Escobedo's MS-20 filter - cutoff frequency?

[Taylor]

I built this today. Seems to work as it should, but the highest possible corner frequency seems fairly low, like around 1khz or even lower maybe. I'm not using this for guitar, but keyboards, so I really do need more bandwidth than that. Is there a way I can raise the frequency a bit, or is this just a limitation of this design?

I have about 9.6v at fully clockwise coming out of the frequency pot, so that seems to check out.

[mrslunk]

i'd try messing with the 100k resistors attached to the pins 1+16.
from what i gather, pins 1+16 are your 'Iabc' pins that control the gain of the OTA and, in this case, the cutoff freq of the fitler.
Correct me if i'm wrong, but cutting them down to say 10k(and thus increasing your current in by 10x) should up your cutoff by the same amount.

Don't feed more than 2ma into pins 1 or 16 of your ota though, You'll fry it. On 9v, that means keep above say 6.8k

[Taylor]

Hey, that works. Thanks a ton mrslunk! I find OTAs fairly mysterious, need to curl up with my Horrowitz+Hill tonight and break through the mystery...

[mrslunk]

No probs.

My limited understanding is that ota's are 'current in' opamps with a 'current based' gain control (Iabc)

Keep us all posted on where you end up with this circuit. Especially if you add diode clippers in the feedback loop, like the note says.

[Taylor]

I did build one a year or so ago, for a bass application so I didn't notice the maximum frequency thing. I installed clipper diodes on a switch in the feedback loop, but I don't think I noticed much of a difference. I don't notice any bad distortion, honestly.

Seems fine as far as 2-pole filters go, but I'm not exactly a filter fetishist. It sounds a little bit resonant even at minimum resonance, though. Kind of weird. I'm just using this one to deal with the high-end hash from my 8-bit sampler, so I won't be using it for resonant sweeps.

[nordine]

this filter works great for bass, specially if you're looking after "dub" bass sounds

what i havent been able to work out is the horrid distortion, diodes didnt work (neither back to back in series, nor shunt to ground)... an active feedback stage didnt work

anyone had luck dealing with the distortion?

[frequencycentral]

Increase the value of the 680pF caps?

[Taylor]

Hmm, I'm really not having trouble with distortion unless I get right on the edge of oscillation with the resonance control. These days I'm using passive basses, I guess that could be significant.

[Taylor]

Hmm, I'm really not having trouble with distortion unless I get right on the edge of oscillation with the resonance control. These days I'm using passive basses, I guess that could be significant.

Scratch that, this thing does distort in a bad way when you hit it with a fuzz in front (a common setup). Maybe just an input attenuator would fix it.

[nordine]

Hmm, I'm really not having trouble with distortion unless I get right on the edge of oscillation with the resonance control. These days I'm using passive basses, I guess that could be significant.

Scratch that, this thing does distort in a bad way when you hit it with a fuzz in front (a common setup). Maybe just an input attenuator would fix it.

i recall seeing a ca3080 based overdrive somewhere, it was a boutique pedals site

there has to be a way to make those lm13600 distort nicely... or at least with dignity 

[Taylor]

When you played around with diodes, was it in the feedback loop only? Maybe a soft clip stage before the OTA instead of in the loop would be a good idea, so if the input approaches the point where the OTA is going to distort, it gets clipped by diodes before that can happen.