So I’ve managed to buy a second case which gave me a chance to rearrange some stuff in my first case. You’ll probably be put off by the Model D in the second rack because of the space it’s taking up, but this is the setup that works for me and I’ll stick with it for now. I know a lot of people utilizing their model d this way and I think it’ll be a good way for me to avoid buying more modules for the time being. I can utilize the case it came in with other modules if I really need to or just sell it. I haven’t installed the modules in the case yet - I want to make sure I have a solid plan first. You’ll notice that i have a lot of sequencers and that’s because I like utilizing different sequencers for different patches….so I planned everything making sure nothing is redundant.

The only modules I DON’T own yet are the Hysteria oscillators, Intellijel Quad VCA, and Polaris. What are your thoughts on the Polaris filter? Also, what kind of module do i need to add glide in my sequences? Is a slew module what I’m looking for?

If you have any suggestions regarding placement of the modules, please let me know. Also would be grateful to know your thoughts on what other modules I could add to this.

Grateful for your time!

First Rack

Second Rack

First Rack

Second Rack


Portamento? Well, probably the best one for the price is Ladik's C-012. That's got two slew limiters with selectable portamento on up, down, or both CV direction changes. Their C-015 loses one slew limiter but adds a gate function so that portamento happens only on receiving a gate, and this also has the voltage direction discriminator. And the big bonus is that either one of those are only 4 hp wide, so they don't take up a bunch of space. And they're CHEAP. And since these don't use audio, you could put one in between the P/S and the Plaits to help keep P/S crud out of any audio modules.

As for the Polaris...I'd suggest adding a second VCF so that if you want to use it as a phaser, you're set. My suggestion would be another SVF such as Doepfer's A-121-2, which gives you simultaneous low, high, bandpass, and notch filtering.

One other point: you'd be better off if you tossed the Doepfer A-138 in deference to a current version Veils. That's only 2 hp larger than the mixer, but it allows you to have CV control over the incoming levels. Plus, since it has a "breakable" mixbus, you could conceivably use it in a 2-2 or 1-3 configuration, depending on what's needed at the time. Also, tossing the Intellijel Quad VCA for a Veils gives you the same sort of variable VCA response and breakable mixbus, costing $10 more but occupying 2 hp less.


I'd agree with Lugia on every single point and add - more utility modules...

think of them as plumbing... copying signals (stackcables or hubs are great for non-pitch signals, otherwise buffered mults)... logic, sequential switches, offsets, sub-mixing, matrix mixing (see my post on the matrix mixer thread for why) etc etc etc... possibly the most important modules you can add - and even better they are generally comparatively inexpensive...

and see my signature for a formula to follow, loosely, that gives you the most versatile modular for the least expense... scales well too - from tiny to massive modular synthesizers...

"some of the best base-level info to remember can be found in Jim's sigfile" @Lugia

Utility modules are the dull polish that makes the shiny modules actually shine!!!

sound sources < sound modifiers < modulation sources < utilities


Hmmm...OK, I went off on this, and I think I've got a better solution AND a rationale for removing the Model D.
ModularGrid Rack
In this version, I have the two cabs side-by-side, mainly to show the division of labor between the two cabs. These could just as easily be on "top" of each other, though, although the left cab should probably be the one on the bottom if this is done.

LEFT TOP: P/S, Black Sequencer, and then the Voltage Block WITH the Varigate 4+ that serves as its memory, controller, etc. After that is a Xaoc Bytom, a multichannel OR + gate/trig distributor...very useful in situations such as this. The last bit there is a XODES bidirectional 4-1 sequential switch, positioned so that it's adjacent to the "voicing" row in the right cab.

LEFT BOTTOM: Pam's expander and Pam's...expander was put on the left to minimize "outside" cables getting confused with patchcords. Then the DOT trigger sequencer, and after it are some basic clock modifiers...a dual pulse delay and a two-channel probabilistic pulse skipper. These support the Boolean logic module, a Frequency Central Deep Thought dual Boolean gate.

If you're wondering what the Boolean logic is for, here's an explanation: Boolean logic uses specific conditional gates which output a gate pulse when certain conditions are met. For example, if two incoming gate signals overlap, Boolean logic can obtain new gates via either an AND (when both inputs are 'high' this outputs a gate), OR (when only one gate is supposed to be "high"...having both or neither inputs hot = nothing) and so on. With a decent Boolean module and some modules to screw with the timing pulses, you can have even MORE clocking or weird conditional states with which to drive all of the sequencers...or anything else that wants a gate!

Anyway, after that are the two Metamorphs, and then I added a Tenderfoot Quad Quantizer for quantizing the sequencers that don't have internal quantization...or to just quantize modulation signals, which can be a really beautiful sound when something like an LFO on sawtooth gets quantized into ascending tessalations.

RIGHT TOP: Voice row. P/S first, then a Ladik dual lag for portamento, etc. Then Plaits...and then, some fun! The two next modules are Klavis Twin Waves mkiis...which are dual wavetable VCOs with internal quantization plus random capabilities, which gives you a total of five (or potentially six with the AUX out on the Plaits) oscillators in this version. Veils goes after that to offer CV control over (most of) the oscillator outputs; outputs that it DOESN'T have space for, though, are easily repurposed as audio-frequency modulators. After that, you've got a dual lowpass gate, Make Noise's LxD, then a way more interesting and complex VCF from Dave Rossum that offers changeable slopes, improved resonance control, sweepable filter types, and so on; this is pretty much perfect as a singular VCF in this sort of build. And at the end is a 4-in, 2-out Tesseract Tex-Mix setup which gives you a VCA input, two FX sends, and panning control across the four inputs, and the output section has the master level, headphone preamp, and I/O for the two FX sends and two stereo FX returns. Got hipped to the Tex-Mix by Jim...and when space is key, it actually is one of the smaller VCA-equipped performance mixers out there.

RIGHT BOTTOM: Modulation, mostly. Batumi (with Poti expander) first, then the ultraversatileness that IS the Maths. After that, you'll notice a Frap 321 and a Happy Nerding 3xVCA...these modules are there specifically to tamper with and alter modulation signals in a large variety of methods, allowing you to derive even MORE (and potentially screwier!) signals from just a few modulation generators. And speaking of those, there's a Quadrax/Qx combo next...the Qx really rips the lid off of the Quadrax by bringing the EOR and EOF signals out, and then these can be used to fire other EGs in the Quadrax; fact is, you could turn this one module, with the use of the Qx, into a cascading AR envelope set with successive EOR -> next TRIG patches which ALSO has CV over the attack and/or release...in short, a totally batshit insane waveform generator! Then the Zadar/Nin is next for "normal" envelope generation...and of course, there's loads of sources to trigger those, ergo the Nin's presence in here. The Dual FX is next (which is perfect for the Tex-Mix's FX bussing: 2 mono sends, 2 stereo returns). And that last sliver is a quad level adjuster, which takes synth-levels down to line-levels. Two, natch, get used for the Tex-Mix outs, which leaves two others to be used as "isolated" outs for a leadline, etc.

So...loads more complex, but majorly improved. Some of the initial modules did get bumped here so that other, better solutions could be pursued...which is also the rationale for pulling the Model D out. It's a good synth, to be sure, but THIS is where you could conceivably get to once it's gone.


Hmmm...OK, I went off on this, and I think I've got a better solution AND a rationale for removing the Model D.
ModularGrid Rack
In this version, I have the two cabs side-by-side, mainly to show the division of labor between the two cabs. These could just as easily be on "top" of each other, though, although the left cab should probably be the one on the bottom if this is done.

LEFT TOP: P/S, Black Sequencer, and then the Voltage Block WITH the Varigate 4+ that serves as its memory, controller, etc. After that is a Xaoc Bytom, a multichannel OR + gate/trig distributor...very useful in situations such as this. The last bit there is a XODES bidirectional 4-1 sequential switch, positioned so that it's adjacent to the "voicing" row in the right cab.

LEFT BOTTOM: Pam's expander and Pam's...expander was put on the left to minimize "outside" cables getting confused with patchcords. Then the DOT trigger sequencer, and after it are some basic clock modifiers...a dual pulse delay and a two-channel probabilistic pulse skipper. These support the Boolean logic module, a Frequency Central Deep Thought dual Boolean gate.

If you're wondering what the Boolean logic is for, here's an explanation: Boolean logic uses specific conditional gates which output a gate pulse when certain conditions are met. For example, if two incoming gate signals overlap, Boolean logic can obtain new gates via either an AND (when both inputs are 'high' this outputs a gate), OR (when only one gate is supposed to be "high"...having both or neither inputs hot = nothing) and so on. With a decent Boolean module and some modules to screw with the timing pulses, you can have even MORE clocking or weird conditional states with which to drive all of the sequencers...or anything else that wants a gate!

Anyway, after that are the two Metamorphs, and then I added a Tenderfoot Quad Quantizer for quantizing the sequencers that don't have internal quantization...or to just quantize modulation signals, which can be a really beautiful sound when something like an LFO on sawtooth gets quantized into ascending tessalations.

RIGHT TOP: Voice row. P/S first, then a Ladik dual lag for portamento, etc. Then Plaits...and then, some fun! The two next modules are Klavis Twin Waves mkiis...which are dual wavetable VCOs with internal quantization plus random capabilities, which gives you a total of five (or potentially six with the AUX out on the Plaits) oscillators in this version. Veils goes after that to offer CV control over (most of) the oscillator outputs; outputs that it DOESN'T have space for, though, are easily repurposed as audio-frequency modulators. After that, you've got a dual lowpass gate, Make Noise's LxD, then a way more interesting and complex VCF from Dave Rossum that offers changeable slopes, improved resonance control, sweepable filter types, and so on; this is pretty much perfect as a singular VCF in this sort of build. And at the end is a 4-in, 2-out Tesseract Tex-Mix setup which gives you a VCA input, two FX sends, and panning control across the four inputs, and the output section has the master level, headphone preamp, and I/O for the two FX sends and two stereo FX returns. Got hipped to the Tex-Mix by Jim...and when space is key, it actually is one of the smaller VCA-equipped performance mixers out there.

RIGHT BOTTOM: Modulation, mostly. Batumi (with Poti expander) first, then the ultraversatileness that IS the Maths. After that, you'll notice a Frap 321 and a Happy Nerding 3xVCA...these modules are there specifically to tamper with and alter modulation signals in a large variety of methods, allowing you to derive even MORE (and potentially screwier!) signals from just a few modulation generators. And speaking of those, there's a Quadrax/Qx combo next...the Qx really rips the lid off of the Quadrax by bringing the EOR and EOF signals out, and then these can be used to fire other EGs in the Quadrax; fact is, you could turn this one module, with the use of the Qx, into a cascading AR envelope set with successive EOR -> next TRIG patches which ALSO has CV over the attack and/or release...in short, a totally batshit insane waveform generator! Then the Zadar/Nin is next for "normal" envelope generation...and of course, there's loads of sources to trigger those, ergo the Nin's presence in here. The Dual FX is next (which is perfect for the Tex-Mix's FX bussing: 2 mono sends, 2 stereo returns). And that last sliver is a quad level adjuster, which takes synth-levels down to line-levels. Two, natch, get used for the Tex-Mix outs, which leaves two others to be used as "isolated" outs for a leadline, etc.

So...loads more complex, but majorly improved. Some of the initial modules did get bumped here so that other, better solutions could be pursued...which is also the rationale for pulling the Model D out. It's a good synth, to be sure, but THIS is where you could conceivably get to once it's gone.
-- Lugia

This is really wonderful and I’m grateful that you took the time to make this for me. You’re awesome! I will definitely look into these modules and see which I can afford to add to my rack.

Also this all is indeed a bit too complex for me right now and as you know I’m still a beginner and learning. However what you showed me is a good way for me to get an idea of what I’ll need in the future as you said. It’ll eventually happen. Just need to figure out my workflow and how I usually use my set up.

I’m definitely going to look into that Tex Mix module and NIN expander for the Zadar for now. I’ll get a couple more buff mults, a matrix mixer probably, some attenuator/attenuvertors, slew limiter, and another filter if I get the Polaris so I can use that as a phaser if I want to. I know I got myself into this modular mess and it’s not looking good for my wallet, but I’m still trying to keep costs somewhat low.

Again, thank you so much for taking the time to do this and explain everything.


All of that sounds right...save for the buffered mults. While it's true that you need those to prevent voltage sag on the pitch CV, I seem to recall that the Klavis oscillators have buffering on the inputs, leaving only the Plaits...which I also think uses buffered CV ins. They're getting more common...thankfully. But as far as mults go, just get some stackcables or inline mults, and that'll be all you need. Again, keeping space open for function is key in a build of this size.

I actually had a certain "synthesizer" in mind when building this, namely: https://www.vintagesynth.com/eml/eml400.php The EML 400/401 was a revelation when I started using the one we had at MTSU in the early 1980s. I'd used sequencers before...the ARP 1604, Oberheim's DS-2 (I think that's it?), and Sequential's 700 were ones I had prior experience with, but the EML really went way beyond those. For one thing, it has the synth built-in (the "400" part), but the 32-step sequencer was just mindboggling. And three CV parameters...plus some gating as well. But the quantizer was totally bonkers...you could go from just a few steps per octave to SIXTY divisions per octave, and with some judicious tweaking of that, you could get the 400/401 to respond properly to 1V/8va CV...pretty cool, given that EML used a totally unique (and effin' annoying) CV at what's basically 1/10th volts per step, meaning that an EML octave was 1.2V/8va.

But yeah...I got this wild hair during the rework to...well, Avesta's into all these sequencer modules, and I can jam a 6U Arturia full of 'em...hmmmm...so why not make it have the general form-factor of the 400/401? After all, that device's form factor was just about perfect for a sequencer/synth combo. Plus, this gives a very clear upgrade path...add two more 6U cabs over time, make the lower left all about clocking, logic, comparators, and various stochastic bits (all super-important in doing truly complex generative work), and the lower right can be all modulation by turning the upper right into all voice. Pretty cool when it works like that! Plus, since you'd have an initial Tex-Mix setup in there, all you'd need to add more inputs to the mixer would be another faderbank of four channels...either mono or stereo. Although, with the master section being all but MADE for that DualFX setup, expanding the Tex-Mix would be far less costly and irritating than having to get Yet Another Device.