Distorted o/p in EXC_OUT Signal from EVAL-AD2S1210SDZ

I am trying to use EVAL-AD2S1210SDZ to read the angle of a resolver. But when I checked the o/p Excitation signal in CRO, I am not getting a sine signal . I am getting a clamped distorted signal.

I am using the op-amp AD8397 with 0.866 gain. 

Can someone please help me to understand why not getting the exact signal at output.

Parents
  • Hi,

    The issue on the excitation waveform resolved.

    I measured across resolver initially. Later got correct waveform by putting a high resistance.

    Now I want to know What is the maximum exciation voltage we can generate using this Board. We have a resolver which needs 4.25 Vrms exciation.

    I am using AD8397 ( LK14 to LK17 in position B) .Can you please let me know What is the maximum output voltage I can get using these configuration and How I can achieve it?

  • 0
    •  Analog Employees 
    on Mar 2, 2021 4:48 PM 1 month ago in reply to tpsamuel

    tpsamuel,

    If your resolver requires 4.25VRMS (differential) then that is about 6Vp (12Vpp).   The RDC swings between 7.2Vpp to 8.0Vpp differentially, natively which means you would need about 1.5x gain so the 15.4K option should be sufficient for your needs (LK10, LK12, LK14, LK16 = B),   As you are using the AD8397 path if you need a bit more gain you can always use OPTION A on LK10 and LK12 and R10 and R30 to increase the gain but you'll need to be mindful of the output limit with regards to headroom and footroom for the amplifier which roughly is 0.5V on either side with respect to the rail at 100 ohm load.  The issue is that you only have 2.5V of offset through that path.  You may be better off going through the AD8662 path to ensure distortion free operation.   The principle would be the same in that if you measured the peak output from each of EXC/EXCB you could calculate the correct gain and implement in the AD8662 stage.

    Sean

Reply
  • 0
    •  Analog Employees 
    on Mar 2, 2021 4:48 PM 1 month ago in reply to tpsamuel

    tpsamuel,

    If your resolver requires 4.25VRMS (differential) then that is about 6Vp (12Vpp).   The RDC swings between 7.2Vpp to 8.0Vpp differentially, natively which means you would need about 1.5x gain so the 15.4K option should be sufficient for your needs (LK10, LK12, LK14, LK16 = B),   As you are using the AD8397 path if you need a bit more gain you can always use OPTION A on LK10 and LK12 and R10 and R30 to increase the gain but you'll need to be mindful of the output limit with regards to headroom and footroom for the amplifier which roughly is 0.5V on either side with respect to the rail at 100 ohm load.  The issue is that you only have 2.5V of offset through that path.  You may be better off going through the AD8662 path to ensure distortion free operation.   The principle would be the same in that if you measured the peak output from each of EXC/EXCB you could calculate the correct gain and implement in the AD8662 stage.

    Sean

Children
  • We have verified the differential output between EXC abd EXC/ and its coming between 7.8 to 7.9 Vpp.

    So that section is working fine. But after that we connected jumpers LK10 to LK14 to position B and selected AD8397 and when we checked the output by connecting load resistance, we are not getting the voltage expected (multiplied by 1.54) . We are getting very less value. Do you know the reason for that or am i doing anything wrong.

    Can you please provide any detail document on this board as I am having only User Guide

  • 0
    •  Analog Employees 
    on Mar 3, 2021 1:00 PM 1 month ago in reply to tpsamuel

    We have verified the differential output between EXC abd EXC/ and its coming between 7.8 to 7.9 Vpp.

    Have you verified that the EXC and EXC/ signals are balanced and anti-phase?   I saw the post you made with regards to a difference in the outputs from U4-A (but for some reason it is not in line here). 

    The reason I ask is that I don't know how to explain the issue with the circuit other than potential damage to that amplifiers output stage perhaps caused by an overload in your initial testing.  What was the original load condition and could you have caused the amplifier to have gone into a short circuit current condition?

    Again,  you might try using the other path to see if you get a better result.

    Sean

  • Yeah I have verified the EXC and EXC/ signals.Initially didn't get EXC/ o/p. But it was probe issue.

    I will detail my requirement.

    We have to measure the angle of a resolver which needs 4.25Vrms excitation. But currently we don't have that. We have another one which needs less voltage excitation. We were able to measure the angle of that resolver using these EVAL board and EVAL-SDP-CB1Z Board.

    Now we want to make sure this will work for the other resolver which needs 4.25 Vrms excitation before purachasing it. So we want to know whether our Eval board is capable of providing that excitation. So we started analyzing the circuit and we want to check the excitation signal can provide enough voltage or not.

     

  • If you have this module can you share us the picture of expected excitation waveform and voltage

  • 0
    •  Analog Employees 
    on Mar 4, 2021 1:05 PM 1 month ago in reply to tpsamuel

    Unfortunately I don't have the equipment or evaluation kit with me as I'm working remotely at the moment.   You can however reproduce the expected behavior using a simulator like LTSPICE or PSPICE as their is a model for the AD8397 on the web page.

    Sean