Will the real MS-20 filter please step forward

[Cliff Schecht]

Are you sure the 13700's are good? IME those are quite easy to blow up during the breadboarding process!

Maybe post your voltages again, everything will be different now that your VCC/2 is correct. You could also try converting your breadboard build to the original dual supply version and see if it works that way.

You said the thing squeals and such, sounds like the resonance circuit is out of whack. Try removing the output of the resonance circuit that feeds into the bottom of the cap after the OTA current output. Instead jump the bottom of this cap to VCC/2 (like you do in the second stage). Doing this removes the resonance circuit and lets you debug just the OTA filters.

[garcho]

I'll give it another go, thanks for the tip about the filter debugging Cliff!

[garcho]

I tried what you suggested Cliff, with grounding the capacitor, but I still got no filter effect, just weirdness. So I put the dual battery back on the breadboard to check the OTAs; they're all fine. I ran out of time, so I'll poke away at it more later this week. Thanks for the help.

[Cliff Schecht]

That point isn't grounded, it goes to VCC/2 (as does the second cap). The circuit won't work correctly if the capacitors are tied to ground instead of VCC/2.

[woolybeard]

After spending the entire day messing with various expo current source circuits, I found an NPN variant that works well with a single supply. It's functionally equivalent to what the MS-20 uses (or good enough at least). Since the NPN's only sink current in this circuit, I mirror the expo current up to a PNP current mirror and source current to the Iabc pin from the second PNP. Works in simulation and should work fine on the bench.

Here is what I have so far (verified in simulation and parts on bench, may need values tweaked!): http://www.aronnelson.com/DIYFiles/up/ms20_9vsupply.pdf

Hey Cliff, I was hoping to breadboard/prototype this, but all of the values aren't there... Is there a more updated version of this schemo that I'm just overlooking? A 9v version of this filter that I could turn into a stompbox is like a dream come true.

[WhenBoredomPeaks]

After spending the entire day messing with various expo current source circuits, I found an NPN variant that works well with a single supply. It's functionally equivalent to what the MS-20 uses (or good enough at least). Since the NPN's only sink current in this circuit, I mirror the expo current up to a PNP current mirror and source current to the Iabc pin from the second PNP. Works in simulation and should work fine on the bench.

Here is what I have so far (verified in simulation and parts on bench, may need values tweaked!): http://www.aronnelson.com/DIYFiles/up/ms20_9vsupply.pdf

Hey Cliff, I was hoping to breadboard/prototype this, but all of the values aren't there... Is there a more updated version of this schemo that I'm just overlooking? A 9v version of this filter that I could turn into a stompbox is like a dream come true.

do a quick google search for the escobedo 9v ms20 filter

btw in my opinion building the filter is not even half of the battle, after having built a few filters (escobedo ms20, "real ms20", minimoog, steiner, meatsphere etc. /btw the ms20 is the last one i would use from this list for guitar/) i think the CV/controlling element is more important. a voltage follower would be enough for guitar use but it is not as flexible as a real synth ADSR. (btw if someone knows a voltage follower with attack and decay/release controls please tell me, the one in the meatsphere is not really good) so i made a real ADSR and a circuit which makes gate signals from my guitar signals but it have it's own problems again because it needs complete silence for a few millisecs to end the gate signal and you actually have to concentrate pretty hard to put a bit of silence between your notes because that is not how you play guitar)

[Cliff Schecht]

After spending the entire day messing with various expo current source circuits, I found an NPN variant that works well with a single supply. It's functionally equivalent to what the MS-20 uses (or good enough at least). Since the NPN's only sink current in this circuit, I mirror the expo current up to a PNP current mirror and source current to the Iabc pin from the second PNP. Works in simulation and should work fine on the bench.

Here is what I have so far (verified in simulation and parts on bench, may need values tweaked!): http://www.aronnelson.com/DIYFiles/up/ms20_9vsupply.pdf

Hey Cliff, I was hoping to breadboard/prototype this, but all of the values aren't there... Is there a more updated version of this schemo that I'm just overlooking? A 9v version of this filter that I could turn into a stompbox is like a dream come true.

Ask and ye shall receive! I updated the schematic (link below) with the changes I mentioned before. Essentially I added an input buffer with some gain (which can be omitted if not needed), a buffered VREF (probably not optional) and some decoupling caps sprinkled around to keep crap out of the power supply with high dv/dt control voltages.

http://www.aronnelson.com/DIYFiles/up/ms20_9vsupply.pdf

[garcho]

Yeah Cliff, it sounds great! Tried out yer latest schematic and it sounds awesome. At first, it started making some fizzy noises and there was no filtering effect (sound familiar?). Then, I realized I was grounding the base of T4 via R15 (1k8) instead of connecting it to VCC/2. I bet that's what I was doing with my previous attempts. Stupid!!!  

I'm using TL074s and 72s and blue LEDs. A 2k resister instead of 1k8, and fudging a little on some cap values. But it seems to work fine. So far I've only used it as a LP filter, not LP into HP or dual or anything yet. I'll take some voltages down in case other folks are having problems. Thanks for the PS help and schematics Cliff. Excited to get this together!

Inspired by this project, I've started putting together something similar with 2 BP filters in parallel. I'll post some clips of that when I get it up and running a little more polished. Right now I've only used it with R.G.'s max1044 bipolar 9V schematic because that's all I know how to do, but it works well.

EDIT: Oh yeah, anyone else having fun with LFOs for this project? I'm trying a few different things for some simple on-board LFOs.

[Cliff Schecht]

Glad to hear it's alive! It is a funky little filter ain't it.

[garcho]

Sure is! I'm gonna try out Ken Stone's Psycho LFO with it, see if I can hear that pseudo randomness.

[digi2t]

Uh... don't know if anyone has mentioned it, if so disregard this.

Has anyone seen the Korg Monotron? Is this thing really running the MS-20 filter? If so, for 50 bux, is it worth building a filter, or can one just gut a Monotron and go from there?

EDIT* - Just found the schematic; http://www.korg-datastorage.jp/Manual/monotron_sch.pdf

and a pic of the guts;

[garcho]

nice, thanks for posting that Dino! but still gonna build one  

I've slowly been working on a single sided PCB for this, I'm sure it will be 2-3 times the size of the entire monotron.  It's Cliff's schematic mixed with Rick's switching and 2 ultra-basic LFOs, CV ins for the cutoff, and a clean blend pot.

I've played with one of those monotrons, and they're a lot of fun. Never plugged it into an amp, though. I've made some small 'travel synth' toys, basically audible Schmitt triggers a la Nicolas Collins and Ray at MFOS, and man, when you plug those in, they can rattle the windows! Next time you're on the road and you want to quell irritating companions, build yourself something like the monotron, just don't let anyone else get their hands on it!

[garcho]

that monotron schematic is pretty interesting, mostly op amps, Schmitt triggers and NPNs. sounds like a good project to start breadboarding!

[digi2t]

I'm just REALLY curious what a guitar going through a fuzz box, then into one of these suckers would sound like.

EDIT* - Think I just found my answer... to some degree. I'm REALLY tempted now.

[WhenBoredomPeaks]

I'm just REALLY curious what a guitar going through a fuzz box, then into one of these suckers would sound like.

EDIT* - Think I just found my answer... to some degree. I'm REALLY tempted now.

fuzz into minimoog vcf: (and some other fx)

http://soundcloud.com/dptp2/primitive-filter-patch

pretty bad recording but it sounds kinda cool

[garcho]

The monotron filter is just LP, right? As opposed to LP->HP, or parallel?

So happy this thread is alive again

[~arph]

This is a nice 9v OTA VC-LPF too:

http://ericarcher.net/devices/diy-lpf/

Btw, all his stuff is pretty cool

[garcho]

wow, killer link, thanks Arnoud!

[garcho]

been thinking about this again, and what would make for a good guitar pedal version of this lovely filter.

9 pots: cutoff x 2, res x 2, LFO rate x 2, LFO/CV depth x 2, clean/wet blend

3 toggles: parallel/series a la Rick Holt's suggestion, LFO ramp up/off/down x 2

4 jacks: in/out, CV x 2 (for cutoff, seems more important than res for pared-down version)

2 OTAs, 12 op amps* (in whatever package configuration is most convenient for the Byzantine layout this will end up being), 8 BJTs, yay 50 resistors, 25 caps.
*8 for the 2 filter sections, 4 buffers (VCC/2, input, parallel/series output, blend)

I'm thinking HP/LP switching isn't so important for a MS-20 clone, and any complication of the LFO is doubled. It seems important to have some kind of on-board LFO if we're gonna make this a guitar pedal project, since most guitarist won't have too many modules or synthy things laying around. I think an op amp with rate, depth and ramping is good enough.

Pipe dream? I want to make this an etchable project, but maybe it's too much. 

[nocentelli]

Guitarists might not have CV-outputting "synthy things" but the (cut-off frequency) CV input can easily be contolled with an expression pedal for a superb wah sound - I've breadboarded the Eric archer 9-volt LPF linked above, and it sounds great as a quasi-wah. Vero layout to follow when it's verified.