• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 81 subscribers
  • Views 323 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.

TIDM-1000: Vienna DC Bus Voltage Limit

chris geraghty
Prodigy 80 points
Part Number: TIDM-1000
Other Parts Discussed in Thread: TMDSCNCD28379D

Hi,

I am trying to scale up the Vienna rectifier design for higher voltage and current operation.

I have built a 440V 15kW rectifier using 1200V SiC Mosfets & Diodes with inductor and capacitor values scaled according to the TIDM-1000 Calculation spreadsheet (higher voltage capacitors obviously).

I am using the same board ("TMDSCNCD28379D F28379D controlCARD for C2000 Real time control development kits") for the control that I used with the evaluation kit, running the same Vienna application, in debug from the laptop 

I have scaled the application to correctly monitor my ac voltage sense and currents (and set my current limit) successfully from the main.cfg page.

The setup has, so far, worked very well as I have, in small steps, increased the applied ac voltage and the dc voltage set-point, with a 48Ω load on the dc. Stable dc operation, <2% THD currents, unity power factor. I got up to 440V input with the VDC set to 700V (10.3kW) without issues

But I believe, in order to regulate with the worst-case (highest) tolerance on my 440V supply, I will need to set the VDC above 700V, so for my next step, I left the input at 440 and set the dc ref to 720V

In this case it would always work perfectly for approx 1.3 seconds, and then just stop: no fault message in the Expressions list, still reading "BoardStatus_NoFault"

I see in the vienna_settings.h file, line 184 reads #define VIENNA_VBUS_TRIP_LIMIT_VOLTS ((float32_t)720.0)

So is 720V a safety limit on the buss? and it does not give a fault message in the debug expressions window? Does my observed 1.3 sec delay make sense? Note that I accidentally ran it nearer 740V one time and it shut down in the same way (no fault message after controlling ok) but this time after less than 0.5sec

So with my higher voltage-rated circuit, assuming my power components, gate drives etc are adequate, if I edit this line of this file to change the 720V value to, say, 750V, will this allow me to run at a higher dc bus? Or is there some extra complexity? I believe I need to generate a buss somewhere around 730V to allow the regulation to work at my worst-case (highest) input voltage.

Thank you for your help

Chris

  • 0
    TI__Genius 15555 points

    Hi Chris,

    Based on your information, I would say it does not seem like over voltage condition because of "BoardStatus_NoFault". 

    However, I would suggest to double check the below register. If there are any other faults, it will set those bits to 1.

    You can check the register EPWM1. TZFLG.OST, EPWM2. TZFLG.OST, EPWM3. TZFLG.OST

    When you say it stops, are you saying the all PWMs are turned off?

    Regards,

    Chen

  • 0 in reply to Chen Jiang ti
    Prodigy 80 points

    Hi Chen, thanks for the response...

    Yes, all PWM signals turn off.

    I just repeated the experiment, and I can confirm that after it turns itself off, those three TXFLG.OST registers are all set to 1.

    Meanwhile, the "BoardStatus_NoFault" remains showing and the measured expressions continue to update.

    (I don't know if this is significant, but I should tell you this operation is for build state 3).

    So I tried again, but with that overvoltage expression in the vienna_settings.h file rewritten as 730V and it appears to run, without shutting down while regulating at 720V.

    So it does look as if the shutdowns are related to the bus overvoltage trip, that it shuts down the PWM operation but doesn't fully halt the program?

    This is not necessarily a big problem at this experimental stage, but I'd like to know if you agree, and whether there might be a problem with raising the trip level that I haven't thought about

  • 0 in reply to chris geraghty
    TI__Genius 15555 points

    Hi Chris,

    That is weird that "BoardStatus" does not show over voltage protection. But it does look like the OVP is triggered. I will suggest to try to setup multiple thresholds(maybe lower than 720 if possible) and if they both turn off at expected Vout, then it is OVP and you are ok to go ahead. I will check the code on my end later to see if there is any bug. 

    Regards,

    Chen

  • 0 in reply to Chen Jiang ti
    Prodigy 80 points

    Thanks Chen,

    I tried your idea in two ways:

    1. I set the trip level to 350Vdc and raised the input voltage from zero to 275V line-line, giving an uncontrolled dc level of 367V: The board had tripped out ("BoardStatus_OverVoltageTrip") by the time I had walked from the variac to the laptop.  
    2. I set the dc reference level to 500Vdc and set the trip level to 450Vdc. Again took the input to 275V L-L. I entered a 1 on Clear Trips and the DC level started to ramp up before going into the same kind of shutdown I described before - with the PWM registers tripped and the board status showing NoFault. I should also say that the shutdown occurred some time after the 450V level was crossed - approximately 160 msec after - by which point the measured dc had just about reached the 500V reference level. I did not do enough tests to quantify this, but it looks like the delay between exceeding the threshold and the pwm shutdown depends on how much the level is exceeded by

    I plan to continue my testing, setting the trip level marginally above the level I intend to operate at, but would appreciate your thoughts - I don't mind at this stage that the shutdown isn't as clean as expected with the GUI, but shouldn't Overvoltage be a near instantaneous trip?

    Many thanks

    Chris

  • 0 in reply to chris geraghty
    TI__Genius 15555 points

    Hi Chris,

    Thanks for your feedback. Looks like it does triggers the OVP. That should be a quick response though we are doing OVP in the ISR. But it should still take much less time than hundreds of msec. The ISR is running in a 10kHz loop which happens every 100us. 

    Regards,

    Chen