• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 48 subscribers
  • Views 655 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.

INA116 Unity gain bandwidth variation

Donald Le Roy
Prodigy 120 points
Other Parts Discussed in Thread: INA116

I am trying to use a pair of INA116 instrumentation amplifiers for a complex impedance measuring system at up to 25 kHz.  I need to determine the phase accuracy of the 2 amplifiers relative to each other at unity gain but the spec sheet only lists a typical gain bandwidth.

Is there a specification for the minimum unity gain bandwidth (lot to lot variation) so I can calculate the potential worst case phase mismatch at various frequencies (particularly the upper frequency requirement)?  I realize that greater than the specified typical gain bandwidth only make things better.

Thanks

Don

  • 0
    TI__Guru 64580 points

    Don,

    The bandwidth of op amps or instrumentation amplifiers like INA116 is set by the internal Miller compensation scheme and is a function of the gm of the input stage and Cc compensation capacitor; BW=gm/Cc.   Since the gm is a function of the current in the input stage (and thus inversely proportional to sheet resistance), any variation from the wafer fab process typical sheet resistance or capacitance (can be up to +/-15%) will change the typical bandwidth of the part (they are NOT correlated in any way).

    The worst case bandwidth variation occurs when the two parameters move in the opposite direction resulting in ~+/-30% maximum bandwidth variation from its typical value.  However, for the units that come from the same lot, or from different lots but manufactured within weeks of each other, matching of the bandwidths is quite good - within few percent - even though the bandwidths' absolute value may have shifted from its typical specification.

    In your calculation of the worst-case bandwidth you may need to take into account that the likelihood of the process shifting to its max or min value represents a six-sigma event and thus it is very unlikely for you to ever see it, let alone for the two fab sheet parameters to be at the opposite limit resulting in a minimum bandwidth (12-sigma event). 

    All in all, it is very unlikely that you will ever see INA116 unity-gain bandwidth lower than 600kHz or higher than 1MHz (+/-25% variation from its typical value) - a 5-sigma deviation or 1 out 1,744,277 units outside such limit.

     

  • 0 in reply to Marek Lis
    Prodigy 120 points

    Marek -

    Thanks for the detailed response.  It is certainly more than adequate for the question posed and I appreciate the detail.  This will certainly allow me to present the worst case phase error due to manufacturing variations to my project leader.

    Don