Mannequins Three Sisters


Hmmmm anybody else have trouble with the trim procedure on this? My Sisters is flat as hell, drops by about a semitone per octave. I tried trimming it but the screw was already at the max, I can only make it flatter…


as an aside, I happened to patch Three Sisters and Mangrove as a layered voice last night and they stayed in tune together for the whole range of the piano!

@smbols, you’re probably doing everything right but just to check: you’re monitoring the low output?


Which out are you using? Only the low is meant to track 1v/octave.

(Edit: wasn’t awake enough to see @alanza had asked exactly the same thing.)


If the screw is at max, turn it back to the middle and use a sequencer to change octaves during the process. You’ll be able to see the accuracy of the tracking.


In case you haven’t seen it, there’s a calibration guide here. If it’s flat at the top of the range, you actually turn the trim pot to make the current pitch flatter - it moves the 0v frequency flatter than e.g. the 2v frequency, so you are actually spreading out the range from 0v to 2v.

As @Jonny suggested, you want to choose an arbitrary root frequency for 0v, send in 2v, check to see if the new pitch is 2 octaves above, if flat, trim flatter, if sharp, trim sharper, then send 0v in again, tune 0v back up to the root frequency, send in 2v, trim again, rinse repeat.


This is great! Thanks for sharing. I’m going to mine this afternoon to try and replicate


Okay, that makes sense. Thanks for explaining what’s going on. That guide is a lot more helpful than the blurb in Sisters’ dedicated manual.


Okay so a few things going on with my 3S that I’ve been meaning to ask about for a while now. Most likely not anything ‘wrong’, just need some clarification on what exactly is going on…

If the frequency of the filter matches the frequency of the oscillator being input then I get a volume spike on the low output (span fully CCW, quality at 12:00). This seems to mean I can’t get a fully clean filter sweep as whenever the frequencies match it overdrives for a very brief moment?

Similarly if I am using the high output, when the frequency of the filter matches the oscillator frequency it seems to overdrive and very slightly low pass filter the signal. Which again makes a clean filter sweep impossible as at some point the signal spikes.

My suspicious is that this is because the 3S is always slightly resonant no matter what the quality is set to? Looking forward to seeing what any of you think though, thanks in advance!


I don’t have a concrete answer for you, but it sounds like your oscillator is pushing the filter into resonance.

What is the effect when you put Quality less than 12:00? Are you using crossover or formant mode?

Also, which input are you putting the oscillator into? I find that if you put it into ‘ALL’ and listen to LOW, that will give you the standard low pass filter. Same with high pass, input into ALL, listen to HIGH. There will be spikes otherwise, I think.


In both instances I’m inputting to ‘all’ and then going out from ‘high’ or ‘low’ in crossover mode.

I’ll have to test changing the quality lower than 12 tomorrow to be sure, but I’m pretty sure when I was testing earlier it didn’t remove the spike. I’ll give it another go though!

For reference I’ve been testing with the square, tri and sine output of the Instruo Ts-L.


I don’t have a 3S, but maybe try attenuating on the way in? You’d still get a volume spike but you could stop it from overdriving.