Monday 11 March 2024

Synthesizer Build part 55: CROSSFADER with VACTROL CV CONTROL.

This circuit lets you fade between two inputs manually or with a Control Voltage by means of two Vactrols. Project fit for Eurorack or Kosmo modulars. 

This was an other project by request and since I haven't built anything like it yet, it seemed like an ideal project for the website, so I decided to look for a good crossfader schematic. 
I came across a schematic from " ©Astro / SYNTHFOX 2020" on which it said 'okay to redocument when properly credited', so I hope this is okay.
I built this project for Eurorack but of course it can just as easily be built for a Kosmo sized synthesizer. In fact with a Kosmo module you have more space to put in more stages so that would be even better.
This project works on both -12/+12V and on -15/+15V without needing any changes.

Here's the schematic I used:

It's a very simple circuit. You have a DC voltage that can swing between the voltagerails coming into the inverting input of an opamp by means of the 'Init' potmeter (Initial Voltage) and a second connection in the shape of a Control Voltage input with a level control, summed to the same input on the opamp. The output of this opamp controls the lighting up of the LEDs in the Vactrols. One Vactrol does the negative phase and one does the positive phase, or one does input A and the other input B. Whichever you prefer. 
The LDR's (Light Dependent Resistors) of the Vactrols are each connected to one input source and the other sides are connected together and go into the second opamp which is simply an output buffer. That's all there is to it.
The CV input comes in via a lower resistance than the Initial Voltage, 22K instead of 100k. This will, in combination with the 100K feedback resistor R4, give the CV input an extra gain. We can calculate that gain with the formula: Gain = (-Rfeedback / Rin) which is -100/22=-4.45 The minus symbol simply means that the output voltage is inverted. 
This is done (I think) to give lower control voltages enough effect to influence the vactrols because a CV voltage can never be as high as the Initial Voltage which covers the full powerrails potential. So if your CV input is not effective enough you can try lowering R2 and so get even more gain, but don't overdo it. Measure first before you start altering things. Make sure this change is actually needed. (I left it as it was myself).

I built one Crossfader stage as seen on the schematic but it would be easy enough to build a few of these so you can mix together more inputs. The circuit, as you can see, only has a few parts so it doesn't take up much room.

With the Init potmeter (Initial Voltage) you can manually fade between the two inputs. This works as long as there's no CV voltage present. When you input a CV voltage that will then take over control of the crossfading the higher you set the CV Level and you can steer the output signal more to one or the other input with the Initial Voltage control potmeter.
   I made the Vactrols myself, in fact I used Vactrols I had already made for the LowPass Gate project, so I don't know exactly which LDR's I used to make them with but I used red 5mm LEDs in them. It is preferable to use red LEDs because they have the lowest voltage drop (about 1,5V) but with the extra gain the CV input gets I don't think this is very important. LDR's do react well to red light. The LDR's I used have a resistance of about 300Ω when the Vactrol is switch fully on. I think it is best to use quick reacting LDRs in the Vactrols so it can handle a wide range of LFO frequencies on the CV input.
   The test results I got after finishing building this project showed that, when using the CV input I still got some bleed through of one signal onto another. In other words the separation of inputs A and B was not perfect when the CV voltage was on full positive or negative voltage. With just the manual control (Initial Voltage pot) I did get a good separation between the signals when the potmeter was fully clockwise or counter-clockwise. Of course the total voltage of the CV input could be the problem here. If it is not as high as the voltage produced by the Init. potmeter then it won't have as much influence. That is why the CV stage has extra gain on it to boost the signal.
So please don't expect perfection from this analog Vactrol design. It's old school and that's what I like about it.

Here are the layouts I made for this project. As ever they are verified. I used them for my build.

Stripboard only:

Cuts and wirebridges. As ever, mark the cuts with a Sharpie. Stick a pin through the marked holes and mark them again on the copper side. Then cut the strips where marked with a sharp hand held 6 or 7mm drill bit.

Here's the Bill of Materials. If you make your own Vactrols then refer back to the Lopass Gate project to get more info on the LDRs I used there. There's a complete paragraph discussing the Vactrols in that article. I did not include any bypass caps in the BOM so if you want them, order two 100nF ceramic caps and solder them from positive to ground and from ground to negative power at the top of the stripboard.

Here are some pictures of the finished module. Mine is 6hp wide (3CM) and about 5,5CM deep which is a lot but it will fit in a Nifty Case, no problem.

The two striped components on the top near the power header are just 100nF bypass caps. You can include them of leave them out. They're not really necessary and I left them out of the layouts and bill of materials. 
In front of the vactrol you can see a little pin header. There's also one on the other side that you can't see. I soldered those in for test purposes. I could connect my multimeter probes to it to measure the resistance of the LDRs in the vactrols.

Here are some screenshots from the oscilloscope. In the three images below the yellow line is the output signal, the blue is input A and the purple is input B.
The first image represents the crossfader with the Initial Voltage potmeter fully counter-clockwise so only the squarewave is let through to the output. There is no CV input present in any of these screenshots.

In the second image the Init potmeter is in the middle position and we get a mix of both inputs A and B.
As you can see here with the blue and purple traces some of the signal is fed back on to the input signals, changing their character a bit but this is easily explained. When both vactrols are fully active the output and both inputs are separated from eachother by only about 600Ω, the series resistance of the vactrol LDRs, so it's obvious some of the input A signal will find its way to input B and vice versa. This is not a problem however and it can not damage your VCO's. Exactly the same happens when you mix signals with a passive multiple for instance, so nothing to worry about.

And in this third image the Init. potmeter is set fully clockwise so only the Triangle wave is let through.

If you connect a sinewave LFO to the CV input and turn the level up, then the result will be a cycling through these three stages. By turning the Initial Voltage potmeter away from the middle position you can emphasize one of the input signals.
The signal that is connected to the CV input needs to be a bi-polar signal, so a signal that has a negative and a positive voltage phase otherwise it will only fade one of the inputs. If you want to use a uni-polar signal, you could try using a capacitor on the CV input socket to turn a uni-polar signal into a bi-polar signal but I haven't tried that myself.

Here is a little test video I made while I was testing the circuit just after finishing it. You hear me discussing a CV leakage problem, that when the CV Level is fully closed it still had some influence, but that turned out to be a grounding issue, as I thought it must be, and once I put the stripboard behind an aluminium faceplate and wired everything up that issue was no longer there. So problem solved.

Okay, that's it for this project. A simple one just like the previous last few projects because I didn't have time to do really big projects in the last few months. However I am working on some new and bigger projects so watch this space :)

If you have any questions about this project please comment below or post your question in the special Facebook Group for this website.


  1. Hi Eddie, I also found that Synthfox schematic, but your stripboard layouts make *much* more sense to me. One question, I have a few spare VTL5C3/2 vactrols, one LED, but two Opto-isolators with a three pin output on that side (the two "middle" pins are tied together). Could those be swapped in here, and if so...any tips? I'm an enthusiastic builder but circuit design is not my lane. Thanks.

    1. If you have a few of them you should be able to use them. You need two separate vactrols, each with its own LED otherwise you can't pan from one side to the other.


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