• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 196 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

INA592: Handheld nanoamp/nA current source

Tom Larter
Prodigy 10 points
Part Number: INA592

Tool/software:

Hello,

I'm working on modifying a 9V battery-powered handheld DC current source that needs to be able to source +1nA-500uA with <1% error. Temperature range is from 65-75 degF (18-24 degC). The circuit is similar to the buffered Howland current pump shown in Figure 9 of TI AN-1515; see below for actual circuit configured for 1nA. An INA592 was used to provide a precision resistor network for the Howland core and a gain <1, allowing for smaller Rs values. A trimmer is integrated into Rs for hand-tuning.

A few questions about this method: 

Do I need to upgrade either of the ICs to work in the nA range? I figure an IC similar to the INA592 but with lower gain could allow for an even lower Rs value and get away from temperature-sensitive GΩ resistors. Perhaps using resistance multiplication like that seen in Figure 11 of TI AN-241 could get around this problem too.

What opamp parameters should I optimize to get best DC accuracy and stability? Also, how can I compute the range of load impedance compliance? Are guard ring PCB techniques necessary for 1nA?

Thanks for your help.

  • 0
    TI__Mastermind 33011 points

    Hey Thomas,

    Unfortunately we don't have a lower gain version of the INA592. For the 1nA range, your load regulation will be dictated by two things, your diff amp swing from rail, and your gain network matching. I've made a circuit in PSpice for TI (so we can run some MC sims) with ideal amplifiers with limited swing from rail. With the current supplies, and using the max swing from rail limits of the INA592 (220mV from either rail) and using ideal resistors, your load regulation is about 4.6pA

    Starting to add in realistic internal resistor tolerances of 0.01% and zooming in a fair bit, we see a fairly minimal change in load regulation.The limiting factor here will likely be your op amp output swing, rather than your internal matching.

    As far as creating an equivalent impedance using a T network, I will run some more sims tomorrow and see if I can get anything useful.

    In the meantime, here is the zip file from my simulation.

    Ideal_Howland_Pump.zip

    Best,
    Gerasimos

  • 0 in reply to Gerasimos Madalvanos
    TI__Mastermind 33011 points

    Hey Thomas,

    Sorry for the delay here.

    Do you need bidirectional current? If not, you may be able to use a different current source topology.

    Best,
    Gerasimos

  • 0 in reply to Gerasimos Madalvanos
    Prodigy 10 points

    Hi Gerasimos,

    Thanks for your analysis! I only need to source (positive) current to grounded loads. If I understand correctly your load-pull technique suggests I can drive loads from 3.53GΩ (4.78V/999.8pA-1.25GΩ) down to 0Ω just fine. Please correct me if I'm wrong.

    I'm open to other topologies. I think it may be best to have a dedicated 1nA circuit/output and then continue using the original 5u-500uA circuit on its own output.

    Thanks!

    Tom

  • 0 in reply to Tom Larter
    TI__Mastermind 33011 points

    Hey Tom,

    This circuit can drive resistive grounded loads (over the 200 MC cases) of about 3.3GΩ to 0ohms, yes.

    Running into some licensing issues with simulating right now, so I can't run PSpice simulations to give a better answer, but I will update this thread when they are resolved.

    Best,
    Gerasimos