Hurray! I got my PCBs via DHL this morning. So happy to receive them and hold in my hand my first professionally made PCB.
I am extremely pleased. I mean, not only because it’s way better than anything I could hope to manufacture on my own, but the boards turned out very well I must say. The pads are tinned all nice, the holes all drilled and plated through, the solder mask matches up well. The only thing wrong-ish I could find with them is that the silkscreen isn’t perfectly sharp but it is quite legible and certainly a pass by my standards. I sat there and tested continuity of all the traces and it’s a 100% pass. It’s going to work… unless I somehow mucked up the schematic, but I highly doubt that.
Of course, the only real problems with the board are the ones I created myself in the design, and even then, thankfully, no show-stoppers.
The biggest win with this board is it is an absolute dream to solder! So much better than those junky perf and protoboards. It was even enjoyable! Quick heat, slip in some solder, watch it wick through the plated holes and secure my device – soldering heaven. So EASY. For anyone who is considering getting their boards manufactured instead of protoboard – DO IT. The difference is night and day.
It kind of reminds me when I was learning guitar. I had this crappy acoustic that was difficult to play and wouldn’t stay in tune half the time. Then when I got a good guitar I was instantly a hundred times better player. It was like that with using a proper PCB. I can make picture-perfect solder joints and have a board assembled in a few minutes.
What I did wrong
Fortunately, nothing serious. I of course went straight to my bench after work and started soldering it up and discovered first off that the 330nF caps for the voltage regulators were just a bit beefier than I had bargained for. Nothing that couldn’t be solved by substituting a smaller 330nF cap, but I didn’t have any. It actually was two problems. One the pin spacing was too small, but that’s no big deal, I could just bend the legs (and I did), the second was the hole size for the leads was too small. Darn it! That one was fortunately solved by a few minutes with my dremel widening the holes. Turned out perfect. I probably don’t have to say this but if you don’t have one buy a Dremel (or other rotary tool) and get yourself one of those sets of bits/grinders/cutting wheels etc. It will save your butt and make building so much easier.
Another issue I ran into was not leaving enough space for my pcb connectors. In EasyEDA, I had used a generic header PCB footprint. This is normally fine except it didn’t have the footprint of the PCB connectors I intended to use. Fortunately, I had left enough space, mostly. Except the DC in connector, which is squashed between an electrolytic cap and where the screw head will be for when I secure the board in. Not the biggest of deals. Though I prefer using proper connectors and not having wires flying off the board, I could get away with it in this case. Or find a connector that will fit in that space which should not be hard to do at all.
I’m also not liking the footprint of the M02 molex connector for the dummy load’s input. It looks kinda small. This is also not a problem. When I get an MO2 molex I’ll check it out, if I don’t like it I can just solder beefy wires to it with proper crumped lugs to screw onto the binding posts. Would be safer and I wouldn’t be introducing any more resistance to that path than I had to.
Lastly, A funny one. Going along with me not planning enough space for the connectors, I found that they coverd up the silkscreen labels I had made to identify them. No big deal, I can just look at another board, a picture of the board, my PCB layout file etc. Just funny
Really, not much. I didn’t mount the MOSFET as yet, as it is attached to a giant heatsink and has wires soldered onto the leads from testing. Just some fiddling about will set that right. Won’t mount the heatsink until its ready to go into its case as it’s heavier than the entire board and all the components on it! It will stand tall once I’ve properly mounted it. Other than that, I think a trip to Spadina and College is in order to get a few stray parts. A 12-pin pcb connector, a smaller 2-pin PCB connector, a M02 Molex connector etc. After that I’ll be able to assemble the board completely and test it. Having tested it a bunch of times I am very confident it will work fine.
I’ll also have to make some changes to two other projects I’m doing the PCB layout for, on account of lessons learned here.
So I poked around a bit, turns out the M02 connector I specified for the load connector is actually of the family I use for all my other connectors and are made by Molex. Now that I know that, I can do future boards with the proper footprints. That connector series is rated for 4A and 250V so I could use it for the input connector since the dummy load has a fixed maximum of 3.34A, and I’d not pull more than 3A practically. That’s 83% of rated, usually I like a 20% margin, but I think I can get away with it. I was planning on soldering the wires directly to the board, until I realized that the hole diameter would probably be too small anyway and I’d have to ream them out like I did for the 330nF caps. Speaking of which, I had noted a trap for young players! When I do ream out a hole like that on a PCB, I get rid of the through hole plating, obviously. So I have to make sure it soldered on both sides of the board. I had to do a bit of fiddly solder work last night to make sure I got all those pads on both sides.