I'm designing a voltage measurement device and I i'm testing it and the front end voltage divider and it has a much lower voltage than expected. Circuit is below.


simulate this circuit – Schematic created using CircuitLab

So I put in 5V (real world) input across R1 (10Meg) and R2 (10Meg) I only see 384.6mV between them?? I expect to see 2.5V. What have I done wrong and how do I fix it?

p.s dont simulate the schematic its just representation


Thank you for the input, I have found the issue and its a novice mistake which took me awhile to find. I've updated the schematic. See im using R2 and R9 to attenuate the input voltage, but I did not take into account the OFF resistance of M1 and M2, which I have measured to be 0.9Mohm, so obviously most of my voltage is being dropped in R9 and M2.

  • \$\begingroup\$ Why are R1 and R2 so huge? \$\endgroup\$ – AnalogKid Jan 7 '18 at 5:29
  • \$\begingroup\$ I beg you to mend your apostrophe key. \$\endgroup\$ – DiBosco Jan 8 '18 at 9:34

The opamp has maximum Ibias of 4 nanoAmps.

What Rin does your Digital Volt Meter have?

  • \$\begingroup\$ Is this an answer or a comment? \$\endgroup\$ – Trevor_G Jan 7 '18 at 15:00

Since you are using huge resistors I also suspect this is a meter resistance issue.

Try measuring the voltage across R1 instead of R2. I'll wager you still get about 385mV with the meter resistance in parallel with R1 instead of R2.

Then measure the voltage on the left of R4. I suspect it will be closer to 2.5V on the output of the op-amp.

  • \$\begingroup\$ Yes, I never thought of this. As im calibrating the ADC to read the exact voltage on top of R2 which should be the input voltage/2. Im using the DMM as my reference tool as the voltage as the voltage wont be exact (resistance tolerances). Difficult to measure because as soon as I hook my DMM across the voltage will drop (10meg + 10meg in parallel), interesting problem. Thanks for bringing this up, would not of foreseen. \$\endgroup\$ – user160063 Jan 8 '18 at 2:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy