Eurorack standard specs

For my A-level electronics project, I’ve been working on making a simple modular synth. So far, I’ve got an 8-step logic-controlled sequencer, clocked my a scmitt inverter osc, driving a 555 timer VCO, which is then run through a 741-based resonant LPF (not very resonant at the moment, as I could only get hold of a 100k rather than 500k pot) and a practivally non-existant VCA (I’ve been using a single TIP141 as an output amp, and I’ve been turning it on and off with the gate OP from the sequencer. I’m planning to add a simple pic-based wavetable, LFO, arduino/attiny-based midi-sync and maybe CV interface, and an envelope if I can make one fit on. At the moment, it’s all laid out on a few breadboards, but I plan to transfer it to PCB later. When I do, I want to try and build each unit as a eurorack module, and use it as a very basic start to a modular synth setup. My problem is, however, I don’t know anything about the eurorack synth standard. I’ve looked, but coulden’t seem to find a lot. So, here’s my main Q’s;

What are the unit dims?
What’s the standard supply voltage/voltages?
What’re the standard connections at the back?
What’s the norm for jack sizes?
How do you callibrate occilators etc. to 1v/8ve?
Is there anything else I need to know?

Thanks in advance.

1 Like

look on the doepfer website or google for the page witch mechanical and electrical specs.

Doepfer DIY info
Doepfer Module Details

that should include all euro relevant construction and power details, and more.
how to calibrate an oscillator depends on its design.

Seen this?


Hope it helps!

Intrigued by what you are concocting, do post pictures, pre-post breadboard!

> What are the unit dims?

From Doepfer

> What’s the standard supply voltage/voltages?

+/- 12V.

Assuming you’ll get exactly a clean +12V / 12V from the rails is not a good practice actually they can be away from this ideal value or they can fluctuate with temperature, etc… Whenever you need a particular voltage (for scaling/offsetting circuits for example) you’d better go with a voltage reference.

If your module needs 5V (for MCUs, etc.), there are two options:

  • either a regulator on your board - taking the +12V as an input. Problems: heat dissipation, digital noise isolation;
  • or use the +5V provided on the bus-board. Problem: not all cases deliver +5V - hence the existence of products like this

> What’re the standard connections at the back?

There is no standard as to where you should put them.

If you only use +/- 12V, you can use a 2x5 pin connector:

If you need the +5V, or the main CV/Gate lines, you can use a 2x8 pin connector:

There are holy wars about the use of shrouded headers. I don’t use them, and explicitly label on the PCB where the “red stripe” (-12V) of the power connector should be.

Everything else (calibration trimmers, expander ports, programmer/JTAG ports…) as you wish.

> What’s the norm for jack sizes?

3.5mm. If you decide to have your PCB parallel to the panel (a wise decision), a commonly used part is this. Another common part is this. Both are total pain in the ass when it comes to industrial production.

> How do you callibrate occilators etc. to 1v/8ve?

Not sure if your question is about:

  • What is the scale standard? It is 1V/Oct, Doepfer uses 0V = midi note 36 but there’s no consensus about that.
  • What is the preferred component for calibration? There is no standard. Multi-turn trimmer, digital calibration procedure, it’s up to you…
  • How to make a circuit follow the 1V/Oct standard? For oscillators and filters this requires the presence of an exponential converter (either as a discrete circuit built with a transistor pair, or through the use of a chip with a built-in exponential response like the 2164) for the “V/Oct” part, and the use of appropriate ratios of resistors at the input of the circuit for the “1” part.

> Is there anything else I need to know?

Gate/trigger level was originally (Doepfer) HIGH = +8V, LOW = 0V - though an increasing number of modules use +5V as the HIGH level - mostly due to the proliferation of digital devices :slight_smile: As a result, your gate/trigger inputs should better have a low threshold (I use a transistor inverter, so my threshold is one diode drop).

Input impedances are expected to be 100k. I rarely deviate from this norm except when the circuit uses some weird normalling paths which can occasionally half it. Corollary: you’ll find it more convenient to put attenuverters (aka “polarizers”) on your CV inputs than mere attenuators.

Your circuit is supposed to behave well when a strong signal source is connected by mistake to one of your outputs; and when a reasonably long stretch of cables is attached to the output. This is often achieved by having a 1k resistor in series with all outputs (protecting from external signals and preventing the op-amp output to directly “see” the capacitance of the cable) - the drawback being that it’ll cause a voltage drop of 1% into each 100k impedance input it is plugged to. Another approach is to have the 1k resistor in the feedback loop of the op-amp (no drop!), with all the associated annoyances.

Your module is supposed to behave well when the power connector is connected backwards. Solution 1 is to have Schottky diodes in series with the 12V / 12V lines. Solution 2 is to use a polyfuse, and schottky diodes to ground in reverse configuration (when power is plugged backwards, all the current is sucked in the path of least resistance through the diodes rather than elsewhere, and they make the fuses temporarily “blow”). Both solutions equally suck in that they cause a voltage drop proportional to the current used by your module that’s why you shouldn’t rely on the rails being exactly/-12V, especially if you make things blink (Use the +5V rail or an onboard regulator to power anything that blinks!).

My modules are ideally designed so that I can run my pinky in the space between knobs without touching them (not always, but I start from there). Modules from other manufacturers sometimes don’t pass this test and I consider them too crowded.

1 Like

Thanks for all the replies! I didn’t expect anything that quick, especially considering the time. I’ll have a further read through of it later, but it the comments look to have answered all my questions for the time being, so thanks.

I am intrigued about your non use of shrouded headers, why so?
cheers and regards

> why so?

No consensus on the way they should be oriented on the bus board and on the way the ribbon cables should be assembled.

1 Like

it seems to be quite standard (judging from how it’s oriented on all offered pcbs that provide room for a shrouded header with silkscreening) that if you look from above with the gap on top then - is on the right and + on the left. i’ve made my bus board that way and haven’t had an issue with any module (besides that my psu is already out of juice damnit)
but anyway, who keeps us from just using a shrouded header if we want to (and there’s room enough)


Yes that seems to be the way, I always knock up my cables to that spec.
cheers and regards.

Hi. I’ve just got back from being on holiday, but I’ve read through the posts a bit more thouroughly now, and the info looks really useful. I can’t really post any photos of what I’ve got so far yet, as I’ve broken up for summer, and the breadboards with my synth so far are 20 miles away in a toolbox in a locked room. I have however slung together a schematic of the sequencer that I’d done, so far as I can remember it. I’ve tagged the PDF on this post. It’s pretty crude, and due for some re-arrangement, but it does work. I’ve yet to add adjustable sequence length, chaining, ext. clock etc., and the correct stuff on the IO for proper modular interfacing. The schematic’s just how it’s laid out at present.
I haven’t bothered to do the diagram for the osc. as it’s just a very basic 555 timer circuit. I’m tempted to try and make a valve oscillator, but I suspect that might be biting off more than I can chew.
Extra question; What do I do about temperature compensation?

Thanks again everyone for the useful replies.

> What do I do about temperature compensation?

From what I understood of your schematics, your thing is a self-contained box with its own oscillator. As such, I doubt people will be using it to reproduce accurate musical melodies, so temperature compensation is not a problem… especially since what I have seen so far doesn’t have any component with a wide dependence on temperature.

Which components will need compensation?

Hard to tell without seeing the schematics of your oscillator!

Usually, it’s the transistors in the expo converter…

Ah. I haven’t got an expo converter or anything on the osc yet, but the schematic I used is here;

I’ll have to tag one on soon.
I found the filter circuit I used, too. It’s shown as a breadboard layout, but it’s here;

It’s probably not the best filter, but it was the only resonant filter I could find that I could build with the parts I had at collage.

How do you plan to put this oscillator under voltage control - for example so that it can be driven by your sequencer?

I’ve got it working fine with the sequencer. I’ve just fed the output on to the CV pin on the 555 through a tuning pot, and it works OK. I’ve just been using a 555 for testing, and I plan to make a better osc. before I make an osc. module PCB.

I am not familiar with this kind of circuits but I doubt it aims at giving precise musical notes. Why would temperature stability be an issue? How is “detuning” a problem to something which is not designed to play musical scales?

As of yet, I’ve just been setting the notes by ear, but I can get it to run through a decent enough bassline sequence hitting the notes pretty well. I’ve come across plenty people using 555s for musical stuff before, such as on the stylophone (arguably musical), or here;