• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 50 subscribers
  • Views 267 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.

INA117: Tolerance of INA117 internal resistors

Ron Christian
Prodigy 10 points
Part Number: INA117

We are depending on an accurate solution to the INA117 scaling as described in TI sboa001 “Increasing INA117 Differential Input Range”. Essentially this is Vout = V3 – V2 + 19xV5 – 18xV1. The INA117A claims accuracy of better than 0.02%. In our initial development this was the case, and we were getting readings with <1% error. Our latest build has shown much greater error and we are coming to believe that the laser trimming does not include all of the internal resistors. We are looking for clarification on the specifications.

  • 0
    TI__Genius 10850 points

    Hi Ron,

    Thank you for your post, I can help you. The internal resistors of the INA117 are precisely matched to 0.005%. Are you using external resistors as well? If so, what are their tolerance level(s)? 

    -Tamara 

  • 0
    Guru 140200 points

    Hi Ron,

    we need a schematic of your circuit please.

    Kai

  • 0
    TI__Prodigy 715 points

    Hi Ron,

    If you're using the circuit shown in Figure 1 of sboa001, I believe I likely understand the problem you're facing.  While the circuit cleverly uses the INA117 to change the input voltage range, the engineer at the time did not consider one thing.  The INA117 precision relies on the fact that the difference in gain between the two inputs is nearly exactly 1.  Nominally, this is 19 and 18 as you highlighted.  However, the INA117 standalone will work just find if 18 and 17, 20 and 19, or even 19.8675309 and 18.8675309.  What the engineer did not consider is that when you add two resistors externally, the absolute magnitude of these resistors matter, not just the difference.  Another customer just recently found the same, and we recently confirmed that the material coming out of fabricated today is offset from 19 and 18.  We just decided to update this application note to highlight this.  What I can say is that if you calibrate those factors, they are precise laser trimmed resistors and will drift very little.  But unfortunately the app note falls short of highlighting the tolerance of the absolute magnitude of the internal matched resistors.  I apologize for this gap in information.  

    Thanks,

    Scott