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.

TMP103: might have conflict issue with FM TX device (KT0806L)

Part Number: TMP103

Hi Support Team,

We have TMP103A and other slave devices connected in the same I2C bus.

Currently we find one of the device (FM TX, KTMicro KT0806L) sometimes will have ACK error during read operation.

for example:

KT0806L READ REG01 register shoud get value 0xdb

but sometimes get ACK error and make unexpected WRITE REG01 procedure (0x22)

After discussion with platform vendor, we did some test listed below

1. Remove all the devices except KT0806L => KT0806L work fine without ACK error

2. Only remove TMP103A => KT0806L and other devices work fine without ACK error

3. Remove all the devices except KT0806L and TMP103A => sometimes KT0806L will have ACK error on read operation

Regarding the test result, platform vendor suspect this might be the conflict issue (which impact the repeated-start procedure) and suggest us to ask for TI's support

1. Is there a deglitch function in the TMP103A?

2. Could you please give us some suggestion how to debug this issue ?

Thanks,

Herry

  • Hi Herry,

    I don't see an ACK error in your pictures. It looks like normal, successful communication with the device at address 6C. Is there some misunderstanding of error condition?

    thanks,

    ren

  • Hi Ren,

    That is also confused me. It is supposed to execute the READ REG01 operation like the figure 1 but the I2C master get a ACK error during this procedure.

    I measure the waveform shown on the figure 2 and observe it become a WRITE operation.Then I try another READ REG01 operation and get value 0x22.

    The only clue I have is to use TMP103A and KT0806L together can reproduce this issue. Hope my description can help understand this unexpected case.

    Thanks,

    Herry  

  • Hi Herry,

    Do you see a change in bus timing when the devices are removed? The devices add some parasitic capacitance to the bus. You could try a smaller pull-up resistor or slower frequency. 

    I'm not sure I can help you with the information you've provided so far. There isn't an error to debug.

    thanks,

    ren

  • Hi Ren,

    Could you please tell me what should I observe about the bus timing you mentioned ? I did the test to reduce the I2C frequency from 100K Hz to 50K/25K Hz, but it didn't solve the problem. (Only TMP103A and KT0806L are connected to the bus) 

    I'll try to increase the pull strength to verify this issue. Thanks for your suggestion.

    BRs,

    Herry 

  • Hi Herry,

    Parasitic capacitance will slow the rise when transitioning from logic low to high. You don't have enough time resolution to say for sure, but your clocks don't look great. Pull-up resistance is the only way to address this after the PCB is designed.

    thanks,

    ren

  • Hi Ren,

    After removing 1K ohm resister from the I2C bus, the FM TX works fine with TMP103 enabled. I'll file another issue if I have further question.

    Thanks for your support.

    BRs,

    Herry