• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 6 answers
  • Subscribers 35 subscribers
  • Views 166 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.

TCA9617A: I would like to inquire about the TCA9617A VOL value.

DONGYEONG KIM
Prodigy 20 points
Part Number: TCA9617A
Other Parts Discussed in Thread: TCA9517, TCA39306, TCA9803, TCA9617B

Tool/software:

Hello, I am currently designing regarding TCA9617A and would like to inquire about the VIL value, especially regarding the B channel.


On DATASHEET, the VIL value of SDAB and SCLB is fixed at 0.4V, but is it fixed at 0.4V regardless of the value of VCCB?
And if correct, why is it fixed differently from SDAA and SCLA?

Since the VOL of the slave IC you're planning to use is higher than 0.4V, it should be checked for compatibility.

And I also used the TCA9517, but it operates based on the VCCA and VCCB supply voltages.

  • 0
    Guru 316510 points

    These values are correct. The TCA0617A requires that other devices on the B side output a lower voltage than its own output. This is necessary for it to be able to determine whether the buffer itself or some other device pulls the line low.

    The I²C specification requires that a device does not output more than 0.4 V with a pull-up current of 3 mA. If your device has weaker outputs, then you might be able to reduce the output voltage by lowering the pull-up current.

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Thank you for your response.

    I have a follow-up question regarding the VIL value. According to the datasheet, the VIL of 0.4V is based on the input threshold of channel B. However, the VOL of the slave device connected to channel B is 0.5V.
    Given this, it seems that using the TCA9617A might not be feasible. From what I’ve seen, most slave ICs have a VOL of 0.4V or higher, which makes it difficult to connect them to channel B. Is my understanding correct?

    Also, would using the TCA39306 for channel B be a viable solution? I’d appreciate your advice

  • 0 in reply to DONGYEONG KIM
    Guru 316510 points

    What is your slave device?

    The TCA39306 is a passive switch; it cannot reduce the low-level voltage, but it also will not increase it by a large amount.

  • 0 in reply to DONGYEONG KIM
    TI__Mastermind 47128 points

    Hi Dongyeong, 

    DONGYEONG KIM said:
    Given this, it seems that using the TCA9617A might not be feasible. From what I’ve seen, most slave ICs have a VOL of 0.4V or higher, which makes it difficult to connect them to channel B. Is my understanding correct?

    You are correct. If the target I2C device has a VIL = 0.4V, connecting this target to the B-side of the TCA9617A is not feasible since the VOLB of TCA9617A is 0.52V (typical) which would be higher than the input voltage of the I2C target device. 

    You would need an I2C buffer that has lower VOL. TCA9803 is another option where VOLB is much lower I think VOLB = 0.26V (max) in its case. 

    If you need only level translation in your application and have no need for buffering, TCA39306 could be an option, however, like Clemens already suggested, TCA39306 is a passive switch and cannot reduce the low-level output voltage, but it also won't increase it that much because it has low RDS_ON between its input and output. 

    Regards,

    Tyler

  • 0
    Prodigy 20 points

    I understand your explanation about the TCA39306.

    Currently, the I²C bus is operating at 400 kHz with 4 slave devices, and the distance between the master and the slaves is approximately 2 meters. Due to this long distance, we considered using the TCA9617 and have already tried the TCA9517, but we are looking to improve performance.

    The master side is not a concern, as it meets all requirements. However, the issue lies in the B-side of the TCA9617: the VIL (input low voltage) between the slave and the TCA9617 is 0.4V, which is problematic.

    Would inserting a TCA9803 between the TCA9617 and the slave help resolve this issue?

    Also, could you explain why the B-side VIL of the TCA9617 is lower than that of the A-side?

  • 0 in reply to DONGYEONG KIM
    Guru 316510 points

    The A and B sides are different because only one side needs a voltage offset to break loops.

    The TCA9803 can help. But it has additional restrictions on the pull-up resistors.

  • 0 in reply to DONGYEONG KIM
    TI__Mastermind 47128 points

    Hi Dongyeong,

    If we are using the following application: 

    Two series TCA9617B's should be okay even with 2 meter of cabling. 

    I usually recommend not having B-side face the long cable side, but with 2 meter of cabling at 400 kHz, this application should work just fine. 

    We can scratch the use of the TCA9803 if we are using series TCA9617A's. 

    Regards,

    Tyler

  • 0
    Prodigy 20 points

    I tried to solve this issue by placing the TCA9803 between the TCA9617 and the SLAVE device, since the VOL of the SLAVE IC is 0.66V.
    I understand that TCA9617A is not compatible with TCA9803, but can I assume that TCA9617B is? I will refer to the datasheet and take the pull-up resistor constraints into account in the design.

    Also, could you explain why it's not recommended to connect channel B to a cable? If this must be considered, the MASTER operates at 1.8V, so additional level shifting would be required.

  • 0 in reply to DONGYEONG KIM
    TI__Mastermind 47128 points

    Hi Dongyeong,

    DONGYEONG KIM said:
    I tried to solve this issue by placing the TCA9803 between the TCA9617 and the SLAVE device, since the VOL of the SLAVE IC is 0.66V.
    I understand that TCA9617A is not compatible with TCA9803, but can I assume that TCA9617B is? I will refer to the datasheet and take the pull-up resistor constraints into account in the design.

    You could do something like the above. THis should work as long as "A" side of TCA9803 is connected to the B-side of TCA9617A/B (only difference between A and B version is powered off high-impedance is supported by TCA9617B). 

    No PU resistors should be present on the B-side of the TCA9803. 

    DONGYEONG KIM said:
    Also, could you explain why it's not recommended to connect channel B to a cable? If this must be considered, the MASTER operates at 1.8V, so additional level shifting would be required.

    It's not recommended because the cable adds parasitic inductance, resistance, etc. 

    Resistance which adds to VOL and trying to maintain a VILC on B-side could become difficult. 

    Inductance which adds undershoots when the buffer drives towards GND strongly. You might need to add a series resistance to dampening the affects of undershoot. 

    Regards,

    Tyler