• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 3 answers
  • Subscribers 33 subscribers
  • Views 498 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

P82B715: Regarding VCC and the pull-up resistor voltages on the L and S sides

hiroshi.yamada
Intellectual 535 points

Part Number: P82B715
Other Parts Discussed in Thread: TCA9617B, LP5018

Tool/software:

I'm thinking of using a P82B715 to communicate between an I2C device master and a slave connected by cable.
The I2C device operates on +3.3V.
For example, I could connect the P82B715's VCC and the power supply for the L-side pull-up resistor to +5V, and connect the power supply for the S-side pull-up resistor to +3.3V.
Are there any problems with the specifications and operation?
Best regards,
Hiroshi Yamada
  • 0
    TI__Guru 55808 points

    Hi Yamada-san, 

    About how long is the cable length? You might be able to use a singular TCA9617B to drive long cabling since its IOL spec can re-drive 30 mA of current. 

    P82B715 is usually used in a pair which means two devices. A singular P82B715 is more expensive than TCA9617B. TCA9617B can also level translate by itself. 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Intellectual 535 points

    Hi Tyler-san,

    Thank you for your comment.

    The cable length is planned to be 3 to 4m.

    I will also check the TCA9617B. Is there any problem with the relationship between the power supply and VCC of the P882B715 pull-up resistor?

    Please leave a comment.

    Best regards,

    Hiroshi Yamada

  • 0 in reply to hiroshi.yamada
    Intellectual 535 points

    Hi Tyler-san,

    Additional confirmation and information.

    The connection is as shown in the diagram below, but on which side should the TCA9617B be placed?

    I'm currently in the design process and the deadline is approaching, so I'm in a hurry, so please comment as soon as possible.

    Best regards,

    Hiroshi Yamada

  • 0 in reply to hiroshi.yamada
    Intellectual 535 points

    Hi Tyler-san,

    It looks like the question you asked earlier can be answered by posting in the forum below.

    It would be best to configure a TCA9617B on both the controller and cable sides, with the A side of both facing the cable side.

    TCA9617A: I2C over long cable - Interface forum - Interface - TI E2E support forums

    Best regards,

    Hiroshi Yamada

  • 0 in reply to hiroshi.yamada
    TI__Guru 55808 points

    Hi Yamada-san,

    Application information in section 9.1 says P82B715 can operate with a supply voltage from 3 V to 12 V, but recommended operating conditions is 4.5V to 12V. 

    The Sx/Lx logic levels are independent of the supply voltage, so Sx/Lx could be at 3.3V while VCC = 5V. 

    The main issue I see with this setup is VIL of LP5018 is VIL = 0.4V. 

    In order for TCA9617B or P82B715 to work, we need to ensure that the VOL at the target is < 400mV otherwise this will not work. 

    TCA9617B should be on the controller side with B-side facing the MCU. The MCU must have a VIL = 0.3 x VCCB where VCCB = 3.3V for this to work. Since VOLB of the TCA9617B = 0.54V, VIL = 0.3 x 3.3V = 0.99V which allows margin for the large output voltage of the buffer. 

    A-side VOLA is much lower (VOLA = 0.23V max) and should reach the LP5018 VIL requirement if A-side faces the target. 

    The bottom implementation should work. 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Intellectual 535 points

    Hi Tyler-san,

    Thank you for your comment.

    I understand.

    If TCA9617B is inserted on both sides, the input conditions of the LP5018 will not be met.

    I overlooked that.

    Thank you.

    I will consider your suggestion.

    Would a twisted pair cable be appropriate?

    Would it be a twisted pair cable such as CAT5, which is normally used in LANs?

    If you have any other cables you can suggest, please let me know.

    I look forward to hearing from you.

    Best regards,

    Hiroshi Yamada

  • 0 in reply to hiroshi.yamada
    TI__Guru 55808 points

    Hi Yamada,

    I am not an expert in cabling, but I know a lot of RS-485, CAN, differential signaling IC's use twisted pair in order to cancel out EMI effects and effects from cross-talk. 

    For I2C, the signaling is slow. Not sure how much added benefit can occur using twisted pair for I2C, but I don't think it would hurt. 

    Twisted pair or shielded cabling should work fine. If you need more rugged communication protocol, consider using RS485 or CAN for its differential signaling. 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    TI__Guru 55808 points

    Hi Yamada-san,

    Another implementation that is valid is to use two TCA9617B buffers but face the B-side towards the MCU: 

    I am a little worried with the implementation. I am not sure how much series resistance 4 meters of cabling provides to the system. There will be a voltage drop across the cable. 

    In any case, whether 1 x TCA9617B or a pair of TCA9617B's is used, you would need to ensure that VILB <= 400 mV at all times. This might tuning the PU resistors so that the rise-time is longer while trying to keep VOL at a minimum. 

    If I had to choose, I think a pair of TCA9617B's in the above orientation is the best option.

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Intellectual 535 points

    Hi Tyler-san,

    Thank you for your quick comment.

    I will consider using one TCA9617B on the MCU side.

    Also, I will consider using twisted pair cable or shielded cable for the cable.

    Also, thank you for your alternative suggestions such as CAN.

    Best regards,

    Hiroshi Yamada

  • +1 in reply to hiroshi.yamada
    TI__Guru 55808 points

    Hi Yamada-san,

    If no further questions, I will close the thread at this time. 

    Regards,

    Tyler