Hello! Katie and I bought a SNES not too long ago and we’ve really been enjoying it! Unfortunately, we’ve been spoiled by wireless controllers for nigh on 9 years now, so even though the SNES controller is one of the most pleasing controllers ever designed…
Tom Servo and Crow T. Cat had a tendency to fly across the room at ludicrous speed, catching the controller cords with their tails. This kills the SNES. That gave us a great idea: let’s take a standard SNES controller and modify it to be wireless! They sell wireless SNES controllers (you can find them on Amazon) but it’s all third party stuff that looks ridiculous, and both of us love the actual controller. We figured out what we’d need to do it, ordered most of the parts off of SparkFun, and then realized we had no way of actually powering the damn thing besides running through a thousand batteries while we figure out what we’re actually doing.
So!
That brings us to our first project! A benchtop power supply! The ones we were looking at were pretty expensive, and there’s not actually a lot that goes into making a basic power supply. Of course, as we bought the last two parts from RadioShack we found a $20 power supply that they don’t list online for some reason, but it was ugly as sin and we didn’t want your stupid power supply anyway.
We found a pretty good tutorial on Instructables that served as a good starting point and made some minor modifications. We got almost all of the parts at RadioShack, along with some assorted resistors and a breadboard from SparkFun (and some acrylic for the case, from Menards):
Let’s take inventory, from left to right and top to bottom (almost all of these are listed in the Instructable as well):
- 10kΩ potentiometer, which we’ll use to control the output voltage
- 0.1µF capacitor, to use as a smoothing capacitor
- 2x 1N4001 diodes for circuit protection
- 1kΩ resistor, used in combination with the potentiometer to get the appropriate output
- SPDT switch to act as a power switch; we picked up a nicer power switch for the final product (not pictured) but this one is easier to use for testing
- 10µF capacitor, which will improve the response speed
- LM317TG voltage regulator, which will let us safely adjust the output voltage
- Connector for a 13.5VDC power adapter, for the input voltage
- Breadboard, because if we build the permanent circuit right away something would go wrong and catch fire
Not pictured: a heat sink to attach to the voltage regulator which will prevent overheating, a small fan to do the same, and assorted bits of wire.
Our design uses a 10kΩ potentiometer rather than a 5k because we thought it looked cooler, and that’s the best reason to make design decisions. The great part about circuits is we can totally adjust to that! We want to get up to 12V output (same as the original design) and our potentiometer has twice the max resistance, so doubling the static resistor gives a good approximation of where we want to be. The closest resistor we have to that is a standard 1kΩ, which should work fine. If we really wanted to get it precise, we could also wire up another 100Ω resistor in series, but we don’t really want to.
We followed the same circuit diagram as the original design, and our end result is a functional (and profoundly ugly) power supply!
We have a working power supply, but we want something a little more… permanent than this. The next step would be to solder everything to a perf board (and attach the heat sink) but every RadioShack within a twenty mile radius was completely sold out of them and we forgot to put them in our SparkFun order. For now, let’s move on to some other fun stuff we can add to the power supply. A variable power supply isn’t very useful if you don’t actually know what you’re getting out of it, so we picked up a cheapo voltmeter that we could tear apart and transplant. 
This thing runs on a 12V, 33mAh battery which is honestly astonishing to me. I hate the idea of having a mobile, convenient voltmeter so the first thing I did was attach leads to the battery terminals so I could power it from the input of the power supply.
I love the way the dial works. You select the output you want and it connects these traces via 16 unique patterns. The way it snaps into place is also pretty neat: 
The metal strips on the back of the dial aren’t just for conduction, they also act almost like springs to press the dial to the face of the voltmeter. The two metal ball bearings rest statically on the dial and fit into grooves on the faceplate so that the dial will “snap” into place. To no one’s surprise, I immediately dropped one of them onto a tall carpet.
All in all, a fun first day of mucking around. We’ll want to start building the permanent circuit and cutting down the voltmeter so we can just use it as a display, but we managed to output a variable voltage and lose several parts, so all in all pretty productive.




