Auxillary Variable Bench Power Converter

After recently completing a simple regulated 0 to 30 volts DC variable power supply for my bench,  See: "A New Power Supply for John"  I realized that the job was only half finished. What about the times when I would need a second voltage for an experiment or if my project would need a split plus and minus voltage supply. Obviously I needed another source of variable power to supplement the new power supply.

I began by taking stock of my resources and planning the new addition around a couple components and concepts. First of all I wanted this unit to be small and portable so that I could move it around. Since I already had 2 non-regulated variable supplies besides the new regulated supply,  I decided to make this unit an auxiliary that would plug into one of the non-regulated supplies. Therefore what I would be building would be a variable regulator and in keeping with the simplicity of the other supply it was decided that I would use the familiar LM317 regulator and a support circuit taken from the data sheet. One component that I wanted to use for this project was an $11 Digital Volt Ammeter that I had purchased from China.

My previous experience with this unit had taught me, with a plume of smoke, that while it would display up to 99.9 volts and 99.9 amperes it would not tolerate more than about 28 volts for its power supply. I decided to include in my circuit design a 7812 regulator to provide power for the V/A meter and that I would also bring this regulated voltage out to the front panel since it was already available. After putting together a couple prototypes on the bench and running tests to see that things would work as I wanted them to I had a good  handle on the final circuit. Here is a schematic of the final version of the circuit.

As you can see there is nothing complicated about the design and it will provide adequate regulation and current for the type of projects that I typically work on. The wiring from the Chinese V/A meter is color coded but you can not trust it. The unit that I received had the large red and yellow wires exchanged but over all it was not hard to figure out. One of the things I like the most about buying things like the Chinese V/A meter is the challenge that comes with circuits that come with little to no documentation or explanation. Furthermore my wife calls this source of gadgets the Chinese Dollar Store, which means that I am only risking minimal dollars if there is a failure.

I began the actual build by mounting the LM317 and the 7812 to a piece of salvaged heat sink. I then used a small universal circuit board to mount the regulators and related circuit components. When I had it built for preliminary testing it looked like this.

The additional portion of the circuit board that was not used was eventually cut off to make for a better fit in the enclosure. At this stage I again set up proto connections to verify that I had hooked things up as planned. I have learned that if I do not do this as I go along I just have to disassemble my project later and fix what I missed.

My next big challenge was to test and calibrate the Volt Ammeter. It was designed to be used with a big honking shunt that would allow it to work in the 100 amp range. My planned circuit would top out at 28 volts and a couple amps. The volt reading would take care of itself as my tests had revealed that it was in agreement with my Fluke multimeter to 0.10 volt. This was within tolerances for my needs. To bring the ammeter into line I had to build a special shunt. I wanted to get as much accuracy as possible in my working parameters of 0 to 3 amps. I took a piece of brass wire and with a known current flowing slid the meter contact down the brass wire until the display of the digital meter matched the known current. The only problem at this point is that the ammeter read 10.0 amps when the actual current was 1.00. At first I planned to do some work on the digital meter's circuit to try to move this decimal point but after seeing that the display, including the decimal point, was being strobed by the driver circuit I decided on a more manual approach. I used black paint to darken the lit LED decimal point segment and painted a white dot on the display where I wanted the decimal point.

I planned to put the circuitry into a plastic case 15cm X 7cm X 5cm. I always make a template out of cardboard or other soft material so that I can have the best chance of properly positioning my peripherals before I start drilling and filing. My tools include a drehmel, files, hand drills, and other simple tools. This is the hardest part for me as I am a bit of a specialist in having drills walk away from my intended drilling site and I also specialize in slipping and scratching what should not be scratched. This time I was fairly lucky and was able to cover up my mistakes with a little file work.

I now began to mount the jacks, meter, potentiometer, fuse holder, and drilled the circuit mounting hole in the back and input wire access hole. If you ever work with one of these Chinese meters it is important to remove the circuit from the bevel before inserting it through the hole as the retention clips will not retract with the circuit installed. I also filed the bevel of the meter so that it was a little bigger as it obscures the lower segments of the display from the supplier. Here is a picture of the internals after the wiring was complete.

Now all that was left was to label the unit and run it through some tests. I connected it up to my old unregulated EICO power supply and loaded it with a 12 volt automotive tail light. I have found light bulbs to make good safe loads with which to test circuits. I do not have the worry about burns or fire that I would have using resistors. I ran the unit for an hour with a load of 0.50 amps and 12.5 volts being driven by the EICO at 20 volts. It got warm and stabilized. As this is was more current and a longer time than I would typically need I was satisfied with the results. Here is a picture of the finished unit.

John