I just completed the core oscillator after lots of hair pulling culminating in the revelation that I needed a bipolar power supply. After building a jenky 12v+/- bipolar power supply out of two 12v wall warts wired together, I got it to buzz. I’m pretty satisfied with the results, the only issue I’m having now is getting it in tune. I’m using a beatstep pro for CV, it plays notes fairly well but the ratio between notes is just a bit off. when I get a given note dialed in, it ends up being a half step off by the time I get to the next octave. I’m not sure if its an issue of troubleshooting or if I should just add the tuner mod to it.
Potential issues that I’ve thought of include my wallwarts being slightly above 12 v. one wall wart reads 12.19v, the other is 12.6. Seems like a pretty slim margin but I’ve been looking into a way to attenuate them to a even 12v. I also noticed that the 1.5m resistor reads 16.66k for some reason. It’s definitely a 1.5m though since after pulling it out of the strip board, it read 1.5m. I put in a fresh resistor which read 16.66k once installed. I assume that something in the circuit is affecting its value. there don’t seem to be any resistors in parallel so I have no clue what’s going on, if it’s supposed to do that, and if it’s affecting the tuning. I’ve considered that using an as3340 threw it off a bit but after browsing this forum it doesn’t sound like that would be the problem .
I believe resistors do measure funny when they are in a circuit as appose to being out of one. I vaguely remember reading that somewhere so I think it wouldn’t be an issue.
If your using Sams design maybe check for the 4.1volts I think it was and your precision resistors for the octave switch.
Thanks for the welcome, glad to have discovered this community. I attempted to calibrate it with the trim pot, unless I’m failing to understand the process, it didn’t seem to help. To try to better explain what’s happening, when I play a C and get it in tune properly with the trim pot, then play a D, the D is pretty close but a few cents off. E will also be close but slightly further off. By the time I get to the next octave, C is a Db. I found some other threads here about similar problems others had and tried a couple of fixes. I was also having issues with the pitch being too high with cv plugged in so I jumpered the corse tune pot to ground which fixed that issue but didn’t help with the tuning. I probed the output of the beatstep while hooked up and it appears to be outputting 1V/O with a possible margin of error of 1mV/O, I’m not sure how accurate my multimeter is.
I’m using sams strip board design for the simple core oscillator, not the 1222 performance oscillator with the tuner. So far the only mods I’ve attempted are the pulse out and pwm, still working on troubleshooting that mod though.
For calibrate, send 2 or 3 ovctaves with your BSP, and with the trimmpot adjust until the 3 octaves are good on your tuner, and all the notes between will be good.
Thanks, turns out I just had to fiddle with the trim pot more like you suggested. Part of my problem was not understanding what the trim pot was doing which makes sense to me now, the other problem is that the trim pot isn’t really a trim pot, it’s just a standard pcb mount 10k. I was able to tune it to be fairly close in 4 octaves but I’ll definitely need a proper trim pot for precision.
My current issue is with the square wave and pwm. I got it to output a square wave but it’s only audible if I run it through a mic preamp with the gain up to sketchy high levels. The pulse width pot doesn’t affect it despite having continuity. the pwm pot works but with limited range and the cv jack does jack all. I’m guessing that a bad solder is to blame, I’ll have to troubleshoot it some more when I get some free time.
No, you do not want to try putting attenuators on power rails. Any substantive resistance on the rail will mess up voltages when you connect loads.
DC wall warts are not the most accurate way of getting power. I’d recommend you look into better power solutions, but any case it shouldn’t affect your ability to tune the oscillator to 1 V/oct.
Peeve: Capitalization of units is important. 1.5m is one and a half milliohms; 1.5M is one and a half megaohms. (Likewise pF picofarads, not PF for petafarads. k for kilo should be lowercase.) Anyway: There are indeed no obvious resistors in parallel, but it’s connected to pin 13 which may have some smaller resistance to the +12 V rail. Generally speaking if you try to measure a resistor in circuit (with power turned off!) you won’t get a value that’s larger than the resistor but you may get a smaller value, and if there are capacitors you may not get a stable value. So it’s rarely useful to try to do that.
Gotcha, capitalization has never been my strong suit , I can see how that would be an issue with measurement units though. Good info about the resistance values being affected by capacitors. I was getting some wonky readouts on the 100K pots as well, that might explain why.
, I can see how that would be an issue with measurement units though. Good info about the resistance values being affected by capacitors. I was getting some wonky readouts on the 100K pots as well, that might explain why.