Post Go back to editing

High Input Impedance op amp

Dear all,

I am measuring the resistance of a large resistor (up to several Mohms) using 4-wire terminal. The resistance is so large that it surpasses the input impedance of my voltmeter (100MOhm). So I am looking for an differential operational amplifier that offers a large input impedance (>2GOhm), that could take input spans several microvolts up to 0.5 volt.

Could you offer me some products that fits into our application?

  • Hi Qian,

    Apologies for the delay of response. We have a lot of amplifiers that has a very large input impedance in our portfolio. To see this, you can go to this link - click the choose parameter and Enable the (Rin  Typical) on the filter options by checking the box. You can also arrange the amplifiers from highest to lowest Rin by clicking the arrows beside the Rin parameter.

    These are the amplifiers that you may want to consider are ADA4505, AD8661 or AD8603.

    Let me know if you need anything else.

    Best regards,

    Emman

  • Hello Emman,

    Thank you for your answer. To have the link to check the input resistance is helpful for me to find the one might fit our use the best.

  • You will also need to consider the effect of bias currents in the opamps (and offsets in these) , if you can reverse the polarity of the current source, then averaging forward and reverse resistance values will remove these effects.

    You may also need to consider

    (a) using teflon wiring

    (b) using teflon PCB's

    (c) using driven guards around the input signals

    (d) careful washing of PCB to remove flux residue, and conformal coating of PCB

    The above measures help with the effects of (surface) leakage currents.

    Traditionally electrometer front ends used a flying capacitor , and a servo-ed voltage supply. The servo adjusts the voltage supply until there is no capacitor current when it is switched from (your resistor) to (voltage supply) .  This circuit produces a single sided voltage that can be easily measured in the presence of large CM noise.

    An EDN article by guru Jim Williams is interesting reading, and has a flying capacitor bridge in fig 26.11 of http://www.edn.com/Home/PrintView?contentItemId=4390838   The advent of high performance AD instrumentation amplifiers has pretty much relegated the flying capacitor approach to the realm of the dodo. I think HP (Agilent) used to make a hybrid flying capacitor module.

    Google " KeithleyLowLevelHandbook_7Ed.pdf  " for a useful handbook on low level measurements, it describes guarding.

    There's some interesting reading about error terms scattered through here http://www.analog.com/media/en/technical-documentation/technical-articles/489168400sscsect3.PDF

  • Here's a tutorial on chopper stabilised opamps :

    http://www.analog.com/media/en/training-seminars/tutorials/MT-055.pdf

    Particularly note figure1 ,  if you were to hang a few gigaohms on Vin , there would be no bias current flowing into the source (as there is no DC connection).  To adapt this circuit to your 4wire measurement, connect capacitors to ground from the top of your gigohm resistor, and the bottom. Move node Z on the input switch from ground  to the bottom of your gigohm resistor, and Vin to the top, (R1 and R2 can be zero ohms)

    With this mod,  the peak to peak voltage at the amp input is exactly equal to the DC voltage across the gigohm resistor.

    This arrangement is open loop, so you will get some second order error terms sneaking in. (so you will measure ~99.9% of the actual voltage)

    I can't find any diagrams on the web to show how to add the servo function, but you basically have the capacitor measuring the difference between the input (from the gigohm resistor) and the output from your amplifier, so current only flows briefly when something changes, otherwise the loop is balanced.

    I was going to suggest googling " autobalance AC bridge "  but it seems some clueless person has used "autobalance" to describe a mundane virtual earth transresistance opamp circuit.

    Hmm maybe something like  the figure in this https://www.google.com/patents/US3454850?dq=servo+ac+bridge+chart+recorder+chopper&hl=en&sa=X&ei=GBVoVf-PIoSI8QX8h4GgAg&…

    (but use another chopper instead of the motor and pot, and you get the whole idea of the motor adjusting the pot, until E1 and E2 are at the same potential)

  • Here's an interesting read on impedance measurement 

    http://cp.literature.agilent.com/litweb/pdf/5950-3000.pdf

    140 pages from the accumulated wisdom of HP.

    Note that with your gigohm resistances,  you will see appreciable effects from stray capacitances, and induced potentials from movements of wires within electric fields (i.e. electret microphonic effects), piezo-and tribo-electric effects of cables rubbing, and people waving their hands around. So gaurding and shielding are important.

  • Hello Bobsbots,

    The information you just gave is very helpful. I will go through the documents and try your suggestions on the measurement.

    Thank you again for the detailed answers.