Pitfalls in mixing Behringer System 100, 2500 & 900 series modules?

[S.Crow]

I’m looking to build a small Behringer System 100 set-up and am completely new to this.
Due mainly to non-availability, I’m wondering what the pitfalls are in building a mixed set-up from 100, 2500 & 900 series modules?

I have noticed that the 900 series support S-trig which would require conversion, so are there other basic compatibility issues when mixing across these three series of modules?

It will be a small set-up due to space constraints, so something similar in size to the Behringer Eurorack 104 which is:
104 HP
Maximum depth: 38 mm
Dimensions (W x H x D): 546 x 84 x 133 mm

Another option is to start with a Behringer semi-modular and then add the System 100 modules as they become available.
The downside of modular seems to be too many choices!

[Arpangel]

Apart from S-Trig there are no other major differences, you can buy the S/V converter so its no problem, but, the essence of the Moog sound is in the oscillators and filters, you can use EG's and VCA's from anyone and not notice a big difference, although Moog mixers do have a sound, but that's nothing to do with triggers etc.
The option of a semi-modular to start from depends on how far you want to go with modular, and what your priorities are, if you plan on going deep into the rabbit hole you may as well go in feet first and start to build up your modular.
I asked all sorts of question when I got started, and then realised its best to just get on with it, and follow your instincts, apart from practical questions like you’ve asked here, there are no rights or wrongs, you just have to work out what you like, theres nothing more to it than that, and be prepared to buy "and" sell, you’ll have a few failures along the way.
Best of luck.

[BigRedX]

You can make a +ve to s-trig converter with a transistor and a couple of resistors which fits inside a standard 1/4" patch cable jack plug. IIRC the circuit was published in E&MM magazine, so a trawl through the Mu-Zines archive should sort you out. I used a have a bunch of these for controlling my Korg MS20 and 50 from a Roland MC202.

[Hugh Robjohns]

This, possibly?
https://www.muzines.co.uk/articles/interface/5957



Or this:
https://www.muzines.co.uk/articles/s-tr ... ject/1608#

NB. The images for the S-Trig converter and a Lead tester project were mistakenly swapped in the original magazine. The Muzine site acknowledges this and says it's fixed... but it wasn't for me today... so the relevant schematic is above.

There's another far more elaborate converter project described here:

https://www.muzines.co.uk/articles/univ ... rface/4314

[BigRedX]

This, possibly?

That's the one. Only one resistor - even better than I originally thought.

[Hugh Robjohns]

Thought so.

It's very crude, but should work. And PW's more elaborate version below is just a more robust and slightly more flexible version of the same thing.

For anyone not electronically confident but interested, the top circuit uses the transistor as a simple switch. When a positive V-trigger input voltage rises above about 1V, a small current flows through the 10K resistor to the transistor's Base terminal (on the left). This causes the transistor to 'turn on' which means that the collector (top) and emitter (bottom with the arrow) terminals present a very low resistance between them.

So they are effectively shorted together, resembling a closed switch... which is what an S-Trigger signal is — a switch contact closing.

When the input trigger voltage falls back to zero the transistor 'turns off' and there is then a very high resistance between the collector and emitter, simulating an open switch contact.

PW's more complex version uses a small relay as the switch element instead of the raw transistor, not least because it galvanically isolates the V-trigger source from the S-trigger destination and thus avoids any potential ground-loop problems.

However, the relay is energised by a transistor doing the same thing as in the first example. When its base goes positive, the transistor turns on, allowing current to flow between collector and emitter. Since that current has to flow from the battery through the relay, it activates the relay too, closing the switch terminals (and the LED shows when it's working).

The V-trigger voltage sends current into the transistor base through a 10K resistor again, but it is routed through a diode bridge rather than going in directly. The reason for that is simply so that the unit will still work correctly if the V-trigger happens to be a negative-going type instead of a positive-going type (which was a feature of some synths back in the day...).

The single diode across the relay is there to protect the transistor from the reverse voltage pulse generated when the relay is switched off.

PW's article described the operation in a little more detail, and in his inimitable style.

[BigRedX]

It might be crude but it worked perfectly for us converting the +ve trigger from Roland synths and drum machines into an S-trig that worked with the Korgs. It also had the advantage that it was small enough to fit inside a standard plastic-bodied 1/4" jack plug so there was no need for yet another bulky interface box with a battery or PSU.

We already had one of those that allowed us to synchronise our Yamaha RX11 drum machine with the Roland MC202 and run both from a single tape sync track.

In those pre-MIDI days, interface standards could be weird and confusing. For some reason the two trigger outputs on the Boss DR55 (one for each step and one for each programmed accent) produced different voltages and while most of the time there was no difference for the receiving device so long as the voltage spike was sufficient to activate the trigger, for some reason when using it to trigger a Kitten or Cat synth, one of them (I don't recall which) would also cause the oscillator pitch to shift upwards by slightly less than a semi-tone.

The other weird one was Korg's insistence on using Herz/volt for the keyboard to oscillator interface but octave/volt for the oscillator and filter modulation inputs, which meant that it was impossible to "play" the filters in tune from the keyboard or to get them to track the keyboard pitch in a linear fashion. On the other hand it did mean that once we had figured that the modulation input was standard oct/volt, controlling the synth from other oct/volt sources like the MC202 was relatively easy and didn't require investment in an expensive interface box. We added a second oscillator modulation input to our Korg MS20 for exactly this purpose.

[N i g e l]

an opto isolator chip could also give an open transistor switch action [ a bit like the MIDI_IN circuit] with ground loop isolation.

high volts on the input gives grounded low volts on the output.

[S.Crow]

Apart from S-Trig there are no other major differences, you can buy the S/V converter so its no problem.

Am I correct in saying that the 2500 series is also V-trig like the 100 series?
I’m generally cooling on the 900 series as they appear to take up more space which will be a potential issue, so the conversion won’t be required.

.. the essence of the Moog sound is in the oscillators and filters, you can use EG's and VCA's from anyone and not notice a big difference,

Great point.
I am looking more closely at what functionality I really need.
I probably don’t require multiple types of oscillators, although I might well want four that output separately from the system to a mixer for a quad channel set-up.
The 100 series has the bonus of offering some compact modules with dual VCO, VCF etc.

The option of a semi-modular to start from depends on how far you want to go with modular, and what your priorities are, if you plan on going deep into the rabbit hole you may as well go in feet first and start to build up your modular.
Best of luck.

Starting with semi-modular is partly about availability as mentioned.
I’ll do more research and try to prototype in software modular before jumping in.
Because I am looking at a quad set-up, I need to get a realistic sense of what I can achieve with a small case.
The idea is to work with relatively simple tones but there’s a risk that might be too limiting.
So if I do a proof of concept in Ableton along with Cherry Audio Voltage Modular, that will point me in the right direction.
More patience required.

I did hear that seemingly there were temperature related voltage stability issues with the 100 series which led to pitch instability.
Any news on that?

[Arpangel]

The 100 series had oscillator tuning issues on the early batches, but that should be sorted out now.
The 100 oscillators are very very good, big, rich sound,
The 2500 filters are amazing, I have both, the Filtamp is my favourite as it has a very smooth sweet response.

[Arpangel]

PS

I’d say the Berry System 100 is the best value modular around at the moment, in every way, it’s very difficult to beat in terms of features, and sound, and it’s ridiculously cheap, but the original was a great design, so they had a great foundation to build on

[The Elf]

The 1047 filter is the cream of the 2500 crop, IMHO. Its notch filter is particularly gorgeous.

The problem with the 2500 modules is their low ratio of functionality to rack width. I can imagine they are amongst the worst offerings in this respect. It's the reason I couldn't add the third oscillator or noise module I would have liked in my NiftyCase. The 1036 module baffles me, so no loss there...

I found a good compromise in the 2600, coupled with my chosen 2500 modules. This pairing cover pretty much anything I'd want to find in a modular system. My 2500 rack is self-contained, albeit fairly humble on its own.

The symbiosis of my 2500 and 2600 is remarkable. I have five oscillators, three ADSR envelopes, high, low, notch and band filters, and the 2600's utility features. This pair are truly great pals.

Here's my 2500/2600 Frankensynth:

[Kayvon]

The other weird one was Korg's insistence on using Herz/volt for the keyboard to oscillator interface but octave/volt for the oscillator and filter modulation inputs, which meant that it was impossible to "play" the filters in tune from the keyboard or to get them to track the keyboard pitch in a linear fashion.

I've been meaning to try using some Hz/V keytracking for things like filter and oscillator mod depths, curious as to how natural it might sound.

I have a Klavis Flexshaper which is much like Oberheim's Tracking Generators n so can approximate some logarithmic/exponential curves. I tried it once on my Korg 700 but it was quite a fickle business. Which led me to think perhaps a more general transfer function might be easier to work with.

Apologies to OP for digressing btw.

[BigRedX]

I've been meaning to try using some Hz/V keytracking for things like filter and oscillator mod depths, curious as to how natural it might sound.

I have a Klavis Flexshaper which is much like Oberheim's Tracking Generators n so can approximate some logarithmic/exponential curves. I tried it once on my Korg 700 but it was quite a fickle business. Which led me to think perhaps a more general transfer function might be easier to work with.

Apologies to OP for digressing btw.

This was over 40 years ago, and all I can recall is that I didn't like the effect I was getting at the time trying to use the keyboard voltage to control the filter cut-off frequency. My biggest problem is that I couldn't use the HPF in self oscillation as another oscillator and get it to track the keyboard with the VCOs. If I was to try it again now I might feel differently about the sounds I could get.

[Kayvon]

I've been meaning to try using some Hz/V keytracking for things like filter and oscillator mod depths, curious as to how natural it might sound.

I have a Klavis Flexshaper which is much like Oberheim's Tracking Generators n so can approximate some logarithmic/exponential curves. I tried it once on my Korg 700 but it was quite a fickle business. Which led me to think perhaps a more general transfer function might be easier to work with.

Apologies to OP for digressing btw.

This was over 40 years ago, and all I can recall is that I didn't like the effect I was getting at the time trying to use the keyboard voltage to control the filter cut-off frequency. My biggest problem is that I couldn't use the HPF in self oscillation as another oscillator and get it to track the keyboard with the VCOs. If I was to try it again now I might feel differently about the sounds I could get.

It's certainly an interesting proposition for tracking, like, which part of the exponential curve might you use? And across what range of your notes? Obvs it's non negotiable if you need the filter to track in equal temperament but in my experience there's a whole raft of elements that need subtle non linear tuning to sound 'right' to one's ear.

The Flexshaper has 5 points so is limited and futhermore I've thought about asking the designer if some type of smoothing might be possible to implement. If you look at the per note offsets some of these piano modelling synths use they are wildly non linear with all types of lumps, bumps, spikes and dykes across the 88 notes. Sometimes peaking and troughing even within an octave. The thought crosses my mind whether 12 points would be especially valid to then scale over multiple octaves.

[S.Crow]

I’d say the Berry System 100 is the best value modular around at the moment, in every way, it’s very difficult to beat in terms of features, and sound, and it’s ridiculously cheap, but the original was a great design, so they had a great foundation to build on

I came to a similar conclusion and seemingly so have many others based on long wait times.
After considering it further, I decided that modular definitely isn’t the way for me for my current project.

Andertons’ have a bunch of Behringer modules across the ranges in stock and some on sale:
https://www.andertons.co.uk/browse/keyb ... =ascending

[Arpangel]

I’d say the Berry System 100 is the best value modular around at the moment, in every way, it’s very difficult to beat in terms of features, and sound, and it’s ridiculously cheap, but the original was a great design, so they had a great foundation to build on

I came to a similar conclusion and seemingly so have many others based on long wait times.
After considering it further, I decided that modular definitely isn’t the way for me for my current project.

Andertons’ have a bunch of Behringer modules across the ranges in stock and some on sale:
https://www.andertons.co.uk/browse/keyb ... =ascending

Those prices are completely crazy, 2 Berry Moog oscillators and a driver for £120! what! and a filter for under fifty quid, please.

The 1047 filter is the cream of the 2500 crop, IMHO. Its notch filter is particularly gorgeous.

The problem with the 2500 modules is their low ratio of functionality to rack width. I can imagine they are amongst the worst offerings in this respect. It's the reason I couldn't add the third oscillator or noise module I would have liked in my NiftyCase. The 1036 module baffles me, so no loss there...

I found a good compromise in the 2600, coupled with my chosen 2500 modules. This pairing cover pretty much anything I'd want to find in a modular system. My 2500 rack is self-contained, albeit fairly humble on its own.

The symbiosis of my 2500 and 2600 is remarkable. I have five oscillators, three ADSR envelopes, high, low, notch and band filters, and the 2600's utility features. This pair are truly great pals.

Here's my 2500/2600 Frankensynth:

Elf, that's more or less my set-up, I’ve got two oscillators, both filters, the dual sample and hold module, and an envelope, they go well with my Gray Meany.
I love the FiltAmp, its such a smooth sounding filter, I use it all the time just for it's silky smooth sweeps.