First, it’s been a while since I last posted. COVID-19 and all kinds of other personal, work, and other stuffs have been making it hard. I hope y’all are ok, well, and dealing as well as you can.
Ok, so after the O2 it all gets a bit hazy. It was around this time a number of projects happened, another Pete Millett amp, I think I built a quick CMOY for some reason, and probably something else – but I remember building a ESS ES9038Q2M dac from a board I bought from AliExpress. The particular board I purchased has lots of variants and there is an extensive discussion of that family of Chinese dac boards on diyaudio.com.
It was my first time actually building a power supply and using a transformer. This was scary. Things like grounding were no longer about eliminating noise but about safety and not being electrocuted. Obviously, this was scary and just reading about what to do on the internet seemed a bit risky.
My day job was working in tech. Much of the company was remote so we all used a chat tool called Slack to communicate. What was great about this is Slack allowed for various channels or rooms that were work related and whatever else you could want. Remote worker water-cooler chats happened in places like “The Danger Room” or “Pancakes”, the latter being a place to post pictures of delicious pancakes. One of those channels was dedicated to music and I had already started talking about amplifiers, dacs, headphones, and all that in there. I was likely being annoying to everyone who was there to discuss Taylor Swift.
But it was through that Slack channel I met Greg. Greg was in to DIY audio as well and was much more accomplished. He’d built speakers and power amplifiers – you name it, he’d done it. He had a crazy awesome home lab complete with measurement gear, a CNC, and loads of other goodies. He was not only soldering things together but milling aluminum enclosures and fabricating the entire product.
He is/was my idol.
Shortly after discovering we were doing sorta the same thing, we started a “DIY Audio” channel and, though I felt silly, started asking him questions about wiring for my dac. Windings, secondaries, center tapped – stuff I didn’t know about and he’d explain to me. I told him I needed 15V AC to the dac board, which apparently had a it’s own DC power supply. He showed me a R-Core transformer that would give me the 15V AC I needed and walked me through wiring it correctly, using a fuse, and where to ground things.
The enclosure I purchased was one from Hammond Mfg. and had some nice wood side panels. I’d take a picture, send it over slack, he’d say “the green wire should go to ground, twist those red/white and black/white ones together”, and it’d do it. Funny enough, if I’d just looked at the wiring diagram that came with the transformer I proably could have figured it out. However, since we’re dealing with AC electricity, I was scared to death. And death is possible.
He also suggested I spend some time building a dim bulb tester. What that does is lets you, relatively safely, test high current gear without a ton of risk of damage. If there is a short, the light bulb will light up and there shouldn’t be a fuck ton of current flowing through the gear. Now, a small dac is not quite super high voltage and testers like this come in to play big time with anything involving tubes, but AC mains are still involved and I didn’t want to risk fire, death, or any sort of bodily harm. I built one out of a wall electrical box and a plugin light bulb receptacle after a quick trip to Home Depot.
There were some things in the middle here – the board had mounted RCA jacks on it and I wasn’t sure enough of my ability to drill correctly aligned holes in the expensive enclosure, so I opted to de-solder them, panel mount them to the enclosure, and run wire back to the board. I did this and ruined most of the solder pads. This lead me to having to trace those pads back to their origins and attempt to solder the wires to those points. I did that, felt accomplished, and re-checked continuity with my multimeter at least 5 times.
Alright, the moment of truth. I’d drilled out the chassis I purchased with my hand drill (which probably deserves it’s own post), wired everything, secured everything, shrink wrapped any wires where something conductive could possibly touch or even move, plugged the dac in to the one end of the dim bulb tester, the tester in to a power strip in the off position, and inserted a light bulb. With fingers crossed and standing a more than a few feet away from the dac, I turned on the power strip.
A small LED lit up on the dac board. No smoke. I was still terrified.
Using a multimeter set to measure AC, I carefully touched the probes to the enclosure to make sure no current was flowing. They said none was. I flipped them to DC, touched one to the enclosure and the other to a spot on the board where I knew DC flowing. It measured the correct voltage. Holy shit.
So, at this point you have a few options. Keep testing various points to make sure things measure as they should or YOLO and plug something in. Because I live on the edge, I plugged in my laptop and beloved Butte. It worked. I heard sound, music, and no noise. Holy shit.
That dac was a lot of terrifiyingly fun to make. It was my first experience with a transformer. Thinking back on it, I could have just bought a 15VAC wall wart and went that route, not even dealing with the transformer, IEC inlet, and grounding would have been a lot easier. But it didn’t even occur to me. At the time I’m not sure I knew it was even an option.
Three things were learned in this build. I learned about transformers, windings, what a center tap is, primaries, and secondaries. I really met Greg and got to know someone who actually knew what they were doing, though I’m sure he’d say he doesn’t know. I also learned that anyone who says a dac makes a significant difference in how something sounds is full of crap. I’m sure there may be exceptions to that, but my experience with this dac, compared to everything else and the onboard dac on my computer, is they all sounded good. Or rather, was just there. Couldn’t hear a difference. I’d learn that as long as the actual analog stages (IV, gain, output buffering) of the digital to analog conversion were properly implemented, it shouldn’t matter.
So, just like Pete, thanks Greg. It’s been a lot of fun to have someone to almost get electrocuted with.