First attempt at patching a useful instrument - Moog Minitaur

#1

This is my first attempt at emulating a real synth using the axoloti. The synth is the Moog Minitaur. I don't actually own one of these so I'm trying to replicate functionality and settings using only the schematic, pictures of the unit and youtube videos of the sound and function.

image.png1574x938 80.7 KB

image.jpg2868x1780 989 KB

And here is my first cut at the incomplete patch:

image.png2488x1586 483 KB

minataur.axp (8.0 KB)

A have a few questions that I think will help me:
- VCAs: I used the multiwave to be able to select between waveforms like on the Miniataur. Are these the right type of objects? The schematic shows the LFO modulating each of the oscillators, what parameter would it be modulating?

  • VCF: It also shows the LFO modulating the VCF, same question what parameter?
  • Modulation: it has dials for the LFO and amounts for the paths to the VCOs and VCF is that just a VCA controlling the amount?
  • VCF: It shows the filter with the cutoff and resonance, but what is EG?
  • VCF: I don't think I have the ADSR connected to the filter properly, any tips?

Thanks!

#2

Nice! One thing up front: the accuracy of a patch like this depends on lot on the particular filter and oscillator implementations you use. One of the biggest keys to the Moog sound is the ladder filter. There are various digital implementations of it around. Off the top of my head I'm not sure if there's an off the shelf object for it in the Axo library.

What oscillator you pick depends on how accurately you want to emulate the Minotaur's oscillators. Just having the same shapes, saw and square, is not necessarily going to sound accurate to the original analog implementation.

Adding some kind of overdrive or saturation to this might help improve accuracy.

On modulation: in a traditional subtractive design like this the LFO is almost always modulation the pitch of the oscillator and the cutoff frequency of the filter.

On EG amount: it looks like on the original this is a "depth" knob that is bipolar. Center is 0. It controls how much the envelope modulates the filter cutoff frequency. So in your patch this would be like a bipolar mixer object in between the adsr and the filter pitch/cutoff freq input. What you have is sort of like having the EG amount cranked up all the time.

The LFO can also modulate the same parameter, the filter cutoff frequency, by some scalable amount, so roughly it's like:

cutoff_frequency = cutoff_knob_value + lfo * lfo_amount + env * eg_amount

The oscillators you're using have the "freq" input for audio rate modulation of the pitch. If you stick with them you can use an audio rate LFO into a mixer for "amount" and then into the freq input.

#3

I actually just ported a 24db Karlsen ladder filter to "Axoloti format" today.

Check it out here:

#4

Thanks for the tips, a couple of updates:

The manual actually does a pretty good job of describing the synth's routing and operation. https://api.moogmusic.com/sites/default/files/2018-02/Minitaur_Manual.pdf

@urklang I'm not trying to match the sound exactly, but getting close would be a bonus. My plan is to try to create a couple of these types of synth clones for use in a live setup and create a hardware case that emulates the knob and button layout and can just plug-in the axo.

On modulation: so are the LFO signals usually just connected directly to the oscs?
On EG: so for routing of the LFO to the filer, would the LFO act as an input into the gate of the ADSR and then that signal connected to the filter cutoff? Right now the gate is connected from the keyboard, but it makes more sense to have that connected to the vca, right?

@jaffasplaffa that filter looks good, but I can't seem to find the code for it in the library or in the forum. Can you paste it here?

I'm going to read through the manual more and make adjustments to try to modify my patch to get close. I'll update this thread with the changes as I go, keep the suggestions coming! Thanks!

#5

Sorry I forgot to sync library, it should be there now, I see it on Github atleast.

Search: jaffa/filt/ladder1

#6

Yup - got it thanks!

A couple of interesting parts, in the manual when it describes the oscilators, it mentions the following with regard to VCO2:

Sets the frequency offset of VCO 2 from VCO 1. The offset range is +/-1 octave. Center position tunes VCO 2 in unison with VCO 1. NOTE: If playing between notes 60 and 72, the pitch of VCO 2 is limited to note 72 (C4) regardless of this control setting.

Not sure how to do this, but probably a big part of the synth sound, as then it will provide a second voice that's linked to the first.

Then the next part is the filter, looks like that filter is perfect @jaffasplaffa.

image.png1076x674 115 KB

And then talking about the panel control:

Adjusts the CUTOFF frequency of the Low Pass Filter from 20 Hz to 20 KHz. As the knob is rotated clockwise, the cutoff frequency is increased, allowing more harmonics to pass through the filter, resulting in a brighter sound. Conversely, as the knob is rotated counterclockwise, the sounds get darker.

So this would mean that I need to have the LFO connected to the ADSR which then has an output to the filter, not sure how to set the objects to get this 20-20k range?

#7

I think that is going to be very hard. Pretty much every single filter in Axoloti world only goes up to 13k.

#8

have you tried jt/filter/vcf4pole2x ? pretty cpu expensive, but sounds nice...and should go up to 18.8khz

#9

There's a bunch of great material out there that goes in depth on the ladder filter.

SuperCollider has a couple of nice digital implementations.

#10

So cool. Honestly the more I get into this the better the axoloti becomes. Thanks again for the filter - it sounds way better. Here is an updated patch with the new filter, the sync/detune for osc1 and osc2. Glide is working but probably needs some tweaking.

image.png2474x2142 571 KB

Things left to do:
- Route the LFO to the oscilators and to the ladder filter through some type of gain control.
- Add some inputs to objects so that they can be adjusted.
- Add midi velocity controls for the VCA (50%)
- probably 100 other things I haven't got on the list yet.

#11

No problem. After resting it for a few days and testing it again, I am pretty happy with it, it has a nice twang. But could be more efficient, it does take a lot of cycles to process around 1000. But it sounds decent :slight_smile:

#12

Do you plan to share the patch? :slight_smile:

#13

Yep, the patch is already included in the above posts, but I'll attach the latest copy here. I guess I should figure out how to share it through the official community patches as well.

minataur.axp (7.7 KB)

Other quick update to this, it looks like the Moog Sirin synth is actually very similar to the Minitaur but also provides for some additional features, I'm going to take a look through that schematic and manual too and see if there's anything else missing.

#14

Thank you nive patch:)

#15

Very nice patch, this .. destined for the jam selection ..

#16

Ok so I have been learning a LOT about synthesis and actually modelling real synths and while my first patch was a valiant attempt at learning the axoloti platform, it wasn't very good at all. I'm going to try to remodel it starting with the basics, and I hope some help from others way more experienced than me along the way.

I'm hoping that this might help others with similar ambitions along the way. What follows is going to be a step by step of my exploration with the modelling effort using the same Moog Minataur synth.

Part 1 - Inputs and Outputs.

Hardware wise, I'm a bit more limited in what I can do with the axoloti hardware, and make it generic enough to be useful for others as well. Thankfully we do have some of the basics that the minataur has:

  • Input, we have a stereo input pair, which is one better than the minataur (mono input only)
  • Output, we have a stereo output pair, again one better than the minataur (mono output only), and the same 1/8" headphone output.
  • In addition the minataur has 4 control inputs which are used for pitch, filter and volume CV and a gate.
  • The minataur only has a midi-in while we have a midi-out/thru port as well, but really usable for inputs.
  • The biggest difference is on the front panel where the minataur has a mixture of 4 buttons, and 17 potentiometers for controlling all of the parameters, and we have none. So we can use midi CC's to control these, and some combination of hardware / software to trigger the midi CC's. I'm going to be using the Axocontrol, and a 16n fader that I recently built (with a non-functioning 9th fader due to very poor soldering skills).

For the patch I'm going to initially build this as a mono in/out synth, but using both channels in and out

In the patch attached I have mapped the controls using object I thing will be best suited for the job, if anyone has input on them please let me know.

  • audio/in stereo for the input
  • midi/in/keyb for the mini input
  • midi/in/cc for the midi cc inputs (probably need 17 of these eventually to match the hardware inputs)
  • audio/out stereo for the output
  • disp/scope (x2) as a utility for the patch

image.png1777x1180 80.8 KB

minataur-step1-hardware.axp (2.8 KB)

Part 2 will be focused on sound generation.

#17

Im confused by some of your use of ‘mini’. I think you mean MIDI?

The MIDI out/through port cannot be used for input.

#18

Part 2 - Sound Generation

With the basics of input and output setup, I've added four oscillators, a saw and a square wave for both OSC1 and 2.
I also connected the midi note to both oscillators, but I added a tuning mixer (mix/mix 1) to use knob 6 to effectively detune the second oscillator.

I then provided a dual VCA to allow for control of each oscillator which then gets fed into another mixer. This could probably be simplified with a 3 input mixer to include the external input. I then routed the signal to the left and right out to test the functions.

The main issue I haven't solved yet is to find away to use the toggle buttons (momentary switches) to switch between the saw and square oscillators. The drj/switch/par_swap_1 controls don't seem to work that way. Looking for someone with the suggestion for this.

image.png1768x1149 190 KB

Next up, hopefully fixes to the toggles, addition of the VCF and glide control for the notes.
minataur-step2-sound.axp (9.1 KB)

#19

I noticed, your connections from the axocontrol to the LEDs are crossed.

I think this is what you're trying to do with toggles, simplified, to verify par_swap_1 operation. I'm using two sine oscillators, tuned to different frequencies. The pushbutton switches them, as expected.

Screen Shot 2020-10-30 at 5.10.19 PM.png1770x1500 246 KB

#20

Hey all,

Also new to this...!

What are the PB0, PB1, etc. for (yellow round ports on the axocontrol object)?

Thanks!