• State
  • Locked Locked
  • Replies 12 replies
  • Subscribers 49 subscribers
  • Views 430 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.

Original question:

Analog input higher than VDD1

ISO224: ISO224B output stability

Allen Du
Prodigy 220 points
Part Number: ISO224
Other Parts Discussed in Thread: LMH6612, , TLV9061, LM6132

Hi Team,

I am testing the ISO224B EVM board to detect high voltage, I am not using the OP-AMP on the EVM board. input of EVM board is resistors divider with 400kohm : 10kohm, EVM board output connects to LMH6612, and 6612 connects to FPGA.

I collected the output data continuously, but looks the output is reducing at the first few minutes, and increasing after that. As the attached picture, Y axis is voltge (Volt), X axis is time (seconds.)

I checked the primary high voltage with oscilloscope it's not changing. the test is done in airconditional room and room temperature is around 23 degreeC, could you advice how to improve the accuracy. thanks in advance.

  • 0
    TI__Guru**** 171076 points

    Hi Allen,

    Is it safe to assume your Y axis is 'volts'?  What is the X axis scale?  Can you provide a complete schematic and details of the ADC that is sampling the ISO224 output?

  • 0
    Guru 140200 points

    Hi Allen,

    what are we seeing? VOUTP - VOUTN? Or only one single output (VOUTP or VOUTN)?

    Kai

  • 0 in reply to Tom Hendrick
    Prodigy 220 points

    Hi Tom, Kai,

    Thanks for your reply. it's VOUTP - VOUTN

    The X axis scale is second, 14323 means 14323 sec.  the data on the chart is calculated from  ADC/FPGA output. I measured the waveform on LHM6612 pin 7 output, the voltage changes as well.

  • 0 in reply to Allen Du
    TI__Guru**** 171076 points

    Have you checked your 2.9V reference and the two outputs of the ISO224 as well?

  • 0 in reply to Allen Du
    Guru 140200 points

    Hi Allen,

    I have totally overlooked your post. Sorry for that.

    The TLV9061 may not want the filter cap being directly connected from the output to ground. This is a simple way to destabilize an OPAmp and can even cause oscillation. You could insert an isolation resistor. But see below.

    Another issue is the LMH6612. In my eyes the LMH6612 is way too fast in this circuit. The ISO224B has a bandwidth of (only) 275kHz but the LMH6612 has a bandwidth of 345MHz. This can cause stability issues, eventually in combination with the TLV9061 (10MHz bandwidth) which is way slower than the LMH6612. It's not good to have a slow OPAmp in the feedback loop of a much faster OPAmp. The output impedance of slow OPAmp can look indutcive at the high frequencies and can potentially add resonances and impedance maxima which can shift the phase in the feedback loop of LMH6612 and cause instability.

    Kai 

  • 0 in reply to kai klaas69
    Prodigy 220 points

    Hi Kai,

    Thanks for your support.

    I have changed LHM6612 to LM6132BIMX/NOPB, but got another problem. the opamp A output looks normal, but output B looks not incomplete, the pulse width is reduced. there was no this issue when using LHM6612, is it because of LM6132 bandwidth too low?

     

  • 0 in reply to Allen Du
    Guru 140200 points

    Hi Allen,

    still the TLV9061 in your circuit? Still the filter cap at the output of TLV9061? And what means 2.9 VREF in your sketch? Isn't 2.9V way too high? Wouldn't this drive the output of OPAmp "B" into saturation?

    Please show a schematic.

    Kai

  • 0 in reply to kai klaas69
    Prodigy 220 points

    Hi Kai,

    TLV9061 still in the circuit, cap at TLV9061 output was removed, 2.9VREF was adjust to 2.6V.

  • 0 in reply to Allen Du
    Guru 140200 points

    Hi Allen,

    are you sure? 2.6V will drive the output of OPAmp B into saturation:

    allen_iso224.TSC

    Kai

  • 0 in reply to kai klaas69
    Prodigy 220 points

    Hi Kai,

    The circuit is different with your TINA file, OP1 -IN should be GND.

    I measured LM6132 OUTA Max voltage around 2.9V(picture 1), and if lm6132 -INB VREF is 2.6V,  then (2.9-2.6) x10 =3V (as picture 3).

    I tried adjust the VREF voltge, the follow picture 2 shows saturation when VREF is too low.

  • 0 in reply to Allen Du
    Guru 140200 points

    Hi Allen,

    this way, according to figure 21 of datasheet of ISO224?

    Ok, but in what range is the ISO224's output signal changing? I ask because there are two possible moments of saturations: Leaving saturation and going into saturation. It looks like that you are leaving saturation. This could explain why the signal is delayed.

    Please tell exactly the range of output signal of ISO224. I cannot see this from your scope plots.

    Kai

  • 0 in reply to kai klaas69
    Prodigy 220 points

    Hi Kai,

    The pulse width is 10us, I think output signal of ISO224 has 2-3 us delay is normal. the LM6132 OUTA has delay but still around 10 us, just not understand why OUTB pulse width is shorter.

    Measure output signal of ISO224 needs differential probe, I will find one to measure it. I think should be close to LM6132 OUTA.