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.

TMS320F28375D: eQEP

Part Number: TMS320F28375D

Hi team,

My customer use TMS320F28375D in their project, and they meet below issues about the eQEP, could you please help on give some suggestions? Thank you!

1. Competitive risk

Set QEPCTL.bit.PCRM=1,QPOSMAX=0xFFFFFFFF,QEPCTL.bit.IEL=1,

When the index rising edge and the edge of A or B arrive simultaneously, the QPOSILAT latch value is sometimes 1 different when both measured in the same operating direction;
As shown in the figure, when the index rises to the positive direction, The value that should be latched is 0, but sometimes it is locked to 1;
When running in the opposite direction, when index rises, the value that should be latched is 2, but sometimes it will be locked to 1;
2.  FIMF: The question of how first index marker flag is defined
As shown in the figure above, assuming the encoder is in position A when powered up, the encoder moves in the order A->B->A->B->C , When measured, the eQEP module will record the reverse motion of QA at D as a rising edge as FIMF, which is not one of the four edges of the QA QB during the period when index is high; if QEPCTL.bit.PCRM=1 is set, QPOSMAX=0xFFFFFFFF,QEPCTL.bit.IEL=3, and the next edge of the latch event will not obviously be this edge at D.
3. Index uses only the quadrature edge after its rising edge as an index event, but there are actually encoders where the index pin outputs a high level when the index signal does not arrive, and the index pin outputs a low level when the index signal arrives. This logic is the opposite of what qep needs; although it can be solved by setting index polarity, it lacks versatility. Is not sure if there is a better solution?
For the three questions, could you please help give some questions? Thank you!
The reference documents are listed below
Regards,
Ivy
  • Hello, 

    Please expect a reply by 11/30. 

    Thanks

    Anu

  • Hi Ivy,

    Response to your questions in order:

    #1 -  Yes it is possible, if Index event (QI) and Quadrature event (i.e. QA or QB) occur at the same time, QPOSLAT value can be off by 1. Reason being the internal delays are different, so it is possible QPOSLAT happens prior to QPOSCNT update.

    Is customer using GPIO in Async or Sync mode? Also, is there a specific concern from customer side? This can be handle in SW where customer on Index flag can read QPOSCNT multiple time, and if there is any update QPOSLAT value can also be updated.

    #2 - FIMF flag is not valid in PCRM=1 mode, it is already mentioned in TRM. Please let me know if there is any specific concern, I am not clear on this issue.

    #3 - Not clear on this question, can you show the waveform of the encoders with QA, QB and QI signals?

    Best Regards,

    Nirav