I'm making a simple op amp circuit to use it as a compactor.

I connect the non-inverting point to a simple circuit and the inverting to a 3 volt dc supply.

the circuit should compare the voltage at the non-inverting terminal with the 3 volt.

Actually the circuit is working very well small resistance. But, when I change the value of R1 and R2 to 500 M ohm the circuit goes crazyenter image description here .The voltage at the point between R1 and R2 is no longer 2 volt it changes to 2.4 volt

I know that the high value of the resistance made that.

What should I do to solve this problem and make the circuit works well at 500 M ohm.

  • \$\begingroup\$ What does "crazy" mean, and do you think it might be important which opamp you use? \$\endgroup\$ – PlasmaHH Oct 13 '18 at 11:36
  • \$\begingroup\$ Crazy >> the voltage at the point between R1 and R2 is no longer 2 volt it changes to 2.4 volt. opamp>> LM324 \$\endgroup\$ – Ahmad Sayed Abdulrahman Oct 13 '18 at 11:47
  • 1
    \$\begingroup\$ You do realize that with resistors of that value that even inconsequential things like a finger print on the PCB or on the part can make a difference to how it acts? Why do you need such extreme resistances anyway? \$\endgroup\$ – JRE Oct 13 '18 at 12:27
  • 2
    \$\begingroup\$ You should also note your multimeter has a 1 to 10 MegOhm input resistance and further perturbs your 500 Meg divider when you try to measure the divider output \$\endgroup\$ – sstobbe Oct 13 '18 at 15:12
  • \$\begingroup\$ I don't want to measure with multi-meter because of the problem you mentioned so I want to use the circuit above @sstobbe \$\endgroup\$ – Ahmad Sayed Abdulrahman Oct 13 '18 at 18:33

Look at the LM324 specification. The bias current is typically 20nA, and in your circuit that comes from your source. With a perfect opamp you would have 4nA flowing through your 1Gohm (total) potential divider, so the bias current upsets that somewhat. Do you really need such a high impedance source?

Such high impedances can lead to problems, particularly from leakage currents. Furthermore, if you must use this circuit configuration, you need an opamp with a maximum bias current of around 40pA - such as TL084 - and great care with cleanliness. My recommendation is to reduce the source impedance.

  • \$\begingroup\$ Isn't the 324 input stage with PNP and wants to output that bias current? In any case, 500Megs is far too much \$\endgroup\$ – PlasmaHH Oct 13 '18 at 15:14

If you can get it, use a comparator chip, not an op-amp. And again, if you can get it, use a FET-input chip.

Even so, I think you'll find that your circuit would be difficult to realize with 5M-ohm resistors -- 500M-ohm is more resistance than a resistor-sized insulator that you've rubbed with your bare fingers.

  • \$\begingroup\$ The problem is the high impedance is not optional as it comes from an external MEMS chip \$\endgroup\$ – Ahmad Sayed Abdulrahman Oct 13 '18 at 18:32
  • 1
    \$\begingroup\$ Ok. That is an important point, and belongs in the question. \$\endgroup\$ – JRE Oct 14 '18 at 10:18
  • \$\begingroup\$ @AhmadSayedAbdulrahman: There's a whole set of lore around making circuit work when you're dealing with such high impedances. Everything gets more difficult -- and I can't help you much, because I've never had to design a circuit for such an application. Even such things as how the board is cleaned and what materials you use will matter, where they don't for more normal impedances. Search on the ADI and TI websites, see if you can find some app notes. \$\endgroup\$ – TimWescott Oct 15 '18 at 3:56

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.