Thursday 5 November 2020

Synthesizer Build part-36: DUAL VOLTAGE PROCESSOR.

This is the Fonik Buchla Style Dual Voltage Processor. A very useful module for altering Control Voltages with five different functions!  Offset, Attenuation, Inverter, Glide or Lag control and, if you follow the tip below the schematic image, it can also be a CV splitter. 
Now also with Eurorack compatible layout.
I wanted a Voltage Processor module in my synth for a long time and I was thinking of copying the ARP2600 VP, but that one is fairly limited in its options and more specialized specifically for the ARP2600 so when I saw this design I thought it would fit much better in my system. This module lets you alter the offset of a control voltage by 0 to +5V or -5 to +5V. It lets you attenuate and invert a control voltage by means of an Attenuverter and it has a Lag control that is a direct copy of the Lag control from the ARP2600, with a 1 MOhm potmeter and a 470nF capacitor (The ARP used a 100nF cap). This alters the slew rate of, for instance, a Squarewave and rounds off the corners turning it into a Sharkfin Wave. In fact it adds a 90° phase shift to the signal. 
If you want that control to behave more like a Glide control to smoothly go between different notes with a 1V/Octave signal then use a 1µF electrolithic capacitor. Try it and experiment. Maybe use a 100K potmeter instead of a 1M one.
Besides control voltages this module can also handle audio signals.

This module will work fine on either a dual 12V or a dual 15V powersupply so no problem for you Eurorack fanatics =).  In fact, there's a Eurorack friendly layout further down the article. One thing though, with a dual 12V supply normally the maximum offset would be 4 Volt instead of 5 Volt but I addressed that issue and fixed it by changing some resistor values.
I guess you could say that the Eurorack equivalent of this module would be something like the TipTop Audio MISO (Mix, Invert, Scale, Offset) which costs a 110 Euro's. 
The circuit is primarily meant for control voltages but it can handle audio signals just as well. Even at very high frequencies it won't distort the signal. With audio you can use the Lag control to turn a Triangle wave into a Sinewave although with less amplitude. It won't be perfect but it's possible. It can also turn a 0V to +10Vpp signal into a +5/-5V signal by adding a -5V DC Offset voltage to it. The other way around works too of course, turning +5/-5V into 0V to +10Vpp signal. Very useful.
The circuit consists mostly of 47K resistors but you can actually alter the value of those and use for instance all 91K resistors. I actually did this as a test with the second part of this dual module and it didn't change the working of the circuit in any way. Just make sure you use the same value for all 7 resistors. Don't make them lower than 47K though. You can also use other quad-opamp chips instead of the TL084. You can use TL074, LM324 or any other, as long as it's a low noise opamp (good for use with audio, which most of them are these days) and they have the same pin-out as the TL084 (and most quad opamps also have that these days).
This circuit was designed by Chris MacDonald and modified by Peter Grenader and then further improved by Matthias Herrmann who added the Lag (Glide) control function. The only thing I did was adding the Offset switch, changing the potmeter values from 50K to 100K, changing the value of the Lag Capacitor from 1µF to 470nF and adding the 470 Ohm resistor before the Lag potmeter to eliminate noise issues, based on practical testing.
The original schematic and a PCB design can be found in this original PDF and I made a new drawing from that schematic which is posted below. Like I just mentioned, they use 50K panel potmeters in the schematic but I didn't have those so I used 100K potmeters. Again, this made no difference what so ever. You must however use a 1 MegaOhm potmeter for the Lag control because this, together with the capacitor, forms a simple lowpass filter and these values are important to get the correct frequency response. The original schematic uses a 1µF capacitor for the Lag control but with testing I found out that this is way too much. So I changed it for a 150nF in the layout but that turned out to be not quite enough. (The original ARP2600 Lag control uses a 100nF capacitor.) In my own build I experimented with different values and I ended up using a 270nF and a 180nF in parallel to make a total of 450nF and that works fine. So I set the capacitor value on the layout to 470nF. I found that this gives the best Lag control response in my case. Of course, if you don't have a cap of that value available, you can use an other one with a value close by. Anything between 300nF and 700nF will work fine and you can put two (or more) in parallel to create the value you want but test it and check to see it works like you want it to. Use an oscilloscope set to DC mode for measuring the output.
If you don't want a Lag control but a 'Glide' control you can use a 1µF electrolithic cap. My advise again is to experiment and use whatever suits your needs.

Setting the trimmer (T1):
The trim potmeters are for setting the attenuverter mid point, but they don't have too much of an impact so you don't have to use multiturn potmeters for those. The normal ones will do fine.  I added a switch to the offset control so you now have a choise to offset a control voltage from 0 to +5V or from -5 to +5V. 
Here's the procedure for setting the T1 trimmer:
- Turn all potmeters to the fully counter-clockwise position.
- Turn the attenuverter potmeter to the 12 o'clock position (mid point).
- set the offset switch to 0 to +5V position.
Connect an oscilloscope or multimeter to the output and turn the trimmer until the output reads exactly zero volts when the attenuverter is at the 12 o'clock position.

A little quirck I found, at least in my build, is that there can be a lot of noise on the output if the Lag potmeters are set fully closed (counter clockwise). Because this was the case with both sides of the Dual Processor I figured this was a fault in the circuit design so I added a 470 Ohm resistor between the Lag potmeter and R6. The value is low enough not to influence the Lag filter and it gets rid of all noise issues that I had.
The schematic drawing doesn't include any de-coupling capacitors but they are included in the layout. Just four 100nF ceramic caps on the power rails as close to the chips as possible. If you experience hum on the audio output you could even put some 10µF to 47µF electrolytic capacitors on the power rails. There's room enough left for that. Make sure they are rated 25V or higher and put one on the +15V to ground (negative pole to ground) and one on the ground to -15V (negative pole to -15V) rails. I leave that up to you but for my module it wasn't necessary to include them. (The electrolytic capacitors are not included in the layout, only the de-coupling caps.)

Here's the schematic drawing which I re-made from the original, from the above linked PDF file. The Dual Voltage Processor consists of two of these circuits side by side with only the Ground as a common link:

I made a Falstad simulation of this circuit. Some of the component values have been alterred a little to make the simulation act more like the real thing. The Lag potmeter value changed from 1M to 100K, the cap from 470nF to 100nF and the 470 Ohm resistor has been taken out.

Here is the verified stripboard layout I made for it. It's the same layout once repeated and mirrored to make it a dual module. This layout was made for the Kosmo format modules but now there's also a more compact layout below for Eurorack size modules.
TIP: Solder a wire from the input-socket of stage one to the socket switch of the input socket of stage two. That way the signal on input one will be present on both inputs and can be processed by both stages and split in two. If you connect a patch cable to the input of stage two, that first connection will be broken by the socket switch and it's back to normal. (Normally we call this normalling a connection ;) Very useful me thinks! 
Of course you need input sockets with built in switches for this but most types have that as standard.

Stripboard only. Beware that some stripboards are sold with 56 instead of 55 holes horizontally. The layout is 55 holes wide. 

Bill of Materials:

As of December 2021 there is now also a new layout for the Eurorack format. The stripboard is 24 by 41 holes. Just like in the Kosmo format layout above, the right part of the dual voltage processor is a mirror image of the left part so I could place all the connections to the potmeters on the edge of the stripboard on both sides. The TL084 has 4 identical opamps in it so it doesn't matter which opamp is used for which part of the circuit. As I mentioned earlier, you can use other quad opamps like the TL074 or LM324 for this without problem.
Here is the wiring diagram:

Stripboard only view:

I built this version on Dec 10th 2021 and everything worked fine except that I had to use a lower value capacitor for the Lag control. The layout has a 470nF cap in it and that works fine in my Kosmo format panel but for this one I had to use a 10nF cap. Not sure why this one is different, maybe it's the fact that this runs on +/-12V instead of +/-15V or it's the potmeters I used I don't know, it's a bit of a mystery. Anyway, it's not important because if you find, when testing, that the Lag potmeter doesn't work over the full throw then you need to lower the capacitance. Just a matter of experimenting. The cap can easily be de-soldered and changed for another one.
The trimpotmeters are for setting the Attenuverter midpoint and as with the previous layout they don't have much influence but you need to set the Attenuverter so that at the midpoint, when the waveform is a flat line, that line is at the zero volt mark. Measure this with a scope and make sure all the other potmeters are turned fully counter clockwise and the offset switch is set to 0/+5V.
Btw, because this module is running on +/-12V the actual offset voltage is plus or minus 4 Volts, not 5 Volts but I only discovered that after I made the panel so I kept the labeling as is.

EDIT: To get a higher offset voltage in this Eurorack version you must change the resistor R6 from a 47K to a 68K. This will actually give you +/-6V offset voltage. R6 on the lefthand side of the stripboard is the 47K between pins 13 and 14 of IC-1 (positions I and J-16). On the righthand side it is the 47K between pins 1 and 2 of IC-2 (positions I and J-27). Change those for 68K's and your offset voltage will be upto +/-6V. If you need +/-5V then use a 56K with a 1K2 in series. You may have to do some tweaking if you want the voltage to be just right for you. You could also use a 100 trimpot to dial it in accurately. 

The 470 Ohm resistor(s) are not in the original circuit schematic. I put those in myself when I built the first version because the Lag control produced some noise when the potmeter was fully counter-clockwise. This resistor sort of prevents that the Lag pot can be fully closed. It has no negative influence on the amplitude or sharpness of the signal so it works fine.

Here's a picture of the finished Eurorack module:

Here's a video with a quick overview of the different functions. 

I watched a demonstration video about the ARP Odyssey and in it they showed the effect that the Odyssey's Lag control had on the filter cut-off control voltage. It made the filter make these 'Wah' sounds. And I'm very chuffed to see that the Lag control in this module has the precise same effect on a filter.

Here are some pictures from the build process:

In the picture of the panel (below) the 'Lag' control is still called 'Glide'. That's what it's called on the schematic but I chose to use the same term that ARP uses in the 2600.  I think it's a more accurate description because it actually creates a phase shift of about 90 degrees (see also the article about the ARP Envelope Follower). So that makes the signal lag behind the original in a small way. 

The picture below shows one side of the dual module wired up and the other side has not yet been wired up. The LEDs of that side are still mounted on the print (which was necessary for testing) instead of in the panel.

When the module is in 'rest' position so to speak, all potmeters should be set fully counter clockwise and the switch set to 0/+5V. That way, any signal you put in will come out unchanged. You can then alter it by turning the controls.

Okay that's an other one done. If you have any questions please put them in the comments below or on the EddyBergman Facebook group. Please read the whole article before asking questions.

If you find these projects helpful and would like to support the website and its upkeep then you can buy me a Coffee. There's a button for that underneath the menu if you're on a PC or Mac. Or you can use this PayPal.Me link to donate directly. All donations go towards the website and projects. Thank you!

DISCLAIMER: The author of this article does not accept any responsability for the correct functioning of this, and any other, module/project on this website. What you build, you build at your own risk. All project layouts are thoroughly tested before publication, it's up to you to replicate them and the author can not be held responsable for any mistakes made.


  1. i just absolutely love your site. It's very user-friendly.

  2. hi eddy, thanks a lot for all the layouts and background history. it's amazing!

    does it have to be TL084 or will TL074 work just as fine ?

    1. Hi Albert! Thanks, I'm glad you like the website so much! I really appreciate it. As for the TL084. A TL074 will do just as well. In fact any low noise quad opamp with the same pin-out will be fine like an LM324 etc.

    2. I made a note of the fact you can use other opamps in the article so people can instantly see it. So thanks for bringing it to my attention Albert!

  3. amazing, thanks a lot and thanks for replying so fast!(:

    1. Hi Eddy! Your site is awesome! I have a quick question. I'm definitely a beginner here... is it okay to use a polarized cap for the 10uf caps on the eurorack layout?


Note: comments are moderated and do not appear straightaway. Your first comment is not allowed to contain any links.