Satelleite Co-Ax

In article , Pet writes

Yes....

From

CT100 and H109 are slightly more robust but still susceptible to kinking. Being stiffer, they are not as good as SATDS (RG6) for looping between equipment. However, losses are lower than SATDS and they survive outdoors better.

probably ok, but this suggests CT100 is better for the long run to the dish.

Sorry, forgot to mention, Probabally looking at connecting to a "multiswitch" in the loft then running to differrent rooms in the house. I presume losses will be insignificant as a multiswitch presumably amplifies signal as well?

In article , a writes

Reckon that its all a bit academic for domestic use where cable runs aren't likely to be that long.

Try it and see, either it will work or it won't but I reckon you'd have to have very long runs for it not to work, and try not to crush it:)

You need to look carefully at the specs.

Many of them have UHF input as well as from the LNB and a separate gain control for the UHF.

In general, they don't make an increase in the satellite signal level as such, but amplify the signals to the extent necessary to compensate it being potentially split N ways. Some switches have an increasing amplification with increasing frequency to compensate in part for the cable characteristics. There may be an adjuster.

Look at

The point is that multiswitches are meant to be more or less plug and play with standard quattro LNBs and CT100 cable

.andy

To email, substitute .nospam with .gl

Thanks Andy. Well, I've already run the trickiest length of the stuff this evening. As it's (RG6) primarily used for transmission of RF rather than just RX I am assuming it should have minimal losses compared to CT100. besides, maximum distance of each run will be around 10 m and Dish to Multiswitch will be about 3 to 4 m max.

Hopefully I'l be OK.

The attenuation of the stuff that the cable TV industry calls 'RG6'[1] is to all intents and purposes the same as CT100 / WF100 etc. - about 30 dB per

In applications like yours, where the cable lengths are fairly short, loss in the cable is the least of your worries - there's about 55 dB gain in the typical LNB and the system will happily stand 20 dB or more loss between the LNB and the set-top- box without any ill effect on signal quality. Other cable properties are more important than attenuation: screening factor and accuracy of impedance are the main two. DVB-S uses high-symbol-rate QPSK modulation and reflections (due to mismatches) in the IF cable will cause inter-symbol interference and a loss of signal quality [2][3]. This is the usual cause of 'missing channels' in systems where there are several of joints and bits'n pieces in the cable run. For the same reason it's vitally important to use good quality (preferably crimp) F connectors and make sound braid connections in the coax.

The other issue you get when combining LNB outputs and off-air UHF on to a single cable - as in a multiswitch or 'loftbox'[4] system - is degradation of the UHF C/N ratio due to the amplified wideband noise present at the LNB output. Multiswitches and loftboxes include sharp cut-off highpass filters to mitigate this effect, but the performance of the filters in different products varies considerably - so filter performance is often a more important selection criterion than gain/loss.

Good quality and F Type connectors are two descriptions which do not belong in the same sentence. Designed to be as cheap as possible F type connectors have succeeded beyond the wildest imagination of their designers. If there is a worse RF connector (other than the TV co-ax plug) I am yet to see it.

On Fri, 27 Aug 2004 18:50:11 +0100, Peter Parry strung together this:

But, for the mass market they're easily available and reasonably foolproof, (well not really but they're easier to put together than co-ax plugs). For 'non-critical' RF installations I use F connectors in preference to co-ax connectors.

On 28 Aug 2004 13:55:19 GMT, Ian Stirling strung together this:

Spot on they are, I've got one on the back of one of the TV's in the van. Works a treat(ish)!

Is that Latin for "not at all"? You only need to look at the damn things and they fall off.

What always amuses me is the various satellite people who witter on about the qualities of one cable versus another (identically specified) one and then stick these abominable terminations on the end.

In article , Peter Parry writes

Which one are U calling the abominable termination? surely not the F plug?....

On consideration I can see your point - to grace the piece of tinplated scrap sold as an "F Plug" with the description termination is unjustifiable.

In article , Peter Parry writes

Well in my 'umble opinion as a radio comms engineer, I reckons that the abomination is the Belling Lee co-ax plug. The F type is better in many ways. So I suppose that all the cable networks that use F types should have used the BL co-ax plug then?......

That's a bit like saying having you arm amputated with no anaesthetic is nicer than having your leg amputated in similar circumstances. The statement may be true but it doesn't paint the whole picture.

No, any one of the proper RF connectors, TNC, BNC even FME would be a vast improvement on either of the apologies for a plug that you mentioned.

In article , Peter Parry writes

So what's the problem with it then?,

And yes, it isn't a N type, but then again their overkill for domestic applications....

No they wouldn't - anyway getting them in 75 ohm versions and sizes to fit the foam and 5-cell dielectric cables used in the TV distribution world would be difficult.

I disagree fundamentally with your premise. The Type-F connector is an example of good engineering, in that it achieves (or can achieve) excellent performance at very low cost. It _is_ a proper RF connector. Perhaps you've only encountered the cheapest nastiest sort. Take a look at the 'F' series products offered by companies like Cabelcon (Denmark), Gilbert (USA), PPC (USA), Tratec (Netherlands) and others.

Good crimp F-connectors give excellent screening factors (90 dB+) and can provide excellent impedance matching (> 20 dB return loss to above 2 GHz). The fact that, for the smaller cable sizes, the cable inner conductor acts as the centre pin of the plug means that the male side of the connection inherently provides a good impedance match and cost is minimised. Matching performance is primarily determined by the internal design of the female connector, which is usually part of equipment. There are two principal types of inner contact design in use - the original flat leaf spring type, which are generally OK up to 860 MHz (top of the UHF TV band) and the round cage jack type which are suitable for use up to 2 GHz + (top of the satellite first IF band).

You do need to be very careful when choosing adapters between F and other connector series, some of which display very poor return losses (high SWR) and others of which are OK.

The 'F' interface is quite respectable - it's standardised in IEC 60169-24 and Agilent supply a calibration kit (85039B) characterised up to 3 GHz for use with their vector network analysers:

It is a good example of cost engineering - designing something for as low a production cost as possible whilst maintaining the minimum level of performance you can get away with. Exactly the same design criteria and standards as the Belling Lee (IEC) TV connector.

The cable pin versions are all cheap, that was the design aim. The consequence of meeting that aim is that they are nasty.

There are variants of the F type with centre pins which are capable of the sort of performance you would expect of a proper connector. The Amphenol pin F type is an example and as they say "...it became increasingly obvious that existing low performance F receptacles could not be used in high speed cable modems and customer interface units"

I seriously doubt if a cable pin F type connector can achieve anything like this figure at 1GHz.

A VSWR of 1:1.22 is not achievable with a plug using a wire centre conductor. Typically a BNC connector will exhibit a VSWR of 1 :1.1 (26dB RL) at 2GHz.

F connectors using the cable core are one plug where you cannot get data sheets because the performance is inherently unpredictable as half the plug is made out of the cable.

It actually means you have a bent copper pin. Copper corrodes and copper oxide acts as a rectifier - just what you don't need in an RF plug. You also get centre conductor nicks, bending, dielectric migration and metal fatigue. None of these help the performance at all.

It is adequate as a higher frequency version of the Belling Lee (IEC) co-ax connector. As something designed down to a price of pennies each its mechanical and RF properties are just what you would expect for the price you pay.

So is the Belling Lee plug so that isn't a terribly high recommendation :-).

It's a cheap and nasty device which is exactly why you only find it in consumer equipment where cheap outweighs nasty.