• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 82 subscribers
  • Views 568 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.

TMS320F28069M: PER UNIT MATH VS FLOAT MATH

gannamraju kumar
Intellectual 900 points
Part Number: TMS320F28069M
Other Parts Discussed in Thread: CONTROLSUITE, C2000WARE

Hi,I am working on a closed loop average mode controller for the buck converter.So i have 2 pi controllers(inner current ,outer voltage loops).Now my doubt is ,as my device is a floating point device,can i use floating point operation throughout the program for the controller calculations?if yes,please can someone tell me how to set the saturation limits for both inner and outer controllers?

what may be the reasons for using per unit math instead of float math.

Please suggest me some documents and any suggestions related to closed loop buck converter are always welcome.

Thanks

 

  • 0
    TI__Mastermind 27200 points

    The "per-unit" approach is useful if you are normalizing controller inputs and outputs.  Normalizing allows you to control the numeric range of the variables and coefficients, and is especially important if you are using fixed point.  The F28069-based buck controller example (C:\ti\controlSUITE\development_kits\BOOSTXL_BUCKCONV) in controlSUITE uses IQmath, so normalizing makes sense.

    With floating-point numeric range is less of a concern, but keep in mind you will have to allow for scaling through the A/D and D/A boundaries.  For example, your 12-bit A/D converter converts a 3V input to 4,095 numeric, so if 3V at the input represents full scale converter voltage/current output, you will have to account for that when your compute your controller gains.  The attached slide may help to guide your thoughts on the subject.

    You might like to look at the floating-point controllers in the DCL library in C2000Ware.  There are several types, and example #5 shows how to convert variables between fixed and floating point in a typical power supply scenario.

    Regards,

    Richard

    C2000 Digital Power Control Workshop - v3-2 - slide 68.pdf