• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 6 answers
  • Subscribers 63 subscribers
  • Views 677 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.

Original question:

CCS/UCD3138064AEVM149: why fusion desinger cannot download ucd3138a64 firmware to evm149？

Compiler/UCD3138A: how can i start a open-loop project of ucd3138A。

dengpeng wu
Intellectual 325 points
Part Number: UCD3138A
Other Parts Discussed in Thread: , UCD3138

Tool/software: TI C/C++ Compiler

now，i want to write a code about LLC half-bridge circuit ，the device i used is ucd3138A， what should i do？how can i start？

  • 0
    TI__Genius 17090 points
    You can download the EVM code for half bridge LLC here:

    www.ti.com/.../UCD3138FW_LLC
  • 0 in reply to Ian Bower
    Intellectual 325 points
    but，it is not open-loop codes，the firmware is closed-loop code。i have tried that code，it didnot work。
  • 0 in reply to dengpeng wu
    TI__Genius 17090 points

    Yes, you're right, it's not open loop.  We've never actually done an open loop LLC code.  It's problematic starting up in resonant mode, because the filter starts out with a zero in it's output and the DPWM never gets triggered, so it never triggers the filter.

    I tried to do it once using preset and using force start on the Filter to put the values in, but I ran out of time before I got it working.  So I'm not aware of any sample code for open loop for half bridge LLC. 

    With the closed loop code, since we start up in PWM mode, it's not a problem.  The PWM mode gets things going so there's a good number in the DPWM when we switch to resonant mode. 

    What do you need it for?

  • 0 in reply to Ian Bower
    Intellectual 325 points

    I have designed llc circuits, I used code from ti，but no pwm output，so i wang tried openloop code to make sure the software no mistakes。

  • 0 in reply to Ian Bower
    Intellectual 325 points

    The LLC circuit i have designed，just like the UCD3138ALLCEVM150，I have used the LLC code from TI, but no pwm output, it didnot work. so i want a open-loop LLC code to debug my PCB.

  • 0 in reply to dengpeng wu
    TI__Genius 17090 points

    If your board is the same as the EVM, it will support PWM mode at light load, so you can use an open loop PWM for testing.  If you look at the UCD3138 video training, you can find a couple of different ways to make open loop DPWM signals.  You will have to adjust the frequency and dead times for your application, but that's not very hard.  In fact, using the first lab that exercises the DPWMs, you can even put out a static simulation of a point on the resonant curve, which is perfect for debugging. 

    training.ti.com/ucd3138-digital-power-training-series

  • 0 in reply to Ian Bower
    Intellectual 325 points

    i  have used the code，it is pwm output is ok，but the topology is HSFB？ i am not sure whether the training code can be used for llc

  • 0 in reply to dengpeng wu
    TI__Genius 17090 points
    The technology is buck, actually. So you would need to modify it for your needs. Lab 2 describes how to use just the DPWM event registers to control the DPWMs with arbitrary timing. So you could set up a fixed timing to match what you need for LLC. You could start with just the primary side FETs on DPWM0, and try various timings to see what happens on your board.

    Then you'd have to add another DPWM to handle sync FETs, if you have those as well.
  • 0 in reply to Ian Bower
    Intellectual 325 points
    thank you very much，what i need to modify is the event time？ if the duty and dead time is suitable for my llc ，the code will be useful despite the topology？if it is true，i will try to modify the training code。actually I have debugged the lab2 、lab 3 to output driver signals。I've been afraid to use the wrong topology。it will be helpful to me。
  • 0 in reply to dengpeng wu
    TI__Genius 17090 points

    You will need to figure out what timing you need.  In lab 2, you can put in whatever timing you need into the event registers.

    I'd start out with a small on-time for both phases, assuming that your primary FETs are on DPWM0, make Event 1 maybe at 200 nsec, event 2 at event 1 plus whatever pulse width you want.  Then make event 3 at half the period plus 200 nsec, and event 4 at Event 3 plus whatever pulse width you want. 

    In this static situation, you could put event 1 at zero, but in a dynamic system, you want to keep it out of the event update window at the start of the period.

    You need to look at the timing on the scope and get it right before you apply power to the FETs. 

  • 0 in reply to Ian Bower
    Intellectual 325 points

    OK,  i will follow your steps to debug,try the code soon，thank you very much,