So, I think I found the issue I am having with my module tester. After reflowing the MCU and the two shift registers, I tested the tact switches. They appear to have an internal resistance of about 23k between either pin and ground when open (like a built-in pull-down resistor), but the switch for the light that’s staying lit is only 13k when open. They all go to 0R when closed. So, either one of the switches is bad and that’s keeping that switch locked, or they’re ALL bad, but that one is worse than the others for some reason. At any rate, it sounds like I’m going to have to source a replacement switch and replace the SOB.
ETA: I looked at the datasheet for these TL1100 tact switches and there’s nothing about an internal residence, so I may have just gotten them too hot when soldering them in. Ooops.
Well, I finally got it sorted out. In the circuit one of the buttons was showing a different resistance than the others, about 10k less. I traced things back checking the voltage between the shift register and the buttons and found that when I metered between one switch and its pin on the register, and then pushed a different button, the 5v signal got cut in half. I tested the resistor array and got a weird result–10k on one pin and 20k on all the others. So, it turns out I had the array in backwards. I’ve never used resistor arrays before and just kind of assumed they were a-polar like regular resistors, but they have a common pin, so I de-soldered the array, soldered it in the other way around, and now everything appears to be working correctly! Testing of the signals in and out will commence, but I’m fairly confident I won’t run into any more issues. YAY NOODLE TOASTER
I just plugged m tester in and it seems to be working a-ok! I am currently printing the enclosure for it and wont really be able to finish it up until it gets done.
I straight up lucked out on the resistor array - I soldered it in and THEN found it that it would only work one way. Guess I had a 50/50 shot. There is a little tiny dot on it, but I would never have guessed that it had anything to do with polarity.
It’s not really even polarity so much as a common ground pin, but yeah–it can absolutely can go wrong way round.
The acrylic case I got is either a little wonky, or the components are bigger than what it’s designed for. It stands away from the LCD and the top of the MIDI port a couple millimeters on either side. Which is fine since I’m planning on putting it behind a panel in a Eurorack kind of situation.
I have noticed some odd behavior with my tester - When plugging my k25 LFO into the audio in, the tester tells me the frequency of the signal. When I plug in any of my VCO’s (either sam’s or my stripboard modules) they do not register at all.
Hmm, the K25 produces a waveform centered on 0 V, while the 1222 uses the waveforms straight from the 3340 chip which are 0 to x. I wonder if the module tester frequency code can’t handle positive-only waveforms? If it works by counting zero crossings…
I’ll see if I can find a way to test this theory, but my rack is nearly empty of modules. I’ll see if I can feed my tester a monopolar and bipolar signal and see what it gives me.
Hope it’s cool to ask; wondering if anyone EU based has an extra PCB kicking around that they might be willing to part with. Looked into getting a batch made but currently more than I can afford. Would cover PCB cost plus shipping of course and/or have some surplus Megacommand pcb’s for example.
Cheers! Even though it’s clearly stated on your site I must’ve missed that last time I checked, or I was still convinced I’d be able to get 5 boards manufactured for 25,- inc. shipping :-
I’ll happily put in my order for pcb, chip and screen after this month’s payday.
This dc-dc converter has been tested by a few people (thanks guys), it’s about 60% cheaper than the original one specced in the BOM, it has about 25% more ripple, it’s 100mv/v rather than 75mv (per volt) Click here for a more ripply but cheaper DC-DC converter
But going there and looking at the datasheet it shows
I’m not sure, it’s possibly a new revision I guess they would change the part number slightly, add a new letter, or put a revised date on the datasheet if they’ve changed anything, or at least shout “now with 50% less ripple!”
