Looking For A New Twisted Pair Video System

[Repeater]

Our TP video transport system from stage to HD projector died this week. We use it for HD computer video at typically 1280x720 resolution. Distance about 400 feet from computer to projector and using CAT 5e. No audio. So we're looking for a new system.

Tell me what you're using these days, and how. If appropriate here, tell me what works well and what doesn't. We want good equipment, and will pay for it. Having said that, we still have a budget (don't we all!).

Thanks!

Laryn

 

[museav]

I've always had good luck with Magenta Research products, they're a leader in that market for a reason. Other manufacturers I typically specify are Extron, FSR, Hall Research and possibly Kramer Electronics. There are a multitude of other brands that may or may not live up to their specifications and that may or may not have good support. There seem to be many distributors or dealers selling off brand products for lower prices that if they don't work will take them back, although usually with you paying for all shipping, however many of my projects can't afford to experiment and hope it works and keeps working.

With 1280x720 resolution and a 400' run length of CAT5e cable, skew could be an issue so you might want to consider a receiver that incorporates skew compensation.

 

[cpf]

I'll second the Magenta Research and Extron suggestions. I've never bought the stuff, but I've used and adjusted both and they work great, it's pretty much plug and play with a bit of knob twiddling to perfect the picture.

 

[Kingcaffeine]

I use Extron and Kramer all the time and they're rock solid. Extron makes a CAT5e skew-free cable, but it's a little pricey. It kind of depends on the end device, but I've never had any problems or noticed a huge difference. Power from two sources can cause some grounding issues, so look into that.

 

[TheDonkey]

I think I saw in some other thread that cat6 isn't compatible with Video systems because of the higher frequency spec or something?

That didn't really make sense to me because I figured the spec specified the maximum frequency, not the only operating one. Could someone explain whether this is true and why?

 

[Chris15]

I think I saw in some other thread that cat6 isn't compatible with Video systems because of the higher frequency spec or something?

That didn't really make sense to me because I figured the spec specified the maximum frequency, not the only operating one. Could someone explain whether this is true and why?

I believe the comment you are referring to is one I made. Yes it is absolutely true. Standby for a full explanation

 

[museav]

I think I saw in some other thread that cat6 isn't compatible with Video systems because of the higher frequency spec or something?

That didn't really make sense to me because I figured the spec specified the maximum frequency, not the only operating one. Could someone explain whether this is true and why?

The issue is that to help with crosstalk, the pairs within a UTP cable each have a different twist rate, in other words each pair is twisted with a different number of twists per unit distance. To improve performance in network applications, CAT5e uses greater variances in twist rates between pairs than does CAT5 while CAT6 uses even greater variances than CAT5e.

The challenge with this for video is that over longer overall cable distances the higher twist rate pairs represent a longer total length of conductor than the lower twist rate pairs. As a result, you start to get individual components of the video carried on different pairs of the cable traveling different lengths and thus arriving at slightly different times. This is skew. When you start getting to higher resolutions and scan rates this difference in arrrival times can result in things likes the red, green and blue components arriving offset to one another and you may see 'ghost' red and green images to either side of the main image.

The solution is to either use 'low skew' or 'skew free' cable or to use receivers that can apply skew compensation, essentially delaying the signals so that they all line up. If you have to use standard CAT cable for audio and video over UTP applications then CAT5 is preferred, followed by CAT5e with CAT6 being a less desirable option.

 

[FMEng]

Belden and Mohawk both make low skew cable for video over twisted pair. It isn't terribly expensive. You do have to be careful where it is deployed because you don't want someone trying to use it for ethernet. It won't work well for data, but it's great for video.

 

[Repeater]

Thanks everyone for good info.

Yes, we've run into serious skew with the 400ft. of CAT5e that we are using. Vertical lines in our test pattern were separated by >1 inch. In our case it was easy and inexpensive to fix. Our projector input is 5 BNC RGBHV. We just mixed/matched appropriate additional lengths for the R and B coaxes by trial and error. Took about 20 minutes and turned out perfectly.

Laryn

 

[museav]

Yes, we've run into serious skew with the 400ft. of CAT5e that we are using. Vertical lines in our test pattern were separated by >1 inch. In our case it was easy and inexpensive to fix. Our projector input is 5 BNC RGBHV. We just mixed/matched appropriate additional lengths for the R and B coaxes by trial and error. Took about 20 minutes and turned out perfectly.

I remember doing component and RGBHV for broadcast and production facilities years ago where we would have to be sure to cut every cable to the same length. Just pulling them together would sometimes result in sufficient differences to matter, so it was often a process of pulling one small cable to get the run length then laying that out in a straight line along with the cables to be run, pulling them taut and cutting the cables to the exact same length.

Another trick for audio and video over standard CAT cable is to use a different jacket color for the cable, get some off color like purple, green, pink or salmon or whatever so that it is easy to differentiate the AV dedicated cable from the CAT cable used for network, building controls, life safety, etc. Off topic, but I knew a freelance tech that brought his own purple tie wraps to every job, it made it easy to look in a rack and see what work was his, a concept that I think can be very useful even if just used to differentiate subsequent changes to a system from the original work or prior revisions.

 

[Chris15]

So I said I'd give a full explanation later, here goes...

One of the biggest impediments to high frequency performance in a twisted paiur cable is crosstalk from adjacent pairs.
If all the cable pairs were to be twisted at the same rate, then they would all couple electrical noise into one another in perfect sync. So to try and mitigate that, different twist rates are used.
For Cat 5e, one source cites twist rates of:
Blue - 64.8 twists / metre
Orange - 56.2
Green - 65.2
Brown - 51.7

As Brad mentioned, different twist rates create different electrical lengths for each pair. Since the speed of light is a constant, longer electrical lengths take longer periods of time to travel. The difference between these travel times is what we call skew.

Let's consider a XGA output running at 1024x768 pixels and a refresh rate of say 75 Hz. That's a 59 MHz signal when you multiply 1024x768x75.
So one pixel's information is transmitted in less than 17 nano seconds...
One Belden Cat 6A cable has a quoted maximum skew of 45ns / 100m. If you had a 70 odd metre cable run, then your colour information could be arriving 2 pixels late. This is not exactly desirable for a good image...

Belden make a product specifically for Video over UTP type applications, called NanoSkew. It has a specced maximum skew of 2.2ns / 100m. To aid in achieving what Brad referred to, it is available only in Maroon jacket...