• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 983 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.

Original question:

TPS3702: Hysteresis

TPS3702: Threshold calculation

Andreas Nestler
Intellectual 350 points
Part Number: TPS3702
Other Parts Discussed in Thread: TPS3703

Hi,

what is the correct way to calculate the thresholds (min / max) for undervoltage and over voltage respectively? If I read the datasheet. It is simply the nominal tolerance +/- threshold accuracy. Taken an example:

nominal threshold +/-4%, threshold accuracy 0.9% -> resulting into a tolerance band of +/-3.1 to 4.9%

If I check the TPS3702 decoder sheet there are two issues:

1) calculation is done as follows: nominal threshold voltage x (1+threshold accuracy), which is basically (1+4%)*(1+0.9%) resulting into a range exceeeding the previous calculated one (+/-3.064% to 4.936%). From my perspective the approach in the decoder sheet is wrong. The deviation is not a percentage from the nominal threshold it is simplky an absolute deviation (same unit)

2) decoder spread sheet uses the released thresholds provided in table 4 of datasheet. In another thread ("TPS3702 hysteresis") I've got already the confirmation that the values are rounded of to 2 digits, so not reflecting the exact thresholds.

Summarizing: from my point of view the correct way to calculate the threshold band is to multiply the nominal monitor voltage by (1+/- (nominal threshold +/- threshold accuracy)), whereby nominal threshold and their accuracy are expressed as percentage.

Can you please confirm the above approach? Addon: this seems to be the correct way if I'm checking the datasheet for a similar device - TPS3703.

Thanks in advance!

Best regards,

Andreas N.

  • 0
    TI__Genius 14630 points

    Hi Andreas,

    Thank you for your question, to calculate for the threshold it is (nominal voltage)*(1 +/- threshold accuracy).

    For example: 

    3.3*(1 +/- 4%)= 3.43, 3.168

    3.3*(1 +/- 9%)=3.597, 3.003

    Jesse 

  • 0 in reply to Jesse Villanueva
    Intellectual 350 points

    Hi Jesse,

    thanks for the answer. However, the answer does not fit to my question I have raised.The question was regarding threshold accuracy (+/-0.9%), so the devaition from the nominal thresholds of +/-4%. The approach in TPS3702 decoder spread sheet differs from datasheet information. There is a (slight) difference if I calculate 3.3*(1+4%)*(1+/-0.9%) as the spread sheet does or if I calculate 3.3*[1+(4+/-0.9)%]. The spreadsheet approach results into (+/-3.064% to 4.936%) instead of (+/-3.1% to 4.9%). This deviation may seems to be negligible, but we are closer to 5%, which is our supply tolerace requirement from component point of view.

    Best Regards,

    Andreas N.

  • 0 in reply to Andreas Nestler
    TI__Genius 14630 points

    Hi Andreas,

    Sorry for the misunderstanding, but yes the correct way is 3.3*(1+/-4%)*(1+/-0.9%)

    Jesse 

  • 0 in reply to Jesse Villanueva
    Intellectual 350 points

    Hi Jesse,

    Thanks for answering. However I don't understand it, sorry to contact you again. If I look at the figure from your first reply this seems not to be the case. The datasheet of a similar device (TPS3703) with tighter threshold accuracy (+/-0.7%) incorporates this figure. According this figure the threshold accuracy (+/-0.7%) will directly be summed to the nominal threshold as percentage (e.g. 4%) as I wrote (4+/-0.7)%. Anyway the deviation between both calculation approaches is only 0.036% in either direction.

    Regards,

    Andreas

  • +1 in reply to Andreas Nestler
    TI__Genius 14630 points

    Hi Andreas,

    For the second equation, 3.3*[1+(4+/-0.9)%], it is missing the (4%)*(.9%) when simplifying the 3.3*(1+/-4%)*(1+/-0.9%).

    3.3*(1+/-.9% +/-4%+/-(4%)*(.9%)). The (4%)*(.9%)=.036% is the deviation that you are seeing.

    Jesse