Adamus, I know that you have checked and checked, but maybe it's time for another set of eyes. There is simply something wrong in the components or board traces.
You voltage measurements on the opamp are:
1= 0
2=2.9
3=2.3
4=0
5=0
6=3.4
7=24
8=0
Pin 3 is the reference input and is the correct voltage. Pin 2 is the sample voltage from the plate divided down by the resistor ratio (1M/30k). It is more positive than pin 3. Since the opamp is inverting the output voltage of the opamp must go negative with respect to the reference. But the O/P, Pin 6, is at 3.4V.
It is almost as though the traces on your board are wired to make the opamp non-inverting. But, of course, this is impossible. So there has to be something else wrong. You say this is the right channel?
The only other possibility is that the opamp is oscillating in your particular setup. Do you have a scope?