ahoy, cannot shed too much light on the mixing question apart from confirming xenus_dad’s observiation. thinking about it, i came across this phenomenon myself but didn’t pay as much attention — eager to investigate, though.

when i started out with analog/hardware synths, i used to have a microbrute, channeling in a second external vco to fatten up the sound. the built in mixer for the input had exactly the same characteristics as described: distorting the sound (much to my pleasure).
these days i mainly mix audio on an external desk mixer or internally on a ac-coupled erica pico a-mix without any distortion whatsoever.
i vaguely have the hunch that using my dc coupled internal mixer (mn moddemix) is causing some ‘distortion’ too. but will investigate.

chris meyer of learning modular has an article on mixer, which might help:
https://learningmodular.com/choosing-utility-mixers-for-your-modular/
from skim reading, a hint might be that dc coupling can introduce offsets and enhance the likelihood of overdriving circuits (i know for example that mangrove can produce quite some dc offset with certain settings, adding another signal can cause a distortion of half the signal, resulting in odd harmonics — ac coupling would filter out those offsets).
but these are all my own layman’s attempts to explain this phenomenon, curious to know more.

ps: the k-mix is ac-coupled, which would support this theory.

I do wonder if this has to do with AC coupling, though I’m really not sure. My sound engineering knowledge is very basic.

I’m going to try eliminating a lot of variables, using known generated voltages & sines and doing spectral analysis of the results. I will get to the bottom of this!

pretty sure right now it is a ac/dc coupling thing. if all your euro-mixers are dc-coupled, this would explain …

(x-post at MW, apologies if you’re experiencing doubles).

This is educational! But still, not sure what’s going on.

So, I’m using a Mordax data to generate two waves, both sines, 200Hz and 300Hz.

I’m running these through the K-mix into Logic, sending all ins to a single mono output, and using Channel EQ on that mono output w/ no actual EQ changes to do the spectral analysis.

If I send these individually to the K-mix, I get spectral readings that seem reasonable:

If I sum these with Shades or Vnicursal, I actually see and hear 100Hz:
Shades, fully open: https://www.dropbox.com/s/l4bbgm2s60r2x9w/Screenshot%202018-07-08%2015.06.37.png?dl=0
Vnicursal, channels at about 10 o’clock: https://www.dropbox.com/s/1uu7h2ag9utgbby/Screenshot%202018-07-08%2015.09.43.png?dl=0

The rolloff of higher frequencies is understandable on the Vnicursal, but the creation of the 100Hz frequency seems odd.

  1. Is this just a measurement error? I’m trying to do the measurements as late in the signal chain as I can, but it’s possible I’m comparing apples to oranges.

  2. If it’s not a measurement error, is there any way in a eurorack mono summing situation to prevent this kind of “subharmonic” generation? I really don’t think this is a clipping/saturation thing at this point, though I could be wrong.

Again, appreciate any feedback. I have a feeling this is a basic signal processing thing I need to learn.

Thanks!

I think the K-mix is doing something weird with the signal. Looking at that, and also going to listen to the mixed signal directly to see if it sounds correct.

H’ok, I think I’ve gotten myself sorted. Lessons learned:

  • Euro signals are hot. We knew this but I think I didn’t properly appreciate it
  • On my K-mix I need to have the trim down all the way or nearly all the way if a mixed signal is in play
  • I cannot run a unity mixer with 2 or more signals into my K-mix at all without attenuating it externally first… It is too hot for the kmix to handle. This was probably the source of my thinking that Euro mixers are inherently muddy.

Thanks for letting me mostly think out loud at y’all.

8 Likes

Helpful stuff, thanks for doing so.

Brief addendum: my Malekko Unity Mix clips with only two signals. This is two sine waves, out of both sides of a DPO:

You can see the signal flattening out. The oscilloscope is set to 5V per square.

Given two signals causes distortion, it makes me wonder what this unity mixer would even be for. Do other folks have positive experiences with unity mixers like this? I got this to have summing on the cheap, and I suppose I got what I paid for, but it seems like it wouldn’t be useful unless I intentionally wanted distortion.

yeah, mixing with unity gain on all channels either forces you to b.y.o. attenuation or use it for CV—unless distortion is something you’re going for.

I’d be curious to see if the difference between a “unity mixer” and a “precision adder” is more conceptual than electrical :joy:

I guess it’s a glorified OR? Works for me I suppose.

well, small difference probably: an OR should choose the peak of two signals, so out = max(in1,in2) (often also including 0 in that list so as only to get positive voltages.) A precision adder should instead perform out = in1 + in2, with clipping at or near the power supply range.

2 Likes

Actually, this gives me a potentially fun idea. I have a ADDAC Manual Latches. 4 of those channels let you pass voltage through rather than just sending 5V.

The latches plus a unity mixer plus some source of steady offset voltages could turn this into a fun playable Beasts Chalkboard cousin. Will experiment tonight.

2 Likes

Hello there friends,

I’d like to have some feedback on my choices for 104hp x 2u rack. Since in the country where I live there is no market for modules or cases, I might have to buy them all over a few months and have them delivered to family members in the US, so I can pick it up, assemble and travel back home.

Do you guys have experience in buying several modules at once?

My objective here is ambient, soundscapes, and very organic techno with the ability to process my field recordings: https://www.modulargrid.net/e/racks/view/698938

I’d love if you pointed out any inconsistency, redundancy or absence in the rack! I’ll use an external mixer, and I have no other modules.

I’m thinking about replacing the BIA with nRings + another 2hp Mix, but I really know nothing about the availability of nRings or its quality. Also, how something like the Nebulae v.2 or the Morphagene compares to the STS?

Thanks so much!

ime, even buying two modules at once can be a bit overwhelming. For example, I bought Teletype and Cold Mac at the same time to finish my case about a month ago and I still haven’t really touched Cold Mac yet because I haven’t had the time to accommodate it in my head.

thoughts:

  • sequencing: a lot of your modules would be unlocked by sequencing, and not just in terms of pitch. do you have a vision for that?
  • buffered multiples: what is the use case you imagine here?
  • mixers: you have four (if you count the Quad VCA). not necessarily too many, but what is your vision for those?
  • if you imagine a patch with this case, what are branches and ears doing?
4 Likes

Thanks so much for taking the time :slight_smile:

I was just thinking buffered multiples were like better multiples haha, 2 of them as general utilities i guess?

Ears is there as an interaction tool, also I’m a guitar player, and it can be used as an envelope follower for my guitar, field recordings, and I just think i like the module idea. I mean, it seems like a nice way of interacting with the system, but hey, idk. If it were not a good match for the kind of music I’ve been digging, i’d love to have extra hp. What do you think?

Mixers were for layering modulation and voices, but maybe it is too many?

Also thought that the OC would provide enough sequencing capabilities, would you advice me otherwise? I thought of the Voltage Block from Malekko before, but ditched it for the OC, looking for more hp.

I’ve been liking branches in VCV for splitting clock signal and random clock, it kind of works with sharp percussion hits too, then maybe it can create a shuffle rhythm between Plonk and the BIA? Random source? I’ve been thinking of replacing it to open space for Rings too.

Then, unfortunately I have no means of bringing modules one by one, the shipping fee is more expensive than some modules, and others don’t even ship to south america. Plus there is already 2K+ for the airplane trip hah

So yeah, the best way for me is buying at once, also having in mind that i’m not afraid to make minor mistakes or to fit my music into the possibilities of the system! I’m a manual worm, so i’d love it to be an overwhelming experience in that sense :stuck_out_tongue:

Thanks again, the planning ahead will be very important since it is a good money to make it work, so i’m truly thankful!

EDIT: I took out ears, excess mixers and multiples, was able to squeeze in Rings, Octone and RCD for greater rhythmic control (and lower price, lol): https://www.modulargrid.net/e/racks/view/699729

Well, if you’re not afraid of minor mistakes, then I won’t stop you! But that said, why are you playing HP golf so hard? If your goal is to make “full” tracks with drums, bass, pads, samples, etc. all in 6U, give up now. [1]

Also, maybe it’s not as exciting to you as putting the whole thing together from different manufacturers from scratch, but you could consider a prebuilt system as a way to avoid some of the headaches you mention. In fact, the Make Noise Shared System Plus has a lot of parallels with what you’re sketching out now—not that I’ve played with one, mind.

ETA – Sorry, I think my pessimism might be a little stronger than is called for here. I will say that it seems like it might be unclear to you what it is that you want your modular to be, and that optimizing for the most functionality you can get while minimizing size and/or cost doesn’t seem like the best way to figure that out.


[1] or accept a lot of constraints, or buy an ER-301, which is I guess the same thing.

3 Likes

Check out perfect circuit for “complete systems”. Some really nice combinations there.

https://www.perfectcircuitaudio.com/systems

5 Likes

I relate to the manual-worm thing and the need to mass-buy, obviously the conventional wisdom being purchasing bit-by-bit isn’t an option. Also I think it’s very cool that you are open to your selection musically leading you somewhere. I would caution that creative output and productivity might slide while you get to grips with things - perhaps attempt learning in small module batches. But be ready to record, learning this stuff (in my experience) can also yield some happy accidents. As mentioned elaborate full tracks will be tough to achieve, but that’s not to say you can’t have evolving, interesting ‘full’ patches for live and studio work.

System wise, for what its worth I don’t think you were too far off on your first post. To me it seems the much more interesting starting point. You have given careful thought to utilities (which is commonly missed) and have some amazing voices (bia/e352/plonk - big combo in my view) for percussive, texture and lead parts. I am in agreement with Alanza regarding sequencing. I don’t think O&C / Disting are great options unless you are happy menu-diving. Malekko VB is a lot of functionality and perfect feature set I think. I also don’t think you need 3x filters (inc disting), bia for instance is so harmonically complex. I suggest switching out the 2x functions for something like Just Friends or an Omnimod, to give you more options for modulation and free up some hp.

‘complete systems’ are a good starter but not for everyone and you wont have anywhere near the flexibility. I say go with your gut, if your wallet can stomach it.

@alanza

Thanks for the heads up, it is definitely not pessimism! I’m fully aware of the limitations, but also, i would not try to build a conventional track in any way, i guess i’m looking for intricate simplicity, and 3 or 4 voices would be cool already. It really helps to have somebody checking in and saying the truth about the system! Thanks!

But again, I might not being fully transparent with my motives since this is not my mother language and my expression is limited in that sense, but I think i do have a clear vision, which is processing field recordings into ambient soundscapes and experimental techno.

I’m working on the sequencers! Thanks!

@steveoath

Thanks, huge inspiration, the glitch series captured my attention. Going to shameless steal some ideas from there!

Thanks again!

@ravel

Thanks for the careful reply! Now you made me hope for happy accidents >.>

I’m going with you and @alanza on the sequencers. And opening space by removing the somehow redundant modules, plus exchanging the STS for the Tip Top ONE, to fit Rings. Soon I’ll appear again with a new sketch :slight_smile:

The wallet hurts, of course, but some good months of planing and saving to open a new musical world? Hey… i’m gonna run that bet hah

Thanks so much you all!

EDIT: Here is a more mature version of the rack, based on the kind input! https://www.modulargrid.net/e/racks/view/699729

2 Likes

I’d ditch Clouds for a uClouds. It’s a lot of hp for a mostly hands-off piece of kit. I’d also recommend Chronoblob over the Tapographic Delay. You’ve already got a tappable clock via Tempi. Consider a Radio Music as an alternative to one of your TipTop ONE’s. It has a very different feel and might be more fun for you to experiment with for your field recordings? Otherwise, this seems like a big improvement on your original post.

I think it’s worth noting that you no longer have a proper envelope generator. Just Friends can do that, but then it’s not doing anything else. In my experience, JF is at it’s best as a kind of wild LFO source. Especially since you don’t really have much else performing that duty.

It’s worth continuing to consider your utility modules. I’ll also say you might consider a Maths still. It does a LOT. Since you’re buying a system whole hog without the ability to easily acquire new tools, Maths could save you a lot of heartache when you suddenly realize you need one of the things it can do for you. (requisite link: https://www.youtube.com/watch?v=XJtpzysqJv8)

2 Likes