I need to build voltage divider to measure the battery voltage (7.3V) using max 3.3V ADC. As the device most of the time spends in the sleep mode (<1uA), the current drawn from the battery should be very low (best <70nA) so the resistance of the voltage divider has to be > 100MOhm.

I know that I need to buffer this voltage but I struggle to design the voltage follower working with such a high impedance source.

My question is: will JFET input opamp (as I know they have high input impedance) will be good for this application?

Maybe there is another method of lowering the voltage which will draw such a small current form the battery.

  • 3
    \$\begingroup\$ Does the voltage divider have to be active all the time? If not, you could disconnect it (with a P-FET) when it's not needed. \$\endgroup\$ May 18, 2021 at 0:46
  • \$\begingroup\$ while it would be possible to take an opamp with pA input current, you would also need to make a resistor divider with many GOhms, which is a bad idea. Better do as others posted: take a cheap generic opamp with 1MOhm resistor divider and keep it disconnected with a pFet \$\endgroup\$
    – tobalt
    May 18, 2021 at 6:50
  • \$\begingroup\$ @JonathanS. Actually it is the best idea. Thank you very much \$\endgroup\$ May 18, 2021 at 9:44

2 Answers 2


You won’t find an Op Amp that meets your spec, so don’t try. Just enable by switching the low side with a PN2222A or an Nch . High side PNP driven by an NPN or a Pch FET with similar driver to a 1M divider when active. Calibrate it with a 10M load R or whatever is offered. Add an RF cap to suppress noise.

This can be done with <20 nA searching for the right low leakage Transistors and kept at room temp or lower in the off state.

  • \$\begingroup\$ If the divider is disconnected from ground, there could still be a parasitic path between the battery voltage and the ESD diodes of the ADC input, through the top divider in the resistor. It may be better to switch out the divider from the high side. \$\endgroup\$
    – user4574
    May 18, 2021 at 1:50
  • \$\begingroup\$ Agreed. @user4574 \$\endgroup\$ May 18, 2021 at 2:56
  • \$\begingroup\$ Yes - I will go that way. \$\endgroup\$ May 18, 2021 at 9:45

Given these two specs...

I need to build voltage divider to measure the battery voltage (7.3V) using max 3.3V ADC


the current drawn from the battery should be very low (best <70nA)

You didn't really say what level of accuracy you need, or what the operating temperature range is. If you want the op-amp to steal less than 1% of the divider current, then you would want the op-amp input bias current to be less than 700pA. There are plenty of CMOS type op-amps that have input bias currents down to 1pA at room temperature.

Just go on one of the big distributor websites like Digi-key or Mouser and search for op-amps that have low input bias current. They have parametric search tools that let you specify input bias current as a selection parameter.

You could also try manufacturers like Analog Devices or Texas Instruments. Both of them make op-amps that would be suitable.


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.