Fullscreen HTML5Med Res (??MB)HTML4

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 How to convert 12V AC to 24V DC I want to add low-voltage LED retro festoon string lights to an existing garden lighting system. All the units that look good and are robust require 24V DC, whereas the garden lighting is 12V AC. Additional cabling would require lifting up pavers & digging up an established garden. So I am keen to know whether I can use a small 230:115V AC transformer in reverse to step up the voltage and then regulate it to DC. The 12V transformer has plenty of spare capacity as originally sized for 12V halogens which have been replaced with 12V LED units. (T. H., Batehaven, NSW) ● It’s possible that the step-up transformer would work; it depends on the details of the transformer, but likely it would give you double the AC voltage. You would still have to convert that to DC. However, there are simpler/ easier ways. To start with, a simple full-wave voltage doubler feeding a pair of 4700µF 16V electrolytic capacitors will give you pulsating DC averaging around 25V DC with a 2A load. Using pairs of diodes in series would drop that to be very close to 24V DC, as shown in Fig.1. The capacitors need to be low-ESR types with a high ripple current rating, ideally at least 2A. There will be a few volts of ripple across the LEDs (6V peak-to-peak according to our simulation). We suspect that won’t bother them, but it depends on their exact design. The output voltage of that circuit is somewhat load-dependent; for a 1A load (say), you would just need to reduce the filter capacitors to 2200µF and eliminate the extra series diodes to get much the same output voltage. If you need the DC supply to be ripple-free, there are a few ways to achieve that. Fig.2 shows a similar circuit with a very basic linear regulator based on an NPN transistor and zener diode. Simulation shows it delivers a smooth ~24V DC output. The transistor should be a type with decent gain up to a few amps (such as the KSC2334Y shown). Q1 will dissipate close to 10W, so it will need a decent heatsink; an IP65 sealed metal case could be used to house the circuit and also as a heatsink, with an additional finned heatsink bolted to the outside directly opposite the transistor. The output ripple with this version is a fraction of a volt. With a linear design like this, it’s hard to avoid dissipating a few watts if you need a mostly ripple-free output. For a more efficient approach, try using a rectifier/filter circuit like those shown here with Tim’s Blythman’s Buck/Boost LED Driver (June 2022; siliconchip.au/Article/15340) to convert the pulsating DC to a smooth, regulated DC. We have a kit for that project (siliconchip.au/Shop/20/6292). Because that board can deal with an input voltage above or below the output, the exact voltage being fed to it isn’t critical. However, it’s better to arrange for the input voltage to be above, or at least close to, the output voltage. That will give maximum efficiency. It can deliver around 5A in this configuration, given sufficient filter capacitance on the output of the rectifier. Dimming for Buck/ Boost LED Driver I ordered the Buck-Boost LED Driver kit (SC6292; June 2022, siliconchip. au/Article/15340) and LED panel from you last week, and the package arrived promptly early this week; thank you. It’s now assembled and running nicely. Compliments to whoever laid the PCB out, as it’s obviously been designed for ease of assembly. Having most of the smaller components lined up across the board’s edges with large pads made it quite easy to place and solder them. One comment about the assembly instructions: nowhere did I see any mention of soldering the ground pad of the LM5118 to the PCB. I heated the via on the underside of the PCB with a hot iron and fed a quantity of solder into the via. It seemed to wick in, so hopefully that worked. Fig.1 (left): the full-wave voltage doubler will give close to 24V DC from 12V AC, but with substantial ripple. Fig.2 (below): adding a very basic linear regulator applies a smooth 24V DC to the load, but with about 10W dissipation. Using the Buck/Boost LED Driver instead would result in much lower losses. 100  Silicon Chip Australia's electronics magazine siliconchip.com.au Getting to the point of my email, Can you suggest a modification to add a dimmer control? I want to make a work light with the LED, but it’s very bright. I’d thought the current limit adjustment pot might work, but it only reduces the current to 1.5A, which is still very bright. (D. S., East Melbourne, Vic) ● Thanks for the kind words about the layout. We must admit that the LM5118 is well suited to a convenient layout. You are correct that we should have mentioned the thermal pad on IC1 – we soldered it on our prototype. We have published an erratum to make readers aware of this. With hand soldering, the process you used is about the only practical way to solder such a pad. Unfortunately, the current control does not lend itself well to a minimum value near zero as it depends on the current rising high enough to overcome the diode threshold above the 1.23V reference voltage. If you are comfortable with a proportionally lower maximum current, substituting a higher-value shunt resistor than 15mW (the one between TP4 and TP6 only) should work. Another option is to apply a PWM signal across JP1, which should effectively PWM the output. We haven’t tested this; you might need to remove (or reduce) the 100nF capacitor labelled C14 attaching to IC1’s pin 7 (7th component from the left along the bottom) to cut out the soft-start ramping. We think a relatively low PWM frequency would give the most linear response. Finally, you could simply add an external voltage control pot. It won’t give perfect control but should let you cut the brightness way down, while the current limiting will prevent any damage to the LED at the upper end of its range. With the current limit at 4A, the voltage pot range from off to full will be about 700W to 400W. So a 500W pot with a 750W parallel resistor between the wiper and one end, plus a 390W resistor in series with both, should give a suitable adjustment range. This combination can be wired up in place of the onboard 5kW multi-turn trimpot. Solar PV (photovoltaic) water heating What is the strategy to connect solar panels (in series or parallel) to a water heater with a resistive element, say about 2.4kW? I expect you need to interface with maybe an inverter and batteries. Perhaps you have covered this previously in the magazine. (F. C., Maroubra, NSW) ● We have covered this previously in the magazine on several occasions. There are various ways to do what you are asking, all of which have problems. For example, see: • September 2013, pages 98 & 99 • September 2014, pages 98 & 99 • October 2017, page 96 • December 2017, pages 4 & 5 The bottom line is that if you’re going to use solar power to heat water, you ideally want an electric water heater with dual elements (a main element and a ‘booster’) so that you can power one from the mains and the other from solar power. That way, you’ll always have hot water. The problem with feeding solar power to an inverter with the output of the inverter driving one of the water heater elements is that it likely won’t deliver any power unless there is enough solar power available to drive the element at its full power rating (eg, 2.4kW). So that approach is generally not going to work well. That leaves the idea of arranging the panels so that they produce around 240V DC in peak sun under load and driving the heater element with DC. This has the advantage that it will provide whatever power is available, even if your panels cannot deliver enough power to achieve the full rated element power. It’s also going to be the most efficient method. However, using DC will promote corrosion, so it is necessary to arrange for a contactor or similar to reverse the polarity periodically (eg, every 12 or 24 hours). Also, it is unclear whether the thermostat will last long if it’s switching DC rather than AC. Finally, you would probably need to get a licensed electrician to do the wiring if the heater is also mains-­powered, and it’s unclear whether your average electrician would want to do this sort of work. In conclusion, you’re probably better off installing a standard solar hot water system as they are designed for the job. Help to find a Coilcraft part I am having trouble finding one of the Coilcraft items for the Precision AM/FM DDS Signal Generator (May 2022; siliconchip.au/Article/15306). Please verify the part number for the Coilcraft 1206CS-121XJEC 120nH chip inductor. It does not appear to exist on either the Coilcraft site or Tricomponents site. (J. S., Avondale, Qld) ● It seems like a valid part number. Here it is on Coilcraft’s website: siliconchip.au/link/abff We don’t think that using that particular part is especially critical; Raspberry Pi Pico BackPack With the Raspberry Pi Pico at its core, and fitted with a 3.5inch touchscreen. It's easy-to-build and can be programmed in BASIC, C or MicroPython. There's also room to fit a real-time clock IC, making it a good general-purpose computer. This kit comes with everything needed to build a Pico BackPack module, including components for the optional microSD card, IR receiver and stereo audio output. $80 + Postage ∎ Complete Kit (SC6075) siliconchip.com.au/Shop/20/6075 The circuit and assembly instructions were published in the March 2022 issue: siliconchip.au/Article/15236 Australia's electronics magazine August 2022  101 many manufacturers have 120nH chip inductors that could be used instead and might be easier to get. Note that element14 stocks the XGLC and XJLC versions of those chip inductors, and they would be fine in this application. Yet another version of the R80 Receiver kit I refer to the review of the R80 Receiver kit in the November 2021 issue (siliconchip.au/Article/15101). I have built this unit but am having problems with the modifications on page 43. The BC548 emitter is connected to pin 9 of the display PCB plug (GND), while the collector is connected to the junction of R18 and D3, not the junction of D2 and CP5 as shown. The BC548 is mounted with the flat towards the board, not away from it as shown in the article. I have checked for errata in later editions but could find nothing on this. Was this a later modification? (J. P., via email) ● Andrew Woodfield replies: your kit appears to be an earlier (!) version than the one we reviewed, most likely V6 (the review was of V7). Most readers who reported discrepancies between their kits and our review had later versions of the kit. We aren’t sure if you need to make the squelch modifications to the V6 kit. It may help, but it will have to be done differently as the design is obviously not the same. It would help if you had a circuit for your version of the kit to compare to the one for V7 posted on the Silicon Chip website (siliconchip. au/Shop/6/5950). Flight level is based on pressure, not altitude I am reading the “Advanced GPS Computer” by Tim Blythman (June & July 2021; siliconchip.au/Series/366), and I hope you’ll indulge me in clearing up an uncertainty. The Computer can show flight levels (FL) as well as altitude, but I missed seeing how barometric pressure is measured (which is necessary to convert altitude to FL). Many thanks for pointing this out to me; I hope I haven’t missed the obvious. (G. M., aircraft museum curator, London, UK). ● The Advanced GPS Computer uses the altitude data from the GPS receiver 102  Silicon Chip module, so it does not calculate an altitude or flight level based on barometric pressure. So it is more correct to say that it can display the altitude in the same units as flight level, but does not display a true flight level based on pressure. We’re considering an update to the project which adds a barometric pressure sensor like the BMP280 so that it can show the proper flight level. Design for density altitude meter wanted Have you published a design for a density altitude meter? (L. B., via email) ● We have not, but we have published hardware designs that could be used as a density altitude meter with some relatively simple changes to the software. For example, the Touchscreen Altimeter and Weather Station (December 2017; siliconchip.au/Article/10898). It has an onboard barometric pressure sensor and temperature sensor. Those provide the two values that you need to compute the density altitude. The BASIC software is available online (siliconchip.au/Shop/?article=10898), so it should be relatively simple to modify the software to calculate and show the density altitude, then upload that to the BackPack. Transducer power for Ultrasonic Cleaner I am interested in building the “Large Ultrasonic Cleaner” featured in the August 2010 issue (siliconchip. com.au/Article/244) but with some modifications. First of all, I am wondering if it’s possible to use a 35W transducer instead of 50W as the larger unit is somewhat hard to get at the moment, and the smaller unit can be had for a very good price on sale. If so, I imagine that some changes may need to be made to the drive circuit to avoid over-driving the transducer. Secondly, to compensate for the lower-power transducer, I was thinking of running two in parallel. I imagine this achieved not by simply running both transducers from one circuit, but by duplicating the drive circuitry (Mosfets and transformer), running both sets from one microcontroller. Is that feasible? Or would two transducers on the same cleaning tank Australia's electronics magazine interact negatively with each other? (A. C., Auckland, NZ) ● You can use the 35W transducer at a reduced power level. We don’t recommend using more than one transducer, even if the driver section is duplicated. That’s because each transducer needs to operate at the correct frequency for the selected output power. Individual transducers will have slightly different resonances, so one transducer will be delivering the majority of the power output if two are used. Note that we published a revised version of that project in September & October 2020 (High Power Ultrasonic Cleaner, siliconchip.com.au/ Series/350). Altronics sell a kit for the newer design, Cat K6022. BWD power supply circuit diagram wanted In the Serviceman’s log column in November 2010, you detailed a repair to a BWD 207B power supply and mentioned that you had obtained several versions of the service manual. I have tried various methods to try and obtain a service manual for this power supply but to no avail. Are those manuals available on your website, and if so, where? (P. A. S., via email) ● We haven’t uploaded the BWD 207B manual or circuits to our website but we can supply them upon request. 2010 DAB+ Tuner has limitations I have had this DAB+/FM Tuner (October-December 2010; siliconchip. au/Series/13) running for many years now (I built it from the Jaycar kit) without any problems. The ABC has recently updated (reassigned bandwidth) to their DAB+ transmissions. They recommended re-scanning the stations, so I did, but the scan did not complete. I turned off the unit and found that it only found 62 stations of the 72 available in Melbourne. I did this at least three times with the same result each time. I did notice that some but not all of the ABC stations were scanned. Once, it left out ABC Melbourne. I have the latest firmware of 7.71. Do you know why it is not picking up all the channels? (A. L., Watsonia, Vic) ● It turns out that the 2010 DAB+/ FM Tuner firmware has a hard-coded continued on page 104 siliconchip.com.au limit of 64 stations in total that it can store. We are unsure of the reason for this limitation as the designer of that project left Silicon Chip many years ago. We assume it is due to the limited amount of flash memory or RAM available. It’s possible that could be changed, but we no longer have a prototype to test new firmware, nor are we confident that compiling the source code with a current compiler will necessarily produce working code without being able to test it. Our much more recent Touchscreen DAB+/FM/AM Tuner design (January-March 2019; siliconchip.au/ Series/330) does not suffer from this problem. It must have an upper bound on the number of stations it can tune, Advertising Index Altronics.................................43-46 Dave Thompson........................ 103 Digi-Key Electronics...................... 3 element14................................... 11 Emona Instruments.................. IBC Hare & Forbes............................. 13 Jaycar.......................... IFC, 5, 7, 21, ..............................24-25, 77, 79, 99 Keith Rippon Kit Assembly....... 103 LD Electronics........................... 103 LEDsales................................... 103 Microchip Technology.................. 9 Mouser Electronics..................OBC Ocean Controls............................. 8 SC Pico BackPack.................... 101 but nobody has run into it yet. Also, the source code is available (it’s written in BASIC), so it could easily be fixed if such a limitation existed. While the parts for that project are somewhat hard to come by, we do have a handful of Si4689 ICs on hand for anyone who wishes to build one, and the PCB, Explore 100 kit and other associated parts are still available – see siliconchip.au/Shop/?article=11369 Consider, though, that this newer project requires some fairly small SMDs to be soldered. But we think that is a better situation than the old tuner, which needed a module that wasn’t available to purchase (it only came as part of the now-discontinued Jaycar kit). Replacement pot for Class-A amp My 20W Class-A amplifier (MaySeptember 2007; siliconchip.com.au/ Series/58) is in need of a new volume pot as it has gone all scratchy. It is an Alpha dual-gang 20kW log motorised pot, previously sold by Altronics as Cat R2000 but no longer available. Do you know of a suitable replacement? Is it possible to retrofit one from another brand? (N. M., Sunbury, Vic) ● We suggest you try spraying some contact cleaner into the wiper assemblies first as that might resolve your scratchiness with much less expense and hassle than replacing the pot (eg, try Jaycar Cat NA1012). If you do need to replace it, the pot value is not so critical (eg, you could use a 10kW or even 5kW dual-gang log pot). The main concern is it fitting on the existing preamp PCB. Bourns Pro Audio PRM162-K420K103A1 is a 10kW dual gang logarithmic pot with dimensions very similar to the Alpha unit Altronics sold. The PCB SC SMD Test Tweezers.............. 63 Silicon Chip Shop.................70-71 Silicon Chip Subscriptions........ 23 Silvertone...................................... 6 The Loudspeaker Kit.com.......... 10 Tronixlabs.................................. 103 Wagner Electronics..................... 73 104  Silicon Chip Errata and Next Issue Silicon Chip Binders................ 103 probably would need some slight modifications to make it fit (due to slightly different motor mounting posts) but we think it would not be too hard to retrofit. Unfortunately, it is somewhat hard to find anyone with this in stock. Verical list 142 in stock at the time of writing and we have purchased from them before, so we think that is a reasonable option (siliconchip.au/ link/abdh). Master Electronics also say they have 147 in stock but we have no experience with them – see siliconchip.au/ link/abdi Stereo Compressor kit wanted Is the Stereo Compressor kit still available? It used to be sold by Jaycar, but they no longer supply it; I hope you can help. (A. B., Christchurch, NZ) ● We designed two different Stereo Compressors that were made as Jaycar kits, one in June 2000 (siliconchip.au/ Article/4328; Jaycar Cat KC5291) and one in January 2012 (siliconchip.au/ Article/809; Jaycar Cat KC5507). Jaycar makes their kits based on our articles; we do not make them for Jaycar. It does look like both Jaycar kits are no longer available. You might be able to get a kit for the January 2012 design from Altronics (Cat K5526). It is still listed on their website; it looks like they are very low on stock, but they have kits in a few stores. Perhaps if you contact them, they can get one back to their warehouse and deliver it to you. Failing that, we can supply PCBs and panels for the January 2012 design, see siliconchip.au/Shop/?article=809 You would need to get the rest of the parts yourself. It looks like they are still available; the critical part is the SA571 IC (in DIP), and they are for sale on eBay. SC Spectral Sound MIDI Synthesiser, June 2022: the orientation of diode D2 in Fig.9 is incorrect. Install it with the cathode stripe facing to the right, as shown on the PCB silkscreen. Digital FX (Effects) Pedal, April & May 2021: Fig.2 in the April issue shows incorrect connections for op amp IC3b. Its pins 5 & 6 are swapped. Pin 6 (−) should be at the top, connected to the 4.7μF capacitors, while pin 5 (+) should be at the bottom, connected to Vcc ÷ 2. The PCB has the right connections. Next Issue: the September 2022 issue is due on sale in newsagents by Monday, August 29th. Expect postal delivery of subscription copies in Australia between August 29th and September 16th. Australia's electronics magazine siliconchip.com.au