A Question about power:

Not since my design technology GCSE have I had to think about power requirements for things.

I'm unclear as to the power capabilities of the Intellijel 7U travel case this design is intended for. The specs for the case are:

Output + 12 V (mA): 3000
Output - 12 V (mA): 3000

The power requirements for this design are: 1998 mA +12V | 1046 mA -12V |

These two numbers are both lower the maximum stated "+" or "-" sides of the case PSU. However, I have a sneaking feeling that these two number added together cannot be higher than 3000. Is that right or am I miles off? Hopefully I'm wrong :-P

Also: the built-in bus board on the Intellijel case has 28 sockets. This design has 34 modules - all of which draw power according to their given Modular Grid specs. Assuming the built-in PSU has the capacity, is is possible to daisy-chain a handful of modules (the mults for example) to one socket on the bus board?

.... Oh and before anyone says anything: yeah- there's no Maths. Get over it


The +12V and -12V rails are separate power busses, they dont' add together - you should be okay with those numbers as stated, though the common advice is to leave some headroom between your usage numbers and the PSU's capacity numbers, so you don't want to push it all the way to 3A on either.

You can't directly daisy-chain modules, but you can either make custom multi-drop power cables yourself or buy some "flying bus" cables which plug into one socket and have more sockets on them for individual cables. In my rack I made a few custom multi-cables just to simplify the cable mess inside! You don't want to chain power-heavy modules this way though, only the "small" ones, my biggest grouped cable is 4 modules totaling ~600ma on the +12v.


.... Oh and before anyone says anything: yeah- there's no Maths. Get over it
-- rextable

ok, but why not? it's a fantastic learning tool for anyone interested in modular synthesis - or do you just want to connect some modules together to make some sounds? have you taken a look at the 'maths illustrated supplement'?

also would be really surprised if you need so many buffered mults - they're only really needed for v/oct signals anything else a tiny bit of voltage droop will be unnoticeable

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

sound sources < sound modifiers < modulation sources < utilities


I don't get the multiple buffered multiples, either. Or the need for two Pam's, as there's more potent methods for manipulating clocks and deriving Boolean gate sources to REALLY complicate the clock functionality.

As for the power question...yeah, this will be OK in that aspect. Note also that Eurorack uses three separate power rails: the + and -12V and the 5V rail which doesn't often come into play but is still necessary for certain digital modules. The only time the 12V rails MIGHT come together would be at a common P/S groundpoint, but they have different supplies on the "hot" side. But the rule of thumb for dealing with current inrush is that the P/S should be capable of supplying 20-25% MORE current than the module complement (roughly) requires, as getting too close to the current limit carries with it the danger that an inrush on switch-on could cause component damage.

One other point: this is really a rather small build, but there's a number of bigger modules in this. Naturally, this limits any further expansion in the case...but right now, it's locking you out of making adjustments to the module complement. This could be done a lot more effectively by trying to replicate the functions of those large modules in smaller ones...this should be possible. But also, some of this is simply redundant, like the thing with the Pams'.


Re the mults:

Short version; I like mults. The end.

Long version; I really really really really reeeeeeally want to avoid stackable cables or any other cable-based means of multiplying where possible. I find them to be very messy and reckon they're just asking to get broken, or to break the things they're plugged into if knocked. No doubt, they'll still end up being used here and there.

In any event, I can always take the mults out to make space, but I can't add them in if my case is full.

Re the Pams and efficiency:

Agreed - inefficient... And somewhat fiddly in use. The same can be said for the O_C. But they are all there for good reason.

Maximum efficiency at this stage is not my aim. Packing in all the functionality I'm inspired to experiment with, is. Between the Keystep Pro, two Pams, OC, Zadar and the various utils, I've pretty much got that covered. Logic, Euclidean rhythms, Turing machine stuff, quantization, complex envelopes etc. It's all in there somewhere and it fits in the case with room to spare. The only small-footprint multi-function module I'm aware of that comes anywhere close to the versatility and footprint of the Pams (other than the OC) is the Disting EX. But the Disting in all its incarnations - unlike the O_C and Pams - is a module that I have not found inspiring.

From what I've read and seen in the intertubez, two Pams synced together equals a sum far greater than its parts. Together they can do more stuff that I wanna do than any other module combination in the same square-footage. I suppose, following that logic, I should swap them out for two more O_Cs (which are even more powerful). But intuition told me that would be unwise. As and when I've figured out my workflow; what functions I want more accessible and which ones not, I can purchase dedicated modules to cater for those needs.

This design has been settled upon after a great deal of thought and research. I have also considered availability of stock. For example, The ever-so-slightly-more compact panning VCA solution Lucia suggested in an earlier incarnation of this design uses modules that seem to be completely unavailable at the time of writing this. The PanMix is only 1HP larger and slightly less flexible. For this same reason I have also since removed the the Freak filter. I just can't find anywhere with any stock.

As a starting point for trying stuff out, this design covers everything I want - albeit in a convoluted way. The next step is to see how it works in the real world and go form there. In any event, Pams is an eternally popular module. I'm sure there will be no problems selling it on if I deem one of them unnecessary.

Re Maths:

.......... .............

Is this some weird running Earorack joke?

Seriously though, with decades of experience with synthesizers, audio processing and signal routing under my belt, I'm no stranger to creating complex modulated patches and signal chains. In other words, I'm somewhat past learning the fundamentals of synthesis. Thus Maths - in the context of this case layout and intended purpose - makes no sense to me. As a learning tool, what's here caters far better to the gaps in my knowledge and methods of composition I have yet to try. If and when I change this design or get a bigger case, sure - I'll get a Maths. :-)

Here endeth the essay

PS I have no idea why half of this post in in italics nor can I figure out how to fix it. Soz :-{


BTW

Does anyone know whether the main L/R output on the PanMix is line level or Eurorack level? I'm hoping/expecting it to be Eurorack level so I can plug it into the Cockpit or Mosaic output module.


I don't know, man... 10 buffered mults? Two Pam's?
The folks that commented before are definitely pointing you in the right direction, especially for a case this size. Everyone is going to do this modular thing differently, for sure, but I can't imagine this is the best use of the limited space you are giving yourself. Are you planning on using the PNWs as your main sequencer, or do you have something external to sequence pitch more effectively?
I'm not a Maths disciple, but since you brought it up twice, it is excellent as Jim said for grasping and accessing the fundamentals of MODULAR synthesis. You may have years of experience with synthesis, but trying to cram all of the functionality of a fixed architecture synth into the small euro case you have selected is very difficult (impossible?), hence the Maths love. Need a mixer? Need a subharmonic generator? Need to attenuvert four different envelopes? Need a basic logic function? Need to create a weird shifting bouncing ball envelope? Maths... It doesn't have to be a complex signal chain. It just gives you all of the stuff you are ultimately going to need more than 10 buffered mults.
The PanMix is euro level. You probably won't need an external output module.
Have fun and good luck!


I don't know, man... 10 buffered mults? Two Pam's?
The folks that commented before are definitely pointing you in the right direction, especially for a case this size. Everyone is going to do this modular thing differently, for sure, but I can't imagine this is the best use of the limited space you are giving yourself. Are you planning on using the PNWs as your main sequencer, or do you have something external to sequence pitch more effectively?
I'm not a Maths disciple, but since you brought it up twice, it is excellent as Jim said for grasping and accessing the fundamentals of MODULAR synthesis. You may have years of experience with synthesis, but trying to cram all of the functionality of a fixed architecture synth into the small euro case you have selected is very difficult (impossible?), hence the Maths love. Need a mixer? Need a subharmonic generator? Need to attenuvert four different envelopes? Need a basic logic function? Need to create a weird shifting bouncing ball envelope? Maths... It doesn't have to be a complex signal chain. It just gives you all of the stuff you are ultimately going to need more than 10 buffered mults.
The PanMix is euro level. You probably won't need an external output module.
Have fun and good luck!

-- farkas

Forget the mults - bloody hell :-D

Version 1.2

ModularGrid Rack

Mults are gone, Freak is gone, 8HP of space to grow into.

Erm, I'm confused. Either you miss-typed or I'm missing something. The PanMix IS Euro level? Surely that means I DO need a drop it down to line level before hitting external gear. No? That was the idea behind the PAnMix and Cockpit combo. My intention was to use the PanMix to for things that don't need to be panned hard left and right by default (eg Rample and Plaits) and then use the Cockpit as a stereo summing mixer for all the effects.

Assuming I'm right about the PanMix, really, between it, the Cockpit (plugged into the Intellijel case output and the additional Mosaic output, the routing options are very flexible.

For all conventional sequencing duties I have a Keystep Pro. I have a Keylab MkII for control and expression purposes. I also have a midi foot pedal by DataLooper that I plan to plug into the Intellijel case midi input. Had I mentioned that before this case might have made a little more sense haha :-)

The two PAms and the O_C are there solely to create and mess around with tempo-locked gates and pitch-quantised self-modulating modulation sources - many of which will be required bring this case to life. If you factor in the number of modules in this case that require clock and pitch and time locked CV and gates, my choice of modulation sources makes sense

... I reckon anyway :-)


Erm, I'm confused. Either you miss-typed or I'm missing something. The PanMix IS Euro level? Surely that means I DO need a drop it down to line level before hitting external gear. No?
-- rextable

There have been quite a few discussions about the necessity of euro-to-line output modules in beginner rack posts on this forum over the past year or two. In general, most people don't need them. I've been at this for a few years, and I still don't use one. I just monitor levels into external gear closely. You can take a look at my rack and those of the other commenters by clicking on their user name. Some users have invested in output modules, many haven't. I do use a Strymon AA.1 for an effects pedal send/return, but otherwise I haven't needed a dedicated output for recording, etc.


Most external gear can cope with euro level - just turn the mixer down - if you have it at full volume it probably will clip - but start without a output module - if it clips then try a pair of passsive attenuators - and if they still clip or you NEED balanced outputs then buy an output module

the argument about mults was mostly about the fact that you don't need buffered ones for anything other than v/oct - but they were a bit excessive - passive mults are cheaper and don't take power headers - trying to stop you spending money on stuff you definitely don't need

the expander for zadar is very useful - more modulation inputs and you can mute the envelopes both of which are pretty handy

as for Maths - it's often left out by people who are willfully defiant due to the fact that it's popular - and as others have said - experience of fixed-architecture synths is really not as applicable to modular synthesis as it would seem - it's great for patching basic east coast style mono-synths and that's about it

personally I'd want more vcas and a matrix mixer

if you want to see my rack - look for this one : All My Eurorack includes DIY backlog and pre-orders, etc if it's upside down it's not there yet!

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

sound sources < sound modifiers < modulation sources < utilities


One other point about Maths: everyone uses it because the alternatives involve taking up even more space (quite a bit, in fact, in Eurorack) and still won't exactly do what Maths is capable of.

I actually DO have some things that can work in the same way...one of them is an EG&G Model 175 "Universal Programmer", another is a JAS arbitrary waveform generator with all analog controllability, and the last isn't workable right now because I need a tech that's not scared off by restoring and modding analog computers, and I've got three Systron-Donner 3300s that need to be refurbed into one fully-working one with synth I/O capabilities.

The latter weighs about 60 lbs and is the general size of a bedroom TV. The EG&G is bigger than most 1970s stereo receivers. And the JAS is SO rare and SO fiddly, I can't recommend it to anyone who's not got a lot of EE chops or doesn't have a soopergenius tech on hand. If you can find one, that is.

This is why Maths gets used all the time. It's not a "sheeple" thing at all...it comes down to the fact that it can do so much in only 20 hp. And if you've not seen it, here's MN's page on it...with what's really just a "basic" overview of the module: https://www.makenoisemusic.com/modules/maths Plus, it traces its lineage right back to Serge Tcherepnin and the famous Serge Dual Universal Slope Generator, so there's about 50 years of engineering and development there. Very hard to go wrong with that!


I don't think it was a bad shout by @rextable to have so many multiples. I know I have one Pams, and I normally have to mult each output using those star jacks to get copies of my signals. I have to admit I find it annoying and its very messy, It's generally cleaner when you just patch into mults and distribute from there. It's how I use mine in my one 1u row which tend to be my first choice even if it means losing the buffered mult for copying pitch.

I also think mults are good when you have a bigger system and you want to get from one module to another using an intermediary mult without having to buy stupidly long cables.

There are some very useful sequencers in the O&C when running hemispheres. However I am finding more and more that the more recently released sequencers really do the leg work in creating complex and interesting variation in sequences with built in quantization that really make it unnecessary to buy specific dedicated modules to create variation in cv. My current goto sequencers are Bloom and Metropolix.


I also think mults are good when you have a bigger system and you want to get from one module to another using an intermediary mult without having to buy stupidly long cables.
-- greenfly

Can’t tell if joking or making a suggestion akin to buying a Mordax Data because of the pretty graphics

--- Voltage control all the things ---


I also think mults are good when you have a bigger system and you want to get from one module to another using an intermediary mult without having to buy stupidly long cables.
-- greenfly

Can’t tell if joking or making a suggestion akin to buying a Mordax Data because of the pretty graphics

-- toodee

I don't know if I would have 10 mults but I understand why having more than one might be appealing. If I think about the star splitters I use, probably 3 are full with cables and 3 more are probably half full. I could see having 5 mults may be viable if you want to distribute cv / audio a bit more efficiently.

Generally it's about the trade-off of what the extra module gives you vs hp consumed so I have 3 one of which is passive. Thats more than enough for me.

You may be a person who likes to have 15'000 sound sources and filters, which would seem daft to me but I just want to say @toodee, that that's ok pet x


I tend to leave mults out of my builds because, in smaller builds, they eat space. However, when I start heading to four rows and beyond (especially if the rows are over 104 hp), a couple of passives or so is a must. At that point, you're not "stealing space" from functional devices...you're adding something that'll make a large build easier to use.

And of course, once your destinations for a CV source go beyond 4-5 devices on the same CV, that's when you stick in a buffered mult, usually up by the VCOs.