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.

TPS23881: through the isolation i2c issue

Part Number: TPS23881
Other Parts Discussed in Thread: TIDA-050026-23881

Hi Sir,

the i2c from the SOC and through the isolation IC, but the i2c signal can not be measure at TPS23881 terminal.

may we know what caused it? attached is schematic, please kindly help check whether the schematic is wrong, thanks.

TPS23881 Schematic.pdf

  • Hi Tommy, it looks like the schematic connections are okay. Can you make sure the I2C aren't shorted to ground on your board? Also is the PSE RESET pin released (pulled up) when you take the measurement? Also, the 10k pull seem high and usually they are 5k like in the EVM. Can you help check these on your board? Thanks!

  • Dear SIR
    At present, I have done some DEBUG, but my PES board still cannot measure the I2C signal at the TPS23881 point of the isolation IC (The SOC point has I2C signals).
    Although you have seen that the circuit is okay, but I want to reconfirm with you the problem of I2C SDA.
    The figure below is a comparison of the I2C design difference between the EVM board and my PES board 
    The I2C SDAI SDAO of TPS23881  is directly connected together in my PES board, Is it OK?

  • Hi Tsai-sen, yes you can connect SDAI and SDAO as long as you are using a bidirectional optoisolator. I've seen many designs that use this with the TPS23881 including our TIDA-050026-23881.

    Did you confirm voltage on 3.3V_PSE is powered up? And is RESET of PSE HIGH? Also did you try reducing the pull up resistance?

  • Dear Sir,

    1. TPS23881 cannot be detected by SOC issue :

       The part you suggest I reconfirm, I have performed a test but I still can’t measure I2C. If it cannot be measured, there is no I2C signal to  respond to the SOC, and the SOC cannot recognize the TPS23881 and execute the automatic test on the TPS23881 command.

        In the current state, the I2C and RESET (about 300nS) from the SOC have signals, but after the isolation IC is connected to the SCL SDA   of the I2C on the TPS23881 side, the RESET only has a continuous 3.3V level.

         Debug process as following :

                @ RESET From SOC:

                       SOC RESET is about 300nS from Low to HI, and the RESET time on the side of TPS23881 through the isolation IC is very steep and it goes up to 3.3V almost immediately

                       The I2C PU resistance value on the PS23881 side has been changed from 10K to a smaller value. It has been changed to 1K. It is still the same and only has a continuous 3.3V level.

                @ TPS23881 Self RC RESET:

                       Disconnect the SOC RESET source, change TPS23881 RESET to RC RESET (R: 1K, C: 0.1uF) -----TEST I2C is reduced from a   continuous 3.3V level to 2.2V

                @ Float the isolation IC on the VDD pin on the TPS23881 side ---The I2C of TPS23881 is still not measured.

                @ Tommy suggests to try BY PASS ISOLATION IC to directly connect the I2C of the SOC with the I2C of the TPS23881. The                               measured I2C only has a continuous 3.3V leve

    2. Design confirmation:

               @ SDAI SDAO You also think that it can be connected in two-way isolation IC

                 The two TI isolation ICs we used (U1: ISO7731FDBQ--for reset, U2: ISO1541DR--for I2C) need to  be reconfirmed the circuit carefully.            Especially ISO7731FDBQ (U1 in our circuit) is used for TPS23881 RESET, there is EN1 EN2, need PULL  UP or Floating?

  • Hi Tsai-sen, looking at the datasheet's table, pulling HIGH or floating gives the same results. 

    Can you try not installing RW5? I see our TI Design that it should be floating.

    If this doesn't work, i will need to see a power up waveform of VDD, VPWR, and RESET of the PSE on the same screen capture just to verify that the power up sequence is met. Thanks!