• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 2 answers
  • Subscribers 48 subscribers
  • Views 202 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.

LM2901B-Q1: propagation delay part to part deviation

Chuqiang Wen
Expert 5060 points
Part Number: LM2901B-Q1
Other Parts Discussed in Thread: TLV9030-Q1, LM339

Hi Expert, 

I would like to check with you the part to part deviation of propagation delay spec. 

Some background, customer have some requirements for the part to part deviation. The input of comparator is PWM signal. They want to keep the part to part deviation of output PWM signal within ±200ns. 

Is it possible? 

Thanks!

Ethan Wen

  • 0
    Guru 272970 points

    This is not specified. I would certainly expect it to be below 200 ns, but there is no guarantee.

    Your largest problem when using this devices with PWM are the slow rising edges of the open-collector output. I would suggest a comparator with push/pull output like the TLV9030-Q1.

  • 0
    TI__Guru 53036 points

    Hello Ethan,

    This family is a low-cost, general purpose device. As such, prop delay is not a guaranteed parameter and is not tested in production. Therefore, prop delay is subject to production process spread.

    For uncontrolled parameters, we would expect up to a random ±30% change over lots and time. Devices from the same lot would have a better chance for matching, but lot-to-lot can have larger variations - so they really need to to account for the larger lot variations.

    See the "Propagation Delay vs. Input Overdrive" graphs starting on page 14. Overdrive and temperature will have the greatest effect. The faster the response, the less variation. Add 30% to the number at the specific overdrive voltage.

    If prop delay variation is a critical parameter in the design, then they should really be using a faster comparator, where the variation is much smaller in proportion to the requirement. That would give more consistent results. If you like to drive fast, would you rather drive a family sedan or a sports car?? Slight smile

    You show 2.5V as the threshold. Just make sure that the maximum amplitude is at least 2V below the comparator supply voltage (Vcc-2V). Exceeding the input limit will cause large variations in the prop delay as the input stage gets cut off and has to recover.

    For some fun reading, see the LM339 family Application note, sections 3.4 and 4.2:

     Application Design Guidelines for LM339, LM393, TL331 Family Comparators

    Assuming you have a input signal greater than 1V above and below the 2.5V reference, and NOT exceeding the limit. the comparator is at "full speed" of 300ns, then the variation will be much, much less than 200ns.

    As Clemens mentioned, the TPLH (low to high) is dominated by the RC time constant of the pull-up resistor and load capacitance. If they want consistent edges, then a push-pull output device would be better (high speed devices are predominately push-pull output) for comparable edge rates.

    What is the input signal amplitude, frequency and the supply voltage? Then we could narrow down the numbers and/or recommend a possibly better device.