Maybe its me, but I pulled cat5e and soldered a s-video tip to it, and it looks okay but crosshatching in the picture. I hooked up composite and it looked better, no noise (well, not as much).

Has anyone had experience in this situation? Is it the wire, the tip, my soldering skills? Do you NOT recommend cat5e for s-video? I only ran it because someone highly recommeded it. Thanks.

Well, I would never recomend it, but in a pinch, I've use Leviton (?) Cat5e to S-Video Quickport Punch Down... I can solder pretty good, but the puchdown guarnatees a solid connection....

Yes, I have used the quick ports also, and they seem okay. I guess I just want to make sure of the proper way to do it. I can solder well, too, but my s-cables and "brand-x" s-cables perform very different, and my solder job looks way cleaner when I look at the job they've done by unscrewing the tip and looking at theirs.

The solder job should be as good as the punchdown, and could easily be better.

IF
The flux in the solder cleaned the surfaces soldered together
The solder flowed over the surfaces
The solder solidified with a glossy surface.

If all those happened, you have not only made a good connection, but you have sealed it against corrosion. A punchdown can be the same good connection, but only the pressure of the punch block against the wire keeps out corrosion, and that can get eaten away much more quickly than a solder connection.


You used CAT5. What did you do with the other two pairs? You should ground them at the end that the signal is coming from, and not connect them to anything at the other end.

You may also find an improvement if you use the video to CAT5 matching transformers that are available. Use two, one for Y and one for C, at each end.

Thanks for the good info, I'll try the grounding suggestion. I didn't think about the fact that the cat5 wire is not shielded, so in theory, grounding the signal end will lead any noise and nastiness AWAY from the input. Thanks.

yeah, that's the reason that some audio cables have arrows on them. The arrows lead away from the grounded shield end.

But CAT5 is good for high frequency stuff because the twist makes sure that any currents induced in a conductor are cancelled out by currents in the opposite polarity induced in the other conductor a half twist away.

I am remembering that shields on these things only drain away electrostatic interference, but suddenly my brain freezes when I try to think about how RG-6 has a shield that is a conductor and we don't see hum induced in it unless it is in a really STRONG field. Is it that RG-6 is a transmission line? Can anyone tell me why audio works so noise-free, and video too fo that matter, on twisted pair, but RG-6 with no twisting is not susceptible to the same kind of interference as, say, lamp cord used for signal wire?

I gather from the preceding posts that the use of CAT5 for the transport of unbalanced audio and video has become somewhat popular. I am surprised that you've all have had as much success as your posts suggest.

CAT5 and other UTP communications cable was intended for use with equipment capable of rejecting any induced common-mode noise by employing a balanced differential receiver or balun transformer. Consumer A/V gear is not properly equipped. Furthermore, video signals (composite, s-video, all common component formats) rely on 75 ohm impedance matching from source to cable to receiver. To use CAT5 to transport audio and video in the manner you've described is a real gamble.

That's not to say that CAT5 can't be succesfully and reliably used for A/V. There are several manufacturers offering receiver/transmitter products that employ either passive balun transformers or active electronics to balance/unbalance the signals and perform the necessary impedance matching. When used with such products, CAT5 can be used to cary signals over very long runs with no appreciale signal degradation. Otherwise, I wouldn't be at all surprised were even a short run of CAT5 to exhibit any of a number of EMI/RFI noise problems. Maybe not at first, but perhaps a few weeks after installation, when a CBer or shortwave radio buff drives by...

For short runs of S-video, I suggest sticking with either siamesed 75 ohm coax designed specifially for the purpose, or two parallel runs of high quality RG-6. (By the way, if anyone knows of a way to neatly terminate discrete coax to s-video, I'd be grateful to hear it. Maybe some sort of pre-fab adapter?)

For long runs using UTP, check out these product lines:
[Link: extron.com]
[Link: kramerelectronics.com]

This message was edited by M_Bruno on 08/09/03 18:59.

Ernie, in response to your question regarding the effectiveness of coax at noise suppression/rejection:

Basically, the outer shield has a low-impedance connection to the chassis grounds of the connected equipment. Induced noise is thus shunted to ground, sparing the delicate signal conductor within. This principal applies to any shielded cable.

By the way, noise induced into an UTP is not of oppostite polarity on each of the two conductors! On the contrary, it is of the same polarity, and idealy the same magnitude and in phase (one reason for the tight twisting). It is the signal itself which is delibrately inverted on one of the conductors during transmission. The signal of each conductor mirrors that of the other, hence the term "balanced." This principal is central to the ability of balanced transmission systems to successfully reject noise. The differential receiver (or balun transformer) is sensitive only to the *difference* in potential between the two conductors. Because induced noise is induced equally on both conductors, it is rejected.

This message was edited by M_Bruno on 08/09/03 18:58.

Tributaries, Synergistic, and Monster are just a few of the companies that produce S-Video "Y" splices for use with Coax cable.

Duh. CMRR. Common Mode Rejection Ratio. Common mode means both conductors have the same voltage in common, and the input rejects this signal -- does not amplify it.

Right.

I am obviously visiting this site too late at night. I should have recognized early on that a balun was not being used at each end of his run, so he will indeed have garbage on the line.

I do this all the time with a Calrad pre-fab adaptor that is anywhere from 6" to 18" long, depending on the stock they have had manufactured. I usually install a five conductor cable with BNCs, using R and G and B for component, and using B and W to carry S video. The adaptors come either with BNC male, BNC female, or RCA connectors. I can get these for you if you cannot find a vendor who carries them. I believe Calrad is carried by A LOT of electronics parts vendors nationwide, and they should be able to order them as well.

Incidentally, I personally discourage using RG-6 for video unless you have the tools and connectors to crimp on BNCs (Calrad has this, too), because you can't solder to its aluminum braid or shield. If you have to terminate RG-6 with an F connector, your connection will sooner or later be flaky with video. RG-59 works well for all current video frequencies. I used to use it for all VHF antenna usage, and that goes up above 200 mHz, so it should be okay for video.

Thanks Ernie. I'll check them out.


True that most coax uses aluminum shield (and copper-plated steel center). Also, by soldering connectors one does not maintain the proper 75 ohm impedance required for faithful transport of video. I use Canare true 75 ohm crimp BNCs and RCAs as appropriate (expensive, but worth it). For applications where the soldering of RG-6 is required, try Belden 9248. It uses all copper conductors (along with aluminum foil). Another benefit: reduced susceptibility to corrosion in salty coastal areas (I first used it on an installation near Montego Bay, Jamaica).

Both RG-6 and RG-59 can be had in sufficient quality to more than adequately cover even high-def video frequencies. Properly installed and terminated, high quality examples of either will handily outperform most "high end" retail-packaged cable at significantly lower cost. The benefit of RG-6 is lower loss due to larger conductor. Over the shorter runs typically encountered in home installations this point is largely academic. Personally, I find it easier to carry one less cable type along with related connectors and tooling, hence my choice of RG-6.

Regards,
Max

(By the way, if anyone knows of

MIT cables makes a good S vid adapter.
uses their one wire system.. Or ya can use RG6
check it out here - [Link: mitcables.com]