• State Answer Suggested
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1938 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Dead time of UCC28950(TABSET and TCDSET)

Oliver Ou
Intellectual 2845 points
Other Parts Discussed in Thread: UCC28950

Our customer calculated the dead time with below equation on UCC28950, and found the real test result is larger different with the calculation dataat fixed delay condition.

For example,

when we use 24K at DELAB and short ADEL to Ground to cancel adaptive delay.  The real test result is 516nS, but calculation result is around 805nS with Equation3 on datasheet.

Can you help us to check if the equation is mistaken or  the error is normal? Do you have the comparing curver between the real teat and simulaiton?

CH1=OUTA; CH2=OUTB

Thanks.

Oliver Ou

 

  • 0
    TI__Mastermind 25570 points

    The TABSET1, TABSET2, TCDSET1 and TCDSET2 values are tested in production and will fall within the ranges given in the datasheet. The equations in the datasheet (eq3 and eq4) are not analytical but are an attempt to curve fit from these data points. Unfortunately they are not very accurate and significant errors are not unusual.

    I would expect that you will get closer results using the equations below, but please remember that these equations are also derived from a curve fitting exercise and there will still be a difference between the calculated and actual values. The best approach is to use the equations to get a ‘ball park’ value for the delay setting resistors and then to iterate to an optimum value.

    TABSET = ((5 Rab)/(0.26V + CS*Ka*1.3)) ns  and TCDSET = ((5 Rcd)/(0.26V + CS*Ka*1.3)) ns

    Where Rab and Rcd are in kOhms, Ka and CS have the same meanings as in the UCC28950 Data Sheet – KA = RA/(RA+RAHI) and CS is the voltage at the CS pin.

    Regards
    Colin

  • 0 in reply to Colin Gillmor
    TI__Intellectual 2845 points
    Hi Colin,
    Thanks for your feedback. We would let customer verify the new equations.
    Thanks.Oliver Ou
  • 0 in reply to Colin Gillmor
    Prodigy 40 points

    This equation 3 is wrong?   ( from "SLUSA16B –MARCH 2010–REVISED OCTOBER 2011")

  • 0 in reply to tu hs
    Prodigy 40 points
    TABSET = ((5 Rab)/(0.15V + CS*Ka*1.46)+5) ns and TCDSET = ((5 Rcd)/(0.15V + CS*Ka*1.46)+5) ns
  • 0 in reply to tu hs
    TI__Mastermind 25570 points

    Hi Tu

    The equation was changed to the one in the current datasheet on the product website page (REVISED NOVEMBER 2016), The newer equation is more accurate but it is still an approximation based on empirical results. You should use the newer equation.

    Regards

    Colin

  • 0 in reply to Colin Gillmor
    Prodigy 40 points

    Thanks

    I calculated dead-time with the newer equation.

    They are so different, then, have any different when the UCC28950's IC be producted in not same time?    

    Regards

    Tu

  • 0 in reply to tu hs
    TI__Mastermind 25570 points
    Hi Tu

    There have been no changes to the UCC28950. The problem was simply that the original equation in the Data Sheet was not as accurate as the new one.

    Regards
    Colin