So, first batch is in the system (top row) and been using it a lot...i'm hooked!

ModularGrid Rack

I common patch looks something like this:
Pams to generate the pitch with different techniques, random from single channel or multiple ones mixed together with logic conditions. This then goes to the pitch of Plaits. Other channels on Pams control the inputs of Plaits either straight in or via Maths. Maths usually helps with this too, but more often I use it to control the filter and VCA. Out from Plaits into Filter and then VCA and out. Beads have not arrived yet..
It's all fun and game, but I feel it's lacking some thing.

Now to the planned new things, row 2:
I use Maths 2-3 for attenuation, but want to free that space up, therefor planning to add a dedicated attenuator in form of Sinc Defero.

I don't find it satisfying to put in LFOs in Pams all the time (usually start using the two in Maths first) and want something more direct, therefor looking at the ochd. Good choice or should I look for something with more exact control?

Lastly the mixer, since I use it alot in Maths and feel like a dedicated one is worth having. Going stero to be used with Beads, but is that overkill? I want to mix both audio and CV, should I get two different mixers?

Three more utility and modulation focused modules, but want to here some suggestion what you would add as a next step? A sequencer? Random generator? More sounds sources?

I would do this a little differently. Instead of the big Intellijel Quad VCA, I would suggest three After Later dual VCAs. They use the same topology as the Intellijel (and, natch, Veils), can sum to mono, all that. Put one after the Plaits, one after the VCF, and one after the Beads and then you'll get a voice path that's got 100% VCA control over amplitude. Maybe put the Quad VCA down in the modulation row then, so that you can have amplitude control (when needed) over your modulation signals. Gonna beat on this for a bit...

LATER: OK, this wasn't all that difficult to bang out, in truth. The voice row was already pretty well-implemented, so the big focus here was to load up the modulation row while still optimizing the voice row above. So...
ModularGrid Rack row has the P/S, then next to that is Ladik's dual slew limiter for portamento, which you can have on either upward CV movement, downward, or both. I then doubled the Plaits, because 1 VCO = good, but 2 VCO = BETTER. This lets you do detuning, VCO sync and so on. After those is the first VCA pair, then we get into the VCFs. I put in a G-Storm Korg Delta VCF here, since this gives you smooth, pad-like responses...which complements the Steiner-Parker clone and its aggro sound. Then the next VCA pair, followed by this multiple-looking thing from Takaab which is actually a unity-gain fixed-pan mixer. The white-ringed jacks are your "left" and "right" outputs, and you use the seven inputs to place signals between them. This sets things up for a stereo feed to the Beads. After the Beads, one more VCA pair, then you've got an Alyseum QMix for your stereo mix. Output and headphone preamp duties are handled by the new version of Happy Nerding's Isolator.

Down below, things start with the Pam's, then a Doepfer Noise/Random/S&H is next. LFO duties get handled by a Xaoc Batumi (with the Poti expander) for four potential "raw" LFO signals. Maths follows, then one of Happy Nerding's 3xVCAs for linear, DC-coupled level manipulation of modulation signals (the Quad VCA was too big), and a Tiptop MISO for altering, mangling, inverting, and pretty much any other -ing you want with the modulation. And then, a Xaoc Zadar (also expanded with the Nin) offers four EG channels. One final addition with HN's new FX Aid Pro for extra processing, and that's it.

Should be easy enough to work with...but at the same time, there's a lot of raw power under the hood now.

Big thank you @Lugia! Knocked another one out the park
I've studied the setup and the different functionality you've added and made a little remix. Fun to mess around with and great start to work from.

ModularGrid Rack

I tried keeping the modules I have and still fit similar functionality and setup as you suggested. One thing to add is that I have an external Electron box I do drums and effects with right now. That's why I removed that effects module from the bottom right.

So, I liked the setup with multiple dual VCAs and two VCOs, but I don't like having the same module multiple time. For me a fundamental thing with modular is the mixing and matching of brands and unique takes to create one system.

1st row: I swapped the dual Plaits with a Twin Waver and two of the VCAs to other brands. The Ladik I couldn't find anywhere close to me, so jammed a Disting for the slew and whatever I might want to use it for, ended up with space for a little sampler. Haven't thought much about it, but might be fun.

2nd row: Added a midi out to help with the drum programming, but that got to tight for the A-118-2, so went with a 2hp one. Not sure if you added the A-118-2 for the S&H or the Noise? The rest of te row is the same except i manage to squeeze in my Quad VCA and output-module I already got.

Did I keep the spirit of your system, or completely missed the mark on some of its features?

So, I liked the setup with multiple dual VCAs and two VCOs, but I don't like having the same module multiple time. For me a fundamental thing with modular is the mixing and matching of brands and unique takes to create one system.

-- Alltvin

Bad idea. So, let's say you have two oscillators, one being a Plaits, the other is a basic VCO. You'd LIKE to get the Plaits to sound really huge and fat, so you...oh, wait! Your VCOs don't have anything resembling the same waveform, so you can't exactly do that, because that detune trick requires (when done right!) two identical VCOs with identical waveforms, and if one's a Plaits, the other should also be a Plaits.

The same applies to the VCAs. Since those are Veils-topology VCAs, they'll all have the same sort of operating method, the same(ish) sweet spot on the variable response, same sort of gain factor, etc. That way, when you set one in the voice chain, the same sort of settings are applicable to each pair. This makes it much easier to get a patch set up, but if you go with all sorts of different VCAs, you won't get that. For example, in your new example, there's one of the After Later duals...but then, the next VCAs are those Doepfers, which DON'T have variable response. So if there's a certain curve you like on the After Later...well, that won't be happening with the Doepfer.

Couple of other things...for one, you really won't find much need for the Pam's MIDI expander. If you want to lock up drums to the modular, you can just use one of Pam's trigger outputs as exactly that. And if you need the drum machine to be the master clock, just send its timing sync from the machine to the Pam's clock input. Solution simple. And as for the question about the A-118-2, yes, the intention was to provide a noise/random source AND the sample and hold. You now have neither, save for the 2hp dual...but without any random sources, you can't use those to S&H noise/random signals to get a random CV output. But worse still, if you want to add a percussive "hit", you're now missing the audio source that you'd probably use for that.

Tell ya what...instead of progressing further with the hardware build, get hold of a copy of VCV Rack, and explore the why behind things like the identical VCOs, VCA behavior, and so on. For the most part, you should be able to put together various bits (I know there's Mutable module clones in the set) that explain the issues. VCV's free as well...

sequencer wise: you could just get some expanders for pams and then plug in a arturia beatstep pro or somthing

dont scrap maths yet till you get a bit further on your journey.

and yea i see miso has jumped in there. good module.
i use noise engineering lapsus os, its great and more performance oriented

and yea i see miso has jumped in there. good module.

-- trigmusicnz

One of my picks, actually. I don't really see it as a "performable" module...rather, I look at it as a "modulation ganger" in addition to its other functions. For example, if you've got a single LFO being messed with via a MISO (or my other fave for tighter spaces, the Frap 321), you can change the LFO's rate and ALL derived signals will change in lockstep with that modulation source.