• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 46 subscribers
  • Views 586 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.

THS4151 CMV out shows some portion of differential signal

Brian youngpyo Jo
Prodigy 20 points
Other Parts Discussed in Thread: THS4151

Hi

I'm using THS4151 as a video differential driver. I use +-12V power supply.

CMV pin is used not to set reference voltage but to apply some ac signal.

CMV ac signal has 500kHz. but I found small portion of video vertical sync is found in CMV.

Is this normal???

 

Thanks

  • 0
    TI__Genius 17905 points

    Brian,

    The vertical sync signal has high frequency content, so seeing the sync couple into the Vocm pin is not surprising. The Vocm pin is a fairly high impedance node (two 30kohm resistors) which will cause it to be more susceptible to crosstalk (any induced current into these resistors will develop into a larger voltage than if the resistances were smaller).

    Also, I do not believe fully differential amplifiers are designed for good isolation from the main amplifier path to the Vocm pin. In most cases, Vocm is a fixed DC voltage and the pin has a bypass capacitor to ground, filtering out any disturbances on the Vocm signal.

    If you haven't already, try adding an external resistor to ground on the Vocm pin to lower the impedance of the node. 50 or 100ohms would be good values to try, but it will depend on the AC signal source drive capability.

    I'm very curious -- why are you applying an AC signal on the Vocm pin? The video output (taken differentially) will still be the same as with a fixed DC reference at the Vocm pin.

  • 0 in reply to Kris F
    Prodigy 20 points

    Thanks Kristoffer

    Two years ago, I designed differential transmitter electronics which could transmit composite video and audio signal through twisted pair cable (up to 2km) at the same time. This one used +-5v differential driver which has Vocm function.

    Due to limited input range (because previous electronics used +-5V, which means only +-3.2V input range.) I contacted your company at that time, and THS 4151 is recommended for this application. Last year I started to design new electronics which has bigger dynamic range. Due to video pre equalization, video input signal would be greater than 1v, considering Vocm, definately I need more input range...

    Idea is same as before, transmitting Composite Video using Differential out, and transmitting other signals using Vocm.

    Now, I have prototype electronics using THS4151.

     

    All works well except onething.

    I transmit FM modulated (200KHz) audio + 20Hz sine wave signal using Vocm.

    When I take out Vocm from the other side of cable (2km) using two resistor summing node.I found there is 60Hz in Vocm which I believe it is Vsync (from differential Video.)

    According to datasheet, Vocm should be good at 200KHz signal without big interference from differential output.

    I tested like followings.

     

    • No video differential input, apply Vocm: Demodulated audio is good. No noise.
    • Differential video input, apply Vocm, Demodulated audio has big noise.

    (Also, I could see modulated pulse signal has small ringing on the edge and has 60hz Vsync interference when I see CMV output using scope math function (and from receiver ).

     

    For Vocm input, I use simple summing amplifier (but resistor value is relatively high (10k, 20k)) compared to 30k to each supply rail in THS 4151 on Vocm.

    I’m curious this crosstalk is nature of this IC.

     

    Do you have any suggestion?

  • 0 in reply to Brian youngpyo Jo
    TI__Genius 17905 points

    Brian,

    That is a very clever idea to send both video and audio over the same differential pair!

    I discussed this with one of our experts here and he does not see any red flags as to what would cause this issue. He ran some simulations and noted that by creating an imbalance in the internal common mode feedback RC divider (see p. 4 of this app note - http://focus.ti.com/lit/an/sloa054d/sloa054d.pdf), then the differential signal could appear on the common mode extracted at the output using another external summing node as you described.

    How accurately matched are the summing node resistors on the other end of the cable? Imbalance here could also cause some of the differential video to appear on the extracted common mode.

    Do you see the same thing when you extract the common mode right at the amplifier outputs? Or do you have a shorter cable you can try?

    Is it possible to input a sinewave instead of the the video signal in the differential path? You could try applying sinewaves of different frequencies to confirm whether it is truly the differential signal getting into the common mode.

    Can you please send your schematic (at least, the section that involves the THS4151 and the receiver circuit) and an oscilloscope capture of the extracted common mode signal at the amplifier output and at the other end of the cable? A snippet of your layout could be useful, too. You can send them directly to me at k.flores@ti.com and we can review them.