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 Sourcing the AX-1000 current transformer I’m gathering the components to build the new Refined Full-Wave Motor Speed Controller (April 2021; siliconchip.com.au/Article/14814). I have run into a problem sourcing the Talema AX-1000 current transformer. Digi-Key has a minimum order quantity of 2,500 with none in stock and an 18-week lead time (probably closer to 26 weeks with the Suez Canal currently blocked). Do you know anyone else that has it? I can’t find it anywhere. Great magazine this month. Well done. (W. M., Sunnybank Hills, Qld) • At the time of writing this reply, the AX-1000 is in stock with RS Online (https://au.rs-online.com/web/p/ current-transformers/7754928/) at nearly $7 each, plus postage, with a minimum order quantity of one. They currently list 18 in stock in Australia, with another 239 overseas. How to obtain capacitors for projects I would appreciate help explaining the capacitors needed for the Arduinobased Adjustable Power Supply in the February 2021 issue (siliconchip.com. au/Article/14741). It lists the codes for the capacitors as 100nF (code 104, 100n or 0.1) and 1nF (code 101, 1n or .001), but I can’t find these at Mouser, Jaycar or Altronics. Can you point me to the right components to buy via online links or Jaycar/Altronics catalog codes? (M. W., Preston West, Vic) • The 100nF MKT capacitors are available from Jaycar (Cat RM7125) or Altronics (Cat R3025B), and the 1nF MKT capacitors are also available from Jaycar (Cat RM7010) or Altronics (Cat R3001B). The easiest way to find these on the supplier websites is to go to the search box and type in either “100nF MKT” or “1nF MKT”. The codes we list in the magazine are the short codes 108 Silicon Chip printed on the parts themselves, rather than any supplier part number. They are necessarily brief due to the parts’ small size and don’t give the full part type code. Mouser (one of your listed suppliers) also have plenty of suitable capacitors, for example, au.mouser.com/ Search/Refine?Keyword=810-FK26X7R1C106K For the 10μF ceramic part, the text in brackets refers to the capacitor’s size (metric 3216 [3.2 x 1.6mm] or imperial 1206 [0.12 x 0.06in]). Searching the website of a supplier like Mouser or Digi-Key for “10uF 16V X7R 1206” (if you want the SMD type) or “10uF 16V X7R radial” (through-hole version) should give several relevant results. Questions about Four Battery/Cell Balancer I’m planning on building the High-Current Four Battery Balancer (March-April 2021; siliconchip.com. au/Series/358) project and am ordering the parts. This Balancer circuit is controlled by a PIC micro, but neither the software nor the pre-programmed controller is currently available on your website. The option that is eventually made available will affect what parts I order from SC and what I order from elsewhere. My current hypothesis is that you haven’t released the software because it is still being optimised, and you don’t want to let it into the wild until it is stable. What’s your intention? Also, I had been looking at various active balance topologies before this timely project. So armed with a bit of background reading, I have two further questions about your design. Only one transformer is active at any time, so the switching could be arranged so that a single transformer is shared. Is the reason you’ve elected to use one transformer per cell that it cuts down the parasitic inductance, or is there some other reason? Finally, I’m not sure what voltage Australia’s electronics magazine should appear on Vstack. Fig.11 hints that it is the series voltage across all cells, consistent with ZD5 having a higher voltage than the corresponding per-cell zeners. On the other hand, the 1:1 ratio of each cell transformer suggests that the stack voltage is close to that of a single cell. I am quite looking forward to the next instalment, showing how this circuit is configured. (M. J., St Lucia, Qld) • It’s actually an ARM-based Atmel micro (ATSAML10E16A-AUT), but it can be programmed using a PICkit 4. We have the programmed chips for sale on our website now; when you wrote in, the second (constructional) article had not been published, and we had not yet received the software from the designer. Regarding only one transformer being active, the firmware currently acts in this way, but a future version could have multiple channels active simultaneously. The hardware certainly supports that. Duraid considered ‘shared primary’ type transformers, but there aren’t many appropriate (high coupling, low DCR resistance, low parasitic inductance) parts on the market. That is the same reason he didn’t use 1:2 transformers; they are more common, but still scarce compared to 1:1 types. This reduces efficiency somewhat, but it vastly increases flexibility in the parts choice. Many different 1:1 transformers will fit on the Battery Balancer board without modification. Also consider that the cost of the additional Mosfets involved in switching a single transformer between the cells would probably be higher than the cost of the extra transformers (highperformance Mosfets are not cheap!). The board layout would also be complicated considerably, as the current from all the cells would have to be brought to a single part, rather than there being separate sections for each cell. Concerning the stack voltage, in the typical battery balancing case, this will be the series voltage across siliconchip.com.au all cells in the battery, because that’s what it will be tied to. Again, the 1:1 winding choice comes with some efficiency cost, but it doesn’t preclude charge being transferred across these different voltages. However, the board has other potential use cases, such as shuffling charge out of a solar panel into cells, charging a battery from a DC bench supply, or even discharging cells into an electronic load. In all of these cases, the Vstack voltage will generally be different from the sum of the cell voltages. BK1198 single-chip radio antenna queries I am building the AM/FM/SW Single-Chip Digital Radio published in your January 2021 issue (siliconchip. com.au/Article/14704), and I have a question about the antenna coil. I went with the ‘coil transplant’ method and found that the inductance value exceeds the nominated 400μH value before the coil is entirely on the ferrite rod. When the coil is flush with the end of the rod, it measures 720μH, almost twice the nominal value. A little bit further on, and it is 2.5 times the nominal value. On another note, there is a discrepancy between the schematic on page 22 and the PCB. The 18pF capacitor C22 is connected in parallel with wirewound inductor L8 near the FM antenna connection on the PCB, whereas the schematic shows C22 shunting to GND. I am assuming the PCB is correct. (S. S., Zillmere, Qld) • The author, Charles Kosina, replies: The BK1198 chip automatically tunes the antenna coil, so the actual inductance is not that important. Because of the ferrite characteristics, the permeability varies a lot with frequency, so the inductance measured by the meter may not be a true indication. It depends on the inductance meter and what frequency it tests at. I measured the original ferrite rod purchased from Jaycar using a Q meter. At 500kHz, it resonates with 175pF (Q=80), which means the inductance is 579μH. At 1MHz, it resonates with 30pF (Q=50), which gives an inductance of 844μH! So the nominal value of 400μH is way off at either frequency. As for 18pF capacitor C22, from an RF point of view, it makes no difference whether it goes to ground or the other end of L8. The 100nF capacitor siliconchip.com.au at the DC input to the coil is a virtual RF earth due to its low reactance at that frequency. Yes, it is a minor schematic error that does not affect the performance of the radio. Advice wanted on CNC machines & laser cutters I read your article in the December 2020 issue about using CNC milling machines to make PCBs (siliconchip. com.au/Article/14672). I am teaching year 9 and 10 electronics at Casino High School in NSW. After much R&D, we were able to produce pretty good PCBs using a laminator and acid etching. However, after reading your article, I wish to purchase a milling machine. Unfortunately, your article did not mention any brand names. Is it possible for you to give me some idea as to which makes are better than others to give me an idea as to what to buy? Thank you very much for a great magazine, keep up the good work. (R. M., Casino, NSW) • The December 2020 article was contributed by Andrew Woodfield. While we have not tried any of these techniques ourselves, we make frequent use of a laser cutter similar to the unit shown on p38 of that issue. Since the article was focused on laser engraving, we did not look too deeply into CNC milling machines. If you want to mill PCBs, we understand the Bantam PCB Desktop Mill is the device to consider, as it is one of the few milling machines designed with PCBs in mind. Core Electronics used to carry these, but they are not available at the time of writing this (see https://core-electronics.com.au/ brands/bantam-tools-australia). In regards to laser engraving, many machines are imports of various degrees of quality, so a specific brand name is unlikely to be helpful. There are several Australian laser cutter/engraver sellers. We suggest that you look up and contact one or more of these, as they will be able to provide you with more detailed information, demonstrate their units and provide better peace of mind (eg, warranty) than obtaining one from overseas. As Andrew mentions, you will need at least 5W of laser power, and naturally, a working area large enough to accommodate the largest PCB you want to make. Australia’s electronics magazine How does wireless charging work? I purchased a new electric razor some time ago, and it came with a wireless charger which took about an hour to charge it. I was wondering how they worked. PS, please bring back the computer articles. (R. M., Melville, WA) • Wireless chargers use inductive coupling, much like an air-cored transformer. The principle is simple; the devil is in the details, such as using resonant energy transfer for better efficiency, and switching off the field when the razor is not ‘docked’. For more details, see the Wikipedia page at https://w.wiki/399W Breaking out Maximite DIL I/O header Has anyone come up with a breakout box for the 26-pin I/O connector on the back of the Maximite (March-May 2011; siliconchip.com.au/Series/30)? I believe this would be very useful. I have a Colour Maximite (September & October 2012; siliconchip.com. au/Series/22), but I have not had a chance to use it. I intend to start soon. I am very interested in electronics (I used to work in the industry) and astronomy. There is a warning in the Colour Maximite kit instructions about the current that can be drawn from the 5V and 3.3V rails (total of 150mA). Should the breakout box have its own power supplies? I am in awe of the Colour Maximite 2 (July & August 2020; siliconchip. com.au/Series/348). Does anyone sell one already built and tested? (R. M., Melville, WA) • It’s a bit difficult to make a “breakout box” for the Maximites because we don’t know what people will be using it for. It’s easy to plug DuPont cables into the connector and plug the other ends into a breadboard or other modules. It’s also possible to plug in a ribbon cable with an IDC header, then connect the ribbon cable to a header on a breadboard or similar. Our DSP Crossover LCD Adaptor PCB (code 01106196) converts a 20pin DIL plug or socket into a 20-pin SIL plug or socket, which could then be connected to a breadboard. You’d be left with six pins unused, but that would still be an easy and cheap way May 2021  109 to connect a Maximite or Colour Maximite to a breadboard. Some breadboards also have their own power supplies, or you can build one of the many 3.3V/5V/adjustable regulators we’ve published over the years to supply extra current. Rictech has a CMM2 board with all the SMDs pre-soldered, but there is still some assembly to be done. See www.rictech.nz/micromite-products Lead-acid vs silvercalcium batteries I was recently stranded with a flat battery in my 1995 Ford Falcon. My local motoring association patrolman arrived promptly, pronounced the battery “dead”, and fitted a new one. He got me to turn the headlights on to high beam and rev the engine while he measured the battery voltage. He told me he got 13.8V, so he said everything was fine, and departed. The next day, I had a good look at the new battery, and noticed that it was a “calcium battery”. Knowing nothing about these batteries, I did some online searching. I learned that these are fairly conventional lead-acid batteries, but with calcium instead of antimony as an additive. This is said to produce several performance advantages, including less gassing, so they are usually made sealed and ‘maintenance free’. However, they require a higher charging voltage than standard lead-antimony batteries (14.8V vs 13.8V). This being the case, I wondered whether my (older) car’s electrical system would be capable of ever charging the new battery fully, and whether this would lead to sulfation of the plates and an early demise of the battery. I contacted the motoring association that had sold me the battery and asked for some advice. Basically, what I got was: “She’ll be right mate, don’t worry about it. You’ve got a 2-year warranty anyway.” This was hardly satisfactory. More online searching produced a confusing array of contradictory information: calcium batteries are not suitable for older vehicles; calcium batteries are suitable for older vehicles etc. Can you shed some (sensible) light on the subject? Will a standard car charging system fully charge a calcium battery? Given that these batteries are sealed and cannot be checked with a hy110 Silicon Chip drometer, I assume that measuring the terminal voltage would be the only way to tell if the battery is fully charged. If so, what should the terminal voltage be, and under what conditions? What should the terminal voltage be of a fully charged calcium battery at rest, with no load? (D. P., Faulconbridge, NSW) • There is some confusion on various web sites between calcium/calcium and silver/calcium lead-acid batteries. Just about every car battery available now is a calcium/calcium lead-acid type. These are sealed and so do not require topping up with distilled water. The older type lead-acid batteries, which had antimony, tin and arsenic added to the lead, are no longer being sold as those heavy metals are toxic. (Of course, lead is too, but it is not generally as troublesome.) Calcium/silver lead-acid batteries are different again and have higher self-discharge rates, and require higher charging voltages. They are not suitable as drop-in replacements for standard lead-acid batteries unless the charging voltage(s) can be adjusted. Typically a car electrical system will charge to 14.4-14.8V. Your 13.8V charge voltage is very low and is generally the float charge value for a lead-acid battery, not the end-of-charge voltage. The alternator’s charge voltage is temperature-dependent, so the measured voltage could be lower than usual under high-temperature conditions. Old Silicon Chip PCB code decoded Greetings to all at Silicon Chip. I have a PCB with the code SC0611287. If this is one of your projects, please advise me of its date of publication. I can then look it up in my library. Many thanks, and please pass on my best wishes to Ann, who has been most helpful in the past. (B. G., Glen Iris, Vic) • You have the Subcarrier Adaptor for FM Tuners (January 1988; siliconchip. com.au/Article/7830). We found this using our Contents Search page (www.siliconchip.com. au/Articles/ContentsSearch). You need to leave off the “SC” from the front of the board number. You can tell it’s an early project since we only used PCB codes with a dash from November 1987 to about October 1988 (ie, the first year or so). Australia’s electronics magazine Recent board numbers are eight digits long and start with a two-digit category code, followed by a disambiguating number (usually 1), then the two-digit month and year codes of intended publication (which could be slightly different from the actual publication date), finished with a single-digit board number within the project, starting with 1. So, for example, the April 2021 Digital FX Pedal PCBs are coded 01102211 and 01102212, with the difference being that one uses a rotary switch to select the effect while the other uses a potentiometer. The project category code is 01 (audio), the disambiguating number is 1, and the month/year code is 0221 (February 2021; delayed due to lack of space). PCB wanted for old ETI project For nearly 50 years, I have been intending to build the ETI 309 Battery Charger. Can you supply a PCB for this project? What are the alternatives to the transistors and diodes listed? I have a suitable transformer and an SCR (although it’s a C220D), and a chassis that I could use. If no to everything, do you have a simple battery charger like this one with similar attributes, for which components might be easier to obtain? (I. S., Glenhaven, NSW) • Sorry, we don’t stock any EA or ETI boards. It would be very difficult to get them made. We don’t have any of the artwork, so all we could do is scan the magazine pages, and the result would not be good enough to manufacture without a lot of extra work (many hours’ worth). In fact, if we did need to get one of these boards made, it would probably be easier to redesign it from scratch on a computer. Generally, we will have published a design much more recently, making the EA or ETI project obsolete. Your best option is to build the newer project for which a PCB is available. In this case, we suggest that you build our Clever Battery Charger Controller (December 2019; siliconchip. com.au/Article/12159). You only need to add a basic charger, which is really just the transformer and full-wave rectifier. Many automotive shops sell these; eg, see www.arlec.com.au/ wp-content/files/BC228.pdf continued on page 112 siliconchip.com.au If you decide to go ahead with the ETI 309, the BC177 and BC178 transistors could be replaced with BC327 types. The 2N3642 could be replaced with a BC337, and the C20D SCR could be replaced with the C122E (Jaycar ZX7012), although new-old-stock items are still available (siliconchip. com.au/link/ab7x). The bridge rectifier can be the MB354 30A 400V type. For D3 and D4, use 1N4004s. The enclosure needed is similar to the Jaycar HB5046 but is only 55mm tall instead of 100mm. The transformer might need to be specially wound. Is my Barking Dog Blaster working? Some years ago, you helped me build the Ultrasonic Cleaner, and it still works perfectly. In January of this year, I bought the September 2012 issue of your magazine for the Barking Dog Blaster project (siliconchip.com. au/Series/28). I built it, but I have a problem. It works, but it seems like it is too weak. You said it would consume about 1.4A with four piezo tweeters, but I measured a maximum of 250mA. I also have a question about the voltage on the drain terminal of the Mosfet. During operation, I measured only about 2V. On the secondary side of the transformer, I measured around 0.5V, but according to the article, it should be around 40V peak. Do you have any idea what the problem is? How can I increase the power? (Simon, Slovenia) • The current drain at 350mA per tweeter adds up to 1.4A; however, be aware that this is peak current and not the current that would be measured us- ing a multimeter. A meter would average out the peak current, so a measurement of 250mA is correct. The exact multimeter reading is dependent on the particular multimeter model and its frequency response. Similarly, a drain voltage of 40V peak will not measure as 40V using a multimeter; it would average it out to a much smaller value. The frequency response of the multimeter would be severely limited at ultrasonic frequencies. An oscilloscope is needed to measure the voltage accurately. We do show how to test the output by using the audible frequency test. This is described under the Testing cross-heading section. Note that the volume level is reduced for this test, as it would otherwise be very loud. So your unit is probably working correctly; the voltages and current you are measuring are likely due to the way the meter measures them, taking into account the difference between peak and averaged values as measured with a multimeter. Where to get a Barking Dog Blaster kit? Do you know where I can buy the Barking Dog Blaster kit? I know Altronics do not have it any more. (J. H., Queenstown, New Zealand) • You are right that the Altronics Cat K4500 kit has been discontinued. That was the only kit for this project. As usual, you can still get the programmed PIC microcontroller and the printed circuit board (PCB) from our online shop; see siliconchip.com.au/ Shop/?article=529 The remaining parts are available from Altronics or Jaycar. SC Advertising Index Altronics...............................75-82 Ampec Technologies................. 11 Dave Thompson...................... 111 Digi-Key Electronics.................... 3 element14................................... 7 Emona Instruments................. IBC GME Australia............................. 6 Hare & Forbes........................... 23 Jaycar............................ IFC,53-60 Keith Rippon Kit Assembly...... 111 LD Electronics......................... 111 LEDsales................................. 111 Microchip Technology.................. 5 Mouser Electronics...................... 9 Ocean Controls........................... 8 PMD Way................................ 111 Silicon Chip Binders................. 19 Silicon Chip RTV&H DVD........ 33 Silicon Chip Shop.................... 37 Silicon Chip Wallchart.............. 63 Switchmode Power Supplies..... 85 The Loudspeaker Kit.com......... 10 Tronixlabs................................ 111 Vintage Radio Repairs............ 111 Wagner Electronics................... 99 Weller Soldering Iron............. OBC Notes & Errata ESR Meter with LCD readout, Circuit Notebook, May 2016: there are some errors in this circuit diagram. Two capacitors were left out: one 470μF electrolytic and one 100nF ceramic or MKT type. They should connect between pins 3 & 5 of IC3, with the electrolytic capacitor’s negative terminal to pin 5 (the -5V rail) and positive terminal to pin 3 (GND). Also, the two 10MW resistors’ connections to pins 2 & 3 of IC4a are swapped. The resistor from the output should go to pin 2 (the inverting input), while the resistor from the wiper of VR2 goes to pin 3 (the non-inverting input). Barking Dog Blaster, September 2012: The initial timer duration differs to that set by the trimpot due to a software bug. Subsequent timer runs after the first are correct. The revised firmware, supplied by reader AJB, is named 2510812B and is available for download from our website. The June 2021 issue is due on sale in newsagents by Thursday, May 27th. Expect postal delivery of subscription copies in Australia between May 25th and June 11th. 112 Silicon Chip Australia’s electronics magazine siliconchip.com.au