Alright here is the deal. We have a client who has a distributed video system consisting of a Kramer VS-44HC 4x4 matrix component switcher, 3 separate displays (1 Toshiba model unknown, 1 Marantz model unknown, and 1 Sony model unkown). There are only two distributed sources, an Escient Fireball and DirecTV receiver. We are using Niles Audio baluns to get our component video over CATV to the displays.

The problem is the video is all over the place. Sometimes it will come in and out, other times it will be really dark, etc, etc.

We have tried re-terminating ends, using pre-made component wires, swapping out baluns, trying powered baluns, changing resolution on both source and display, swapping out the Kramer, testing the new Kramer in a different location (which worked fine without baluns). Nothing works. All the displays are affected differently (different video problems), but none work correctly.

I am really at a loss here, we have been fighting this system for a few weeks now to no avail. I need some help guys, anyone been through anything remotely close to this or have any advice??? :confused:

This happens with all sources? Because Macrovision can do this. When I got my first DVD player, I was in my teens, and I hooked it up to my VCR, and the picture would fade in and out and it was unwatchable.

All sources. Different results on each. Escient seems to be saturated, while the DirecTV seems very dark.

Total guesses: electrical power issues and/or heavy EMI/RFI interference from something.

Does the Kramer use sync transitions in the source video to determine the "switch" boundary? If it does and macrovision is present, it could be faking out the Kramer sync detection. Most of the cheap component switches don't switch on sync. BTW, what is CATV?

I think he means coaxial cable, the type of cable that "cable TV" (CATV) uses.

I figured it out:

CATV = CAT-5 (twisted pair)

That may be your problem. A balun is a 1:1 impedance device. The video terminations are 75 ohms (unbalanced) which implies that the balanced side also face 75 ohms. CAT-5 twisted pair is roughly 110 ohms, which means that you are misterminating the system. With the "wrong" lengths of CAT-5 cabling, you could be experiencing severe waveform distortion - enough to cause the type of problems you are describing. You need to get an oscilloscope on the video outputs to see what is happening.

A balun is a balanced/unbalanced converter. It does not necessarily have a 1:1 impedance ratio. You may be right about the CAT5 being what he meant though.

Yeah, I think the OP is talking about component video cat5 balun. The fact that component video is 75 ohm and cat5 is 110 ohm is a non-issue - the balun takes care of that. Another possibility for the problems: crap TP wiring. Is it really cat5? it might be cat3. If it is cat5, that means nothing if it was not installed right - and no - plugging a computer into it and not seeing any problems does not mean it is ok. Ethernet connections use error correction and packet re-transmission, marginal or just down-right bad wiring can still work "ok" for a LAN, but use a balun with an analog signal path on bad TP wiring and you could see the kinds of problems described.

EDIT: get a long cat-5 cable and use it to bypass the building wiring for one display. If the problem disappears, then it is the wiring.

CAT-3 and CAT-5 exhibit the same basic impedance (110 ohms). The primary difference between them is that CAT-5 has a tighter controlled twist to reduce high frequency crosstalk. Real baluns have a particular winding structure which is based on integral turns ratio: i.e. 1:1, 1:2, 1:3 .... Impedance ratio is the square of the turns ratio, so for baluns, the impedance ratios are: 1:1, 1:4, 1:9 .... All of the baluns except the 1:1 ratio have a winding shorted to ground and can't be used for component video. Some vendors also sell RF transformers as baluns, but they are not the real thing. Remember, you need a 30 Hz to 30 MHz frequency response to transmit HD component analog video. The ONLY thing a 1:1 balun will do is convert 75 ohm imbalanced to 75 ohm balanced. If you connect to any of the CAT-x cables to this configuration you are creating a mismatch which will result in waveform distortion. Distort the waveform too much and the device at the end of the line won't respond to sync pulse. Component analog video belongs on 75 ohm coax cable - period.

Oh, when you're right, you're right. And you? You're always right!

Seriously, though, a balun designed for this purpose is going to have impedance matching built in, though you are right that the term "balun" does not necessarily imply that.

Does the Kramer use sync transitions in the source video to determine the "switch" boundary? If it does and macrovision is present, it could be faking out the Kramer sync detection. Most of the cheap component switches don't switch on sync. BTW, what is CATV?

Kramer engineers say it is a direct pass through.

I figured it out:

CATV = CAT-5 (twisted pair)

That may be your problem. A balun is a 1:1 impedance device. The video terminations are 75 ohms (unbalanced) which implies that the balanced side also face 75 ohms. CAT-5 twisted pair is roughly 110 ohms, which means that you are misterminating the system. With the "wrong" lengths of CAT-5 cabling, you could be experiencing severe waveform distortion - enough to cause the type of problems you are describing. You need to get an oscilloscope on the video outputs to see what is happening.

Yeah sorry bout the confusion there. Cat-5. The lengths vary depending on the display. The closest display (the main) is only roughly a 20 foot cat 5 run. Shouldn't be a problem with that short a run. I could see length being an issue on longer runs and switching to an amplified balun, but in this case it should work just fine. Will definitely try the oscilloscope though and see what I come up with.

Yeah, I think the OP is talking about component video cat5 balun. The fact that component video is 75 ohm and cat5 is 110 ohm is a non-issue - the balun takes care of that. Another possibility for the problems: crap TP wiring. Is it really cat5? it might be cat3. If it is cat5, that means nothing if it was not installed right - and no - plugging a computer into it and not seeing any problems does not mean it is ok. Ethernet connections use error correction and packet re-transmission, marginal or just down-right bad wiring can still work "ok" for a LAN, but use a balun with an analog signal path on bad TP wiring and you could see the kinds of problems described.

EDIT: get a long cat-5 cable and use it to bypass the building wiring for one display. If the problem disappears, then it is the wiring.

Have tried using pre-made and custom wiring on both the component and cat 5. Neither solve the problem. And yes it is cat 5.

Well this one is puzzling. Is it possible that both sources are bad? I would try this: get a known good source (dvd player) - for the display that is 20' away - connect directly to display via short component cable. Then try 20' component. Then try 20' component + Kramer + DVD player. My guess is EMI/RFI and/or power issues.

Are you sure that you didn't get polarity reversed along the line somewhere? Always a potential problem with balanced interconnects. Video has to maintain absolute polarity sense to work properly.

So here's something even more confusing to add to the mix:

As we were troubleshooting this 'little' problem we ran into an interesting phenomenon. With only two baluns hooked up, one to the main display (ROOM 1) and another to another room (ROOM 2), we would disconnect the feed from the switcher to the outgoing balun on ROOM 2. As we tried to re-connect it we noticed that just touching the tip from the component feed of the switcher to the balun would cause adverse effects on ROOM 1's display. Over and over, touch tip to balun and it would screw up the picture.

So with this phenomenon repeatedly occuring, we tried it with different outputs and it is the same on all outputs. (Switched ROOM 1 to ROOM 3, etc) We tried all sorts of scenarios and it seems to be across the board.

We ordered a different switcher and are going to try that, it seems there is something weird with the switcher that is just not jiving with the baluns.

** UPDATE **

So after the second swap out of the Kramer, it finally resolved the problem. After opening it up and taking a look inside, it looks like the solders on the video inputs would fall apart when applied with a small amount of heat.

Seems as if in this application, the heat of the rack was breaking apart the solders causing the video interference. This would explain why, when taking out of the application, the video problems could not be re-produced.

Good times. Thanks for your input guys.