hi
i build schmitzbits ms-20 filter and it’s working, but the output signal is extremely loud. from a 10vpp input i got a 27vpp output
im using exactly the same values as in the schematic, and a lm13700 instead of the 3080.
any clue what could be? i read in another thread that some people attenuate the input signal, but i still get this extremely loud output
thanks!
some updates: tried not only attenuating but also using another synth with a 5vpp output and the same, i got a almost 28vpp output.
with a really low input, like milivolts, i can get a 10vpp output from the filter. the resonance doesn’t seem to affect the signal, but with nothing connected some times there appears to be something like a self oscillation.
also de cutoff acts more like a lpg, where between 0 and around 1/3 of the range of the knob i have no output, then i have some filtering in a small range, and for the rest of the range there’s just the original signal.
i double checked every component and connections, everything seems fine. only thing i don’t have at hand right now to try is another lm13700, could a faulty one be causing this issue?
I would definitely want to switch out that 13700. Are you sure it’s a genuine 13700? I know I’ve used some fake 13700s to try on this exact circuit without success. Genuine National Semiconductor is definitely desired for a good result. Probably could get away with cheap opamps, though. I’d also try a different pot for the cutoff. Sounds as though it may be faulty.
Could be the cutoff pot is too large (like 10x the correct value) or the 100k resistor connected to it is too small (like 1/10 the correct value) — that will make the voltage near +15 V (or +12 V, whatever) over a big range, then a quick change to near -15 V, staying there over the rest of the range.
In Schmitz’s design the OTA input voltage divider is 10k/220 giving voltage at the pin of 220/10220 = 2% of the signal amplitude, so a 10 Vpp signal becomes 200 mVpp at the chip. That’s way too big; even 100 mVpp gives you about 8% distortion, about 50 mVpp would be better. (It’s possible Schmitz was designing for a smaller input signal amplitude.) Which is why it’s a good idea to attenuate your input — or just replace those 10k resistors (on the inputs to both OTAs) with more like a 39k.
thanks for your reply. I tried with another por for the cutoff (a 50k and a 100k) and the same issue with both, only a short usable range.
regarding the 13700, I’ve had this one for a long time and I’m not sure it’s a genuine, might be a fake one. I’ll try to get another one and see how it works.
There are sources on this site for good lm13700. I have successfully made this circuit with chips from Jameco. I’ve heard Tayda Electronics is also a good source. Also, I would suggest attenuating the input signal with the successful circuit when you figure the real problem out. I personally think it’s simplest implementation in the circuit gives you so much more manipulation of the sound you might desire. I actually have two of these and with one I built the attenuator into the panel and the other one I built with no attenuation at the input. I almost always want the attenuator so I have to patch it in.
thanks! i’ll measure what i got from the cutoff pot and see if changing the pot/resistor value i get a more usable range.
i tried attenuating the input with no results, so i’ll try the 39k resistors and see if that improves the response of the filter.
apart from that, you think this kind of behavior could be due to a bad lm13700?
measured the pot and i get an even output from -15 to +15.
so considering all this, my final guess is that i had a bad lm13700 (i got it locally a few years ago, it wouldn’t surprise me if it’s a fake one)
another thing i didn’t mention is with no input and the resonance all the way up i don’t get exactly self oscillation but some weird output that’s also almost 28vpp
when i can get another one i’ll update the results.
thanks all for replying
another thing that may be worth mentioning. i’ve seen on many build on youtube how the leds work whe you open the cutoff depending on the level of the resonance. well, many don’t behave the same. with the resonance pretty high both light up and flicker when the cutoff is around the middle position. if i move it pass or below that point, they stop emitting light, dim, or one of the two turns off.
maybe this can help troubleshoot it a bit more. maybe something wrong in the opamp?
also reading eddy bergman’s blog, he talks about not using ceramic caps, saying polystyrene and silver mica sound better. i only had ceramics at hand. again that may affect the audio but doubt they’re giving me this problems
It almost looks like the output opamp A2 is giving you extra gain. Are you sure that connection from the inverting input to the output is good. If that’s faulty the opamp will give you a ton of gain.
thanks for your reply. looking at the schematic and the board i made, looks like i made a mistake with the 10k going to the inverting input of O2 and ground through a 220R. the connection from the inverting input to the output and the capacitor before the filter out looks good though. i’ll fix in the module and update
