• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 47 subscribers
  • Views 354 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.

INA196A-Q1: application question

Sherry Liang
Genius 10255 points
Part Number: INA196A-Q1


Dear team,

Could you please help answer below questions? Thanks!

1. For V+ selection, my customer has two option, +12V and +5V. Could you please help analyze the effect of the +12V or +5V selection on the application circuit/chip? What scenario we recommend 12V? And what scenario we recommend 5V?

2. Could you please help clarify what the function of T1,T2 are? When VCM is negative, A1 is active; when VCM is positive, A2 is active. What is the difference between A1 and A2? Why do we use two amplifiers here?

3. In my understanding, 30% error comes from internal resistor itself, and this has no relationship with our device, right? If so, we choose 30% tolerance resistor is considering low cost? After all, 30% is a large error.

4. What factors result in the gain error in below table?  How do we evaluate the total output error? Do we have the rough formula?

5. Could you please tell me the reason that we divide into five situation when considering the accuracy? Can we provide each case's accuracy formula?

6. I didn't see diagnostic protection description in the datasheet. This device has no protection?

Thanks & Best Regards,

Sherry

  • 0
    Guru 140200 points

    Hi Sherry,

    the 30% "error" or better say "deviation from the regular 5k" of 5k resistors only comes into play, if you want to add some additional resistances in series to them to provide a low pass filtering. Concretely spoken, the 30% error is no error but manufacturing tolerances from chip to chip or from wafer to wafer. The two 5k resistors on each individual chip are very precisely matched and there is no big error between them. But both 5k reistor together can show a deviation from the regular 5k by 30%.

    Kai

  • +1
    TI__Mastermind 35530 points

    Hi Sherry,

    1. For V+ selection, it is mainly determined by input/output signal levels. For 12V, the output can swing higher. However downstream circuitry needs to be able to process 12V signal.
    2. T1 and T2 works as a current mirror, which directs the output current produced by A1 and T1 to the load resistor RL. A1 and A2 are made with PNP and NPN input pairs respectively. Depending on Vcm, one of them works – negative, A1; positive, A2. Neither by itself is able to work for the entire common mode range. This is the reason why both are needed. A3 is only a buffer and doesn’t participate in signal amplification.
    3. As Kai mentioned, the absolute value of the resistor can vary up to 30%, but the device accuracy is not dependent on the absolute accuracy. Rather it depends on the ratio of resistors. For example, gain depends on RL/5K. This ratio has excellent accuracy even without any trimming.
    4. Gain error is mainly affected by the quality of the opamps (A1, A2) and resistance matching (see #3). For error calculation, please refer to online videos. Here is an overview, and you’ll find the complete series on the same TI web page. We also have an online calculator that enables fast accuracy evaluation.
    5. The reason the device has different modes of operation is due to reasons in #2. A1 and A2 don’t turn on and off sharply. However this is a tradeoff in order to accommodate wide common mode range.
    6. The only protection is ESD diodes, with device ESD ratings listed in the datasheet

    Regards, Guang