• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 82 subscribers
  • Views 699 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.

Choice of C2000 micro

Prasanna Nambi
Prodigy 240 points

Other Parts Discussed in Thread: TMS320F28335

Hello experts,

Iam developing applications requiring 850KHz high res driving PWM resonant converter and sampling 6 analog channels including 3 realtime feedback.

Do you have  any recommendation for the choice of micro controller that can handle this high speed sampling as well as PWM generation + PID computations.

 

  • 0
    Guru 192875 points
    Hi Prasanna,

    I would like to know more about your cost sensitiveness. For the above application TMS320F28335 would be an excellent choice which is well equipped with 150MHz clock speed. Do have a look at the datasheet: www.ti.com/.../tms320f28335

    Regards,
    Gautam
  • 0 in reply to Gautam Iyer
    Prodigy 240 points
    This is not cost sensitive, a preference is a smaller package size .

    Trying to understand the CPU overhead in high speed PWM operations and analog sampling. Have you tried the controller at these PWM rates and have any performance related recommendations.

    Iam evaluating the choices between TI , other silicon vs custom FPGA/ASICs
  • 0 in reply to Prasanna Nambi
    Guru 192875 points
    I've not tried at this high rate... but F28335 is very well capable of switching at MHz.
    You can check this link: e2e.ti.com/.../1398914

    Regards,
    Gautam
  • 0 in reply to Gautam Iyer
    Guru 27285 points

    Gautam Iyer said:
    I've not tried at this high rate... but F28335 is very well capable of switching at MHz.

     Hi Gautam, also never tryd over 100KHz, when it is at just one bit resolution can also EPWM work? Never used at upper limit.

  • 0 in reply to Roberto Romano
    Prodigy 240 points
    Driving higher PWM frequency is not the issue , the processor supports that and its a mere config. My question is from a architecture standpoint if the pwm , adc driver and other math intensive computational overhead like PID loops can be handled and any one have experience operating at these rates
  • 0 in reply to Prasanna Nambi
    Prodigy 240 points
    Also note in high resolution mode, the MEP is not active for 100% of the PWM period is going to be 3 / 6% min or the max which is a big limiting factor, I would like to understand this more from TI experts
  • 0 in reply to Prasanna Nambi
    TI__Guru 57255 points

    Hi Prasanna,

    Based on the fact you are talking about space constraints, I might recommend looking towards the Piccolo family of devices.  This family offers more integration on the chip: internal LDO, internal oscillator, etc.

    Based on previous experience I have very strong confidence that you will be able to do what you want on the F2806x series.  This family offers 90MHz of processing power on the CPU core as well as an additional 90MHz from the CLA.  Depending on what you expect the MCU to do, it may make sense to run the control loop on the CLA and leave the CPU to do auxiliary functions (like communications/host control/etc).  You could also have the CLA be responsible for updating the PWM at a variable rate (which changes with the resonant converter) and use the CPU to run the control loop at a fixed rate.

    Obviously I don't fully understand your system, but I think there's also a good possibility that you can run this on the F2803x series.  This series runs at 60MHz (for both the CPU core and the CLA).  On the F28035 device, I have personally run 6 separate 150kHz buck control loops on the main core and in a different project run these same 6 control loops on the CLA.  I don't have the numbers readily available, but there was at least some bandwidth spare for safety margin (in both cases).

    With regard to your question about HRPWM limitations:
    You are correct in your understanding about the 3 cycle limitation in the HRPWM.  In many fixed frequency applications, we can set it up such that the limitation is set up as the maximum duty cycle we will allow and limit the duty cycle via software.  With a variable frequency system, this limitation affects us in a different ways.  However, since duty cycles are roughly 50% in a resonant system, I think this limitation shouldn't affect you too much.

    Please note that we do offer a LLC Resonant EVM, based on the F28035, which runs in the 100-150kHz region.  This may be a decent starting point for you: TMDSHVRESLLCKIT


    Thank you,
    Brett