CV inputs seem to have voltage present when no patch cable attached

[depaffect]

This is a bit of a strange one I've noticed for a while

If I automate a parameter using the the CV Input (via the Automation), it acts as though a small amount of voltage is present even when something isn't patched into the CV Input.

Eg, I "override step value" with CV1, and even with nothing patched into CV1 the pattern will often return to the top before finishing, and sometimes get stuck around step 1 for little periods of time.

I've had my Nerdseq in multiple eurorack cases where this happens, so it's not a dodgy power supply or anything.

This happens with all 4 CV inputs. 

If the CV Input is automating the "CV" value (ie an oscillator's pitch), it's not noticable

I wonder if there's never *truly* 0V in the system, so maybe it's responding to tiny little bits of residual voltage or something? Perhaps there could be some sort of threshold so that it ignores anything too close to 0V?

Video to demonstrate : https://youtu.be/dVTIJXakstU (happens at start and around 30 sec mark)

[XORadmin]

The override step value function is experimental only and that for a reason ;-)

Also the CV input -> Automators are not filtered at all and so the tiniest changing will affect the destination. For example, if the ananlogue input is between exactly 2 values, it might change from one to the other all the time because of the common noise floor.
On a future version i will add more functions with the CV input -> Automators so you can change the level of filtering, quantisation, range etc.
Not sure which version you are using, but also with the latest release candidate the CV input got a new calibraton option. More to come.

[depaffect]

Ah! I see. Excellent!