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: TI Designs: TIDA-060013: Design for I2C range extension + SMBus battery

Part Number: P82B96
Other Parts Discussed in Thread: TIDA-060013, TCA4311, , TCA4311A, TCA9517


We have implemented the suggested design that is recommended in TIDA-060013

As following:

But also added the I2C buffer TCA4311 between MCU and P82B96.

The communication with I2C Slave is working fine, but with battery (SMBus) there are errors on I2C MCU and I can't communicate with it.

the defined I2C clock is 50KHz.

My guess why the battery is not working its because P82B96 minimum Low-level output voltage Sx/Sy pins is 0.8V and the SMBus uses fixed voltage levels to define the logic 0 (max 0.8V), it that correct?

When I bypassed the TCA4311 and connected 6 to 7 and 3 to 2, the communication with the battery started to work properly.

Is it really valid to add TCA4311 to P82B96? or this design won't work well for SMBus batteries? what solution you advice to make it work?

Waiting for your advices

Best regards


  • Igor,

    Our I2C expert has been notified of this thread and will respond accordingly. Thank you for your patience.


  • Hi Igor,

    "Is it really valid to add TCA4311 to P82B96?"

    The TCA4311A has rise time accelerators and certain ViL requirements which likely doesn't play well with the P82B96 because its VoL is much higher than the TCA4311A's ViL thresholds. You likely aren't resetting the internal logic of the 4311 correctly (it goes through a few checks to make sure the RTAs are ready). The device also has a missing ACK event which may also play an issue into why the 4311 isn't working with the P82B96.

    "what solution you advice to make it work?"

    Is there any reason why the TCA4311 was designed to sit in series with the P82B96? The P82B96 already is an I2C buffer. My only guess as to why someone may do this is to try to lower the VoL from the P82B96, if thats the case then maybe using a TCA9517 and have the A side face the P82B96 may work better. The MCU would see 9517's B side which has a VoL of about 500mV instead. ,


  • Hi Bobby,

    Thanks for your suggestion.

    I will try the TCA9517 to see if it solves the problem.

    Best regards,