When Pedro approached me asking if I could mod their 921s, I started toying around with the 921 and the smaller 921B. Originally, I was reluctant for firstly, AMS Synths has this wonderful resource, and there is further really good material online about various Moog clones and originals. Secondly, the iconicity of these circuits can be a bit intimidating. Be this as it may, let happy little accidents rule, so here’s some stuff…
Proper tracing of differences will hopefully come when I find more time, but some differences here already. LM3046s are replaced with a coolaudio V3046, the big single turn trimmers are all 100k, and, most importantly, the OTA and FET buffers are ditched here (Moog Q3, CR3 and IC7 are missing, see schematic below). Maybe the latter was on account of tuning instabilities? Moreover, the original IC used in the VCO sync setup is replaced on the Behringer module with a MC1496 – a substitution described somewhere a while back by Kevin Lightner on the Moog forums in a different context.
Tuning-wise, the two 921Bs I have are stable but needed calibration per scale trimmer adjustment. One unit was off by 40cents over 4 octaves, the other a whooping 100cents. Once scaled, both play at D when the frequency knob is on zero. Super too-quick judgement: if space is no issue, I prefer the bigger 921 over this one.
For the 921B, I like three additional modifications well: alternative VCO sync, manual PW, and manual Sine controls. You can implement these modifications using an additional panel and wiring the panel behind the scenes to your 921B. Or, with some more effort (and at the loss of some jack/controls), so as I do with the modules pictured above, if space is an issue.
Here’s an overview of what happens where in the 921B schematic – NB this is from the old Moog service notes, so part numbers do not correspond with the Behringer module.
Alternative VCO Sync
The original sync circuit does something really interesting – the incoming sync signal is compared with the internal VCO signal and depending on a set threshold, the internal pitch is pulled to that of the incoming signal (in a phase locked loop setup). The sound of this sync is nice, but I was hankering for something more ripping, so I discovered this…
Inspired by the Minimoog sync modification, I tested how to best reset the integrator cap and found that the spot marked ALTSYNC above works super well with a variety of input signals when you add this little passive circuit here:
What happens here is that an incoming VCO signal approaching zero volts pulls the base of the NPN labelled ALTSYNC IN to ground. As the transistor closes, the adjacent NPN opens and this, in turn, opens Q1 (Moog), which makes capacitor C1 (Moog) reset. Cs1, a 100nf ceramic or film box capacitor, debounces the sync process a little – NB this is between the sync in and ground. Rather than working directly at the integrator (like the Minimoog sync mod) going this somewhat roundabout way is more compatible with various sync input signals, at least I found so while testing. Here’s a row of different VCO inputs (3340, discrete, various level spans) and even self-oscillating filter inputs:
When you implement this modification, make sure that the pin lug of your ALTSYNC input is not just dangling freely around. A safe way of avoiding random glitches, is to use a switched socket with the switch lug tied to +12V via a resistor (I used 100k, works well). (Pulling the ALTSYNC in to ground, on the other hand, prevents C1 from charging and stops the VCO).
Not sure how sacrilegious all of this is, but hey… I like the sound of it. As for finding the right spot on the Behringer PCB, there is, quite conveniently so, a through VIA you can use rather than soldering directly onto an IC pin.
Kind of a wasted opportunity not to have manual pulse width control here (yeah, tHe OriGInAl aLsOO diDDNtttt – sorry), so I just did that here. One way, of course, would be to repeat the circuit from the 921A, but in my version below you don’t need to think about voltage divisions bla etc. but simply replace the rectangular width trimmer with a potentiometer and two resistors.
On the Behringer, VR5 is a 100k trimmer. so I used a B100k pot. You need a 100K resistor between the PCB wiper terminal and your pot wiper and a 33k resistor between the ground terminal of that trimmer (see pic above) and pin3 of your pot. This way, you get 50:50 PW on just about 12 o’clock on your potentiometer. The PW input, of course, remains functional with this modification, so PWM is still possible.
Manual Sine Shape and Symmetry
These two are super easy and fun! Simply replace the trimmers for sine shape and for sine symmetry with an B100K potentiometer each (Behringer uses 100k trimmers while ‚Moog schematic indicates 1k). By turning up SHAPE, you the sine gains more pulse like qualities and symmetry tilts the waveform in musical ways. NB that in some extreme settings, the sine sound can disappear entirely, but neither are these mods harmful nor will you be unable to find the “that’s my sweet sine” setting again.
Apologies for not providing any further updates/ or pending mod descriptions for this project any longer. I describe many mods here (more than the classic mods for the original), including mistakes and spelling glitches, and feel I rather want to spend time on other synth modding/development projects.