For reasons, I took a break from synth building for a few months. Some design work but no building. With (meteoroligical) fall here, it’s time to get back into it. So what’s up?
One of my top priorities: Modifying YASH. This sample and hold works okay, but the droop rate is pretty high. From the LF398 datasheet, it should improve more or less proportionally to the size of the integrating capacitor, so bumping that up from 1 nF to 10 nF should make it a lot better. But that also means it will take ten times as long to charge. So the cap in the trigger processor also needs to go up a factor of ten. The charging gate as it stands is only 3 µs; making that 30 µs should be fine, still too short to be audible.
Kosmodrome Phase III isn’t particularly well-defined, but it’s mostly drum/percussion modules. I have two NLC Bong0 circuit boards — two because they’re cheap, I can use two of them, and I can mess up one and have the other as fallback. Not that I expect to mess one up, but it’s a surface mount design and my SMD skills are rudimentary. It’s time to work on them. In fact I can try different techniques with the two boards and see what I like better.
I want to try a Kosmo adaptation of Thomas Henry’s Bass++. Someone else on the LMNC Discourse was working on doing the same a long time ago — they were posting regular updates on their progress, and then the updates just stopped. I heard they ran into some problems, put it on the back burner, and never got back to it. I’m surprised no one else has stepped up. So I guess I will. I’ve drawn up the schematic in KiCad, but before going further I need to breadboard the module to test some modifications.
I also have a design I did for a major upgrade of the Barton Analog Drum. It adds a white noise generator feeding into a VCF and VCA. Breadboarding this is the next step here as well.
The ring modulator I built is okay, but I’d like to try out a 4-quadrant amplifier based on the AD633N. What I have in mind is similar to the Befaco A*B+C, but single channel because the chip’s so freaking expensive and I’m not sure how much I’d use both channels, and with the input section reworked because their attenuverter design strikes me as just strange.
A quickie project, or it should be, is another power display module — I want to do one that has a power header on the front panel, for plugging in modules under test or troubleshooting. That’ll be another mostly SMD project. So will a distortion module based mostly on the Music Thing Mini Drive, but with some stuff adapted from their London Drive.
A while ago I bought a Kassutronics Slope PCB set, thinking I’d put it behind a Kosmo front panel, but on further thought that would be fairly hard. Meanwhile Ben Rufenacht on the LMNC Discourse did a PCB design for a Kosmo version of the Slope, so just building Ben’s version — with a few small changes — seems the better way to go.
I have a design for a little utility module, a very simple thing for converting control voltages from one range to another.
And finally, I want to try making vactrols again. I haven’t had much use for vactrols to date, and the couple times I have used them I’ve gone with commercial ones, but I want to do a Kosmo version of the Music Thing Vactrol Mix Expander. That requires eight vactrols — and I’m not about to pay $13 each for the specified VTL5C10!
That should keep me busy a few months, right?
First step: Parts. I’ve already ordered some LDRs and flat top LEDs from AliExpress for the vactrols. For the rest, I used my spreadsheet to come up with a pair of shopping lists, one for Tayda and one for Digi-Key. Then I decided to check on how much my spreadsheet was lying to me about what I have in my stash. Answer: It wasn’t lying… much. But there were a few things in shorter quantities than it claimed. And I discovered I’d put a linear pot in my cart thinking it was audio. And I discovered I’d assumed an entire SMD module was THT, and no, I did not have many of its parts in 0805, so I had to add them. I finally convinced… no, let’s be accurate, persuaded myself I wasn’t overlooking anything, and sent the orders. And a day later I still haven’t thought of anything I missed, so maybe.
Now I’m doing likewise with a JLCPCB order. Uploaded Gerbers for two modules last night but I’m sitting on them until I convi persuade myself I haven’t forgotten a JLCJLCJLCJLC or uploaded the wrong Gerber version or anything.
Then it’s back to the breadboard.
<– 2023-09-02_hub-to-lab — 2023-10-19_ac-coupled-square-waves-are-less-than-obvious –>