• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 48 subscribers
  • Views 2236 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.

Fully Differential Fixed Gain Amplifier?

Jay He
Intellectual 280 points
Other Parts Discussed in Thread: THS4520, INA132, INA128, INA129, PGA2505, ADS1259, PGA280, INA331

Hi, I'm trying to design a low noise amplifier circuit using one of the TI's FDA such as THS4520 etc. This amplifier will be in a EMI intense environment. So we chose FDA. But then I realized that in order for the FDA to function properly, the feedback resistors have to be matched almost perfectly, otherwise the noise imunity or CMRR will suffer.

I also realized that there are so called instrument amplifier that does not have the negative feedback like in a typical FDA. It seems to me this is like a FDA with factory matched feedback builtin.

But I don't see a fully differential instrument amplifier with fixed gain. Is there such a thing?

Thanks,

Jay

 

  • 0
    TI__Guru 64530 points

    Jay,

    Some instrumentation amplifiers, like INA132, are fixed-gain difference amplifiers with well-matched internal resistors (see below).

    INA132

     

     

     

    Some other instrumentation amplifiers, like INA128/INA129, have well-matched resistors in both signal paths (resulting in superior CMRR) but absolute gain is a function of the external, RG, resistor (see schematic below).  Therefore, the gain depends on the value of external resistor but not the CMRR.

    INA128/INA129

     

    For fixed gain (digitally selectable) products, like PGA2505 (below),  look in PGA (Programmable Gain Amplifier) product line. 

     

  • 0
    TI__Intellectual 2620 points

    Hello Jay,

    there is indeed a differential-in and differetial-out instrumentation amplifier with fixed gain options available. Actually the gain is not exactly fixed, it is rather programmable. Please have a look at the PGA280. This in-amp has a +/-15V capabale differential front-end as well as a fully differential output (up to 5V supply. Gains can be adjusted from 1/8 to 128V/V. The ADC companion chip is the ADS1259, which also has a fully differential input. In addition we have added EMI filtering inside the PGA280 to make it robust against EMI noise.

    Best regards

    Frank

  • 0 in reply to Marek Lis
    Intellectual 280 points

    Marek,

    Thanks a lot for the detailed information. I've since done a lot of research along the line.

    Since we need to have some gain, INA132 is not going to work. INA128/129 is good, but 1) the BW is a little low. We'd like to have a solid 10MHz at G=20X. 2) it does not seem to be a fully symmetric FDA. Does that (not being fully FDA) matter (for CMRR)?

    INA331/332 is in the same class as 128/129, and cheaper, but has even less symmetric schematic layout. So that leaves me with only the PGA type for fully differential instrument amplifier. However, at this point, we want to limit the wires going into the sensor area. So we get back to the regular FDA.

    For the regular FDA, I found the THS4520 to be an overall good choice. But there are still many similar op amps and I hope I won't make a big mistake by sticking with 4520.

    Since our sensor has pretty high internal impedance, we'll need two buffer op amp in front of the FDA (which circuitry seem to me have intrinsically low input impedance.)

    Question here, does the buffer op amp need to be matched, like the feedback resistors?

    Another question, we also need to put some capacitors in the feedback. Could we use a variable capacitor to tune it?

    And still one more question, going back to FDA with internal ADC, is there any "back" EMI from the digital nature of the ADC, or other part of PGA, to the analog signal, ie, degraded noise performance due to digital signal?

     

    Thanks,

    Jay

  • 0 in reply to Winnie Puuh
    Intellectual 280 points

    Hi Frank,

     

    See my reply to Marek. Yes, the PGA type will be the one we might consider later to convert the signal directly to digital and go from there. Right now, we're using the op amp to be a pre-amp and bring the stronger signal to our control box. But if the footprint of the PGA is small enough and the EMI is not an issue for the non pre-amp'd signal (is that 100% the case?) we might just stick it right to the sensor and getting the digital signal to our control box.

     

    Thanks,

     

    Jay