• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 21 subscribers
  • Views 1278 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.

Using TXS0102 for DDC level translation

Benson Lam
Prodigy 30 points
Other Parts Discussed in Thread: TXS0102, PCA9306

Hello,

I am considering of using the TXS0102 to perform level shifting for our DDC signals. The DDC signals coming out from our processor is 1.8V and we will need to drive it up to 5V before reaching the HDMI connector. I have few questions and I hope someone can enlighten me.

1) Between the HDMI connector and an external monitor, there is a cable and the cable length is unknown since this depends on the cables the end users use. As a result, we have no idea on the bus capacitance. There is a statement in the spec saying: "Both PCB trace length and connectors add to the capacitance that the TXS0102 output sees, so it is recommended that this lumped-load capacitance be considered to avoid O.S. retriggering, bus contention, output signal oscillations, or other adverse system-level affects". We are worried by the words O.S. retriggering. Do you have a range of bus capacitance that will guarantee the One shot not to retrigger if we keep the bus capacitance within the range? Furthermore, how does the One shot know a retrigger is needed?

2) Could you recommend any voltage translator (1.8V to 5.0V and vice versa) specific for DDC application? Ideally for driving long cable as well.

Thanks

  • 0 in reply to Chris Cockrill
    TI__Expert 4705 points

    Hi,

    Please see below feedback for your concern.

    (1)For the TXS0102, it has a one-shot circuit,which will accerlate the rising edge. the one-shot circuit will be open for approximately 30 ns or the output comes to 95% of the input edge, whichever occurs first.

    What the datasheet said, retriggering means , one-shot close and then open, when triggering longer PCB trace, which means lumped-load capacitance is higher, which leads the delay from TXS0102 to the receiver is very long, if the round-trip delay (from TX0102 to receiver, and receiver back to TXS0102) is larger than 30ns, which will leads the retrigerring of the one-shot circuits, and cause the oscillation issue.

    So when you use it, you need to calucate the round trip delay. usually, one inch PCB will cause 200ns delay, you need to calculate the cable delay. and leave enough margin for the design.

    (2)  if the TXS0102 can not meet your requirement, you can refer to the PCA9306 which is the I2C level shfiter and can drive higher capacitive load up to 400pF.

    BR

    Junjie

  • 0 in reply to Junjie Mai
    Prodigy 20 points

    Hi


    for same application we try to use the TXS0108. Already with very short cables we have the oscillation issure.
    Is there no way to use it with high capacitive load? For the HDMI standard the maxium cable is 700pF.


    BR

    Werner

  • 0 in reply to Werner Rohde
    TI__Expert 4705 points

    Hi, Werner

    For the TXS0108 applicaiton, could you please provide the waveform for the oscillation issue? then I can double check the root cause, also better to clarify the detail applicaiton.

     

    BR
    Junjie

  • 0 in reply to Junjie Mai
    Prodigy 20 points

    Yes, of cause. Sometime the oscillation will continue even he DUT at the end of the cabel will be removed.

    BR

    Werner

  • 0 in reply to Werner Rohde
    TI__Expert 4705 points

    Hi, Werner

    Could you please provide the application detail?

    is this the issue for TXS0108 you talked while has the short cable?

    please lists the detail, for how long the cable is, and how much equalivent capacitive load is?

    based on the waveform you provided, the oscillation happens, as I said before, two scenarios maybe here.

    1. The long pcb trace which make the round trip delay larger than the one shot duration time

    2. if the trace is not long, must the equalivent capacitive load is very heavy which makes falling or rising edge can not reach to low or high enough when one shot opens.

    maybe you can also provide a enlarge waveform for the oscillation.

    Also, please provide the schematic. have you add the enough decoupling capacitor for the TXS0108?

    BR
    Junjie