• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 82 subscribers
  • Views 259 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.

TMS320F28379D: Three phase 3 gpio synchs to 3 ePWMs

Wynand Du Plooy
Prodigy 10 points
Part Number: TMS320F28379D

Tool/software:

To all the power control experts,

I have an application where I need to synchronize 3 independent zero crossings to 3 independent ePWM modules. I have gone through all the available materials on similar matters and cannot find something that will look like it can work for me.

There are only 2 external synchs available on the C2000 series for all ePWMs and even the eCAPs.

It will be possible to do this if I can have the ePWMS working in one shot mode, but the synch signals cannot be shared.

Is there anyone out there that possibly have an idea on whether this can be be done with a single C2000 controller and how. ANy advice would be greatly appreciated.

  • 0
    TI__Mastermind 23280 points

    Hi Wynand, 

    Can you perhaps provide a quick diagram or sketch to help me understand the sync scheme you are wanting? The EPWM modules are linked together in groups within the sync scheme as shown in the TRM snippet below. Are you trying to use 3 independent external sync signals?

    Best Regards,

    Allison

  • 0 in reply to Allison Nguyen
    Prodigy 10 points

    Good day Allison,

    I have a three phase full wave thyristor rectifier bridge that needs to be controlled very accurately. 

    There is a lot of detail missing in the picture, but what is important to me is detecting the zero crossings at VA,VB and VC and the switching on the appropriate thyristors at the exact instant it needs to.

    The challenge with this is to minimize the dc voltage ripple down to 0.1%.

    I spent a year on this with an ARM M7 only to realize that it was the M7 creating the additional ripple. This is because of the 6 stage pipeline of the core. 

    My only solution is to use a dedicated core like on the F28379D to ensure I know the exact frequency and zero crossing of every phase. 

    As you mentioned here the ePWM has a synch option, but this will allow me only two legs of the bridge and thus I am short of one. This is for an industrial application and therefore I cannot assume that the phase shift is perfectly 120 degrees between the phases at all instants. To give you an idea 100us inaccuracy creates an additional 3V ripple where I am only allowed 200mV.

    Sorry for the long explanation, but the essence of it all is I need to synchronize exactly to the three voltages VA,VB and VC.

    Please help.

    Kind regards

    Wynand

  • 0 in reply to Wynand Du Plooy
    TI__Mastermind 23280 points

    Hi Wynand,

    Thank you for the context and information - please allow another day for me to review your post and formulate a response. Thanks!

    Best Regards,

    Allison

  • 0 in reply to Wynand Du Plooy
    TI__Mastermind 23280 points

    Hi Wynand,

    Unfortunately the built-in hardware only allows for the two EXTSYNCIN source paths. Another thought is would you be able to use a digital compare event to issue a sync pulse instead?

    Please take a look at the TRM section (15.11.4 Operation Highlights of the Digital Compare Submodule) for more details on using the digital compare events (DCxEVTs) and whether this seems feasible for your application.

    Best Regards,

    Allison