Split coax signal from aerial to Set Top Box?

[dave]

I'd like to feed two STB's from the one aerial on the roof. Is it ok to just splice the coax's that feed the STB's near the aerial? Or will that result in a poor signal to the STBSs? Thanks

[Peter Crosland]

It depends on the signal strength at the aerial. Ask Bill Wright in uk.tech.digital-tv for more details for a more detailed answer

Peter Crosland

[e-mail]

[meow2222]

no, use a splitter. That will deal with the impedance issue, thus avoiding unnecessary signal loss and distortion due to reflections etc. Theyre only 50p.

NT

[The Natural Philosopher]

and are lossy..

better is a small two port distribution amp. Probably 20 quid or so.

[Lurch]

On Sun, 29 Jul 2007 15:55:20 -0700, [e-mail] mused:

Mybe, depends what the signal is like in the first place.

As ever, a completely incomplete answer, basically just bollocks.

[Graham.]

Hope you are taking notes Peter, HI HI de G3ZVT

[Peter Crosland]

I refer the honourable gentleman to the answer I gave earlier!

Peter Crosland

[e-mail]

[Mark Carver]

You should *never* split any RF feed by electrically paralleling it. You'll create a mismatch, and risk generating standing waves in both of the down leads. This could give all sorts of problems, such as some channels noisier than others, and garbled teletext reception for analogue. For DTT you could end up with sections of the CODFM signal in one or more muxes notched out, effectively rendering the whole signal unusable.

You must use either a splitter, or some form of distribution amplifier.

[The Natural Philosopher]

Not the risk, the actuality.

However in many cases the riposte 'so what?' elicits the actual answer, 'er, not much really'

You MIGHT. Then again you might not. Mismatching a cable with a - say -

38ohm double load instead of 75ohm may actually give you a few -6db nulls, bit not much worse than a passive attenuator/spillter does to ALL signals.

And degrade ALL signals 6dB IIRC..

or some form of distribution amplifier.

which is, as I said, the best solution. But costs more.

[Mark Carver]

You can get sharper nulls than 6dB, and I've also seen teletext reception totally obliterated on one or more channels, whilst the accompanying TV pictures look perfect. But like you say it's a lottery.

Yes, but in a uniform manner. If you've got gallons of signal, it's no problem, and even desirable to avoid overloading the tuner(s) if you're very close to the transmitter.

Indeed.

[The Natural Philosopher]

With an open circuit sheath,yes..

Not if you are where I am yo.

[Andy Wade]

True up to a point, but since the aerial and receivers/tuners also present significant mismatches, there's all the more reason to use a reasonably well-matched splitter. SWR ripple tends to build up alarmingly when there are several mismatched devices on one run of coax. (Read up on the flow-graph technique and work out some examples for yourself.) Decent hybrid splitters are readily available[1], so there's no excuse for not doing the job properly.

The dips can be a lot more than that. The worst case comes when one of the split outputs is unused and is left open. A simple parallel connection provides no RF isolation between the outputs, so at frequencies where the open leg happens to an odd multiple of an electrical quarter-wavelength long the junction will see a reflected short circuit giving, in principle, an infinite notch. It's not quite that bad in practice, because of the finite loss of the cable, but you can get nulls of 10-20 dB.

A proper hybrid (aka "inductive" splitter) gives typically 20 dB isolation between the output legs, and largely overcomes this objection. It also helps reduce mutual interference between receivers resulting from local oscillator leakage from the aerial inputs.

About 4 dB in practice for a 2-way splitter.

If (and only if) there is enough signal, a simple passive splitter is the better option.

[1] Such as which I will declare having a vested interest).

[tony sayer]

In fact this week we made up a filter to get rid of Band one frequencies for a studio to transmitter link for a local radio station. The TX system was overloading the aerial distribution amp and a couple of open circuit notch stubs measured at 30 dB down WRT to band 2 88-108 !..

Complete with some harmonics up to UHF!...

[meow2222]

not sure what your issue is really.

A splitter is the usual good practice way to split an ae feed. If the feed needs amplification, and if we're assuming a standard domestic system, then the amplifier should go as near the ae as possible, which is not generally at the split point.

Of course things get a fair bit more complex if you're looking at large scale distribution, but that is going beyond standard domestic ae systems.

NT

[Phil B]

provided it doesn't overload the signal. I had to fit an attenuator on the input of the amp here about 10km from my local transmitter.

Phil