• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 49 subscribers
  • Views 2047 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.

Op amp gain vs. operating temperature

Matthew Dzikowski
Prodigy 10 points

Other Parts Discussed in Thread: OPA4354

I have an application for which I need a quad op amp that has almost perfectly flat gain across its operating temperature.  Unfortunately very few data sheets contain this information.  My operating conditions are 3.3V single side supply, gain bandwidth product does not need to be very high ( 1 MHz + ).  The best part I could find in my search is OPA4354AIPWT, however my tests with this part are still showing more gain drift with operating temperature than I had hoped.  I made sure to use +/- 10 ppm/C resistors in the op amp network, currently configured as a non-inverting amplifier.  If anyone has personal experience or knowledge of an op amp product line that has excellent gain vs. temperature performance, please let me know.

Regards,

Matthew Dzikowski

  • 0
    TI__Guru 74561 points

    Matthew,

    The close-loop gain of an amplifier is only a function of the accuracy of the external resistors used (initial accuracy=?, TC=+/-10ppm/C) and not op amp itself - the open-loop gain, AOL, of OPA4354 will only matter if your close-loop gain approaches the minimum AOL value of 50,000 (94dB).  However, you have to be careful how you measure the close-loop gain - to measure the gain properly, you must use two-point measurement (single reading would not calibrate out input voltage offset).

    Apply two different input voltages and measure two respective output values:  Gain=(Vout1-Vout2)/(Vin1-Vin2).  

    As you set the close-loop gain with two external resistors, one may be drifting +10ppm/C while the other may be drifting -10ppm/C thus changing the gain by maximum 20ppm/C - this would result in 2000ppm (~0.2%) gain error over 100 deg C change in temperature due to resistors TC drift.

  • 0 in reply to Marek Lis
    Intellectual 405 points

    Hi Matthew

    Instead of using separate resistors why not consider using a resistor network? The temperature tracking between the 2 resistors in a resistor network is generally much better than that achieved using individual resistors.

    regards

    Steve