Op-amp inputs max ratings on Branches, Stanges, etc

Hi there. I’m trying learning how to properly implement a CV input with offset knob in digital modules, so I looked at Branches schematics:


The power supply for the op-amp is 0-5V, if I’m not wrong. This means that the output to the MCU is correctly limited inside the supply range, but the op-amp input (pin 2) is not. For example, with the knob fully clockwise, a voltage near 5V will saturate the op-amp (which would be no more in linear region), and then near 10V the inverting input will be outside the power supply range. This shouldn’t be fine, right?

A similar schematic is seen on Stages, for example, where the power supply is even smaller.

What am I missing here? Thanks.

I don’t follow your analysis of the Branches circuit. When R15 is fully clockwise (ie, the left side of R7 is grounded) and a CV of 10V is applied, the op-amp output will saturate at 0V (it cannot go to -10V). Thus, V- will be at 1/3 of Vin, still within the rails.

Anyway, I would recommend the more modern technique of separating pots and CV inputs on different MCU ADC pins - it’s much easier then to introduce special control curves in software, alternate functions, etc. for the knobs… and the offset and clipping is just a little bit of code.

Regarding Stages’ CV inputs, what is the problem?

Even when the output saturates, the input pin does not move very far from ground.

Also, remember that the CV is never directly exposed to the op-amp pins, there’s always this 100k resistor, so the currents flowing are in the 100µA range. In the (extreme) scenario when the theory would predict that an input pin would fall outside of the rails, the currents at stakes are so low that the ESD clamping diodes present on most op-amp’s inputs do their job.


My bad, my observations came from testing on a breadboard, where I messed up with the knob wiring.

So, you suggest using inverting op-amps with a negative offset (as if the Branches knob is fixed at -5V) for bipolar CV inputs, and then simple resistor voltage dividers for knobs (like I see in Stages), right?

Thank you very much for your help.

Yes. I don’t think of it as a “negative offset”, just what is necessary to map the wanted input range (bipolar or unipolar) to 0V…3.3V.

Yes, wiper of the pot to the ADC and to a 100nF cap, one end to 0V, the other to VCC.