• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 48 subscribers
  • Views 3876 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA111: Picoamp current measurement using instrumentation/operational amplifier

Sakthikumaran Panneerselvam
Prodigy 30 points
Part Number: INA111
Other Parts Discussed in Thread: INA331, INA818, INA121, INA332, , LMP7721, TINA-TI

Hello all,

I am in the process of developing a circuit to measure the device current, the minimum device current in the order of picoamp. The shunt resistor is fixed to 50 Ohm, and I don’t have any flexibility in changing the shunt resistor. The measurement is from the low side, making my common-mode voltage to be nearly zero.  The following are my requirements

  1. Low side current sensing
  2. Vcm=0V
  3. pA current measurement
  4. unidirectional current measurement
  5. analogue voltage output

Bandwidth and slew rate is not a constraint as I am interested only in the steady-state results.

The instrumentation amplifier should have a bias current of 10fA to measure 1pA current reliably with an accuracy of 99%. On searching the list of available instrumentation amplifiers meeting the above requirements, I find only offering from bias currents of the 20pA, making the minimum measurement limit to 2nA. The following are the top3 in my list

INA111, INA121, INA818 (with Vs up to 36V support). INA331/321 and INA332 were good too and are single rail supply designs. So with this, I have the following questions

  1. What are all the instrumentation amplifier choices for the above requirements?
  2. Instrumentation amplifier offers parasitic resistance rejections, which opamps don’t show, but specific opamps have lower input bias currents. So should I prefer opamps over ins-amps, or should I stick with ins-amps?
  3. Any other constructive suggestions or views I should incorporate?

Thanks

SK

  • 0
    TI__Guru 64950 points

    SK,

    To start, you cannot directly measure 1pA current with 50ohm shunt resistor as this would require you to be able to measure 50pV voltage (50*1e-12) - this is impossibe considering that the integrated noise of any IC across a reasonable bandwidth would be much higher.  Also, in order to measure 1pA current with an instrumentation amplifier (INA), as you pointed out, you would have to have INA with an input bias current of 10fA or less, which we do not have.  However, we have operational amplifiers like LMP7721 with the typical IB of +/-3fA, and maximum of +/-20fA, that could be used in the transimpedance configuration as shown below.

    Using RF of 20Mohm, the first stage converts 1pA current into VF1 of 20uV, which then gets gained up by gain of G=101 of the second stage into 20mV.  Assuming required range of the IG1 current of 1nA, this would result in the output of -2V for IG1=1nA - see below.  CF and C1 are used to filter out resistor thermal noise before the signal in fed into second stage. If needed, the circuit may be re-configured to give you a positve voltage. I used here +/-2.5V supplies for convenience.

    Attached please find the Tina-TI schematic for your own simulation.

    SK Low-Side current measurement - LMP7721.TSC

  • 0
    Guru 140200 points

    Hi Sk,

    I fully agree with Marek. If I'm understanding him correctly, you could open the ground connection of your 50R shunt and connect the ground side of 50R shunt to the input of his TIA, as shown below:

    marek_lmp7721.TSC

    And by using the good old chopping method (periodically turning-on and -off your your decive and by this turning-on and -off the device current to be measured), you can easily correct for any offset voltage error and even offset voltage drift error at the output of second OPAmp. You measuring signal would be the Vpp value of output voltage, or by other words 27.88mV - 25.81mV = 2.07mV in the above example.

    Kai

  • 0
    Prodigy 30 points

    @Marek Lis Thanks for your solution. I Will certainly look into the solutions carefully and will update soon.

    @Kai was thinking of the same configuration as you have mentioned as a backup option. Will simulate and study the 2 circuits closely, analyse them, and update the observations in the same thread. Thank you Kai

  • 0 in reply to Sakthikumaran Panneerselvam
    Prodigy 30 points

    Hi. I forgot to include the most important design requirement, the current to be measured in the range of 1pA to 60 μA. I mentioned (my fault) only the minimum limit for the setup in my design requirement descriptions as I was using the minimum current value for zeroing on the amplifier (since the minimum current and input bias are almost the same for many amp) and forgot the max limit.

    On studying the circuits provided, stage 1 is designed as a trans-impedance amplifier, and I used this as my reference guide for designing/understanding. Following it, the second stage is an amplifier.

    So if the measurement current range is such a wide variation, Can I still use the same configuration?  Or I should go for a separate similar circuit for each range, say 1pA to 100nA and 101nA to 60 μA? I am now studying closely the circuits will update soon of my inferences on the design suitability.

    Thanks,

    SK

  • +1 in reply to Sakthikumaran Panneerselvam
    TI__Guru 64950 points

    SK,

    In order the maximimize the output voltage for IG1 input range of 1pA to 60pA, you would need to increase the gain of the first stage by 10x and gain of the second stage by 2x  - see below.