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.

LMH0303 SDI Driver output swing & rise time isues

Other Parts Discussed in Thread: LMH0303

Hi all,

We are using LMH0303 as SDI driver in our design .

We are able to see the output video for SD , HD & 3G SDI formats. But the video looks to be out of standard with respect to Output swing & Rise time .

Below are the observations : 

  • For HD-SDI

             Output swing - 710mV (Typical value - 800mV)

             Rise time - 550ps (Standard value - 270ps )

  • For 3G-SDI

             Output swing -  510mV (Typical value - 800mV)

             Rise time - 222ps (Standard value - 135ps )

This reduction in swing happens only if the data pattern is alternating 1s & 0s . If the data contains long runs of 1s followed by long runs of 0s then the swing will be  near 800mV .

We have tried to increase the input differential voltage swing , but this is not having any effect on output swing . 

We have used the same reference circuit present in the datasheet in our design . 

How can increase the output swing ? Please suggest .

Thanks & Regards ,

Suresha.N.S.

  • Sureshna

    The LMH0303 is characterized using standard SMPTE test signals, and using those signals it will meet all of the relevant specifications for rise/fall, amplitude etc in SMPTE 259, SMPE 292 and SMPTE 425.   The alternating 1/0/1/0 pattern is not a valid SMPTE signal.

    The output swing can be adjusted to compensate for your particular board using the Rref resistor - nominally this is 750Ohms

    The slew rate is switched between two different values with the HD/SD pin

    Mark Sauerwald

     

  • Mr . Mark Sauerwald , 

    To isolate the above issue , we terminated the SDO line just after the capacitor . By doing this we were able to remove the PCB trace path .

    We measured the waveform across this resistor , there is significant improvement in the SDI output signal .

    For example 

    • For HD-SDI

                 Output swing - 780mV (Typical value - 800mV)

                 Mean Rise time - 290ps (Standard value - 270ps )

    • For 3G-SDI

                 Output swing -  670mV (Typical value - 800mV)

    Is this improvement is because of SDI PCB trace isolation ? Please share your thoughts .

    Also , what is the relationship between Rref resistor value and the output amplitude . We are not sure what value to be placed instead of 750E currently mounted .

    Can you please share any document in this regard ? 

    Thanks & Regards ,

    Suresha.N.S.

  • Suresha

    What signal are you using to measure the waveforms with - the input signal should be SMPTE colorbars, if you are still using a square wave to measure the amplitude, the results will be in error.

    Mark Sauerwald

     

  • Mr. Mark Sauerwald , 

    Thanks for your response .

    We are using Xilinx FPGA as the source for the SDI driver LMH0303 .

    We have measured the SDI output with following video sources : 

    • Internal Color pattern generated by FPGA
    • Video input received from HD Camera 

    We are unable to see any difference between the output swing in these two cases .

    As mentioned earlier output swing reduces whenever there is a 1-0-1-0 pattern in the SDI output data during color pattern display or Video display .

    Does the values mentioned in the datasheet such as output swing , rise time correspond only to SMPTE standard color bars ? Will there be any difference in characteristics if some video is sent out from LMH0303 ? 

    Thanks & Regards ,

    Suresha .N .S.