• State Resolved
  • Locked Locked
  • Replies 24 replies
  • Subscribers 34 subscribers
  • Views 5422 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.

TXS0102 - Does it support the "HDMI DDC " ?

Sadanori Kato
Guru 36490 points
Other Parts Discussed in Thread: TXS0102, TCA9517, TCA9517A, TCA9617A

Hi,

We are considering to use TXS0102 for 1.8V <--> 5.0V(the connector side) level shifting on the HDMI DDC(VCCA=1.8V, VCCB=5.0V).
So, is TXS0102 able to be applied to the HDMI DDC although TXS0102 supports the I2C bus ?
We are concerned the phenomenon such as the oscillations and especially whether passing the HDMI DDC compliance test.
Could you please suggest us your recommneded parts for the HDMI DDC based on your experience ?

Best regards,
Kato

  • 0
    TI__Expert 6335 points
    Kato-San,

    I would recommend using an I2C specific level translator like the TCA9517.

    Best Regards,
    Nirav
  • 0 in reply to Nirav Patel
    Guru 36490 points

    Hi Nirav-san,

    Thank you for your quick response.
    I understood and will suggest our customer to use TCA9517 for the HDMI DDC.
    By the way, can both the A-side bus and B-side bus on TCA9517 pass the HDMI DDC compliance test ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Expert 6335 points
    Kato-San,

    Sorry for the delayed response on this. I read the portion of the HDMI Compliance Test Specfication (CTS- Verson 2.0) related to DDC testing. I would expect the TCA9517 to pass. Most of the tests involve expected read/write responses by an HDMI source and sink. The one test that was different was support for clock stretching, which the TCA9517 does support.

    I am moving your request from this forum (Voltage Translation/Level Shifters) to the correct forum (I2C) for proper support of the TCA9517. Thanks.

    Best Regards,
    Nirav
  • 0 in reply to Nirav Patel
    Guru 36490 points

    Hi Nirav-san,

    Thank you for your strong support.
    I understood.

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points
    Kato-san,

    The TCA9517 has been used in the past for HDMI DDC channel.

    My only recommendation is to do 1 of the following:

    1) Connect the A side of the TCA9517 to the HDMI cable if possible (static voltage offset has caused weird issues with compliance test, and doing this will avoid those issues).

    2) Have a detection scheme that will pull EN to GND while doing compliance testing.


    The compliance testing would normally be done while TCA9517 is enabled, which will send false data on I2C bus unless EN is pulled to ground. this false data has caused issues with other customers.
  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for sharing your experience.
    I understood and will inform our customer of your information.

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    Guru 36490 points

    Hi Nirav-san and Jonathan-san,

    I have obtained an additional question from our customer.
    Could you please tell us the reason that TXS0102 isn't recommended for the HDMI DDC ?
    In addition, how does TXS0102 work if applying it to the HDMI DDC ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points

    Kato-San,

    The reason is because the TXS part is not a true buffer, and uses rise time accelerators on the bus. I think it would be possible to use it with HDMI DDC channel, but compliance testing would likely be a problem due to the rise time accelerators.

  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for your quick response.
    I understood and will contact you if I obtain additional questions from our customer.

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    Guru 36490 points

    Hi Jonathan-san,

    I have an additional qestion from our customer again.
    Could you please tell me the more detailed reason that TXS0102 cannot pass the HDMI compliance testing due to the rise time accelerators ?
    What is the testing item which TXS0102 cannot pass ?
    (e.g. : Test ID 8-9 DDC/CEC Line Capacitance, etc.)

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points

    Kato-San,

    I did not mean that TXS can't pass HDMI compliance, only that it may face a similar set of challenges as the TCA9517.

    The rise time accelerators will likely make the capacitance measurement small (if it uses a step function measurement), but it is also possible that capacitance could read high because the RTA will mess with the measurements during the procedure.

  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for your support.

    What is the meaning of the RTA ?
    Does it mean that TXS0102 cannot pass the compliance testing regardless of the rise time accelerators ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 17205 points

    Hello Kato-san,

    RTA stands for Rise Time Accelerators.  If you are curious about how the RTA's work please refer to the app note:

    http://www.ti.com/lit/an/scea044/scea044.pdf

    It doesn't;t mean it can't pass, but rather great care needs to be taken with respect to the layout of the DDC traces.  The Rise Time Accelerators do just that, increase the speed of the rise time which can be problematic for radiated emissions because the faster rise times means that the frequency content get moved to a higher frequency which means a the quarter wavelength gets smaller and generally that gets closer scale of the trace.  That means more radiated energy. 

    Please let me know if what I said makes sense.  I can try and create diagrams to illustrate what I am trying to explain if that would help. 

    Sincerely,

    -Francis Houde

  • 0 in reply to fhoude
    TI__Genius 17205 points
    If this is a true concern, then they could go with the TCA9517 but they would have to make sure the A side of the device is connected to the slave/cable side of the device. The TCA9517 doesn't have Rise Time Accelerators, which generally helps with radiated emissions. We would be more than willing to review schematics and layouts offline if this would be helpful.
    -Sincerely,
    Francis Houde
  • 0 in reply to fhoude
    Guru 36490 points

    Hi Francis-san,

    Thank you for explaining so politely.

    Unfortunately, I didn't understand the relations of RTA and radiated emissions.
    So, could you please explain it by using the diagrams and tell me how to make the layout of the DDC traces in detail to pass the compliance testing ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points
    Kato-san,

    The RTAs can be used on the DDC channel.

    Francis is saying that due to the very fast rise time from the use of an RTA, the trace should be as short as possible to help prevent ringing, or noise that gets emitted to other traces.
  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for your response.

    I understood that both of TCA9517 and TXS0102 can pass the HDMI DDC compliance test.
    Then, could you please suggest your recommended parts to me about the following use case ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points

    Kato-san,

    For the configuration shown, I would recommend the TCA9517 and connect A side to HDMI connector and B side to host controller. This will help with capacitance measurement and the voltages you are using are ok.

  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for your quick response.

    The design requirements are mentioned on page 12 of TCA9517 data sheet as below.
    So, is the A-side able to be connected the HDMI connector although the power supply voltage on the A-side is higher than one on the B-sinde ?

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points

    Kato-san,

    The portion of the datasheet you are referring to is the design requirements for an example system. It is not stating the requirement of the part, rather than describign the example system setup. I am adding to our back log to have this confusing note removed or corrected in the datasheet.

    The portion of the datasheet you want to look at is the recommended operation condition section.

    It is ok to have VCCA > VCCB for the TCA9517/TCA9517A. The TCA9617A/B cannot do this, though. Please see below snippet from datasheet. You must still meet the power supply requirements for each voltage rail (B side cannot go less than 2.7V, even though the A side can)

    Update: Note that you will need to make sure that the control has to be able to accept the 0.6V VOL as a valid low, because the VOL on the B side is 0.6V max.

    I would recommend having the HDMI connector be on the A side to avoid the compliance testing issues that I have seen in the past. The reason this device can have issues if you put the B side to the connector is because of the testing procedure used, it is not good for any buffers with static voltage offset:

    1) Input signal from compliance tester swings between 80% and 20% VCC to test Capacitance. The test also requests that the device be powered, this means that the TCA9517 has VCCA and VCCB, and is enabled.

    2) When the input on B side (connector side) goes low, it will trip the TCA9517, which will think an external slave is pulling low (voltage is below 30% VCC), and so the A side will pull low to propagate this signal

    3) WHen the A side goes low, it will also trip the A-side input low threshold (Because A side is now below 30% of VCCA). This will cause the B side internal pull down (of the TCA9517) to activate to propagate a signal back through.

    4) When the compliance tester releases the pin to see how long it takes for the voltage to rise (my measuring this rise time, it can calculate capacitance), the TCA9517 B side pull down is still active, and holding the B side low. However, the voltage will be above the VILC value, so the A side pull down will be released and start rising. Once the A side voltage rises above 30%, the B side pull-down will be released and then the B side will rise.

    5) The compliance tester will see this additional propagation delay as capacitance load, and it estimates the capacitance on the line to be ~20 nF (even though this is not even remotely true).

    You can avoid this entire wrap-around delay if you put the A side on the HDMI connector side.

  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for explaining so politely.

    I understood why TCA9517 cannot pass the HDMI compliance test if the B-side is connected to the HDMI connector side.
    I greatly appreciate your cooperation.

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    Guru 36490 points

    Hi Jonathan-san,

    I have an additional question.
    Could you please suggest your recommended parts to me about the following use case again ?
    I have changed 3.3V to 1.8V on the power supply voltage of the host controller and need to pass the HDMI DDC compliance test.

    Best regards,
    Kato

  • 0 in reply to Sadanori Kato
    TI__Genius 16430 points

    Kato-san,

    In this case, either part could be recommended.

    Both should pass compliance testing, but care needs to be taken.

    If TCA9517 is used and B side is connected to HDMI connector, you will need to ensure that EN is pulled to ground during capacitance compliance testing, otherwise the measurement equipment could get an invalid measurement due to the reasons I stated above. If the tester is more intelligent (and uses small signal analysis to do capacitance measurement instead of 20% to 80% VCC swing), then this is not a concern and will pass capacitance test.

  • 0 in reply to Jonathan Valdez
    Guru 36490 points

    Hi Jonathan-san,

    Thank you for your support.
    I understood and will suggest TCA9517 to our customer.

    Best regards,
    Kato