• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 240 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCA9803: Use case with ISO1541

Allen Lu
Intellectual 1670 points
Part Number: TCA9803
Other Parts Discussed in Thread: ISO1541, TCA9800, TCA9509

Hi team:

 My customer use TCA9803 cooperates with  ISO1541,

There is a problem in the debugging, which is shown in the following diagram, which is mainly realized as the SDA signal, and the 1 side of ISO1541 cannot pull down the ACK of SDA to the normal level;

1-- We analyze the data sheets of the two chips, the minimum IILC of TVA9803 is 1000uA (1mA), while the IVOL of ISO1541 side 1 is 0.5-3.5mA,

2--Suspect the 1 side of ISO1541, when outputting VOL, the current limit does not meet the IILC of TVA9803;

3--Please kindly help to review this case according to the current hardware design scheme? PCB has been mass-produced;

 Tks~

Block diagram:

Spec details:

  • 0
    TI__Mastermind 23843 points

    Tks,

    The IILC parameter is the minimum amount of current required that the external device must sink from the TCA980x in order for the TCA980x to accept the LOW on the B-side. 

    Since the ISO1541 device has a range from 0.5mA - 3.5mA, it might be an interesting test to see if switching to the TCA9800 would solve this contention issue, since it has a lower minimum IILC value.

    Another test could be to remove the ISO1541 device, remove the 4.7kohm pull-ups from the ADC, operate at a slower data rate,  and see if the system works without the isolation unit. This may also help us to pin point the problem. 

    One more suggestion. In place of the DNI resistors on the B-side of the buffer, try adding a high resistance pull-up >40kohm  to see if they helps the issue. When adding pull-up to the B-side of the TCA980x, it is important to respect the IEXT-I parameter (<100uA). Adding too strong of a pull-up resistor on the B-side could cause issues with passing a LOW to the A-side. 

    Regards,

    Tyler

  • 0
    Guru 244020 points

    The TCA9803's restrictions are listed in section 9.4.1 of the datasheet:

    • VIL (= 0.99 V): This must be larger than the ISO1541's VOL (= 0.8 V); it is.
    • IILC (1 mA): The ISO1541's output must be able to sink this. The electrical characteristics (section 6.9) show that it does. (The recommended operating conditions in section 6.5 describe something different. They say that your circuit must source between 0.5 and 3.5 mA; the TCA9803's 3.3 mA current source does.)
    • RILC (150 Ω): The ISO1541's output impedance must be lower. But it is 0.8 V / 3.5 mA = 229 Ω, which is too large.

    In general, all I²C buffers have some special outputs (current source on the B side of the TCA980x, voltage offset on side 1 of the ISO1541), and you should not connect special outputs to each other.

    Try replacing the TCA980x with the TCA9509, which has the voltage offset on the A side.

  • 0 in reply to Tyler Townsend
    TI__Intellectual 1670 points

    Hi Tyler：

     Tks for your comments, as i see ,the IOL1 of ISO1541 can output 0.5~3.5mA, and the IIIL of TCA9803 need input at least 1mA. So i think the ISO1541 have the capability to handle the sink current from ISO1541. Am i right?

     By removing the ISO unit, I will check with the customer, it may have some restriction for isolation purpose.

  • 0 in reply to Clemens Ladisch
    TI__Intellectual 1670 points

    Hi Ladisch:

     Tks for your feedback.

     Do you think the major reason is the RILC is not match with the design requirement of TCA980X and cause this problem?

  • 0 in reply to Allen Lu
    Guru 244020 points

    The RILC mismatch is the only obvious problem.

  • 0 in reply to Allen Lu
    TI__Mastermind 23843 points

    Hi Allen,

    Please let me know if you have any more questions. 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    TI__Intellectual 1670 points

    Hi Tyler:

     Just check with customer, the test bench is in Japan, hard to modify by their end customer ,so i will try to build a bench TCA9802EVM + ISO154XEVM + customer MCU for test purpose.

    Allen Lu said:
    Tks for your comments, as i see ,the IOL1 of ISO1541 can output 0.5~3.5mA, and the IIIL of TCA9803 need input at least 1mA. So i think the ISO1541 have the capability to handle the sink current from ISO1541. Am i right?

    regarding to this parameter, am i right for this comments?

  • 0 in reply to Allen Lu
    Guru 244020 points

    Yes, the currents match.