Hi guys,

I’m working on an attenuvertor to include on a new module and used the one in Ripples as inspiration. I was just wondering a couple of things… is there something special about the use of a 10K pot? I can imagine it having to do with the susceptibility to noise interference because the current lines to both opamps are relatively longer (normally you keep the input resistor of an inverting op-amp very close to the inverting input for this reason).

Another thing I was wondering is whether the non-inverting input of the attenuvertors (first) opamp could still be used as a summing node for other signals. I simulated this a bit in spice, and found some interactions between both inputs in this case (the one going through the pot, and the one going in through a 100k input resistor). With a larger pot (100k) the interaction is smaller, but still there. I’m wondering whether this is fundamentally impossible/unwise to do…


  • I use 10k pots everywhere to reduce the number of references in stock.
  • You can use the second op-amp’s inverting input as a summing node.

Thanks! Do I understand correctly than that I cannot use the first opamp for summing?

Why would you want to do so?


  • if you have let’s say CV1 going to the attenuvertor pot (swapping it’s operating direction) and
  • if CV2 is summing at the first op amp
  • you feed a circuit that already start with an inverting op-amp (e.g. an expo convertor)

That would save you an opamp…

ah I see… I think there could be issues with clipping.

Hi Olivier,

Have you ever had problems with offsets in these attenuvertor circuits? I was suprprized to find a 100mV offset occuring in one instance where I used 10K pots and a TL072 opamp. Didn’t expect this to occur with a TL072.


I don’t remember the details, but yes, this sounds familiar to me. Problem went away by using a 50k pot.

Ah, glad to hear. In that case I’ll do another parts hunt. I like the Alps knob-shaft pots very much, but I have only seem them in 10K.