• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 47 subscribers
  • Views 1142 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.

AMC1200 poor gain accuracy

Nicolas PLUMEREL
Prodigy 160 points
Other Parts Discussed in Thread: AMC1200, AMC1301

Hello,

we use the AMC1200 precision isolated amp to measure the AC input voltages in a an active 3-phase rectifier (as well as the rectified output voltage).

To be able to measure negative voltages of the AC inputs we shift VINn to the mid voltage of (VDD1 - GND1) with a voltage divider followed by an op-amp in a follower configuration.

The AC input voltage is scaled down to a suitable range using a differential amplifier and shifted using the output of the follower described above, then provided to the VINp of the AMC1200.

Our measurements show that the input voltage of the AMC1200 is as expected but the ouput shows a gain error much greater than what it should be (in the order of several %). We perform our mesaurments at ambiant temperature in a lab, using labaratory power supplies (clean supply voltage).

We have no idea what is wrong with our circuit. Can you help?

Thanks and regards,

Nicolas

  • 0
    TI__Guru 55795 points
    Hi Nicolas,

    AMC1200 belongs to Precision Data Convertors and they would be right people to give you best comments, hence I am moving your post to Precision Data Convertors E2E Forum.

    Regards,
    Koteshwar Rao
  • 0
    TI__Guru**** 180751 points
    Hi Nicholas,

    Can you share your AMC1200 schematic with us?
  • 0 in reply to Tom Hendrick
    Prodigy 160 points

    Hi Tom,

    herebelow is the piece of interest of our schematic:

    Note that since yesterday, we found that adding a differential RC filter at the AMC1200 inputs helps getting better results. This may be due to the fact the AMC1200 input stage has some kind of switching device (modulator) which may interact with the output of the op-amps and generates ripple/noise. This may affect the gain accuracy of the AMC1200. Do you confirm?

    If yes, can you recommand values for this RC filter knowing that we need a bandwidth of at least 10 KHz.

    Regards,

    Nicolas

  • 0 in reply to Nicolas PLUMEREL
    TI__Guru**** 180751 points
    Hi Nicolas,

    Thank you for the schematic snippet. I'll take a closer look at that over the weekend. You are correct in that with the AMC1200, you are essentially looking into a switched capacitor that may have some impact on the amplifiers. Have you looked at the voltage going into the AMC1200 before you added the differential filter? Have you taken a look at the AMC1301? That might be another option for you as it has an input structure that is closer to a true diff amp. If you used the AMC1200 in the DWV package, the AMC1301 would be a direct replacement and I believe you may see better results without the need for the differential filter.
  • 0 in reply to Tom Hendrick
    Prodigy 160 points

    Hi Tom and thank you for your quick response.

    We indeed tried to look at the differential input voltage with a scope to see how noisy the signal was. But with the long wires of our differential probe and with such low voltages, it was dificult to assess if the noise we were seeing was real or caused by our setup. We will try again early next week.

    Regarding the AMC1301, it looks like an interesting option. I will try to procure several samples to try it and compare the results against the differentiel filter.

    Anyway, I am still interested in getting your opinion regarding the R and C values I should ude for the filter.

    Thanks and regards,

    Nicolas

  • 0 in reply to Nicolas PLUMEREL
    TI__Guru**** 180751 points
    Hi Nicholas!

    My appologies here! I was going to ask if you had gotten any samples of the AMC1301 to try in your circuit and realized that I never got back to you on the filter components. I must not have hit 'reply' before closing out my browser. To start with, you can try the R and C values listed in the AMC1200 datasheet, Figure 32, page 15.
  • 0 in reply to Tom Hendrick
    Prodigy 160 points

    Hi Tom,

    we ordered several AMC1301 from our TI provider.

    In the meantime, we tried a couple of RC networks starting with what we found in the datasheet as you just suggested. But this did not bring any benefit.

    So we tried larger values of resistors and caps and ended up with 2x 100 Ohm + 3.3 nF which gave better results (less than 1% gain error). This is still far from the component specification (0.05% nominal!) but I am not sure we can do better than that. But other suggestions are welcome, of course!

    I also noticed the linearity is neither as good as it should be (probably because the noise more affects lower Vinp-Vinn values). I still need to do a couple of tests to asses if this may be a problem in our application.

    The AMC1301 may overcome those issues. I hope I will get the parts very soon.

    Regards,

    Nicolas