• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 344 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.

TCA9548A: SCL rise time

K.Hamamoto
Guru 12155 points
Part Number: TCA9548A


Hi,

We got some question from the customer and there are some abnormal phenomenon. Can I ask some question ?

The situation,
1. The rise time of SC6 is almost 2.5us. (The input SCL is normal and also the fall time of SC6 is normal. The "normal" means these time are around a few 100ns.)

2. The other SCx signal has same phenomenon as SC6 as well.

(The customer condition is SCLK is 50kHz.)

(Question)
#1 I think the slow rise time such as above value is abnormal. Because it does not meet the I2C standards.
       Standards , Rise time ; <1000ns (max) at 100kHz.
     Is my understanding correct?

#2 We already had changed the pull up resistor but the phenomenon did not changed. (10k -> 2.2kohm)

   Is there any other points to check ?

Best Regards,

Koji Hamamoto

  • 0
    TI__Guru 53575 points

    Hi Koji,

    2.5us is a very long rise time indeed. This time is not a characteristic of a single device, however, and is dependent on the I2C bus design. A slow rise-time on an I2C bus is normally caused by a bus capacitance being too large for the pull-up resistor used. Will you check the following to help diagnose the source of the problem?

    1. Do you have a large number of devices on the bus, long bus trances, or other sources of parasitic capacitance present on the bus? This shows up on all SCx lines. Do all lines have similar typologys/capacitive loads?

    2. Did you see no difference in rise-time between 10k and 2.2k pull-up resistors? Or were the differences only slight?

    3. How are you calculating the values needed for pull-up resistors? This calculation should take into account Vcc, Vol, Iol, t-rise, and C-bus to ensure proper low values and rise times are met. Info on these calculations can be found here:
    http://www.ti.com/lit/an/slva689/slva689.pdf

    Regards,

    Eric

  • 0 in reply to Eric Schott1
    TI__Guru 62920 points

    Hello Hamamoto-san,

    I wanted to see if we could still help to better understand the root cause of the abnormal behavior.  Have you reviewed Eric's questions above?  Or, have you found some explanation for the issue already?

    Thank you,
    Max Robertson

  • 0 in reply to Miguel Robertson
    Guru 12155 points

    Hi Max-san, Eric-san,

    Thank you for your support. I have been investigating this issue.

    And I borrowed the customer board. I am going to check some behavior.

    Anyway, I summarized this issue for your question. Could you give me any advice if you have any concerns?

    1.The cables are not connected. The trace of the bus is around 2cm.
    2. We did not see the waveform actually. But the customer says there is no differences.
    3.They just added a few kohm resistor because they do not know C-bus value.

    Now I also got the waveform with the pull up resister is 5kohm and Vcc=3.3V. Please see the following waveform.

    I have the schematics of the customer's board . But I can not post the schematic on E2E site. Because of the confidential information.

    Please give me your email address if you need some information.

    For more detail, I would like to talk on private message or Email if you mind.

    Best Regards,

    Koji Hamamoto

  • 0 in reply to K.Hamamoto
    TI__Guru 62920 points

    Hamamoto-san,

    I'll mark this thread resolved for now based on the e-mail discussion.  If you have further questions please feel free to let us know.  To summarize the findings, the slower rise time was found to be due to probe capacitance.

    Regards,
    Max