V rail are expected to be no more than 20mA, even with the I2C Controller connected, so this widely available part will be fine. IC2 and IC3 are the motor driver ICs described earlier. Their supply (Vmotor) is bypassed by a 4700μF capacitor that is charged from the 12V rail via a 1W 100W resistor. The large capacitor allows brief bursts of high current to be provided to the motor drivers, while Fig.2: the Servo-type Decoder has a switchmode supply to provide ample current to power servo motors. The low-voltage circuitry (such as the microcontroller) is powered via a diode from this supply. 72 Silicon Chip Australia's electronics magazine siliconchip.com.au the resistor limits the rate at which the capacitor charges between uses; the charge time is around one second. Since LED2 is connected across this capacitor, it will show point motor activity by dimming briefly. Two 10kW/1kW dividers with 100nF smoothing capacitors monitor the voltage upstream and downstream of the resistor. Thus IC1 can detect when the 4700μF capacitor is fully charged. IC2 and IC3 each have a 1μF local bypass capacitor and have 3.3V supplied to their Vref pins by the DAC output noted earlier. The 0.1W shunt resistors on the Isen pins set the current limit for IC2 and IC3 to 3.3A, just below their 3.7A maximum. The pairs of IN1 and IN2 pins are driven by IC1 to control IC2’s and IC3’s outputs, which connect to CON1 and CON2, respectively. For cases such as slow-motion point motors where lower loads are driven and the burst capability is not needed, the 100W resistor could be replaced by a link and the 4700μF capacitor reduced in value to, say, 100μF. Servo-type Decoder specifics With most hobby-type servo motors operating at around 6V and typically drawing a few hundred milliamperes, we need something more capable than a 78L05 to provide the low-voltage rail on this variant. REG1 is an MCP16311 switching regulator that’s used instead. It is configured for an output of 5.3V and the circuit here is much the same as that used on the DCC Base Station. The Servo-type Decoder (the board shown directly to the right) could be used for other applications such as level crossing booms and semaphore signal arms. siliconchip.com.au Table 1: Accessory Decoder CVs CV number Purpose Default Notes 1 (513) Low address byte 1, 2, 3, 4 All values (0-255) are valid. 3 (515) Output duration 25 In steps of 10ms; the default gives 0.25s and the maximum value of 255 gives 2.55s. A value of 0 gives an unlimited duration. 9 (521) High address byte 0 Only three bits are valid (ie, values 0-7). 33 (545) Servo time thrown 100 Servo pulse length in steps of 10μs (100 = 1ms). 34 (546) Servo time closed 200 Servo pulse length in steps of 10μs (200 = 2ms). 35 (547) Number of outputs available 2 (Snap) 4 (Servo) When the Snap decoder’s CV35 is set to four, there are four open-drain outputs available. This only has an effect on the first decoder output of each board. Apart from the SMD parts that surround it, it has a 100μF capacitor on its input and output. The power rail for the microcontroller is supplied from the 5.3V rail via schottky diode D6, giving close enough to 5V; this rail also has a 100μF capacitor for bypassing. The diode ensures that bursts of current from the motors do not cause brownouts on the microcontroller. LED1 is fed from this 5V supply, so it should be steadily lit. Since servo motors are quite simple to control, the remaining circuitry is straightforward: the four servo connections at CON1-CON4 consist of threeway headers with ground, 5.3V and a digitally generated signal from IC1. The four 470W resistors help to isolate IC1’s I/O pins from any noise or surges from the servo motors. These two compact boards allow you to control different points motor types in a DCC model railway. The Snaptype Decoder (the boards shown above and to the left) can also be used to control lights, such as signals. Here we’ve shown it with and without the 4700μF capacitor attached. DCC accessory decoder details We should briefly explain some of the terminology related to DCC accessory decoders. Accessory decoders might also be called stationary decoders, in contrast to the mobile decoders found in locomotives and the like. Accessory decoders have a separate addressing scheme to mobile decoders, so locomotive #1 and accessory decoder #1 are distinct and will not be confused. The packet structure and data contents are different, too. The current standard promotes a flat addressing system that ranges from 1 to 2048, although earlier standards used a segmented sub-addressing system. We tested our Decoders with a DigiTrax system along with the JMRI software; both work with the linear system, so that is what we are using. Unlike mobile decoders, stationary decoders have just two outputs; these are known as ‘closed’ and ‘thrown’, based on US railway terminology for points set to the straight (or default) route or curved (non-default) route, respectively. The common Australian equivalents are ‘normal’ and ‘reverse’. For example, the DigiTrax unit uses the abbreviations ‘c’ and ‘t’ to describe the outputs, and the JMRI software has buttons labelled “Closed” and “Thrown”. July 2026  73 Each output can be activated, which will deactivate the other if it is active. There is also a duration setting, which determines how long an output is activated; this is the duration of the brief pulse when a snap-type motor is activated. If the duration is set to zero, the output runs indefinitely. This brings us to the CVs (configuration variables) used by accessory decoders. Table 1 shows the CVs that are supported by our Accessory Decoders. CV33, CV34 and CV35 are custom CVs whose purpose is not fixed by the standards. Not all of these CVs are used, but CV3 is commonly used as a duration setting and is available for all outputs. The CV numbers are given as two different addresses (that differ by 512) since these were created under a different numbering scheme, which has also been simplified. We treat the CV numbers the same in software by ignoring the upper bits. Firmware operation Like the locomotive decoders, these Decoders monitor the DCC signal via 100kW protection resistors. When a relevant packet is detected, it triggers one of the outputs or programs a CV as needed. The Accessory Decoders also check if an I2C Controller is connected and interact with it if it’s present. This Controller has no processor of its own, so the Decoder must provide a display driver and menus for allowing settings to be made. The driver keeps a character buffer, not unlike an older 8-bit personal computer. It updates one character at a time from the buffer, which Snap-type Decoder assembly The Snaptype Decoder attached to the I2C Controller. takes about 2ms. Since DCC packets take about 5ms to receive, this means it is very unlikely for the Decoder to miss a packet. The Snap-type Decoder can also delay activation of an output if the Vmotor rail line is low from a previous activation. The threshold used is 90% of the 12V rail or 6V absolute minimum. A round-robin counter ensures that only one pulse output is activated at a time. Construction We’ll describe the construction of the two Decoders separately, followed by some common operational features, then the unique aspects of each. How to use the I2C Controller with the Accessory Decoders is described in its separate article. You’ll need SMD assembly gear, since there are a number of SMD parts. This should include flux paste, tweezers, a magnifier and solder-wicking braid. Illumination and ventilation will also help. Fig.3: start by soldering the exposed pads on the undersides of IC2 and IC3; there are large holes in the PCB to allow access from below. There is nothing smaller than SOIC or 1206-size parts on this board. The large capacitor has not been fitted in this photo. 74 Silicon Chip Australia's electronics magazine Make sure that you have the PCB coded 09111254; this and the board for the other Decoder are the same dimensions and have the same mounting holes. Refer to the Fig.3 overlay diagram in this case. Start with IC2 and IC3, since they have large underbody pads that need to be soldered to the PCB. We find it easiest to temporarily hold these parts in place using a high-­ temperature tape, such as Kapton. Make sure the pins are over the correct pads and the part is orientated correctly, then add flux and flow a generous amount of solder through the hole from the underside. Check that the chips are aligned and firmly held by the solder before attempting to solder the remaining pins. Apply flux to the pads for the remaining SMD parts and fit IC1 next, ensuring that its pin 1 aligns with the marker. Tack one lead to start. Check that the remaining pins are aligned with their pads and flat against the PCB before soldering them. Next, solder the SMD capacitors; there are three 100nF parts and two of 1μF. The latter are adjacent to IC2 and IC3. Note that the 100nF marking on the silkscreen near REG1 indicates two parts, one to its left and one to its right. The larger 0.1W resistors are also near IC2 and IC3, so solder these next. The remaining SMD parts are the 11 smaller M3216 (1206) size resistors. Below the pair of 100nF capacitors, there are pairs of 10kW and 1kW resistors as shown. With the SMD parts complete, you clean up any flux residue on the board (eg, using alcohol and a lint-free cloth). Now you can move on to the through-hole parts. The five 1N5819 diodes all face the same way, with their cathodes to the right. Similarly, the two LEDs can be fitted flush against the PCB with their cathodes to the right; LED1 is green and LED2 is red. Snap the two tactile switches into place and solder them. The 100W 1W resistor can be spaced slightly (about 2mm) above the PCB to help with air movement for cooling. Join the three-way screw terminals CON1 and CON2 via the moulded dovetails before slotting them into place. The two-way screw terminals (CON3 and CON4) are fitted separately. Make sure that the cable entries to the terminals are accessible from the edges of the PCB. Next fit REG1 and the two 100μF capacitors near it. Solder the jumper header JP2 in place and leave the shunt off for now. The remaining headers (CON7, CON8 and CON9) might not be needed, so fit those as needed and to suit. Finally, solder the larger capacitor in place. If you are using a 4700μF part, bend the leads over and lay it over the top of the remaining components, towards CON5. If you are using a smaller part (for example, to power slow-motion point machines), it can be fitted vertically. Servo-type Decoder assembly This version is assembled on the PCB coded 09111255 and with the help of overlay diagram Fig.4. Apply flux to the SMD parts and put the tiny MSOP-8 REG1 in place with its pin 1 marker at upper left, near the REG1 designator. Tack one lead, verify that the placement is still good, then solder the remainder. If you get a bridge, add extra flux and use the braid to draw away the excess solder. Follow with IC1; its pin 1 is also at upper left. The remaining SMD parts are passives. There are two 100nF capacitors and three 1μF capacitors, plus 13 resistors. The single inductor might need a bit more heat to solder properly since it is larger and has more thermal mass. It’s best to put some flux paste on its pads before placing it. These components are all individually marked on the PCB; none of them are polarised. Clean off any excess flux siliconchip.com.au * servo wire colours may vary Fig.4: the switchmode regulator is in a tiny MSOP8 package and should be soldered first. You should find the remaining components straightforward after that. Check the pinout of your servo motors before connecting, since we have seen some that use a different pin order. before fitting the through-hole parts and allow the solvent to evaporate. Next, solder the six through-hole diodes and two LEDs. Apart from D6, all these components have cathodes facing to the right of the PCB. D6’s cathode is towards the top of the PCB. Snap in the four tactile switches and solder them, then follow with the three 100μF capacitors. Now mount both two-way screw terminals, CON5 and CON6, making sure that the cable entries face away from the PCB. The jumper header (JP3) and three-way headers (CON1CON4) should be fitted next, followed by CON7, CON8 and CON9 if needed. Leave the shunt off the jumper for now. Microcontroller programming If you have purchased a chip or kit from the Silicon Chip shop, microcontroller IC1 (for both boards) will already be programmed with the correct firmware. Make sure you choose the correct variant at the time of purchase. You can skip forward to the section on testing. To apply power, use CON5 since it is followed by the bridge rectifier (D1D4) and the polarity will not matter. A 9V battery is a fairly safe option and should cause LED1 to light up when connected. Connect a programmer (Snap, PICkit 4, PICkit 5 or PICkit BASIC) to CON8; be sure to align pin 1 (with the > marker) to the matching marker on the programmer and use the Microchip IPE program to upload (program) and verify the appropriate HEX file (see the parts list for the code). Testing The pushbutton controls mean that both boards can be quite thoroughly tested with little more than a suitable Australia's electronics magazine power supply. It doesn’t even need to be a DCC system. If you want to be cautious, try a 9V battery or a current-­ limited (100mA) 12V supply such as a bench PSU. Connect it to CON6, observing the polarity. On either board, LED1 should light up to indicate when power is applied. You can probe for the other expected voltages relative to circuit ground (eg, the – terminal of CON6). You should see 5V (4.9-5.1V) on pin 2 of CON8, the ICSP header. This pin is next to the one marked with a chevron. If all is well, you can now connect a more powerful supply (or your DCC system) and your motors. Fig.5 shows wiring examples for different types of point motors. The most typical connection will be to wire the main DCC track output from a base station (“MAIN” on our DCC Base Station) to CON5. Since a DCC signal is effectively AC, the polarity is not important. Pressing one of S1, S2, S3 or S4 should cause the corresponding (CON1, CON2, CON3 or CON4) output to activate for a quarter of a second. Pressing the same switch a second time should activate the alternate action. For example, repeated presses on S4 should cause a servo motor connected to CON4 to toggle between its two pre-programmed positions. From this, you can see that the Accessory Decoders are quite useful, even without a DCC system connected. Bridging the respective pins on CON9 to ground should have the same effect. DCC operation By default, the DCC accessory addresses correspond to the connectors. So CON1 will respond to address 1, CON2 to address 2 and so forth. For our testing, we used our DigiTrax system and the JMRI software, as well July 2026  75 as our own DCC Base Station. JMRI is an open-source project that works on Windows, macOS and Linux – see www.jmri.org There are many options for hardware to interface JMRI to a layout, including commercial systems that have a computer interface. In our January 2020 DCC project (siliconchip.au/ Article/12220), we used the DCC++ BaseStation sketch. It can be found at https://github.com/DccPlusPlus/­ BaseStation It’s also possible to use a bare Arduino Uno (programmed with the DCC++ BaseStation sketch) to generate logic-­ level DCC signals. We used this to quickly test JMRI’s operation with the Accessory Decoders. To do the same, install JMRI and configure your programmed Uno as the DCC interface by setting the connection name to DCC++ and the serial port to that allocated to the Uno. Fig.6 shows the wiring needed to feed the signal into an Accessory Decoder and also supply it with power. You will need a suitable DC power supply that is capable of sharing ground with the Uno and thus your computer. Either Accessory Decoder can be used this way. The JMRI DecoderPro program provides a few useful windows under the Actions Menu. The Turnout Control window can be seen in Screen 2. Enter the address number (1-4 by default) and then press Thrown or Closed to operate the outputs. When a command is received (that the decoder should respond to), LED2 will flash for 200ms. If the activation is delayed (due to the capacitor charging or round-robin sequencing), LED2 will emit another very brief flash when the output is ultimately activated. The Single CV Programmer (Screen 1) can also be used to set the configuration variables. Set the lower radio button to “Ops Accessory Byte” and the upper radio button to “Accessory Address”. Enter the Accessory Decoder address (lower text box) and fill in the CV and Value fields before pressing the “Write CV” button. We have locked out the ability to program addresses (CV1 and CV9) to ensure that they cannot be inadvertently changed. Since these CVs also need to be accessed through their own address, it can be messy to do it this way. We will discuss how these can be changed shortly. 76 Silicon Chip Fig.5: We found snap-type motors (a & b) to be the most difficult to get working. You may need to adjust the motor and points to ensure that they are moving as freely as possible. Points with integrated motors (c) worked well. Many slowmotion motors (d) have extra switch contacts, which can be ignored or used for other purposes, such as operating signal lights. For loads like LEDs, make sure that the polarity is correct (e). If you have a Servo-type Decoder, we recommend changing CV3 to 0 so that the outputs are always driven. Then the servos will immediately respond to changes in CV33 and CV34 if the output has already been set to thrown or closed, respectively. CV3 should also be 0 for slow-motion motors that can be constantly powered. For a Snap-type Decoder, CV3 should be set long enough to ensure that the points are thrown, but not so long that the coil overheats. The points motor manual should provide guidance on this. If you have your own DCC system, it should have instructions on how to work with accessory decoders. CV35 on the first output only can be changed to configure a Snap-type Decoder to have two full-bridge outputs or four open-drain outputs. The values are 2 for the full-bridge outputs and 4 for the open-drain outputs. The Snap-type Decoder will not accept any other values for CV35. With four open-drain outputs, the outputs follow the numbering shown next to CON1 and CON2, with outputs 1 and 2 coming from CON1 and outputs 3 and 4 coming from CON2. The output is on (sinking current) while the “throw” output is active, so it can be turned off by setting the “close” Screen 1: DecoderPro’s Single CV Programmer (also under the Actions menu) can be used with both mobile and stationary decoders. Select Ops Accessory Byte as the mode and then Accessory address before entering the address, CV and value. Screen 2: JMRI’s DecoderPro program has several tools for interfacing with accessory decoders. The Turnout Control (found in the Actions menu) opens the window here, which can be used to manually operate accessory decoders. Australia's electronics magazine siliconchip.com.au Fig.6: an Arduino Uno can be used to generate logic level DCC signals, which can then be passed to the Accessory Decoder using this wiring. The DC power supply needs to be capable of sharing ground with the Uno and computer it is connected to. output. It can also switch off due to the timer expiring. Due to the operation of the DRV8231, these aren’t true open-drain outputs. If, for example, output 1 is on and output 2 is off, output 1 will be driven low (to ground), while output 2 will be driven high (to the voltage on the COM+ pins). However, this shouldn’t be a problem for loads like lamps or LEDs. We did note a small leakage current from the outputs, so sensitive loads like LEDs might benefit from a resistor across their leads to shunt this current. Inductive loads like relays should also be fine, since the DRV8231 has internal clamp diodes. DCC Base Station software update We’ve updated our DCC Base Station from January 2026 (siliconchip. au/Article/19558) to allow control of accessory decoders. There is an extra screen (accessed from a new AC button on the main screen) that can be used for operation. An extra button has also been added to the CV programming page. Copying the file 0911125A.UF2 to Pico 2 on the DCC Base Station will add these features. Note that loading a different firmware (new to old or old to new) will invalidate the Base Station’s configuration. Thus, it’s a good idea to record the calibration parameters from the Settings page before reflashing the Pico so they can be easily reinstated afterwards. If you find it difficult to access the BOOTSEL button, try connecting to its USB serial port (with a terminal program) at 1200 baud; this is the method the Arduino IDE uses to enter the bootloader. If you run into problems after loading the new firmware, try clearing Screen 3: the updated version of our DCC Base Station Firmware has a page for controlling accessory decoders; its simple interface is shown here. siliconchip.com.au the flash with the flash_nuke.UF2 firmware image, then load it again. Screen 3 shows the new screen. It is quite simple and just contains a button to select the accessory decoder address. This opens a numeric keypad for number entry. The buttons for THROW and CLOSE will activate the corresponding outputs on the addressed decoder. We have tested this with our own Decoders and also a commercial Rokuhan decoder; all worked as expected. Screen 4 is the updated CV programming screen. A new ACC button has been added to provide programming for accessory decoders. This works on the MAIN track output (operations mode) of the base station and uses the accessory decoder address entered on the AC screen. Being on the MAIN track means that there is no readback – only writing is possible. To avoid corrupting addresses, the accessory decoder addresses cannot be set via their CVs. Instead, you should use the I2C Controller board or follow the instructions in the next section. Our Accessory Decoders also support being reset (all CVs to default values) by programming a value of 8 into CV 8; this will, of course, change the address back to its default value. Addressing The default addresses will work fine for testing, but may need to be changed if you are using more than one Accessory Decoder board, since they would all be on the same addresses otherwise. Fitting the shunt to the jumper sets up address programming mode; the section below assumes the shunt is fitted. Screen 4: the new ACC button on the programming page of the DCC Base Station allows CV programming of accessory decoders on the MAIN track output. Australia's electronics magazine July 2026  77 Parts List – Accessory Decoders Snap-type Accessory Decoder 1 42 × 70mm double-sided PCB coded 09111254 2 3-way 5-5.08mm/0.2-inch pitch screw terminal blocks (CON1 & CON2) 2 2-way 5-5.08mm/0.2-inch pitch screw terminal blocks (CON5 & CON6) 1 4-way 2.54mm/0.1-inch pitch polarised header or similar (CON7) 1 5-way 2.54mm/0.1-inch pitch right-angled pin header (CON8; optional, for ICSP) 1 3-way 2.54mm/0.1-inch pitch right-angled pin header (CON9; optional, for external switches) 1 2-pin 2.54mm/0.1-inch pitch header and jumper shunt (JP2) 2 6 × 6mm tactile switches (S1, S2) mounting hardware to suit installation (eg, 3mm machine screws and spacers) glue to secure the 4700μF capacitor Semiconductors 1 PIC16F18146-I/SO microcontroller programmed with 0911125P.HEX, wide SOIC-20 (IC1) 2 DRV8231DDAR motor driver ICs, SOIC-8 (IC2, IC3) 1 78L05 regulator, TO-92 (REG1) 5 1N5819 schottky diodes (D1-D5) 1 green 3mm LED (LED1) 1 red 3mm LED (LED2) Capacitors 1 4700μF 25V electrolytic (optional for Snap-type motors) 2 100μF 25V electrolytic 2 1μF 25V X7R SMD M3216/1206-size MLCCs 3 100nF 50V X7R SMD M3216/1206-size MLCCs Resistors (all SMD M3216/1206-size ±1% ¼W except as noted) 2 100kW 5 10kW 1 3kW 3 1kW 1 100W ±5% 1W axial 2 0.1W 2W SMD M6331/2512-size Servo-type Accessory Decoder 1 42 × 70mm double-sided PCB coded 09111255 4 3-way 2.54mm/0.1-inch pitch right-angle pin headers (CON1-CON4) 2 2-way 5-5.08mm/0.2-inch pitch screw terminal blocks (CON5 & CON6) 1 4-way 2.54mm/0.1-inch pitch polarised header or similar (CON7) 1 5-way 2.54mm/0.1-inch pitch right-angle pin header (CON8; optional, for ICSP) 1 5-way 2.54mm/0.1-inch pitch right-angle pin header (CON9; optional, for external switches) 1 2-pin 2.54mm/0.1-inch pitch header and jumper shunt (JP3) 1 22μH 1.3A 6 × 6mm SMD inductor (L1) [eg, NRS6028T220M] 4 6 × 6mm tactile switches (S1-S4) mounting hardware to suit installation (eg, 3mm machine screws and spacers) Semiconductors 1 PIC16F18146-I/SO microcontroller programmed with 0911125V.HEX, wide SOIC-20 (IC1) 1 MCP16311(T)-E/MS buck regulator, MSOP-8 (REG1) 6 1N5819 schottky diodes (D1-D6) 1 green 3mm LED (LED1) 1 red 3mm LED (LED2) Capacitors 3 100μF 25V electrolytic 3 1μF 25V X7R SMD M3216/1206-size MLCCs 2 100nF 50V X7R SMD M3216/1206-size MLCCs Resistors (all SMD M3216/1206-size ±1% ¼W) 2 100kW 1 56kW 4 10kW 2 1kW 4 470W 78 Silicon Chip Australia's electronics magazine The pushbutton switches also influence this mode and will not operate the outputs while the shunt is fitted. In this state, the Accessory Decoder will record the address of the first accessory packet that it sees three times in a row. The first output (which defaults to address 1) will take on this address. The second output will take on the next address and so forth. Addresses wrap above 2048, so if 2046 is sent to the Servo-type Decoder while the shunt is set (and S1-S4 are not pressed), CON1 to CON4 will be set to respond to addresses 2046, 2047, 2048 and 1, respectively. Holding one of S1-S4 will program just the corresponding output to the address seen on the DCC bus. When the jumper is set, LED2 will light up for a second when a valid action occurs. Pressing S1 and S2 together while the shunt is in will force a reset of all outputs to the CVs and values shown in Table 1. For our DigiTrax system, we had to push the accessory button twice to ensure enough packets were sent, since it only sends two packets per action. Note that this arrangement means that you do not need to know if the addressing used by your system is linear or otherwise, since the bit patterns are all that is matched. Mounting The Accessory Decoders have four mounting holes to suit 3mm hardware, and we expect many readers will mount the Decoders underneath a baseboard or control panel. You might like to use the bare PCB or the overlay diagram (which is to scale) as a jig to mark mounting holes. The centres are at 36.5mm and 64.5mm spacings. Summary If you wish to use the I2C Controller to monitor the Accessory Decoders and change their CVs, there is further detail (including screenshots) in that project article. With this article, we now have a fairly complete DIY DCC system, including mobile decoders, stationary decoders, a base station and numerous other items! We plan to round that off in the future with a miniature destination display that can be installed within model trains and controlled by our previously described microDCC SC Decoder. siliconchip.com.au PRINTED CIRCUIT BOARDS PRINTED CIRCUIT BOARD TO SUIT PROJECT IDEAL BRIDGE RECTIFIER, 28mm SQUARE SPADE ↳ 21mm SQUARE PIN ↳ 5mm PITCH SIL ↳ MINI SOT-23 ↳ STANDALONE D2PAK SMD ↳ STANDALONE TO-220 (70μm COPPER) 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) DATE DEC23 DEC23 DEC23 DEC23 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 For a complete list, go to siliconchip.com.au/Shop/8 PCB CODE 18101241 18101242 18101243 18101244 18101245 18101246 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 Price $2.00 $2.00 $2.00 $1.00 $3.00 $5.00 $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 PRINTED CIRCUIT BOARD TO SUIT PROJECT ROTATING LIGHT (BLACK) 433MHZ TRANSMITTER VERSATILE BATTERY CHECKER ↳ FRONT PANEL (BLACK, 0.8mm) 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 DATE APR25 APR25 MAY25 MAY25 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 PCB CODE Price 09101251 $2.50 15103251 $2.50 11104251 $5.00 11104252 $7.50 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 ADJUSTABLE ULTRASONIC CLEANER MAIN PCB ↳ FRONT PANEL CONTROL PCB SNAP-TYPE DCC ACCESSORY DECODER ↳ SERVO-TYPE I2C CONTROLLER JUL26 JUL26 JUL26 JUL26 JUL26 04105261 04105262 09111254 09111255 09111256 NEW PCBs $7.50 $5.00 $3.00 $3.00 $3.00 We also sell the Silicon Chip PDFs on USB, RTV&H USB, Vintage Radio USB and more at siliconchip.com.au/Shop/3 By Tim Blythman μDCC Decoder Accessory Decoder I2C Controller 2 I C Controller Destination Display for DCC Accessory Decoders Programming accessory decoders can be tricky, even though the standards provide for it, since not all base stations provide the capability. This small board makes it easy to program our DCC Accessory Decoders. It only requires power and two signal lines on an I2C bus, so it could be a handy device for other projects that need a simple user interface. A s we noted in the DCC Accessory Decoders article, it can be tricky to program accessory decoders. They are usually permanently wired to the main track circuit, so they can only be programmed in ‘operations mode’. There are means for bidirectional communication in operations mode (to read back and verify programmed values), but that requires extra hardware, both on the decoder and on the base station. Providing a display and user controls for each Accessory Decoder would make this much easier, but that would be a waste as it would sit idle most of the time and you’d need one per decoder. It would also make the Decoder larger, harder to build and more expensive. This I2C Controller is a simple removable device that provides a display and some buttons. One Controller can be used to set up several Accessory Decoders, one at a time, then be put aside for future use. We also think that the I2C Controller might come in handy for other projects that require a similar interface. In this article, we’ll describe the I2C Controller and its construction. We’ll also detail how it can be used with the DCC Accessory Decoders described elsewhere in this issue, including programming their CVs and monitoring their operation. Circuit details The circuit of the I2C Controller is shown in Fig.1. CON1 is a four-way header that has the SCL and SDA I2C signals, plus ground and power. This connects to a matching header on the DCC Accessory Decoder (or other device that needs a control interface). The Decoder provides power and acts as an I2C master. LED2 is simply a power indicator LED with a 1kW series resistor. The I2C Controller provides several I2C slave interfaces that can be controlled by the master. The first of these is an OLED module, MOD1. It contains a display controller IC, a 3.3V regulator (for the display controller) and pullups from the SDA and SCL lines to 3.3V. IC1 is a PCF8574 (or PCF8574A) I2C IO expander (IC1); this also connects to the I2C bus at its pins 14 and 15. IC1 is powered from the supply at pins 8 and 16. These are bypassed by a 100nF capacitor. Three pins (1, 2 and 3) are pulled low to set the I2C sub-address of DCC PROJECT KITS Snap-type Accessory Decoder (SC7685, $40) includes the PCB and all onboard parts, including the electrolytic capacitor Servo-type Accessory Decoder (SC7686, $40) includes the PCB and all required onboard parts I2C Controller (SC7690, $30) includes all the parts in the parts list overleaf 80 Silicon Chip Australia's electronics magazine IC1 to zero. This means that a PCF8574 will appear on address 0x20 and a PCF8574A on address 0x38. Eight of the remaining pins are designated as quasi-bidirectional I/O pins. In practice, these are open-drain outputs with weak internal pullups. The chip also implements a pullup accelerator that briefly applies a stronger pullup as the open-drain output is switched off. Pin 13 (INT) is an output that is triggered on an I/O state change and is not used here. The open-drain arrangement means that the pins can be safely pulled to ground without causing a conflict. IC1 has commands to read its I/O pin state and to control its open-drain outputs. Switches S1-S4 are connected between the I/O pins and ground, so a switch closure is detected as a low level by IC1. LED1 is a two-pin bicolour LED, with each lead connected to another of IC1’s pins. They are also pulled up to the supply by 1kW resistors. When one of LED1’s leads is taken low by IC1, current flows into the resistor on the other lead and out via the grounded pin. Current is also wasted on the other resistor, which is now directly connected across the supply. Still, by taking one or the other pins low, either the red or green element in LED1 can be lit up. Taking both pins low means that they are at the same potential and no current flows through the LED, holding it off. Helpfully, the wiring of CON1 is the same as the I2C OLED modules that we use, so the two can be interchanged to a degree. Be careful, though; we have seen some I2C OLED modules with siliconchip.com.au Fig.1: this board does little more than break out an I/O expander IC connected to some pushbuttons and a bicolour LED along with an OLED module. swapped power and ground connections! For example, an OLED module can be plugged in where an I2C Controller is expected, and the display will work as intended. The default display is a status information screen that can be viewed without the pushbuttons needing to be pressed. Since the pinouts are identical, other projects or devices that use an I2C OLED module could be reprogrammed to access the I/O expander chip to add extra inputs (S1-S4) to an existing interface. Having said that, note that the PCF8574 and PCF8574A parts have different I2C timing requirements to the OLED module’s display controller. So the I2C Controller may not work in place of a bare OLED module unless the firmware takes account of this. Options We have marked the PCB to suit a polarised header so that the I2C Controller and Accessory Decoders can be connected by flying leads without a risk of reversed connections. However, you might prefer a different plug and socket arrangement. For example, it might be just as easy to hard-wire a flying lead (with plug) directly to the Controller, since there will be little need to detach the lead from the Controller. Our early prototypes used a simple header plug and socket arrangement. pin 1 indicator lines up with the mark on the board. Tack one lead and check that the other leads are on their pads, then solder the remaining leads and after that, refresh the first lead. Next, solder the three resistors and one capacitor; none of these are polarised. Clean off the flux residue and allow the board to dry. Now fit the four switches, making sure their bodies are flat against the PCB, then mount LED2 with its cathode to the left as shown. LED1 has the red element’s cathode as marked. Use a multimeter on diode mode to check the orientation before installing it. When this LED lights up red on a tester, the negative lead (usually black) is connected to the cathode of the red element. When trimming the leads of the LEDs, put the offcuts aside. Attach the pin header to the OLED module if necessary, then solder it in place, being sure to leave a small space between it and the parts below. A piece of card could be used as a temporary spacer. Finally, use the offcuts to secure the lower mounting pads in the OLED module to the PCB below using solder. You can test the I2C Controller by applying 3.3-5V between the G (negative) and V (positive) connections. If LED2 lights up, everything is working as well as can be tested without an external microcontroller driving the I2C bus. Wiring harness The photo overleaf shows how the harness is assembled; we recommend it be no more than 10cm in length. Remember that I2C is short for inter-­ integrated circuit and is designed to cover short distances within a PCB. Both ends are wired the same, and the colour code we have used matches black for GND and red for V, so there are also some visual cues to ensure it is not wired up incorrectly. Fig.2: both the decoders and I2C Controller are marked with their pin layouts, but we’ve chosen a polarised cable to ensure that the boards are always connected correctly. The OLED module is fitted last and sits over the other components. We designed this board to interface with the DCC Accessory Decoders (elsewhere in this issue), but it could be a handy addition to any project that needs a simple user interface. Construction Referring to Fig.2, the overlay diagram, apply flux paste to the pads for the SMD parts. Place IC1, ensuring its siliconchip.com.au Australia's electronics magazine July 2026  81 Use the headers and lead offcuts to space the OLED module off the PCB and clear of the parts below. We used the colour coding shown here (matching our prototype) but it isn’t critical since the plugs are polarised. The two ends can be interchanged without any problems. How to use it As shown in Fig.2, connect the I2C Controller to CON7 of the DCC Accessory Decoder (based on a PCB coded 09111254 or 09111255) and power up your DCC system. Both LEDs and the OLED screen should light up. LED1 should be green after a second or two, and the OLED will display something like Screen 1; it will be slightly different on the Servo-­ type Decoder. In general, buttons S3 and S4 (left and right) cycle through the screens, while S1 and S2 (down and up) edit the values on the screen. Pressing S3 and S4 together resets the OLED display controller; try this if the display is corrupted. Screen 1 is a status screen, useful for monitoring the Accessory Decoder’s operation. As mentioned before, this will appear if you just plug an OLED screen in too. The second line is only present on the Snap-type Decoder and shows the voltages on the motor supply before and after the 100W resistor. The screen will show VM OK and a green LED1 if the 4700μF capacitor is fully charged, or “VM --” and a red LED1 if it is charging. The servo motor supply rail voltage is shown by the Servo-­ type Decoder. The DCC text shows if DCC packets are being received, while the last line shows the address of the most recent accessory packet, or dashes if none have been seen in the last five seconds. These should allow you to check that your DCC base station is sending packets to the expected addresses. Pressing S4 will cycle to Screen 2, showing the main mode. This screen only affects the Snap-type Decoder since the Servo-type Decoder is fixed at four outputs. When two outputs are selected, they are full-bridge types, while the four outputs are open-drain types. S1 or S2 can be used to change this setting. Screens 3-8 are repeated for each output (1-4) and the output number is shown at top left. These mostly allow direct editing of the CVs (configuration variables). On these screens, S1 GPS-Synchronised Analog Clock with long battery life ➡ Convert an ordinary wall clock into a highlyaccurate time keeping device (within seconds). ➡ Nearly eight years of battery life with a pair of C cells! ➡ Automatically adjusts for daylight saving time. ➡ Track time with a VK2828U7G5LF GPS or D1 Mini WiFi module (select one as an option with the kit; D1 Mini requires programming). ➡ Learn how to build it from the article in the September 2022 issue of Silicon Chip (siliconchip. au/Article/15466). Check out the article in the November 2022 issue for how to use the D1 Mini WiFi module with the Driver (siliconchip.au/Article/15550). Complete kit available from $55 + postage (batteries & clock not included) siliconchip.com.au/Shop/20/6472 – Catalog SC6472 82 Silicon Chip Australia's electronics magazine siliconchip.com.au and S2 increment by one, unless the other button (S2 or S1) is held down; this will increment by ten instead, allowing you to reach the desired setting faster. Screen 3 is the address, which can be set from 1 to 2048. Multiple outputs can be set to the same address so that they respond to the same commands, although this will not work with CV programming packets. Screen 4 sets the output runtime in multiples of 10ms, so the top-right value is the raw CV3 value, while the lower value is the actual runtime. A value of zero means the output runs indefinitely, at least until its counterpart is activated. Screens 5 and 6 set the pulse width that is applied to the servo motors in the Servo-type Decoder. If an output is active and its CV3 is set to zero, the servo position will update the output in real-time to allow it to be easily fine-tuned. Note that these screens are not present on the Snap-type Decoder since they would serve no purpose there. Screen 7 allows the manual operation of an output. This means that it’s entirely possible to operate the DCC Accessory Decoders using just the I2C Controller, without a DCC base station. Screen 8 is used to reset the selected output’s CVs back to the default values. The Accessory Decoders have been programmed to recognise when the I2C Controller is connected, but we recommend that it is only connected or disconnected while the power is switched off to the decoders. This should help to avoid damage due to stray voltages while the connection is being made or broken. Screen 1: the first screen shows status information, meaning that even a bare OLED module can be used to check and monitor an Accessory Decoder. Screen 2: the Snap-type Decoder can be configured to have four open-drain outputs on this page. All changes are made by pressing S1 or S2. Screen 3: like the other pages, the values can be quickly changed by holding down S1 while pressing S2 (or vice versa). Screen 4: the runtime applies to all decoders. When it is set to zero, the output is on until the other output is activated. Screen 5: this screen sets the servo pulse width for the THROW output; values between 1000μs and 2000μs (1ms and 2ms) are typical. Screen 6: the servo pulse width setting for the CLOSE output. The defaults should give about 90° of movement, but most constructors will need to trim the values to suit their installation. Screen 7: this screen allows manual operation, giving yet another option (besides the onboard pushbuttons) for using the Decoders without a DCC system. Screen 8: each output can be set to its defaults by pushing S1 or S2; this includes the address, run time and servo pulse widths. Parts List – I2C Controller The I2C Controller is little more than a module with a few I2C devices on it, so it should be easy to connect to microcontrollers like Arduinos and Micromites. As we mentioned, the PCF8574(A) has different timing requirements to the OLED modules, so you might find you need to tweak your bus speed to suit. Using an I2C scanning program is a good way to test whether the bus is working correctly. The PCF8574(A) should be seen on 7-bit addresses of either 0x20 or 0x38 (32 or 56), while the OLED modules are typically set SC to 0x3C (60). 1 57 × 40mm double-sided PCB coded 09111256 1 4-way 2.54mm/0.1-inch pitch polarised header or similar (CON1) [Jaycar HM3414, Altronics P5494] 1 1.3-inch OLED display module (MOD1) [Silicon Chip SC5026, SC6511] 4 6 × 6mm through-hole tactile switches (S1-S4) Semiconductors 1 PCF8574 or PCF8574A I/O expander IC, wide SOIC-16 (IC1) 1 red/green 3mm bicolour LED (LED1) 1 green 3mm LED (LED2) Capacitors/resistors 1 100nF 50V X7R M3216/1206-size SMD MLCC capacitor 3 1kW ¼W M3216/1206-size SMD resistors Cable 2 4-way 2.54mm/0.1-inch pitch polarised header plugs with crimp pins [Jaycar HM3404, Altronics P5474 + P5470A] 1 10cm length of 4-way ribbon cable OR 1 40cm length of light-duty hookup wire siliconchip.com.au Australia's electronics magazine Using it with other projects July 2026  83 Review by Tim Blythman ALTIUM DESIGNER 26 With each new year comes a new major version of the Altium Designer EDA (electronics design automation) software. Now that we have had the chance to use Altium Designer 26 for a while, we will detail what we have found. L ike much modern software, Altium Designer sees frequent updates, so the changes that we see are often incremental improvements. For this review, we are using Altium Designer version 26.2.0. Our last review, in the June 2025 issue, was of version 25.0.2 (siliconchip.au/Article/18307). New minor versions appear about monthly, so some of our observations in this article will be of features that first appeared after that review, including versions after Altium Designer 25.0.2. In the Altium Designer 25 review, we noted that Altium had been acquired by Japanese semiconductor company Renesas. EDA (electronics design automation) software is best known for its ability to design PCBs, but Altium Designer has many features that go beyond this, such as PCB CoDesign and MCAD (mechanical computer-­ aided design) integration. CoDesign refers to being able to collaborate with other engineering disciplines; this is mostly managed through Altium 365. MCAD typically relates to project aspects such as enclosures and similar hardware. Over the last few years, we have also seen the introduction of the Harness Designer (for the design of wiring harnesses) and support for 3D-MID designs (three-dimensional mechatronic integrated devices). The 3D-MID manufacturing process uses a custom 3D-printed substrate in place of a traditional 2D fibreglass substrate. This means that the substrate which forms the PCB can also be shaped to work as an enclosure, making designs 84 Silicon Chip more compact. We published a feature on 3D-MID and similar technologies in the April 2025 issue (siliconchip.au/ Article/17936). Altium Designer 25 introduced the ability to create designs using bare COB (chip-on-board) silicon dies and bond wires. Wire bonding allows COB dies to be incorporated as components and connected to the PCB by bond wires. The Constraint Manager has also been added to provide a unified, hierarchical way to manage the design constraints of a project. Currently, it is offered as an alternative to the older PCB Design Rules. Existing projects can be converted to use the Constraint Manager. Many new features are introduced as optional features available only to registered beta testers. Beta software is mostly complete but may still have minor bugs. After ‘beta testing’, the features are made widely available but can be disabled by an option switch in the Advanced section of the Preferences menu. If you can’t immediately find a new feature, check to see that it is not hidden as a beta or optional feature. Some features may also be unavailable depending on your license inclusions. Licensing changes Probably the biggest change to coincide with Altium Designer 26 is the way that subscriptions are being licensed and bundled with other features. Roughly speaking, the new Altium Develop product is comparable to a Standard or Pro license and is intended for smaller teams. It includes the Altium 365 cloud platform. Australia's electronics magazine Above this sits Altium Agile, consisting of the Teams and Enterprise subcategories. Ultimately, the core Altium Designer software remains much the same, but the different product lines include different features and tools for data management, such as Altium 365. Fig.1 shows the web page at www.altium.com that highlights the different options. We are a small team developing relatively simple designs, so we do not need all of Altium Designer’s advanced features. Some advanced tools are aimed at things like compliance, traceability and security in heavily regulated fields such as medical electronics; these are well above our needs. Logging into an Altium account is now via a ‘unified login’ in a browser, which also provides access to other services such as the Altium 365 Workspace. Depending on your chosen access method, this could include a simple email address and password, SSO (single sign-on) scheme, or twostep verification. Altium Discover Yet to launch at the time of writing is the recently announced Altium Discover. This tool will take in a set of project requirements and analyse reference designs from manufacturers to provide high-level design suggestions for components, such as chips and modules, to help fulfil those requirements (see www.altium.com/ discover). The web page at www.altium.com/ capabilities/requirements suggests that an AI assistant will be part of this process, being used to analyse and siliconchip.com.au Fig.1: the new AD26 licensing scheme. Altium Develop and Altium Agile both incorporate Altium Designer and Altium 365. The upcoming Altium Discover will suggest project solutions from design requirements. Fig.2: the Harness Designer now shows break points in harness diagrams where necessary. summarise the vast quantities of information that need to be reconciled in developing a complex design. With Altium owning the Octopart parts database, such a tool would have access to information such as availability, pricing and technical specifications to feed into these analyses. The Manufacturer Part Search within Altium Designer makes use of the Octopart database; we often use this to import new components to our library. In our series on How to Design PCBs, published in December 2025 to February 2026 (siliconchip.com. au/Series/453), we work through the steps needed to design and manufacture PCBs. In particular, we use Altium Designer’s Schematic Editor and PCB Editor for the design stages. at www.altium.com/documentation/ altium-designer/new while upcoming features are highlighted at www. altium.com/altium-designer/coming-­ soon In our Altium Designer 25 review, we mentioned that the underlying software was transitioned to the opensource .NET 6 framework. Starting with version 26.10 of Altium Designer, this has been updated to .NET 8, allowing Altium Designer to benefit from newer features in the framework, as well as promising a performance increase. This update means that Altium Designer can no longer run on Windows 7 or Windows 8. As has been the case in the past, newer versions of Altium Designer can also be installed alongside older versions. AD26 overview Improvements We have previously noted Altium’s focus on continuous improvement, and Altium Designer 26 follows this trend. The latest features can be found Small but helpful updates have appeared in Harness Designer, particularly in relation to the creation of harness manufacturing drawings. This siliconchip.com.au Australia's electronics magazine includes automatic annotation of bundle lengths and the embedding of 3D models. For bundles that cannot be displayed at their true scale length, a break symbol indicates this – see Fig.2. Altium Designer has the ability to import designs from other EDA tools, which we find handy, since we occasionally need to process contributed designs that have been developed using other software. There are a number of improvements noted in the importing tools since our last review. Fig.3 shows the Import Wizard and the options that are available in Altium Designer. Solder mask expansion rules We are also seeing continuous improvement in the capabilities of PCB manufacturers, and it makes sense that these should be reflected in Altium Designer’s behaviour. In particular, PCB manufacturers can achieve smaller feature sizes and tighter tolerances. This will generally mean that July 2026  85 Fig.3: the Import Wizard can handle the file types shown here, and there are many more options in the native File → Open menu. It’s also possible to add other importers for files produced by tools such as KiCad. Fig.4: these pads on an SSOP (small shrink outline package) IC with 0.65mm lead pitch show the advantages of reducing solder mask expansion in line with modern PCB manufacturing capabilities. The default 0.1mm expansion at left does not allow solder mask between the packs, while the zero solder mask expansion at right provides the best defence against bridging pads. smaller designs and closer spacings become more achievable (or cheaper!). Over time, we have been able to update our design rules with tighter spacing and clearances where this is necessary or helpful. Since we often present our projects as kits, they are intended for manual assembly, and having appropriate solder mask coverage can be helpful for avoiding bridging between narrowly spaced leads. Previously, some pins (especially on small surface-mount parts) were so closely spaced that it was not possible to provide solder mask in the gap; the so-called solder mask sliver between the pads would have been thinner than what the PCB manufacturer could reliably produce. This is compounded because the tolerances of older processes dictated a narrow margin between a pad and its solder mask opening; this is the ‘solder mask expansion’ and Fig.4 shows how this reduces the available space between close pads. It appears that many PCB manufacturers now use a LPI (Liquid Photo Imageable) solder mask process, which requires negligible solder mask expansion; this expansion is the border left around a pad to ensure that the solder mask does not encroach upon the pad due to tolerances in the process. So-called zero solder mask expansion is now possible and is the default in IPC-7351B, the industry standard for surface-mount device land Fig.7: being able to add QR codes and other 2D codes allows more machinereadable information to be printed on PCBs. This can include tracking codes and product identifiers. 86 Silicon Chip patterns (footprints). As of Altium Designer 26.1.0, the default solder mask expansion rule has been set to zero. Of course, this can be changed in the rules if needed or manually set for individual pad requirements. ActiveBOM The bill of materials (BOM) for a project is an important document, and Altium’s ActiveBOM is a tool for managing this. We often use a simple spreadsheet-based BOM (which Altium Designer can export) to ensure our parts lists and kit listings are correct. The BoM CoDesign feature is intended to allow collaboration with the purchasing and procurement departments in relation to the PCB BOM. This can provide live information about parts availability, alternatives and lead times from the Octopart database. Figs.5 & 6 show a typical BOM and its supply chain status. QR codes For a while now, Altium Designer has been able to create 1D (linear) barcodes as part of a PCB document. These would typically be applied to the silkscreen layer to act as machine-readable identifiers for stock and part management. It’s now possible to add QR codes and DataMatrix codes; these are two different types of 2D codes that can be Australia's electronics magazine siliconchip.com.au Fig.5: the BoM CoDesign tool can be used by procurement teams; there is a more detailed supply chain view that can highlight potential issues with specific line items. Fig.6: the BoM CoDesign tool presents a range of information relating to part options and supply chain availability. used to encode more data than a linear barcode. Fig.7 shows a QR code being instantly generated to encode the PCB code on one of our PCBs. We delved into QR codes while reviewing a tiny QR code reader module in the February 2026 issue (siliconchip.au/ Article/19663). Free stuff Altium Designer still offers a free trial at www.altium.com/altium-­ designer/free-trial/roadmap although this page also notes that the website is being updated, so this may change in the future. Altium CircuitMaker (www.altium.com/circuitmaker) is also free to use. CircuitMaker is an EDA tool aimed siliconchip.com.au at hobbyists and makers. It is built on the same engine as Altium Designer and allows designs to be easily shared with other CircuitMaker users. We reviewed CircuitMaker in January 2019 (siliconchip.au/Article/11378). Even if you don’t use Altium Designer, they have a trove of resources relating to PCB design, including a guide to getting started at www.altium. com/documentation/altium-designer/ tutorial and the Altium Academy YouTube channel at www.youtube.com/<at> AltiumAcademy Summary It’s not surprising that the Renesas purchase of Altium has seen some changes to the way the product Australia's electronics magazine is delivered, with the new Altium Develop and Altium Agile products being introduced. The Altium Discover product sounds like it will provide an interesting addition to the Altium repertoire. Altium Designer 26 continues to improve; the advances in the Harness Designer and BOM tools continue the trend of Altium’s tools gaining a wider scope beyond simple PCB design and layout. It’s good to see contemporary updates to the PCB Editor, such as QR codes and alignment with modern PCB manufacturing standards. For more information on the software, see www.altium.com – it can be downloaded from www.altium.com/ SC products/downloads July 2026  87 SERVICEMAN’S LOG Batteries, monitors, lights and audio Dave Thompson is once again busy doing whatever he does when he disappears: fighting an alien invasion, creating a new sheep hybrid, or practicing his haka – we aren’t sure which, but it must be one of those. Anyway, in the meantime, Bruce Pierson has been very busy in his shed fixing everything that comes across his bench, so here are some of the things he has repaired lately. To start things off, my son asked me if he could borrow my battery charger because his had stopped working. I asked him to bring it over and I could see that it was a switch-mode type, which are sometimes tricky to fix, but I said I would a look at it. I started by removing the four Phillips-head screws on the bottom and split the case apart. Thankfully, they didn’t use ‘tamper-resistant’ screws as is so common these days! I had a good look over the circuit board. All the electrolytic capacitors looked OK, with no bulging tops, so I turned it over and examined the copper side with a magnifying glass to check for dry joints, but there were none. In fact, it looked as if it were quite new, probably because it was. It was a change from the older items I usually work on. I got out my ESR/Low Ohms Tester from Electronics Australia, February 1996, that I built from a Jaycar kit. It indicated that all the electrolytic capacitors were good. This device has been very useful over the years for finding faulty electrolytic capacitors that showed no external signs. The MK II version of this tester was featured in the Silicon Chip March & April 2004 issues (siliconchip.au/Series/99). 88 Silicon Chip Next, I got out my In-Circuit Transistor, Diode and SCR Tester to start testing all the two- and three-legged semiconductors. I started with a diode, but the tester was not working correctly. It was flashing very dimly and slowly, then stopped flashing, with one LED lit dimly. It was a sure sign that the battery had gone flat since I last used it. I built it from a Jaycar kit of an Electronics Australia project from September 1983, and it has proved very useful. When I tried to remove the 9V battery, the negative terminal came off the battery. It took some effort to remove this detached terminal from the battery connector, but I eventually got it off. With a new battery fitted, I could continue testing. Nothing showed up as faulty. I noticed that near the battery leads there was a component that was likely some sort of SCR. I wondered if it might be faulty. I tested it, but no matter how I connected the leads from the tester, it came up as an open circuit. Well, that must be it. I used my 20W soldering iron to remove it from the board and tested it again with the same result. Now I had to identify it so I could order a new one. The writing on the device was almost unreadable, but I thought I could make out IRF724N. I looked on eBay for this and found nothing suitable. A Google search for the component gave many hits for IRFZ24N, so I guess what looked like a 7 must have been a Z. I searched again on eBay and ordered five from China for just over $5. It took only 12 days to arrive. This is a 55V, 17A Mosfet so I was surprised the original failed as it seems quite robust. I soldered it to the board and clipped the excess leads off. It mounts in the top-left corner of the circuit board, on the left side of the battery cables. I decided to test the charger before reassembling it. I set the charger on the bench with a spare 12V SLA battery connected, plugged it in and switched on the power. After a couple of seconds, the charging LED came on next to the power LED, so I knew I could reassemble it and give it back to my son. I checked the Supercheap Auto website to see if this charger was still available, but it was not. Similar chargers cost between $60 and $100. My son has had this charger for several years now (it came with the caravan he bought). His caravan battery is charged by two solar panels, with the charger being used to top up the battery when necessary. Australia's electronics magazine siliconchip.com.au The internals of the Samsung SyncMaster 2253BW monitor. The remains of the lizard were very clearly present when the back was taken off. It cost just over $1 to repair the battery charger, a considerable saving compared to buying a replacement charger. Saving a ‘new’ SLA battery I was sorting out some things in my shed when I came across a new UPS box that had a brand new UPS in it. I didn’t remember having this, so it must have been in the shed for a very long time. I could tell by the weight of the UPS that it had a battery in it, which was a bad sign. These SLA batteries must be kept charged because when the voltage drops below a certain point, they will refuse to charge again. In that sense, they are worse than flooded lead-acid batteries. Of course, car batteries should never be in a situation where they are allowed to become dead flat, as it shortens their life. Still, usually you can recharge them if they go flat. I took the SLA battery out of the UPS and tested the voltage with my multimeter; sure enough, it read 0V. That was a waste of a new battery. I was just about to recycle it when I thought I would connect my SLA battery charger to see if there was any chance it would charge. It wouldn’t hurt to try. I left the battery connected to the charger for a couple of hours, but when I came back, the battery was still dead flat. Just as I was about to put the battery into the scrap pile again, I had a thought. I had nothing to lose, so I decided to connect my car battery charger to it. I came back after 15 minutes and the ammeter on the charger showed that the battery was charging. That was unexpected, but it was a good sign that the battery might be able to be salvaged. I didn’t want to leave the battery connected to the car battery charger, so I swapped it for my SLA charger, and the charger’s LED turned red, indicating that the battery was charging. After several hours of charging, I switched the charger off and got a 55W quartz halogen globe and tested the battery with it. The globe lit up at full brightness, showing that the SLA battery had come good. I had another old SLA battery in the scrap pile, so I thought I would try charging it with the car battery charger, but it did not work, no doubt due to the age of the battery. Also, it had bulging ends, indicating that the plates had deteriorated to the point of no return. An SLA battery is only worth a few cents as scrap, but siliconchip.com.au a replacement battery is around $40. So this experiment was worthwhile. Sometimes in a situation like this, with nothing to lose, it pays to experiment. You never know when things might turn out better than expected. Samsung monitor reptile removal I’ve had this Samsung SyncMaster 2253BW monitor for many years. Previously it was in use in my back shed where I used to do electronics repairs before I got my new electronics workshop finished last year. I used to work more on computers and monitors, but now I mostly work on laptops. A few years ago, I went to switch on the monitor and it tripped the safety switch. I unplugged it and put aside. I reset the safety switch, got another monitor, and continued doing what I had been doing at the time. I suspected that a gecko might have gotten inside the monitor and shorted out something on the power supply board. I had this happen to an Asus monitor a long time ago. In that case, the fuse had blown and some tracks were damaged from arcing, but I was able to repair it. This monitor is a bit tricky to disassemble. I started by unscrewing the stand/base, then I unclipped the front screen surround. This gave me access to the two screws that hold the stand stem on, so I could remove it. Then, with the monitor face down, I could remove the back shell. With the back off, I unplugged the front control cable. There is a metal screen to remove, plus four plugs for the high-voltage supply. I put some dots on them so I would know which one went where, as I thought this might be important. I now had the metal shell with the power supply board and the video board loose from the rest of the Items Covered This Month • A shopping list of repairs • Lights out on a receiver • Fixing the fan bearings in a gas heater 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 July 2026  89 The garden lights at night and one of the working glass solar panels. monitor. And there I found what was left of the gecko. After removing the gecko and cleaning the board, I could not see any damage. There were no burnt tracks or any other signs of damage, but there was a small area on the back of the screen with a black mark where the high voltage had arced onto the gecko. Maybe the gecko shorted things in such a way as not to cause any actual damage. Here’s hoping! I placed the metal shell with the boards in it on the concrete floor, plugged in a power cable and switched it on. Nothing happened. There was no smoke or any other sign of anything being wrong. Maybe I got lucky with this one. I put the monitor back together loosely so I could test it before reassembly. With it facing up, I connected it to power and pressed the power button on the front screen surround. The screensaver came on, indicating that the monitor still worked. I reassembled the monitor and then took it into the house and connected to my Linux laptop with a VGA cable. I got a good picture, so the monitor had been restored to working order. This time I didn’t have to actually repair anything, just remove the cause of the safety switch being tripped. Good thing we have a safety switch! Garden light repairs We have 50 garden lights around features in the front of our house and on the side of our driveway. Some of these lights need maintenance from time to time. I usually check them every so often to see if there are any that aren’t lit when it gets dark. It had been a few months since I last checked them, and I found that 30 weren’t working. The most common reason they fail is 90 Silicon Chip that the rechargeable AAA cell needs to be replaced. This was the case with 27 of the non-working lights. I only had 21 rechargeable AAA cells in stock, so I had to order some more cells on eBay. I usually get 24 at a time so that I have a good supply on hand. Another point of failure is the YX8018, which is a specialised, low-voltage solar LED driver IC used primarily in solar-powered garden lights, lawn lamps and fairy lights. It is a single-chip solution that manages battery charging during the day and drives the LED at night, operating efficiently at the low voltages typically supplied by a single 1.25V NiMH rechargeable cell. As with the cells, I was able to obtain replacement YX8018 ICs on eBay. Another point of failure is the colour-changing LED, which I occasionally have to replace if one colour goes out. I was also able to obtain these on eBay. It’s not very often that I need to change one, though, as they last a long time. The final point of failure is the solar panel, which is a 40 × 40mm glass panel. If this fails, the garden light is not repairable, as I cannot find glass solar panels of that size anywhere. Similar plastic solar panels are available on eBay but they cost more than a replacement garden light, and in any case, plastic solar panels are lucky to last a year here in the Queensland sun. The glass panels can last 10 years or more. As for the other three garden lights that were not working, two had a failed solar panel and one had a failed YX8018 IC. In the case of either component failing, the light will either not work at all or be on all the time in bright light. To find out which, I disconnect the solar panel from the circuit board, set it under a bright light and connect my multimeter to the wires. These small solar panels deliver around 1.8V at 20-100mA to charge the AAA cell during the day. They produce voltage even out of direct sunlight. They comprise three individual cells that generate around 0.6V each. I replaced the two lights with failed solar panels and repaired the last with a new YX8018 IC. Whenever I buy a new garden light, I take out the AAA cell that comes in it and I replace it with the much better one from eBay as these cells keep the light lit for longer at night and also have a longer life expectancy. In the process of repairing the garden lights, I had to reseal some holes where the wires from the solar panel Australia's electronics magazine siliconchip.com.au Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column in SILICON CHIP? If so, why not send those stories in to us? It doesn’t matter what the story is about as long as it’s in some way related to the electronics or electrical industries, to computers or even to cars and similar. We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. Left: the standing spotlight, which was rewired so that it had a longer cable, and did not need an extension cable. enter the light, as some were not sealed correctly, allowing water to enter the light. This resulted in a couple of battery connectors needing to be replaced; I used parts from old lights that had failed solar panels. In over 10 years, I have amassed 20 lights with failed solar panels. I also replaced two glass bodies, as the water had rusted the bottom screws, which caused the screws to expand and break the glass. Sometimes a light may get broken from another cause as well. Simple headphone repair My son brought me some headphones that had the cord ripped off. He pointed out that there was a tiny piece of red wire still attached to one speaker terminal so I’d know which wire went where. I got my 20W soldering iron out and plugged it in to heat up while I prepared the wires for the repair. The wires appeared to be cotton-covered, multi-core enamelled copper wire, so I used a lighter to burn off the cotton covering and the enamel at the end. I was then able to tin each wire, ready to solder them back to their respective terminals on the speaker. I carefully soldered each wire to the correct terminal and put the speaker back onto its bracket. This was an easy repair that only took me a few minutes. I gave the headphones back to my son and suggested that they should be stored where their cat could not find them again, as it was fortunate that the headphones had not been ripped to pieces. If the cat had another go at them, it could end up a lot worse than just the wires ripped off. Rewiring a standing spotlight I’ve had this standing spotlight for several years, but hadn’t used it. It had a short cable with a footswitch, which may have worked well inside a house with the power siliconchip.com.au points near the floor, but it was impractical in my workshops with power points above the benches. I had been using it with an extension cable, but I decided to rewire it to have a longer cable without the foot switch. I looked on Jaycar’s website and found the exact cable I needed, sold by the metre. I worked out that I needed six metres, as the stand was around 1.5m high and that would give me around 4.5m of cable to reach a nearby convenient power point. I knew I had a plug to suit this twin-core flat cable, as I had planned to use it to repair a fan from the local Tip Shop that someone had cut the plug off, but at the time I could not find it. In the meantime, I had found it and lost it again, but after a quick search, I located it and put it on my workbench until the cable arrived. When it did, I started by removing the two screws securing the shade and lamp holder to the stand. Then I fed some cable up through the stand so I had working room to disconnect the cable from the lamp holder. With that done, it was time to remove the old cable and run the new cable through the stand and connect it at both ends. After removing the lamp holder and the lampshade, I tried pulling the cable through the stand from the bottom, but it got stuck and I could not pull it through. I suspected that there might be a join in the cable, so I unscrewed the segments one-by-one and, sure enough, there was a join in the cable. I cut the joint off the cable and I was then able to extract the old cable through the segments and then through the bottom of the stand and out of the base, ready for the new cable to be installed. I fed the cable into the stand base from the bottom, Australia's electronics magazine July 2026  91 The JVC receiver, which needed tracks to be bypassed with wire, due to corrosion. disintegrate is a mystery because they are on opposite ends of the PCB. On close examination, it looks like the tracks have oxidised in numerous sections. There is no visual indication of burning or lifting as if it had occurred from overcurrent heat stress. Editor’s note: while those supply tracks carried an AC voltage, there would likely have been a small DC voltage. That, combined with moisture and acid from decaying insects, is likely to lead to electrolytic corrosion. The solder mask is porous and doesn’t prevent the copper tracks underneath from corroding away under these conditions. Paul James, Kanwal, NSW. Braemar gas heater fan bearing repair but I found there was a plastic sleeve at the top of each section that the cable was getting caught on. I fed the cable through from the top instead, which proved to be successful. I screwed the segments together and then onto the base as I fed the cable through. Having connected the lamp holder and reassembled it, I screwed the lamp holder and shade back onto the top flexible part of the stand. The last job was to wire up the plug. A regular plug is not suitable for this two-core flat cable because the cable hole in the outer section of the plug is too big, so it was lucky I still had this one that suited the cable. With the rewiring finished, I tested the spotlight, and it worked as expected. JVC RX 5032-VSL receiver repair I was asked if I could do anything about the lights on an amplifier/receiver so it could be used properly. The amplifier had no display and therefore couldn’t be readily controlled because there was no indication of volume settings, radio stations or any other functions. I completely dismantled the receiver and found that numerous insects had entered the casing and had died there or been scorched. On tracing the power supply circuit for the vacuum fluorescent display, it was clear that no power was getting to it at all. The AC voltage was coming out of the transformer at 5V but not getting to the ribbon cable pins that supply the display. Testing with a multimeter on the PCB track from the transformer and the pins showed an open circuit no matter where tested along the PCB track. The same thing was happening on both outputs from the transformer. The PCB track appeared to be a different colour from the others and was unstable. Further testing on the track revealed that it was all open-circuit. The PCB track had to be bypassed with some wire to get the transformer output voltage to the ribbon cable, maintaining the resistor in one leg of the supply. During this process, I obtained a circuit diagram that showed that the transformer was supposed to be supplying 15V AC, but all my measurements showed 5V AC and the display worked after bypassing the faulty tracks, so I guess that must be correct. Apparently, the circuit I found is incorrect or for a different version because the transformer pinouts were different as well. What really caused both of the supply tracks to 92 Silicon Chip When purchased and installed in 1982, our Braemar gas wall heater fan was equipped with plain bronze bush bearings. By 2021, not only had the bushes worn out, but the bush at the fan end of the motor had worn a shallow groove into the motor shaft. Because there is only about a 10 thou (~0.25mm) clearance between the rotor and stator, the wear caused the rotor to rub against the inside of the stator and emit a rhythmic scraping sound, which steadily grew louder. By the winter of 2022, the friction of the rotor against the stator became sufficient to prevent the rotor from spinning. Ball bearings should have been fitted when the motor was manufactured. Provided that they were a reasonably tight fit on the shaft, the ball race would rotate with the shaft, and there would have been no shaft wear. Clearly, something needed to be done, but what? Braemar stated that they had no spare parts, but for approximately $3000 they would provide and install a whole new heater. Needless to say, that was never going to happen for the sake of a couple of measly ¼-inch bearings. I explored other possible sources of new motors or rotors, but couldn’t find any with a sufficiently long shaft. The options then were: Australia's electronics magazine siliconchip.com.au 1. Find and fit a new shaft. Perhaps a length of ¼-inch (6.35mm) diameter silver steel rod or a drill blank would do the trick. 2. Manufacture a new shaft from a suitable length of hardened steel. 3. Purchase a replacement motor of suitable dimensions but with a short shaft, and extend it using a collar. 4. Fill the grooves on the existing shaft, perhaps by metal spraying. 5. Turn some new bushes from bronze bar or rod. 6. Fit ball bearings and fill the shaft grooves with epoxy resin. 7. Fit ball bearings and fill the shaft grooves with epoxy resin reinforced with fine steel filings. Such mixes are available commercially, but anyone with some scrap metal and a file can make their own (I once drove from Oslo to London after repairing a chewed-out rear axle spline in this manner). Option #6 seemed to be the least expensive and most expedient. Even though epoxy is a relatively soft material compared to steel or chrome, it wouldn’t be subjected to wear so long as the shaft and inner race rotated as one. Because option #6 was likely to involve a delay of several days while suitable bearings were delivered, I attempted a temporary fix using the existing bush bearings by filling the groove worn into the shaft with epoxy mixed with graphite. In theory, the graphite would provide a low-friction bearing surface while the epoxy held it in place. Unfortunately, the epoxy collapsed after about four days, bringing the fan to a halt once more. Curses! By the time the ball bearings arrived, I’d refilled the groove worn in the shaft with plain epoxy and, using nothing more than a fine file, reduced the filler to the same diameter as the shaft. Voilà! The bearing slid on neatly with a small interference fit. Moreover, the ball bearing width proved to be equal to that of the original bronze bushes, so the shaft position and end play were no different once the original spacer was installed on the shaft between the rotor and fan end bearing. Having measured the bearing outer diameter, I made ready to turn a suitable spacer ring to support it inside the motor housing. Then I had a brainwave. Had my old friend Ian not recently gifted me a box of various-sized O-rings? Maybe the box might yield a couple sized suitably for this purpose. Indeed, I found four. There was only space for one per bearing, leaving the possibility of the O rings migrating to the outer end of the bearings and releasing them. I needed some annulus-shaped packing spacers that would slip over the bearing yet fit inside the housing. After rummaging through my box of washers and finding nothing suitable, I decided to experiment by cutting some thin cardboard packing washers. After all, there was no more danger of the cardboard burning than of the wiring or plastic formers within the motor. I could always make metal versions later if this worked. Work it did, and wonderfully so. The fan is silent and rotates with a vigour as never before. Whereas we habitually ran it at the medium speed setting, we now use only the low speed setting. It has been running this way for a couple of years, and I’ve never needed to replace the O-rings or cardboard washers with anything more substantial. SC Ron, via email. siliconchip.com.au Australia's electronics magazine July 2026  93 Vintage Radio National R-72 “Toot-a-Loop” radios by Ian Batty Is it a musical instrument? Is it a telephone? No, it’s a radio! This quirky transistor set comes in a unique, colourful plastic case that’s sure to attract attention. Inside, the circuitry of this six-transistor set hid some surprises. W ho doesn’t remember the 1960s? The Beatles, the Vietnam War, moon landings, Mao’s Great Leap Forward, Woodstock, and cars with massive tailfins. Fashion designers shook free the drab aesthetic of the 1950s, releasing more and more flamboyant, colourful, exciting designs in a frenzy to capture the new postwar economic boom. By the late 1960s, transistor radio engineering had pretty much settled on the standard set: three radio frequency (RF)/intermediate frequency (IF) stages, a volume control, an audio driver and a push-pull Class-B output stage, likely powered by the then-­ ubiquitous 9V PP6 battery that’s still in use today. Matsushita’s National brand, unable to use that name in the USA due to an existing company of the same name, had rebranded as Panasonic. They put forward several remarkable offerings; the R-72 Toot-a-Loop (appearing at the end of the 1960s) is one of the most distinctive. It’s another example of National’s successful marketing strategy: 94 Silicon Chip visually attractive radios using sound electronic design. Like the R-70 Panapet (March 2025; siliconchip.au/ Article/17800), the Toot-a-Loop is unique. Even if you have no idea of its provenance, you’ll be impressed by its styling. As an early ‘wearable’ radio, it’s a standout. The quirky design was complemented by bright colours to create a radio rivalled in its overall effect only by an identical offering from RCA subsidiary Japan Victor Corporation (JVC). The Toot-a-Loop came in white, red, blue, yellow, orange and lime, with the last two options being specific to the Australia/New Zealand markets. Our models were badged National JIS transistor coding Prefix Type 2SA High-frequency PNP BJT 2SB Audio-frequency PNP BJT 2SC High-frequency NPN BJT 2SD Audio-frequency NPN BJT 2SJ P-channel FETs (JFETs & Mosfets) 2SK N-channel FETs (JFETs & Mosfets) Australia's electronics magazine Panasonic and were advertised as a “Sing-O-Ring”. Between my yellow and red sets and what’s online, I’m aware of three different circuits for this radio. Let’s look at the R-72S yellow set (serial number 88009) first. Ernst Erb’s Radiomuseum has the circuit, showing a classic six-­transistor set; my redrawn version is shown in Fig.1. It runs off a battery of two AA cells, giving a nominal 3V supply. The six transistors comprise one converter, two intermediate-­ frequency gain stages, one audio driver, and the transformer-­coupled Class-B output pair. My example matched this with a few exceptions. The audio driver transistor, rather than a metal-can germanium PNP 2SB475/AC125 type shown in the original circuit, is an epoxy silicon NPN type, the 2SC828. The RF/IF section varied even more; the converter uses an epoxy silicon NPN 2SC829 transistor, while the IF strip has only two IF transformers, one coupling the converter to the first IF amplification stage, with the second siliconchip.com.au coupling the second IF stage to the demodulator. It has two IF transistors, both ceramic-cased silicon NPN 2SC920s, with resistance-capacitance coupling. The red R-72 set (serial number 11878) runs from a PP6 9V battery. The RF/IF section is similar to that in the original R-72S circuit, with three metal-can PNPs (2SA102, 2SA101 & 2SA101). These are drift types, with typical ft values in the 25MHz range, an improvement over the preceding alloy-junction OC44 with its typical 15MHz ft. It’s a very similar circuit to that of the previously reviewed R-70 Panapet. Does that mean my red R-72, apart from the supply voltage, is similar to the R-72S? Well, no. It shows a notation for a 2SC829 silicon NPN converter with the correct symbol, but it’s wired into the circuit with the correct polarities for a PNP device. It’s odd that it says 2SC829 (OC1044), since the OC1044 (2SA101) is definitely a PNP germanium type, which is what I found installed. Also, the audio section’s transformerless design uses two epoxy-cased NPN transistors and one PNP type. Online searching revealed the Philips 20RL012 long-wave-only set using PNPs for the converter and IF amplifiers, with a complementary output stage using NPN and PNP transistors. It’s an unusual design, as long-wave broadcasting had begun declining by the time the Toot-a-Loop arrived. So I ended up with two chimeras – not quite the classic lion’s head, goat’s body and snake’s tail, but close enough. Finding no authoritative circuit for the yellow set, I resorted to tracing it out as-built. This was complicated by the extreme compactness of the design and by almost all the resistors being printed onto the circuit board. Where I would usually lift one end of a resistor to measure it, I had to apply my ohmmeter with both polarities and take the higher reading (to prevent transistor junctions giving a false reading) or, bravely, short-­circuit bias resistors to ground, measure the short-circuit current, then apply Ohm’s Law. You may not want to try this at home! The resulting circuit for the red set is shown in Fig.2. Both sets are built on double-sided PCBs with most of the resistors siliconchip.com.au printed directly onto the phenolic substrate. This would reduce the amount of mechanical assembly, as resistors don’t need to be placed prior to wave-soldering the PCB. It’s also a great way of reducing the overall size of the set, and would be highly reliable. It does make circuit tracing harder, especially when the as-built set differs as much from available circuits, as these two do. So we have three different circuits for just two sets. Let’s look at the RF/ IF sections first. Circuit details The 3V (Fig.1) front-end uses Q1, an NPN 2SC829 converter transistor. With a typical ft of 230MHz and being recommended for “RF amplification, oscillation, mixing, and IF stages of FM/AM radios”, it’s similar to the more familiar BF115. The base bias for Q1 is supplied from the decoupled supply via resistive divider R1/R2, with emitter resistor R3 stabilising the circuit, bypassed by capacitor C4. I’ve never seen a tuning gang returned to the emitter (or cathode) of a converter before, but this does conform to the single-point grounding technique. Be aware that the tuning gang’s ‘cold’ RF connection is above ground and must not be used as an earthing point during testing. Q1 is configured as a self-­oscillating converter, with the usual ‘Japanese-­ style’ oscillator feedback from LO transformer L2 via 10nF capacitor C3 to the base. Since connecting a signal generator to the base stops the oscillator, I used a low-value series capacitor to inject to the top of the tuned antenna circuit. This confirmed the high sensitivity borne out in testing (more on that later). IF injection is also reliable. The converter feeds the tuned, tapped primary of the first IF transformer, IFT1. Its secondary feeds first IF amplifier Q2, a 2SC920 transistor with a typical ft of 250MHz. This stage’s emitter goes directly to ground, eliminating the usual emitter resistor and its bypass capacitor. It gets weak Fig.1: both of my sets had a different configuration from the ‘standard’ published circuit. This is how my yellow set was built. It uses many different transistor types from the standard circuit and has one fewer IF transformer, with RC coupling taking its place. Australia's electronics magazine July 2026  95 forward bias via R4, and gain-control voltage from the demodulator via R10. Load resistor R5 measured as 1.3kW. The signal is then resistance-capacitance coupled to Q3, another 2SC920, via 22nF capacitor C8, working with fixed bias. Voltage divider R6/R7 measured as 30kW and 16kW, while emitter resistor R8 measured at 210W, bypassed by 22nF capacitor C9. IFT1 was confirmed as the first IF transformer (converter circuit) by its yellow adjusting cap, and IFT2 as the final IF transformer (demodulator circuit) by its black adjusting cap. The usual second IF (white cap) was missing, confirming (i) only two IF transformers and (ii) R-C coupling. My familiarity with two-stage audio preamplifiers initially suggested a direct-coupled design, but the application of AGC would presumably have disturbed DC operating conditions excessively. Q3 feeds final IF transformer IFT2’s tapped, tuned primary, shunted by 100kW resistor R9, presumably to provide some damping and broaden the IF bandwidth. IFT2’s secondary feeds demodulator diode D1, a miniature allglass silicon diode. Jim Greig’s January 2025 article in Radio Waves magazine on the Sanyo RP-1250 shows a circuit that’s very similar to the R-72. The IF circuit, on pages 50-51 of that issue, is very similar to that of my Toot-a-Loop. Given the relationship between Panasonic and Sanyo, a shared design makes sense. That circuit notes the silicon rectifier diode as a BAY41. A silicon type is needed, rather than the usual germanium type, as its forward conduction voltage must match that of silicon transistor Q2 for proper AGC action. As usual, the diode is weakly forward-biased by the controlled IF amplifier’s bias circuit, in this case, 22kW resistor R4. The rectified IF signal is filtered by RC network A1. The demodulated audio is sent to volume control potentiometer VR1, and the rectified DC component is sent back to the first IF stage via 7.3kW resistor R10. Audio stages The as-built circuit of the yellow set’s audio stages was close to the published R-72S circuit, with a few oddities. The audio driver, rather than a metal-cased PNP 2SB475, was an epoxy-cased silicon NPN 2SC828 type. The R-72S showed an adjustable resistor of some kind in the lower end of the 2SB475’s base bias divider, but no emitter stabilising resistor. This would be consistent with laser-­ trimming, where the set would be put on test and the resistor element carefully vaporised to give the correct circuit voltages/currents. I was unable to discover any evidence of laser-trimming, though. As the 2SC828 is an NPN type, its emitter returns to supply ground, and it gets base bias via resistor divider R11/R12. Its collector feeds the primary of driver transformer T1, with its ‘cold’ end going to the battery’s positive terminal. Top cut is applied to the audio signal via collector-base feedback capacitor C12 (10nF). T1 phase-splits the audio signal and applies it to the bases of transistors Q5 and Q6 in anti-phase. Both are metal-can germanium PNP 2SB475s. They get around 200mV of forward bias via the R15/R16 divider and temperature compensation via 240W thermistor R14. Output transformer T2 feeds the 8W speaker via the earphone socket, with the usual disconnection of the speaker when the earphone jack is plugged in. Red set front-end My circuit (Fig.2) may seem unusual but, as it uses a positive supply, the audio section is easily understood. This does ‘invert’ the all-PNP RF/IF section, with the IF transformer primaries going to ground and transistor emitters returning to the decoupled positive supply. The RF/IF stage is a completely ordinary all-germanium circuit, which should be similar to that of the R-72S. My set differs from the R-72S in that the latter shows a 2SC829 silicon NPN converter wired in-circuit as PNP! It does correctly show its equivalent as a germanium PNP OC1044, however, which is equivalent to the 2SA101. The equivalence is confusing; the OC1044 is described as a ‘junction’ type (with a typical ft of 15MHz), while the 2SA101 is a drift-field type, with a typical fαβ of 25MHz (not an identical specification to ft but usually close). As built, converter Q1 gets bias via divider R1/R2 (5.6kW/33kW), with emitter stabilisation via 1.2kW resistor R3, bypassed by 22nF capacitor C4. R2 is one of four discrete resistors, probably used because printed-circuit types (marked on the 20RL012 circuit as “imp”) could not give sufficiently high (R2/R12) or sufficiently low (R14/ R15) values. Note, though, that the 20RL012 circuit shows R2 as a printed type. LO transformer T1’s tuned, tapped secondary feeds back to Q1’s emitter via 4.7nF capacitor C3. The tuning gang uses a cut-plate LO section, so R14 R9 R15 C13 Both variants (R-72S, left; R-72, right) are built on a double-sided PCB, with only two components on the bottom side. 96 Silicon Chip Australia's electronics magazine siliconchip.com.au there is no padder. The bias voltage on the converter transistor is only some 100mV, confirming the Class-B operation needed for the conversion process. Converter Q1 feeds the tapped, tuned primary of the first IF transformer, IFT1. Its untuned, untapped secondary feeds the base circuit of the first IF amplifier transistor, Q2, noted as a 2SA101/OC1045. Although the drift construction method improved high-frequency performance over that of alloyed-junction devices, the resulting collector-base capacitance was still significant, so Q2 is neutralised by 2pF capacitor C8. Q2 gets weak forward bias via 100kW resistor R4, bypassed for audio by 10μF capacitor C7. Q2 feeds second IF transformer IFT2’s tapped, tuned primary, with its untuned secondary feeding second IF amplifier transistor Q3. Unusually, transistor Q3 is also gain-controlled. This did not give much better output constancy with changes in signal strength, but it did give a very early onset of gain reduction, demanding an abnormally high input signal to give the standard 50mW output. More on this below. Q3 feeds the tapped, tuned primary of third IF transformer IFT3. Q3 operates without neutralisation, possibly due to demodulator diode D1’s loading of IFT3 giving a lower gain in this second IF stage. Demodulator diode D1, an OA70, feeds the IF filter block C11/C12/R7. Audio is fed, via 1kW resistor R9 and 1μF coupling capacitor C14, to the volume control. D1’s DC output is fed, via 10kW series resistor R8 and 10μF audio filter capacitor C7, to the bases of Q2 and Q3 for AGC, as noted above. Red set audio stages The red set’s audio section appears complicated, but it became the design of choice and remains so to this day. You’ll find its principles everywhere, from the LM386 audio chip to highpower amplifiers in the kilowatt range. Its circuit stability surpasses that of previous designs, while the removal of transformers and most capacitors allows manufacturers to deliver any level of performance, from low-power AM radio quality up to hifi systems of previously unmatched fidelity. Transistors Q5 (NPN) and Q6 (PNP) form a complementary pair. Feeding an AC signal to their bases will see Q5 siliconchip.com.au biased on for the positive half-cycle, with Q6 coming on for the negative half-cycle. They’re both configured as emitter-­ followers, so they provide roughly unity voltage gain. They do, though, provide considerable current gain, with high input impedances. A highgain driver stage can take the millivolt-­ level signal from a radio’s demodulator and amplify it up to speaker levels, with the output pair matching the low-impedance speaker load. In detail, audio from the volume control enters the circuit via coupling capacitor C16, arriving at the base of preamplifier/driver transistor Q4. Q4’s collector current, flowing via D2 and R13, becomes the driving voltage for Q5/Q6. Their emitters (via R14 and R15) connect together to drive the speaker via C20. The circuit is able to deliver almost the entire supply voltage (as a peak-topeak AC signal) to the speaker, around 8V peak-to-peak in this circuit. A quick calculation gives a potential output power of around 200mW into this set’s 40W speaker. It’s important that the circuit is biased correctly. This demands a quiescent (idling) current of a few milliamperes in Q5/Q6, temperature compensation to ensure that Q5/Q6 do not enter destructive thermal runaway at high temperatures, and that the Q5/ Q6 emitters set close to half the supply voltage, to allow maximum undistorted output, ie, equal magnitude positive and negative half-cycles. The quiescent current is set by the base-to-base voltage of Q5/Q6. Diode D2 is designed for a breakdown voltage of just about 1.2V, which is roughly twice the normal Vbe for silicon transistors. D2 also has a negative temperature coefficient (NTC). As the ambient temperature rises, D2’s forward voltage will fall, compensating for the 2 × -2.5mV/°C fall in Q5/Q6’s total Vbe. For the emitter voltages of Q5/Q6, we need to look at the DC feedback path via 560kW resistor R12. For example, a rise in the emitter voltages will supply more bias current (via R12) to Q4. This will raise Q4’s collector current, drawing its collector voltage down and lowering Q5/Q6’s emitter voltages. Fig.2: the circuit of my red set is different again, with most of the transistors having different polarities than in the yellow set! Australia's electronics magazine July 2026  97 A fall in the emitter voltages of Q5/ Q6 would result in less bias for Q4, allowing the circuit to send the Q5/Q6 emitters higher. It’s a simple feedback loop that stabilises the entire amplifier’s DC conditions. The final problem is to get enough voltage swing at the bases of Q5 & Q6. They need several milliamperes of base current to deliver full current into the speaker at signal peaks (the speaker current divided by their hfe figures). Pulling the bases down to switch Q6 on hard is easy; Q4 can readily pull Q6’s base to ground and supply many milliamps of base current. Pulling up seems harder. Let’s send Q4 to cutoff. Now, Q5’s base is fed via 820W resistor R13. Assuming we need about 2mA base current to bias Q5 fully on, we’ll get a drop of around 2V across R13. If only we could supply R13 from a higher voltage than the battery. That’s the job of R13’s connection to the speaker. With no signal, this point will be at around 9V. As the Q5/ Q6 emitters start to swing positive, so will the speaker voltage. But the speaker is already at 9V, so the positive half-cycle will see the speaker connection increase above 9V on the signal’s positive peaks. In theory, this point can get to The R-72 and R-72S share the same dial and case. Apart from the colour, the only other external difference is on the nameplate. around 13V. So that means that the voltage drop across R13 is fairly constant at around 4V. It’s known as a ‘bootstrap’ circuit, based on the principle of pulling oneself up by one’s bootstraps! This particular design offers circuit protection; if the speaker is open-­ circuit, there’s no DC supply to the top of R13 and the circuit simply fails to operate. Restoration Confusingly, a very recent search on Radiomuseum turned up an R-72S circuit that shows the RF/IF correctly, but retains the PNP audio driver! My advice is to always check any circuit against the as-built equipment. Both sets came in acceptable condition, although the red set’s coin slot had seen excessive force and was a bit mangled. They both cleaned up nicely with a spray wash and some automotive polish. I’ve had the yellow set for some years now. When I first tried it back at Harcourt, it failed to impress. I was Fig.3: the original circuit; it seems like it may be representative of only a minority of the R-72 sets that were manufactured. 98 Silicon Chip Australia's electronics magazine siliconchip.com.au able to pick up Melbourne stations, but at less than full gain. I put the lack of sensitivity down to the small ferrite rod antenna and the oddball resistance-capacitance coupled IF channel. But when I started to test the set, I found that the solder tab connecting the top of the ferrite rod’s tuned winding had broken, open-circuiting the connection to the tuning gang and leaving the antenna circuit completely untuned. The low sensitivity was one clue, but I also couldn’t get a peak at the 600kHz alignment point. If the antenna circuit was not being tuned, it would not resonate at 600kHz, so adjusting the LO to maximise the 600kHz sensitivity would be fruitless. As the antenna circuit would just be acting as a simple untuned pickup coil, the set would work about as poorly no matter the LO frequency. With that fixed, and with a quick tweak, 3WV Warrnambool rocked in at full volume; not bad for a station over 200km from my previous place at Rosebud. The red set was dead, though. No output, nothing. Connecting my monitor amp to the earphone socket got it going, and the lack of output was found to be an open-circuit speaker. Not expecting to get it rewound (does anyone repair/rewind three-inch speakers?), I got a replacement online. Its diameter was a little smaller than the original, but I turned a collar using a circle-cutter on my bench press drill from an old ice-cream container. That done, it was onto the test bench for alignment and performance analysis. We’ve just moved to Malvern, where the local levels of EMI even intrude on 774 ABC Melbourne’s powerful signal. Taking both Toot-a-Loops for a walk in the park, though, I was easily able to pull in my favourite 3WV at full speaker volume, albeit with a bit of noise. How good is it? I was able to test the yellow set at the standard audio output of 50mW. The audio response from volume control to speaker is around 220Hz to 4.6kHz; from antenna to speaker, it’s some 160Hz to 1.1kHz. Driving the output stage to clipping gave 150mW at 10% total harmonic distortion (THD), while the 50mW output (4.5% THD) and 10mW output (2.8% THD) are creditable for a Class-B push-pull output circuit. siliconchip.com.au Volume control Demod Output transformer 2nd IFT Driver transformer 2nd IF Output 1st audio 1st IF 1st IFT Oscillator coil Converter 3rd IFT 2nd IF Volume control Demodulator 1st audio D2 Output 1st IF 2nd IFT Converter 1st IFT Oscillator coil R2 The insides of both the R-72 and R-72S with some of the components labelled. Australia's electronics magazine July 2026  99 Some Panasonic history National Panasonic’s founder, Konosuke Matsushita, was born in 1894 to a family that fell on hard times. Young Konosuke was forced to leave school at age nine to find various jobs until he made his first invention, a light socket. While his product surpassed many others in quality and cost, he found marketing this product very difficult. The experience showed him the need to find marketing outlets for his products. He formed a strategy that reduced the energy put into manufacturing in favour of the establishment of a sales force that led to a retail store network. His next outing seems equally humble today: a bicycle lamp. Still, it was battery-powered, making it far superior and much more convenient than his competitors’ candle- and oil-based offerings. Matsushita was in danger of being removed as president of National after the end of WWII. One of General Douglas MacArthur’s strategies to rebuild the Japanese economy involved breaking up the Zaibatsu (large national corporations). A 15,000-strong petition from National employees changed MacArthur’s mind and, in 1947, Matsushita gave brother-in-law Toshio an unutilised manufacturing plant to manufacture bicycle lamps. This company eventually became Sanyo Electric. It’s tempting to cast National as the also-ran to Sony, especially given Sony’s remarkable rise from the ashes of WWII. Against this, we can consider Sony’s vulnerability as industry leader and the VCR wars of the 1970s and 1980s, which saw Sony’s Betamax outsold and finally obsoleted despite continuing improvements that ultimately delivered CD-quality audio. The competitor to Betamax, Video Home System (VHS), had been developed by Japan Victor Corporation (JVC) and was strongly supported by National, with Mitsubishi, Hitachi and Sharp adding their marketing power to the VHS push. It’s fair to say that Sony’s innovative energy and design flair would always be challenged by products that, while perhaps not the cutting edge of technology, were sound, reliable and well-marketed. And the leader of that group was Matsushita’s company, National. The Panasonic “Toot-a-Loop” radio could rotated into a ring, which was designed to be wrapped around your wrist. And all it needed was a single 9V battery or two AAs. 100 Silicon Chip The set needed 140μV/m at 600kHz and 110μV/m at 1400kHz for signal+noise:noise (S+N:N) ratios of 18dB and 13dB, respectively. For 20dB S+N:N, the levels were both 190mV/m. AGC control was as expected for a single stage: a signal rise of around +30dB gave an output rise of +6dB. Selectivity was ±2.3kHz for -3dB down and ±44kHz for -60dB. The skirt selectivity, especially, is very wide for a transistor set of this era, confirming the reduced selectivity expected from the use of only two IF transformers. Selectivity reduction is also increased by R9 shunting of ITF2’s primary. As for the red set, its unusual two-stage AGC showed very early onset, needing an artificially high signal to get to the Australia's electronics magazine standard 50mW output. Accordingly, I tested at 10mW output, so all measurements on the diagram are for a 10mW output. The audio response from volume control to speaker is around 125Hz to 2.7kHz; from antenna to speaker, it’s about 65Hz to 1.7kHz, although the unusually low bass response is wasted with the tiny speaker. The top end of just 2.7kHz seems low for an output-transformer-less (OTL) design, but 22nF capacitor C18, connected from Q4’s collector to ground, gives a significant amount of top cut. Driving the output stage to clipping gave 150mW at 10% total harmonic distortion (THD), while the 50mW output (2% THD) and 10mW output (3% THD) show the value of a well-­ designed OTL circuit. RF performance was also creditable; noting that I tested at 10mW audio output, the set needed 75μV/m at 600kHz and 120μV/m at 1400kHz for S+N:N ratios of 8dB and 7dB, respectively. Without the early-onset AGC, these should have equated to about 170μV/m and 270μV/m, respectively, for 50mW out. In reality, the red set needed 450μV/m and 550μV/m to achieve 50mW. Despite the two-stage AGC, the input change for a +6dB rise was only about +30dB, the same as for the single-AGC stage design. Selectivity was ±1.6kHz for -3dB and ±12kHz for -60dB. Would I buy another? I could keep going and get the complete set. It’s a striking example of packaging; take a popular commodity that’s gotten a bit ho-hum and wrap it inside an exciting, attractive case that makes it stand out from the pack. My research turned up the R-72, R-72S and Wadley RF-72 FM version from South Africa. There’s also the identical AM R-720 from Citizen Electronics. For a vividly ‘interesting’ design, hop onto Radiomuseum and look for the JVC Balance (8008). Special handling When replacing batteries, take your time and be kind to the coin slot. For more information on this series of radios, visit the following Radiomuseum links: R-72(S): siliconchip.au/link/acb1 RF-72: siliconchip.au/link/acb2 8008: siliconchip.au/link/acb3 SC 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 Do finished PCBs need a coating? I hope this isn’t a dumb question, but I recently saw a YouTube video where the guy said that once you were happy that you had soldered all the components onto the board and the circuit did what it was supposed to do, you should clean all the solder joints and, when dry, coat them with a protective lacquer. Examining most of my assembled projects from years past, the poor condition of the soldered joints suggests he could be correct. I cannot remember any mention in any of the project assembly instructions to do this. Have I missed this in the instructions, or is it so basic that it’s assumed I would know it? If he is correct, what cleaning fluid and coating compound should I use? Would they be available at the usual electronics suppliers? (C. K., Oxley, Qld) ● Actually, that is a good question. Generally, if using solder masked (coated) printed circuit boards (PCBs) and the correct solder for electronics, where the flux is non corrosive (such as rosin flux), you shouldn’t need to clean off the flux and coat the PCB with a protective coating. When bare copper PCBs are used, these will require a protective coating to prevent the copper tracks from corroding. If the PCB is used in corrosive conditions, such as in boats or coastal locations, coating is advisable after soldering. Both Altronics and Jaycar sell circuit board lacquer. If the flux is removed to make the PCB look cleaner, you can use isopropyl alcohol or methylated spirits for a less clean job. Specific flux cleaning products are also available, such as Chemtools Deflux It, available from Jaycar (there are two versions; we particularly like the G2 type). We have prototype boards that are over 20 years old and they are still mostly in very good condition, but that may be because they were stored in containers and/or in dry conditions. How important is Liion battery protection? The Human Comfort Indicator project (June 2026 issue; siliconchip.au/ Article/20362) specifies a lithium-ion rechargeable cell with built-in protection. I have looked online and in the Jaycar catalog, but the 14500-size, 3.2V batteries available don’t seem to have protection. Could you tell me if that is strictly necessary and, if so, where I could obtain one? (B. D., Mount Hunter, NSW) ● We purchased the cells used in that project a few years ago. We found the following cells online, which appear to be suitable, but we have not tested these: Klarus 14GT-92UR: siliconchip.au/ link/acc7 Elemex 14500: siliconchip.au/link/ acc8 Most of the real hazards (thermal runaway and fire) associated with Obtaining precision resistors for the SmartProbe I am intending to construct the SmartProbe (July 2025; siliconchip.au/ Article/18515). The parts list calls for ±0.1% 0805 SMD resistors (2 × 2MW, 1 × 680kW and 1 × 51kW), which seem to be difficult to obtain. Could you please advise where I might be able to get them? (J. A., Townsville, Qld) ● We found some options from DigiKey, Mouser and element14. For example: DigiKey: A119974CT-ND (2MW), P680KDACT-ND (680kW), P51KDACT-ND (51kW) Mouser: 279-CPF0805B2M0E1 (2MW), 754-RG2012P-684-BT5 (680kW), 754-RG2012P-513-BT5 (51kW) element14: 2992247 (2MW), 2337965 (680kW), 1670250 (51kW) Those three retailers should also sell a lot of the other parts required for the SmartProbe. siliconchip.com.au Australia's electronics magazine lithium-ion cells occur when they are over-discharged and the internal chemistry changes, and then the degraded cell is subsequently recharged. LiFePO4 (lithium iron phosphate) cells are generally considered more stable but can also be damaged by over-discharge. So if you use an unprotected cell and accidentally over-discharge it, you should not attempt to recharge it. Of course, that means you have to pay enough attention to know what has happened. The requirement for protection really comes down to your own circumstances and what sort of risks you are willing to accept. We wouldn’t use (or suggest) an unprotected cell if we thought it could be used by someone who didn’t understand those risks. In summary, we suggest you purchase one of the cells linked above for this project so you don’t have to worry about what will happen if something goes wrong. It’s easy to accidentally discharge a battery! Trouble with circuit built on protoboard I’m having trouble with the Simple LC meter on page 62 of the May 2026 issue (siliconchip.au/Article/20235). I used AVRDUDESS to program the ATtiny85 since my copy of Extreme Burner V1.0 doesn’t support that chip. I used an AVRMk2 programmer. I couldn’t program the Lock Fuse (LB) to 0xFF. It seems to default to 0xFD. AVRDUDESS doesn’t have the facility to change EEPROM bytes, but since my 1000pF capacitor is within 1%, I left that for now. I must state that I built it on some proto PCB material, the type with a grid of holes in a 2.54mm pitch. My capacitor is a mica type. Inductor L1 is a 100μH RF choke from my stock. 74HC04 is a DIP part from TI. I have mounted two terminal posts for the unknown part, rather than flying leads in the article. It is running from a 5V DC bench power supply. July 2026  101 Using the MPPT Solar Charger for charging smartphones I built the MPPT Solar Controller (February & March 2016; siliconchip.au/ Series/296) from one of the last available Altronics kits. I was just wondering if I could buy a cheap 120W panel from Jaycar and charge a 12V/7.2Ah Jaycar ($36) rather than 80Ah battery mentioned in the article. I don’t want to use it to control any lighting but to charge mobile phones and similar devices. I was thinking I could just use an alligator clip to cigarette lighter port connected straight to the battery. Would this work, or do I need to connect to the LAMP connector instead? (E. M., Kew, Vic) ● Yes, you can use a 7.2Ah battery. It will charge quicker than a larger battery. For use with charging mobile phones, connect the mobile phone charger across the battery directly. The mobile phone charger must be suitable for use with a 12V supply to deliver the 5V power for charging. Presumably, you will use a cigarette lighter USB adaptor for charging. I measured the 8MHz clock at TP2 using my 200MHz oscilloscope. It’s as close to 8MHz as the scope can measure. There seem to be some problems, though. When it powers up, the logo is displayed on the OLED screen. Pressing the CAL switch clears the screen, then shows the logo again. After about 25 seconds, the screen then shows: Calibrating Short test leads Then I see: Frequency: 3375532Hz Calibrating Short test leads If I measure at TP1, it’s actually 496KHz. If I add a nominal 47μH inductor, the frequency displayed changes to 323634Hz but my ‘scope measures 410KHz. So, the LC meter frequency shown is way off. Has the software been changed recently? I used the posted software from the 2nd of May. (G. P., Narre Warren, Vic) ● Building the Simple LC Meter on a prototype board is possible, but reliable operation may require small changes to the values of the 4.7pF crystal oscillator feedback capacitors, since protoboard can add unwanted stray capacitance. These capacitors should typically be 3.3-10pF. Programming the fuses and EEPROM with the correct values is critical for correct operation. While it isn’t the software we use primarily, we have some experience with avrdude/AVRDUDESS and it can definitely program fuses and the EEPROM. It’s odd that you say it can’t program the lock fuse to 0xFF, since that should be the default for this chip per its data sheet. Have you selected the right chip in AVRDUDESS (ATtiny85)? Once the fuses and EEPROM have been programmed correctly, the frequency reading on the screen will be right, and everything else should work as expected. Digital LC Meter library isn’t the expected one I’m in the middle of building the Wide-Range Digital LC Meter (June 2018; siliconchip.au/Article/11099). I downloaded the “Wide Range LC Meter software v1.zip” file and, upon opening the sketch, I see it includes the header file “LiquidCrystal_ PCF8574.h”, which corresponds to the text of the article. However, the zip file I downloaded includes the library “LiquidCrystal_ I2C”, which is not the same. This is not a problem if you follow the article, but it could cause confusion. Thanks for a great project. (R. S., Wellington, NZ) ● You are right, and there has often been confusion in relation to these I2C LCD libraries, especially as there are some variants that have very similar names. It’s a while ago now, but we believe that this project was developed around the time that the Library Manager was becoming a reliable choice for installing libraries for the Arduino IDE. We suspect that we intended to use the LiquidCrystal_I2C library (hence its inclusion in the software downloads), but we were able to switch to the LiquidCrystal_PCF8574 library instead, meaning that both necessary libraries could be installed via the Library Manager. As you say, the instructions in the article are correct; let’s hope that the Library Manager continues to work! SC200 and general amplifier questions I just purchased the SC200 Amplifier kit and I have some questions (January-March 2017; siliconchip. au/Series/308). It states that the PSU should be ±57V nominal. Will I be able to drive the amplifier to its full power using a 48V 10A power supply? 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. 102 Silicon Chip Australia's electronics magazine siliconchip.com.au MARKET CENTRE Advertise your product or services here in Silicon Chip KIT ASSEMBLY & REPAIR FOR SALE PCB PRODUCTION 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 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 KEITH RIPPON KIT ASSEMBLY & REPAIR: * Australia & New Zealand; * Small production runs. Phone Keith: 0409 662 794 keith.rippon<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 Silicon Chip Binders REAL VALUE AT $21.50 PLU S P&P PMD WAY offers (almost) everything for the electronics enthusiast – with full warranty, technical support and free delivery worldwide. Visit pmdway.com to get started. Order online from www.siliconchip.com.au/ Shop/4 or call (02) 9939 3295. 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. What modifications do I need to make? Could I use a 43-0-43V toroidal transformer? Do I need to change how I connect the amplifier module to the power supply module if I am using two transformers to power two channels? (W. L., Singapore) ● Taking your questions one at a time, sinewave power delivery is approximately V2 ÷ 4R, so for ±57V and 4W speakers, that’s 57V2 ÷ 16W = about 200W, or 100W for 8W. Assuming the 48V power supply you have mentioned supplies ±48V, with 4W speakers you will get 48V2 ÷ 16W = about 144W, or 72W into 8W. Make sure it’s a split supply; you can’t use a single-ended 48V DC supply to power a standard Class-AB amplifier. If it is ±48V, we don’t think any modifications would be required as that is close enough to the specified supply rails. siliconchip.com.au Regarding your second question on using a 43-0-43V toroidal transformer, it depends on the speaker impedance. We don’t recommend increasing the supply voltage for 4W speakers. For 6W or 8W speakers, a small supply voltage increase might be acceptable. Still, there are risks. Just a few volts can make a big difference to whether the amplifier is operating in the safe operating area (SOA) or not. We would expect ±61V DC rails with 43V secondaries on the transformer, although it could be a little higher depending on mains regulation, tolerances etc. Change all the MKTs to 100V rated (not hard to find) since the 63V ratings could be exceeded. You will need to replace the 100μF 63V electrolytic with an 80V or 100V type. Just make sure it will fit on the board. The small-signal transistors will be running close to their ratings – the Australia's electronics magazine BC546 and BC556 transistors are rated at 65V. They should be OK, but only just. If you want to up-rate them, we suggest switching to the KSC1845FTA (NPN) and KSC992 (PNP). We haven’t tested them, but they should give similar performance and they have 120V ratings. The bottom line is that if the speaker impedance is high enough, you could try it, but you will need to up-rate some of the parts (especially the capacitors) for safety. If you want to drive 4W speakers, you’re better off using a lower-­ voltage transformer (eg, 35-035V). You won’t lose much power and it’ll run cooler with much better margins. With each SC200 module powered via a single transformer and power supply module (that includes the bridge rectifier and filter capacitors), you would use the Fig.13 wiring diagram that is shown on page 79 of the July 2026  103 Advertising Index Altronics.................................43-46 Blackmagic Design....................... 7 Dave Thompson........................ 103 DigiKey Electronics..................OBC Emona Instruments.................. IBC Hare & Forbes............................... 9 Jaycar.................. IFC, 26-27, 36-37 Keith Rippon Kit Assembly....... 103 LD Electronics........................... 103 LEDsales................................... 103 Microchip Technology.................. 5 Mouser Electronics....................... 3 PCBWay....................................... 11 PMD Way................................... 103 SC Ideal Bridge Rectifiers........... 68 SC GPS Synchronised Clock...... 82 Silicon Chip Binders................ 103 Silicon Chip PDFs on USB......... 42 Silicon Chip Subscriptions........ 25 Silicon Chip Shop................ 69, 79 The Loudspeaker Kit.com.......... 93 Wagner Electronics..................... 10 Errata and on-sale date Simple LC Meter, May 2026: the initial batch of PCBs sold, including in kits, were inadvertently based on an old version of the board that had the USB power input pins shorted together on the bottom ground layer. All PCBs we supply have now been replaced with the correct boards without the short circuit. Also, the parts list called for M2 × 5mm machine screws, but that is too short. The correct screw length is 10mm. Our kits now include the longer screws. Next Issue: the August 2026 issue is due on sale in newsagents by Monday, July 27th. Expect postal delivery of subscription copies in Australia between July 24th and August 12th. 104 Silicon Chip March 2017 issue. This is for wiring up a single-channel SC200 to the mains power supply. Use similar wiring for the second stereo channel. Only join the grounds of the two modules at one point, ideally at a star Earth point. Are power amplifiers inverting? Firstly, thanks for the excellent magazine and projects. I have a question regarding your Studio 350, Ultra-LD and SC200 power amplifier designs. Are these non-inverting amplifiers, ie, are the output signals in phase with the corresponding input signal? I want to use them in an active threeway speaker system utilising a pair of the Studio 350s on the bass drivers and two pairs of Ultra-LD 200W amplifiers for the midrange and tweeters. I need to make sure there aren’t any phasing issues at the crossover frequencies. The speakers are Jamo R909s. The ULDs are superb amps, quiet with a neutral and detailed sound. (J. M, Auckland, NZ) ● Those are all non-inverting amplifiers. We don’t think we have ever published an inverting power amplifier, although some of our amps have been bridged types with both outputs being actively driven (so the – outputs are effectively inverting). It generally isn’t important whether a power amplifier is inverting or non-inverting, since you can almost always swap the speaker connections if you want to reverse the phase. While it’s possible to design a power amplifier to be inverting, it usually results in a more complex circuit for no real benefit. If you needed an inverting power amplifier, you could use one of our Amplifier Bridge Adaptors (May 2019; siliconchip.au/Article/11626), ignoring the non-inverted output. Air conditioner controller wanted I have a problem that I think may be common to other owners of reverse-­ cycle AC units. My nearly new Daikin unit has difficulty tracking the setpoint that I put into the remote. This is particularly the case when used with cooling. If I set it for, say, 21°C, the unit just keeps running until the room is very cold, many degrees under the set point. Australia's electronics magazine Setting the temperature higher has little effect, and the actual room temperature will fluctuate depending on the ambient conditions. I understand the difficulty for the manufacturers – the temperature sensor is located on the wall unit, instead of somewhere else in the room. I don’t understand why they don’t make a remote that communicates constantly with the wall unit to maintain a steady room temperature. I was considering cobbling something together by combining a thermostat with an IR transmitter to simply send the on/off code to the wall unit from the other side of the room and wonder if you have ever published something that would do the trick. Perhaps your Circuit Notebook entry on the Micromite-based Air Conditioner Remote Control (December 2017; siliconchip.au/Article/10914)? I lodged a service call under warranty and was called back by a company engineer who basically told me that that is just the way it is. (K. W., Newport, Vic) ● This does seem to be a design flaw in many split-system air conditioners, especially considering that IR sensors that can read the average temperature in a room are available and are not expensive (like our Thermopile-based Heater Controller from the April 2018 issue; siliconchip.au/Article/11027). We don’t have a great solution for this since there are too many different kinds of air conditioners to support. As you suggest, a remote sensor that could send commands to the unit via infrared would seem to be an ideal solution, but it would have to be tailored to the specific brand and possibly model. The December 2017 circuit you mentioned should work if you can get access to an on/off switch. Arduino libraries are available for controlling Daikin air conditioners (eg, https://github.com/danny-source/ Arduino_DY_IRDaikin). Temperature sensor interface libraries are also easy to find. We also have some Daikin air conditioners to test with, so we will consider putting together an Arduino-based remote controller with an onboard temperature sensor that can either switch the unit on and off, or adjust its settings, to provide better control of the room temperature. If this works, readers could potentially expand the code to work with different brands SC and models. 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