• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 35 subscribers
  • Views 265 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.

DSLVDS1048: LVTTL 50 ohms

Louis Thuilliez
Prodigy 70 points
Part Number: DSLVDS1048
Other Parts Discussed in Thread: DSLVDS1047,

Hello,

I need to convert 100 ohm LVDS signals to 50 ohm single matched LVTTL in both directions. I'd like to use the DSLVDS1048 and DSLVDS1047 dirvers. However, I'm having trouble ensuring that the 50-ohm adaptation will be performed on the LVTTL side. I'd probably have to position a termination resistor on the receiver side and one in series on the transmitter side of the LVTTL. Can these components meet my needs?

Thanks,

Louis

  • 0
    Guru 257880 points

    The DSLVDS1048's outputs are not designed to drive a load of 3.3 V / 50 Ω = 66 mA. But a characteristic impedance of 50 Ω is not the same as a 50 Ω resistor; you probably do not need to place a 50 Ω load at the LVTTL receiver. See [FAQ] What happens when I connect a logic device's output to a 50 ohm transmission line?

  • 0 in reply to Clemens Ladisch
    Prodigy 70 points

    Thanks for your feedback.

    I Think he driver current would not be 3.3V / 50 but 3.3V/100 (50 ohms in series on the output + 50 ohms termination resistor. This would make 33mA, but I still don't think the DSLVDS1048 driver accepts this.
    Thanks for the link. However, I'm afraid that connecting a 50 ohms termination device to the LVTTL output of the DSLVDS1048 won't work very well.

  • 0 in reply to Louis Thuilliez
    Guru 257880 points

    If you had two 50 Ω resistors, you would need to drive the line with 6.6 V in order to get a valid LVTTL signal at the receiver. In practice, you never terminate at the receiver (so the receiver's impedance is near infinite); source termination is enough.

    Please note that the source termination resistor adds to the output impedance of the LVTTL output; their sum should match the trace's characteristic impedance. And the DSLVDS1048's output impedance already is higher than 50 Ω, so you should not use any resistor on the LVTTL line.

    Why do you need a controlled 50 Ω line in the first place? You should place the LVDS receiver near the receiving device so that the LVTTL trace is short enough that its impedance does not matter.

  • 0 in reply to Clemens Ladisch
    TI__Guru 65052 points

    Louis,

    Clemens is correct. Our TI experts for this device are out of office until the new year, but we will try to get an official response to you by the first week in January.

    Regards,

    Eric Hackett 

  • 0 in reply to Clemens Ladisch
    Prodigy 70 points

    Hello Clemens,

    Thanks for your detailed feedback. I need a controlled 50Ω to drive measuring instruments through a cable and connector assembly. What would happen if a measuring instrument configured for 50Ω were connected to the DSLVDS1048 output?

    For the DSLVDS1047 (LVTTL input to LVDS output), can we simply place a termination resistor on the LVTTL line (input of DSLVDS1047 and output of a measuring instrument) to make it suitable for a 50Ω configuration ?

  • 0 in reply to Louis Thuilliez
    Prodigy 70 points

    I think I could wire several channels of the DSLVDS1048 to get the 33mA (3V3 / 100 ohms). However, I'm having trouble finding the rated output current information. I can find the typical short-circuit current at 47mA, but no real rated output current (without short-circuit).
    -> Is it a good idea to put several channels in // ? How many should I use?

    Thank you in advance for your help ;)

  • 0 in reply to Louis Thuilliez
    TI__Guru 65052 points

    Louis,

    Thanks for your patience, our LVDS expert is out of office but will be back next week.

    Regards,

    Eric Hackett 

  • +1 in reply to Louis Thuilliez
    TI__Mastermind 45577 points

    Hi Louis,

    I am sorry for the delay here. 

    So I think you are going to have a real hard time achieving this without additional circuitry. 

    The output impedance on the DSLVDS1048 is variable and can go above 50 ohms; generally speaking this part is meant to connect to high impedance inputs and be on short enough of a trace where impedance matching in unnecessary.  (our VOH is tested with only 400uA of current and short circuit conditions could come into play at as low as 15mA).  

    For you direct question - what would happen if you connect the output to a 50 ohm characteristic port? 

    Ideally you would connect the output of the LVDS receiver directly to the port and the distance is short enough where impedance matching doesn't matter. Based off a really quick rough estimate looking at the rise/fall time of the LVDS receiver - it would probably be around ~10mm where impedance matching would start to become important (it will depend on the physical properties of the trace - but this should be an okay estimation). Assuming the port isn't actually drawing a lot of current. If the port is equivalent to a 50 ohm resistor to ground and not the characteristic impedance - you won't get valid levels. 

    If for whatever reason you cannot get the LVDS receiver closer to your target - using a 50 ohm buffer in the signal chain - i.e. LVDS receiver --> buffer --> 50 ohm port.  This does have the tradeoff of adding delays to the datastream and depending on actual port setup it may still be tricky to get the correct values.  So honestly I'd try to get the LVDS receiver closer if at all possible to where you can more or less ignore the impedance matching concerns 

    Just for your final question - do not short outputs together on the receiver as it could compromise devices behavior because you are adding capacitance and additional circuitry to the output which was not intended by the designers. 

    Please let me know if you have any other questions. 

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 70 points

    Hi Parker,

    Thank you for your detailed reply