• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 47 subscribers
  • Views 353 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.

TLV3601: Prop Delay and Edge Rate

Tyler Holmvik
Intellectual 771 points
Part Number: TLV3601

Hi, 

I have a customer looking at using the TLV3601/3 for a TOF application.

We are trying to understand the how consistent propagation delay is. The datasheet only lists typical values and max values, but a few more clarifying questions:

- Is there a reason why is the min spec not listed?

- How does the prop delay change over time and temperature?

- Are there any methods for reducing the amount of variation in prop delay? Customer wants to filter our digitally, so a consistent prop delay is critical.

As for the edge rate, I understand this is the fastest edge rate push-pull comparator that TI offers. A couple questions:

- Only typ values are listed for the tFall and tRise. Is there a reason no max / min value is listed?

- How does this value change over time and temperature?

- Are there any methods for reducing the amount of variation in edge rate? Customer wants edges as sharp as possible.

Thanks,

Tyler

  • +1
    TI__Mastermind 24550 points

    Tyler

    Thanks for your post.  You have posed some interesting questions.  I will try to answer them and we can take this offline if more is required.

    I think it is best to explain the sources of prop delay variance.  Some are certainly silicon related (process variation), but the majority of prop delay variance comes from external sources.  As shown in typical performance curves, temperature is a big factor. The other main contributor also highlighted in curves is overdrive.  When we provide max values for prop delay, it is at a very specific test condition and thus does not tell the entire story.  However, it does give customers a way to benchmark different parts.  Not sure how to explain not including a min value for these specific test conditions but in general we have not done that.

    In regards to variation over time, when we run our qualifications, we check parts for large shifts in performance from time t=0 to t-final.  We do not see significant shift in performance.

    Trise and Tfall and very much dependent on capacitive loading.  Designers aiming for very fast response times know to minimize parasitic capacitance on the outputs and inputs of a high-speed comparator.  Our TLV3601EVM shows some common techniques for minimizing parasitic capacitance on inputs and outputs.

    The faster the device, the less propagation dispersion (prop delay variation with overdrive).  This is why most customers who seek minimum dispersion go to our LVDS output devices.   Unfortunately, gain controlled amplifiers that produce a constant output voltage aren't fast enough and there is no lookahead signal to warn about an input signal that is lower in ampliture.  

    Since overdrive dispersion is a variable for all high speed comparators, we have seen customers use dual channel devices so one channel can have a low voltage threshold and the other a higher one.  This way they are aware if a prop delay increase is caused by small overdrive or if the object was merely a long ways off.

    I think I will stop here for now and we can follow up offline if that would help.  Your questions cover a fair amount of ground and each requires probably more than I have provided here.

    Chuck

  • 0 in reply to Chuck Sins
    TI__Intellectual 771 points

    Thank you. This is plenty to work from for now.

    Tyler