What are the considerations for PCB layout, and connector selection for video signals. Is it necessary to follow 75ohm impedance criteria in a closed system. The pipeline starts at Sony FCB IX 11 A camera core which outputs VBS 1V p-p NTSC signal with no impedance specifications, goes to a processing board (TI processor based), and outputs as a same specification video signal. I do not have real estate budget to accomodate 75 ohm RF connectors in the design and was wondering if I could work with Micro D connectors, and treat video as just another analog signal. The output is not to be transmitted over lengths more than 2 feet.
4
-
\$\begingroup\$ You can get microminature coax connectors; I have some on my board here that are about 3mm square. \$\endgroup\$– pjc50Commented Dec 17, 2012 at 13:40
-
\$\begingroup\$ This really depends on the bandwidth of your monitor and the performance you need. It should not be hard to run some experiments with intentionally "evil" cabling and get a sense of the consequences. \$\endgroup\$– Chris StrattonCommented Dec 17, 2012 at 15:39
-
\$\begingroup\$ "RCA" connectors re commonly used for composite video. Take a look at the back of any DVD player or TV. \$\endgroup\$– tcrosleyCommented Jan 24, 2013 at 20:23
-
\$\begingroup\$ NTSC is only about 3MHz wide. It was very common to see amplifiers made with discrete components at one point, and I'm reasonably sure that the signal path didn't maintain an even 75 ohms all the way through that kind of thing, i.e., with gaps of varying widths between component leads, etc. Impedance matching for video is primarily about keeping signal reflections down, which show up as 'ghosting' and other distortions on the monitor. And I'll second @tcrosley with regard to RCA connectors. \$\endgroup\$– JustJeffCommented May 24, 2013 at 10:26
Add a comment
|
2 Answers
\$\begingroup\$
\$\endgroup\$
For 2 feet, I believe trying 'evil' cabling and result is likley not noticable on screen. For better result, microminature coax connectors, used on Wifi, etc. tends to be 50 ohms. Although not 75 ohms, it is better than Micro D connectors.
\$\begingroup\$
\$\endgroup\$
The amount of video interference due to impedance mismatches will be strongly related to the distance between the most widely separated points of mismatched impedance.
If a composite video signal is terminated at both the source and load (with the source outputting a double-strength signal which it expects to be 50% reduced by the double termination), then one can have a single impedance mismatch or cluster of close-together impedance mismatches somewhere without visibly affecting the picture (since any signal reflected within the cluster will either die down quickly or leave the cluster, and any signal that leaves the cluster will be absorbed before it's re-reflected).
If a signal is only terminated at one end, then even a single impedance mismatch which is a significant distance from the non-terminated end may cause visible interference.
The extent to which an impedance mismatch to a video connection will affect the picture is in large measure a function of what impedance mismatches, if any, already exist.
