• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 81 subscribers
  • Views 840 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.

F28377D CMPSS behavior for zero or negative inputs

Panos
Prodigy 160 points


Hi,

I need some help with the CMPSS .  I am using F28337D controlcard and I have configured some comparators to trip the PWM modules.

My issue is that when the comparator input is zero or negative (some mVs due to an offset of my rest circuit), either the comparator's trips are not stable (the resultant duty cycle is not constant) or it doesn't trip at all. If I increase the input value to 10mV, everything works fine. The ADC result in both cases is as it has to be (close to zero). My question is how the comparator behaves in such cases, when the input is either close to zero (with some noise) or negative? I would expect the same behavior with that when the input is positive. One possible reason may be that the output of comparator is forced to gnd in that case.

I am sure that someone can answer my question because I know that I am missing something.

Thank you very much.

Regards,
Panos

  • 0
    TI__Guru 62975 points
    Panos,

    Can you elaborate on your setup? How are both comparator inputs driven? Which pin(s) have zero? Do you have hysteresis enabled? Have you considered the input referred offset error? How is the comparator output being forced to GND? Is there a reason you are working at <10mV rather than centered >10mV?

    -Tommy
  • 0 in reply to tlee
    Prodigy 160 points

    Hi Tommy,

    Thanks for your interest.

    I use a CMPSS where the positive input comes from my current tranducer whereas for the negative input I use the internal DAC with the slope compensation. I control a bi-directional converter so the output of my current transducers is positive or negative (centered at zero). I have amplifiers to amplify and invert the signal in case it is negative and I have a signal selector before the CMPSS positive input in order to feed it with only positive inputs.

    When there is no current, at the start, the positive input of the CMPSS is almost zero (or with a small offset less than 10mV) and with some noise. In my control, I increase the maximum reference of the slope compensation ramp and I expect the PWM module to be tripped by the comparator. If my converter is under operation, the current increases (it is not anymore zero, or close to zero) meets the ramp and everything works as it has to. When the current flow is reversed, everything again is fine.

    However, when the converter is not power supplied, the current and so the output of the current transducer stays all the time close to zero. In this case, I observed that when the input is zero or maybe with a low negative value, the comparator never trips the PWMs. It seems like the output of the CMPSS is grounded no matter the inputs. The "problem" is that at the start of the converter operation and before I turn on the power supplies, I would like to check the pulses to be sure that everything works fine. Since my maximum reference of the ramp is low, I would expect to see short pulses. Instead of that I see sometimes short pulses, but sometimes no pulses at all. I thought then that this was happening because the output of the current transducers is sometimes zero, or negative due to the noise.

    So far my control seems to work fine. I just wanted to know what happens (the input referred offset error could be the reason of that) and If a negative voltage (<|10mV|) to an input can damage the controlcard. By the way, hysteresis is not enabled.


    Thank you once again,
    Panos

  • 0 in reply to Panos
    TI__Guru 62975 points
    Panos,

    There is nothing inherently bad with applying positive voltages near zero aside from the CMPSS offset errors that vary between devices. In your case, you would also need to consider the internal DAC will contribute its own offset error as well.

    The device can tolerate short exposure to small negative voltages (see the Absolute Maximum Ratings table in the datasheet), but prolonged exposure is not recommended because it will wear down the transistors.

    Since you already have amplifiers in the system, you may want to explore the possibility of introducing a positive offset to the transducers so that the entire range that is seen by the CMPSS module is positive. You can avoid the offset error and might potentially eliminate some of the rectifier/switch components.

    -Tommy