• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 35 subscribers
  • Views 570 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

AM26C31-EP: Part to part and gate to gate skew.

Gerald Horgan
Prodigy 100 points
Part Number: AM26C31-EP
Other Parts Discussed in Thread: AM26C31, DS26C31T

What is the part to part variation of propagation delay, when the parts are at the same operating temperature and voltage?  Is there data on the part to part and gate to gate skew?

  • 0
    TI__Mastermind 28815 points

    Hey Gerald,

    Our expert is currently out of office until monday, so I will refer to him.

    Thanks,

    Daniel

  • 0
    TI__Genius 17125 points

    Hello Gerald,

    On a sample size of 31, the worst case standard deviation for propagation delay was 0.075 ns and for skew was 32 ps.  These data are for part to part for the same channel.  

    There is a more difference channel to channel than part to part.  Calculating standard deviation for all channels from the 31 units;

    Propagation delay:  0.16 ns

    Skew:  60 ps

    Kirby

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 100 points

    That is a small sample size.  It doesn't seem correct that the standard deviation for propagation delay is only 75 ps part to part when the typical delay is specified as 7 ns and max delay is 12 ns.  I also would not expect channel to channel to be worse than part to part.  I would expect that all channels on a die (single part) would tend to be close, and die to die (part to part) would have larger variation.  Can you explain why channel to channel is worse than part to part, and why the standard deviation is so small compared to the specifications?

    I am also looking for the IBIS model for the AM26C31-EP to do signal integrity on the PWB.  Can you provide the model?

  • 0 in reply to Gerald Horgan
    TI__Genius 17125 points

    Hi Gerald,

    This is an old design (late 1980s) and we do not have much info on the design or characterization data.   I would speculate that internal metal routing and wire bonding are not exactly the same for each channel resulting in the small variations.   Process control across a wafer is fairly tight.   There may be more lot to lot variation but we do not have data for that.

    We do not have an IBIS model for the AM26C31, which is of TI origin.   We do have a model for the DS26C31 which came to TI from the National Semiconductor acquisition in 2011.  They are fairly close to each other but the timing specifications are not exactly the same.

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 100 points

    Where can I find the IBIS model for the DS26C31?

  • 0 in reply to Gerald Horgan
    TI__Genius 17125 points

    Any available models can be found in the product folder on ti.com in the Design & development tab:

    http://www.ti.com/product/DS26C31T/toolssoftware

    We also have this link to find all models for all products:

    www.ti.com/.../midlevel.tsp