• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 558 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.

INA821 or is there a better choice ?

Gerhard Kreuzer
Expert 2600 points
Part Number: INA821
Other Parts Discussed in Thread: ISO124, OPA2141, , INA188

Hi,

I have a DC voltage of +-120mV to monitore. I am searching for some INA with very low 1/f noise (we measure DC), a very low initial offset, but we can remove some offset by (software) calibration. The full scale output voltage should be +-12V, fed into a ISO224B. The source impedance is < 2k. Supply voltage is +-14V, very clean, made locally by low noise LDOs.

The input signal is switched on/off and changes polarity, so the amp should have sufficient step resonse, the ADC sampling rate is 15ksp.

A simulation gives 1.6µV output noise for 100 Milliherz, but how much rms noise will I see in a bandwith of 7.5kHz starting at DC?

The signal ends up in a 24Bit ADC with 15ksp.

Thanks for helping.

Current solution: A OPA2141 pick up the signal, g=100, passes it to 2 x ISO124 (which formed an isolated full diff amp). With this circuit we get -95dB SN when we analyze the output of the ADC.

The new circuit should be simpler, cheaper and should have a better SN (we hope).

With best regards

Gerhard

  • 0
    TI__Guru 54111 points

    Hi Gerhard,

    INA821 seems to be a good choice for the application. Please also take a look at INA188 with zero-drift feature, since you mentioned that you are able to remove unwanted voltage offset via SW. INA188 will eliminate the 1/f noise concern, which is better than INA821. 

    Here is voltage noise spectral density vs. frequency comparisons:

     

    The link below is our precision op amp search tools, please take a look for other selections. 

    https://www.ti.com/amplifier-circuit/instrumentation/products.html?pqs=paqs&familyid=500#p480=1&sort=p770typ;asc

    If you need further assistant, please let us know. 

    BTW, INA188 Tina pspice model consists of noise models, and we are able to simulate the total noise in rms for a given BW once you finalize a circuit. 

    Best,

    Raymond

  • 0 in reply to Raymond Zhang1
    Expert 2600 points

    Hi Raymond,

    yes, but the slew rate of the INA188 is not very high, so the switch off event we have to record will not be very precise .. We sample with 15kSp, so we have 66.6666µs between samples ... The INA821 is much faster, probably too fast ==> ok, maybe there is one type with a slew rate/GBW little mor higher than INA188 and good noise figure?

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 54111 points

    Hi Gerhard,

    With 15 ksps, which is sampled at 66.7 usec/sample. 

    Also, you mentioned that " I have a DC voltage of +-120mV to monitore". If this is the case, the step voltage will be <240mV across the sensing resistor. with G=100 and Slew rate of 0.17V/usec, it will take .240V/0.17V/usec = 1.412 usec to complete from switch transition in the worst case. 

    Yes, 0.17V/usec is not high, but it seems that it is adequate for the application. If this is not good enough, I will find you an alternative. 

    Best,

    Raymond 

  • 0 in reply to Raymond Zhang1
    Expert 2600 points

    Hi Raymond,

    we have to measure DC current which is switched on/off and it is bidirectional. We switch from +I to zero or from -I to zero, never from +I to -I.

    The desired voltage across the shunt resistor is 120mV (or -120mV). This signal should be amped to +-12V, so g = 100.

    Sample distance is 66.67µs.

    So, we have a step of 12V at the output: 12V / 0.17V/µs = 70.6µs. Ok, it fits so ..

    With best regards

    Gerhard

  • 0 in reply to Gerhard Kreuzer
    TI__Guru 54111 points

    Hi Gerhard,

    I do not see that other instrumentation op amp is better than INA188 with zero-drift features. (or use INA821).  If you'd like to use INA188, you may have to lower the Gain slightly or increase the shunt resistor value to compensate for it, if you want to keep under 66.67 µs.

    Best,

    Raymond