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.

P82B96: P82B96

Part Number: P82B96

Dear Support team,

Good day.

I have a problem of I2C communication error with P82B96 on our board.

It seems to I2c receiver can't receive I2C address & data

due to Low voltage floating(≒1.08) of SCL/SDA outputted from P82B96. ※with 4.7~2.2kΩ pull-up

In general, I think I2C Vth(Low) spec of receiver is 0.3 Vcc as I2C standard,

If Vcc = 3.3V then Vth(Low) is 0.99V.

However P82B96's Vol spec - Max = 1V can not meet 0.99V as above calculation.

In addition, pull-up current also pulls up the I2C-Low voltage,

I think this Vol spec is barely below the level for normal I2C communication.

Could you help me to answer the following questions.

(1)My understanding is correct or not about Vol spec?

(2)Solution to fix my I2C communication error.

  • User,

    Our I2C expert has been notified of this thread and will get back to you with a response.

    Regards,

  • Hi User,

    Do you have a schematic and I2C block diagram you could share?

    Do you have an o-scope shot of the VoL from the P82B96?

    The value of 1.08V as a VoL is a bit higher than the typical we spec in the datasheet at 3mA. Are you doing some kind of temperature testing? Is there any kind of series resistor on the bus?

    -Bobby

  • Dear Mr. Bobby,

    Good day,

    Thank you for your reply.

    I will send a I2C block image and actual o-scope shots.

    Different Low voltages  between Input and Output is confirmed as below.

    Input from Master :0.2V

    Output  from P82P96 with our pull-up R :1.08V

    ->Our Slave IC can not receive 1.08V as I2C Low voltage.

    We confirmed this result under 25℃ condition,

    and if remove 4.7kΩ(Output pull-up) then Low voltage still 1.08V.

    Could you provide a solution to fix my I2C communication error

    with exchanging for discrete part, for compatible your product ,,, etc

    Best Regards,

  • user6534594 said:

    Dear Mr. Bobby,

    Good day,

    Thank you for your reply.

    I will send a I2C block image and actual o-scope shots.

    Different Low voltages  between Input and Output is confirmed as below.

    The different low levels from the Sx/Sy side and the Tx/Ty/Rx/Ry side is a design feature meant to prevent the device from locking itself low. 

    Input from Master :0.2V

    Output  from P82P96 with our pull-up R :1.08V

    ->Our Slave IC can not receive 1.08V as I2C Low voltage.

    It looks like your using a 22 ohm series resistor between the pull up resistors and the device's Sx/Sy, this generates a higher VOL which seems to be why we are above the 0.99V. 

    Have you confirmed that the \I2C controller is able to talk (receive an ACK) if you jump wires from the left side to the right side? 

    We confirmed this result under 25℃ condition,

    and if remove 4.7kΩ(Output pull-up) then Low voltage still 1.08V.

    Could you provide a solution to fix my I2C communication error

    with exchanging for discrete part, for compatible your product ,,, etc

    We do not have a pin to pin device to replace this device with. It also seems like the I2C system is set up improperly as this device is meant to be used in pairs and not by itself.

    Best Regards,

    Can you try to short the 22 ohm resistors (make them 0 ohms) or short the left side to the right side and confirm the issue is VoL>ViL?

  • Dear Mr. Bobby,

    Good day.

    I confirmed your 2 proposals as below.

    Could you advise how to fix this issue is best?

    (1)Short 22 ohm resistor (make them 0 ohms)

    → I2C Low Voltage : 1.08V (No ACK) → 0.98V (No ACK)

      [Opinion] Not changed "No ACK" issue , Not enough Low voltage improve.

    (2)Short the left side to the right side

       (2-A)If short between input and output with Vcc power supply.

    →Master can NOT output I2C waveform.

        Please explain why this behavior occur?

       (2-B)If Short between input and output without Vcc power supply(Hi-z).

    →Master can communicate I2C with Slave1 & Slave2 normally. (← I think this solution is best so far.)

      [Question1★] If use P82B96TD without Vcc power supply(Hi-z) as below,

    Is it right that "Sx,Sy" and "Tx,Rx,Ty,Ry" are to be absolute Hi-z condition?

                              And are there any side effect used P82B96TD in case of this condition?

    Best Regard

  • (2)Short the left side to the right side

    (2-A)If short between input and output with Vcc power supply.

    →Master can NOT output I2C waveform.

    Please explain why this behavior occur?

    [Bobby] The device's Tx/Ty driver is basically driving the Sx/Sy when Sx/Sy sees a low. You get stuck in a active low due to the output seeing the input and the logic cycles back and forth. Its the same concept as why we add a large VoL on the Sx/Sy side, to prevent a lock up like this.

    (2-B)If Short between input and output without Vcc power supply(Hi-z).

    →Master can communicate I2C with Slave1 & Slave2 normally. (← I think this solution is best so far.)

     [Question1★] If use P82B96TD without Vcc power supply(Hi-z) as below,

    Is it right that "Sx,Sy" and "Tx,Rx,Ty,Ry" are to be absolute Hi-z condition?

    [Bobby] When the device isn't powered, we technically do not guarantee anything. I suspect there may be some leakage current, especially since the Vcc pin is floating and could act as an antennae and pick up noise.

    And are there any side effect used P82B96TD in case of this condition?

    [Bobby] I believe just leakage current.

     

     

    Thanks,

    -Bobby

  • You mean, the (2-B) solution has some side effects unfortunately,,,

    If so, is it better to short In-Out and remove the entire P82B96TD IC from PCB

    than Vcc power less use-case?

    Best Regards,

  • Yes since it also seems like the P82B96 isn't being used properly based on the block diagram.

    -Bobby