• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 1067 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.

INA226: Need a part recommendation for nA measurement

Roy Hsu
Genius 9350 points
Part Number: INA226
Other Parts Discussed in Thread: INA148, OPA189

Hi Team,

Could you please help find a suitable current sensor for the operating conditions below?

  • Vcm = 18-48V
  • Irange = 40 nA to 5 uA
  • Sensing needs to be done at high-side (Feel free to share the low-side solution too if there's any)
  • Analog or I2C output is fine

It's pretty hard to find one since it requires low input offset voltage and input bias current.

If you have any solution in mind, please let me know!

Best regards,

Roy Hsu

  • 0
    TI__Mastermind 26335 points
    Hello Roy,

    Thanks for reaching out on the forum. As you noted, parts with sufficiently low input bias current for your current range are hard to find. Unfortunately, all of our current sensing devices that can handle a 48V common mode have input bias currents in the uA range. Therefore I am going to pass your post over to someone in precision amplifiers. I think they may be able to help you find something that gets you closer to your goal.
  • 0
    TI__Mastermind 23050 points
    Hi Roy,

    Can you share a few more things about this application?

    1) What are the available supplies for the amplifier?

    2) Does this really need to be high side? Sensing such low currents in the presence of a high common-mode is a pretty difficult task. You need high input impedance, but you also need a device that can operate at high input voltages while maintaining accuracy. A low-side measurement would be significantly easier.

    3) Speaking of which, what is the accuracy required over the dynamic range?

    4) What is the maximum shunt resistance that can be tolerated?
  • 0 in reply to Zak Kaye
    TI__Genius 9350 points
    Hi Zak,

    Thanks for the response! Let me talk with the customer then get back to you.

    Best regards,
    Roy Hsu
  • 0 in reply to Zak Kaye
    TI__Genius 9350 points

    Hi Zak,

    Please find the response below.

    1. ±5V, ±15V, 24V or 48V.

    2. Although my customer said it has to be high-side, please feel free to suggest low-side solutions if you think they are beneficial.

    3. While an accuracy of 1nA is preferred, 10nA is also acceptable.

    4. 100 kohm.

    Thanks!

    Roy Hsu

  • 0 in reply to Roy Hsu
    TI__Mastermind 23050 points

    Hi Roy,

    I've spent some time going over this and unfortunately I do not have anything to offer you for a high side solution unless the customer has a power rail that is greater than the maximum common-mode voltage. The problem is that we don't have a part that can operate accurately at that high of a voltage. Ideally you would use an instrumentation amplifier, but with such a high common-mode this is not an option. The other option is to use a difference amplifier, but the best one we have for this, INA148, only has an input impedance of 1MOhm, which is not sufficient to sense nanoamps of current. The error this produces will shift with the common-mode voltage and this will dominate the output signal.

    If your customer is willing to sense low side, then this problem becomes quite easy to solve. Since the currents are so low, you shouldn't have much issue with ground bounce so you could use a simple non-inverting op-amp configuration as shown below. For this I assumed a 5V full scale range into an ADC. In this circuit with the OPA189 as an example, 1nA of input current translates to 1mV of output voltage. With a maximum offset of 3uV (0.4uV typical), this translates to 3mV of error at the output. With filtering as shown the total integrated noise is ~320uVrms , or 1.92mVpp. This means you effectively get accuracy down to 5nA worst case, assuming you have the necessary resolution on the converter.

    Low-Side current monitor_40nA to 5uA.TSC

  • 0 in reply to Zak Kaye
    TI__Genius 11135 points
    Roy

    We haven't heard back from you so we assume this answered your question. If you need more help, post a reply below, or create a new thread if this one has timed-out.

    Thanks
    Dennis