It’s been a week of tweaking and assembly, mostly. This is both the part I love the most, and dread the most. On the one hand, it’s awesome to see the project come together and not just be a dangerous collection of bits on my bench. On the other hand – drilling into aluminium is such a chore!
Anyway, here it is thus far:
Waiting as always
I am so very close to declaring this construction complete except for two missing pieces. First, and most importantly, is the SIL-pad and associated hardware for mounting the IRF3205 to its heatsink. For those unaware, watch when mounting TO-220 (or other large package) transistors/MOSFETS. Though it has the three pins Gate, Drain and Source nicely separated from each other, the tab and back is also hooked up to the drain! This could mean that your heatsink, which is inevitably mounted to the case, will be the drain also, and touching it could be a very bad idea!
The solution is to use a mounting kit, which consists of a SIL pad, which is thermally conductive but electrically isolating, a nylon bushing to fit in the mounting hole of the TO-220 package, this isolates the M3 bolt and other hardware from the heatsink. Anyway, I have a kit of these mounting kits on order from Amazon.
The other missing part, which is less critical, is the markings on the front panel. Lettering and labelling controls etc. The operation is very straightforward, but it is always nice to be complete. This is a simple matter of me going to get some laser printer adhesive transparency sheets and printing it out. I already have a design done.
A Look Inside
Here’s a look inside the tin:
Looking pretty neat I must say and actually it went together quite well. The worst part, as I said earlier, was drilling the holes. I’ve always had issues with this, and used to be far worse at it! I also got to use my panel-nibbler tool for the first time and I both love it and hate it lol. It can be a very easy to use and precise tool but its tricky to get used to. The square holes turned out okay, if a little uneven. The round hole for the fan at the back looks a bit meh. I sort of gave up on having it be perfectly circular.
The best advice I could give to those new to working with aluminium project boxes is this: be patient, use the right tools, and be careful.
Always measure and lay out your intended holes and cutouts and ensure that the parts to be mounted don’t collide with each other or stick out to touch the pcbs and hardware inside.
Mark each hole with a punch at the centre, this prevents the drill bit from skating off centre when you start drilling. Start with a small bit, I used a 1/16″ bit to drill a pilot hole to start. Use a drill press where possible, this keeps your holes nice and even and gives you good control of the drilling. Go slowly and let the bit do its work, don’t use downward force. Use a slow speed and lubricate the bit as necessary. For holes larger than 1/8″ or 3mm, use a step bit. They are far superior at cutting even holes than your larger basic drill bits and will not grab the piece you are drilling and fling it at you (this does happen with standard drill bits above a certain size).
Needless to say but always: measure twice, drill once AND wear safety equipment (safety glasses a must). Always keep your eyes and attention on your work and your paws away from the sharp pokey bits.
Wiring wiring wiring
It was amazing to me how much space all the circuitry and wiring actually consumed inside this relatively large case. The PCB for the main brains of the operation is pretty small, as you can see, but routing all the wires and the power supply board took up quite a bit. As did the large heatsink and mounting it such that it is directly cooled by the fan. Keeping the mains AC wiring properly isolated and tied off also was a bit tricky. Definitely all heatshrunk and properly crimped here. I realize in looking at the picture I forgot to provide a chassis earth, but that’s easy to do, I’ll just add a lug and screw to the back.
I was mostly able to route and tie down all the wiring without a problem, but as usual, I made some of them too long. This will add some unwanted interference possibly, note the bundle of wires heading to the switch is hanging out over the switching transformers… not ideal. I think in future projects, I’ll handle such switching on board and use a hardware switch in a more noise-immune way. Still it seems to function fine so I can’t really complain.
So apart from mounting the MOSFET properly to the heatsink, and adding a label to the front panel, she’s done! So happy to see a project finally come together. Also happy to not have to drill into metal for a while, tedious and messy that.