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.

TMP112: I2C tHD_DAT Timing

Part Number: TMP112

Hi,
For the TMP112 I²C timing spec, the datasheet lists t(HDDAT) (data hold time) as 100–900 ns(as table in Section 6.6).
I would like to ask: Is there any risk if t(HDDAT) is below 100 ns?
Thanks.
image.png
  • attach waveform for reference

  • Hello Wills,

    Could you please tell me more about your setup? How many devices will you be using and are you using Fast Mode (Standard Mode) or High-Speed Mode? Depending on the setup there could be some risk if the data hold time is lower than 100ns in Fast Mode since performance cannot be guaranteed below the recommended hold times.

    Regards,

    Harry

  • Hi Harry,

    Our I²C bus (PMIC + Power IC) runs in Fast-mode (400 kHz).
    For the other slaves on the bus, tHD;DAT(min) is not specified in their datasheets(UM10204 Rev7.0).

    However, the TMP112 datasheet specifies a tHD;DAT requirement of 100–900 ns in Fast-mode.

    As shown in the attached waveform, we measured tHD;DAT on our bus and obtained approximately 66.79 ns and 74.18 ns.

    Since the I²C Fast-mode specification does not mandate a minimum data hold time in the same way, could you please clarify:
    Can the TMP112 reliably accept a tHD;DAT in the range of 65–75 ns, or is there any risk when operating below the 100 ns minimum stated in the datasheet?

    We would like to understand whether the 100 ns value is a strict requirement or a recommended margin.

    Thanks.

  • Hello Wills,

    Could you please share the date/lot code from the packaging of the devices in your possession?

    Regards,

    Harry

  • Hi Harry,

    We are not able to retrieve the date code or lot code for the TMP112 units, as the original packaging is no longer available.

    Also, may I confirm: the tHD;DAT behavior should not be related to device date/lot code, correct?
    I assume this timing parameter is defined by the device design/specification rather than a manufacturing-lot‑specific issue.

  • Hello Wills,

    I'm going to move this post to email and will follow-up with you soon. Please keep an eye on your inbox in the meantime.

    Regards,

    Harry