• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 1777 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.

Original question:

TCA9517: TCA9517 and TCA9617A

TCA9617A: TCA9517 and TCA9617A

Richard Yang1
Intellectual 1220 points
Part Number: TCA9617A
Other Parts Discussed in Thread: PCA9517, TCA9517

Hello team:

   My customer has used the following Wrong i2c buffer serial connection in NXP's PCA9517 before and no issue is reported, but recently, they tried to use TI's PCA9617A to replace it, but found the communication problem. if used the correct connection way, TI solution is ok.  so customer want to know the reason why this  issue occurs only in TI socket, could you help to elaborate the reason and give your comments?

     

  • 0 in reply to Richard Yang1
    TI__Guru 73061 points

    Hello Richard,

    PCA9517 and PCA9617A are completely difference devices and use different logic.

    I also believe the NXP version uses a different triggering mechanism than we do.

    The customer should also avoid B to B connections anyways as both NXP and TI do not recommend using the device with B connected to B because of how the logic low detection works with the static voltage offset. Our device also uses a fall time accelerator which shoots under the ViLc temporarily. This could temporarily trigger a logic low to A side and immediately trigger a logic high due to the offset. To me, this could potentially could ringing/oscillations. In most cases I would suspect that he master would not be able to send signals to the slave because VoLB is larger than ViLc. I am not sure if NXP's requires the same or uses the same logic low detection.

    Thanks,

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 1220 points

    i2c waveform in the two i2c buffer.pptxBobby:

        thanks for your verification and suggestion, Customer will produce a notification to correct their circuit, but before this, we may need to provide some details to them.

       I will add below information in the report, can we help to list more detail in 9517 behavior (Like TI 9517 IC internal circuit logic diagram why port A ACK can’t pass to port B) when we use B to B connection? It will easier for  our customer know this issue root cause.

  • 0 in reply to Richard Yang1
    TI__Guru 73061 points

    Thanks for the waveform Richard.

    Can you tell me which device is U26 and which is U27? (which one is TCA9617 and which is TCA9517?)

    What is VccA? and VoLA(U27) at the 9th clock cycle?

    What I expect to happen/see is U27 B side goes low but U26A side remains high but that doesn't seem to be the case. Something else is going on here...

    Thanks,

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 1220 points

    Bobby:

       since customer need to make the circuit and layout to A-B, but this is big change because the previous design is not like this, so , we need your help to elaborate this to convince more of the customer members, I attached the circuit and customer's response here.schematic.pptx

  • 0 in reply to Richard Yang1
    TI__Guru 73061 points
    Hey Richard,

    What is the VoL on the 9th clock cycle on U27 SDA A side? We may want to try shorting R396 (33 ohm resistor) and seeing if we still see the same results. I wondering if VoL >ViLc which may be causing some kind of bug? (This is the first time I've seen this before)

    Also what is the clock frequency?

    Thanks,
    -Bobby