• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 366 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.

INA282: Measuring 8mV at output with REF's grounded and no current through shunt

Jeff Beran
Prodigy 10 points
Part Number: INA282
Other Parts Discussed in Thread: LM7705

Below is the circuit I'm trying to test. Without a load I'm measuring 8mV at the output. I've built 6 copies of this circuit and all behave the same. In 2 instances I measured 0V at the output, upon the initial power up, but as soon as I applied and removed the load the output would only measure 8mV with no load. If I place a load on any of the 6 circuits I will measure the calculated output for that particular load as long as the output exceeds 8mV. For example a 250Ω load would yield 55mV at the output.

  • 0
    TI__Mastermind 18780 points

    Dear user,

    Thanks for using E2E!

    The issue you are most likely having is that you are driving the device into the swing to GND limitation. Like all amplifiers, the INA282 cannot be truly driven to zero in single supply configuration, and a headroom exists that the output must stay above to remain in linear range:

    A potential solution you could try is to provide split supplies (+/-2.5V, etc.) with the REF pins referenced to GND so that the amplifier can swing in both directions around zero, but in single supply configuration, the lowest the amplifier will be able to drive is to 15mV typically. 

  • 0
    Guru 140200 points

    I see two issues:

    1. The INA282 (as any other amplifier) suffers from unavoidable input bias currents. Because of this the input bias currents should always see a current path to signal ground. This means that running the INA282 without any load is no good idea. At least provide a dummy load for the input bias currents.

    2. Referring to what Carolus said, another option could be using the LM7705.

    Kai