• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 46 subscribers
  • Views 784 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.

TVP5154 Conducted Immunity: Loss of syncs

BigToast
Prodigy 50 points
Other Parts Discussed in Thread: TVP5154, TVP5154A

Hi, I'm EMC testing a product which uses the TVP5154 ( not the TVP5154A )  to decode PAL video and it seems particularly susceptible to conducted interference on the video input/output and mains cables at multiples of 15.625kHz ( most notably 156.25kHz ) 

The test uses a current injection probe clamped around the video input coax cable to induce a sinewave interfering signal into it:  At low amplitudes of interfering signal the video image becomes heavily patterned and then the TVP5154 loses the line sync well within the amplitude limits of the test signal.  

It seems pretty clear that the interfering signal is messing up the line sync detection in the TVP5154 when it is a multiple of the line sync rate. 

The video input connector and and the mains earth are both well bonded to the product chassis - this normally helps protect the video signal from interference applied in this way but does not help in this case.

Are there any particular factors I could look at which could help increase the immunity of the sync detection in the TVP5154?

  • 0
    TI__Genius 16810 points

    Hi,

    Do you observe any difference in performance if TV/VCR mode in REG02h is forced to TV mode or VCR mode?

    Have you been able to quantify the effectiveness of the coax shield at these frequencies?

    Have you tried different coax cables to see if performance is affected?

    With the video cable connected to an o-scope and terminated with 75-ohms, how much induced noise is observed?

    Can you supply a schematic of your TVP5154 input circuitry?

     

     

  • 0 in reply to Larry Taylor
    Prodigy 50 points

    Hi Larry,

    I normally use it in Auto Mode, if I force it to TV mode there is no change in response to external interference. If I force it to VCR mode then the picture 'wobbles' from side to side in normal operation.  The break-up of the picture in response to the external interfering signal is the same as in TV mode though.

    I'm using a good quality 75Ohm coax cable with properly crimped ends and have tried swapping it for a 'very good' quality coax cable used by the EMC test house with no change in performance

    I've established it takes as much as 400mV peak-peak ripple superimposed on the video at 156kHz (measured across the 75Ohm resistor) in order for the TVP5154 to lose lock on the video ( Interrupt Status Register A bit 7 Lock state interrupt changes from 1 to 0)  - so the noise induced must reach at least this level because I can observe loss of lock during the test.( the control microprocessor reports it).

    Im attaching the input schematic, please let me know if you have problems viewing it

    8424.Video Inputs.pdf

     

  • 0 in reply to BigToast
    TI__Genius 16810 points

     

    We are not sure if this is the culpret, but we do have a few concerns about your input circuit.  The TVP5154 clamp circuit is designed for a 0.1uf coupling cap and 37.5-ohm source impedance ( or less).  As a test, you might want to try bypassing the opamps and wiring in a 37.5-ohm-37.5-ohm resistor divider and 0.1uf TVP coupling cap to see if performance improves.   With the current circuit, the clamp circuit may be having trouble with the noise spike.

  • 0 in reply to Larry Taylor
    Prodigy 50 points

    Thanks, I've done as you suggested - I've bypassed the op-amps  and fed the input video into the top of a 37.5Ohm-37.5Ohm potential divider , the signal for the TVP5154 comes from a 0.1uF capacitor attached to the centre-point of the divider.  The performance has definitely improved a bit - I can now couple a 600mV 156kHz sinewave into the video before it starts to lose video lock which is an improvement on the previous 400mV. 

    The video is still visibly twitchy at these levels of interference, even when the video is reported as locked -  occasional groups of tv lines slide a little bit right or left momentarily. 

    Do you have any more suggestions for improvements to prevent this kind of break-up for as long as possible?