Home analogue RF network: splitters and combiners

[backintime]

Hi everyone,

I finding conflicting advice online and hope you could help. I have three analogue TVs and three media sources (with modulators on different frequencies). I'm hopefully going to add teletext via a Raspberry Pi too, but will work that out later! I want to combine the signals and then split them for the TVs. My question is: Do I need to combine / split the signals with an amplifier, and could I have one line out which is splits into three, or should I look for a splitter has multiple lines out, individually going to the television sets?

Apologies if this is a dumb question!

Many thanks

Will

[OscarFoxtrot]

You might not need an amplifier, depending on the signal levels of the modulators. You'd need a combiner, which could be passive, and then a splitter, which could also be passive.

[kan_turk]

It depends to a degree on signal levels outputed by modulators - in fact some modulators have a loop through facility obviating the need for a combiner - it would help to know details of your modulators (pics would be good) - no such thing as a dumb question - we all had to start somewhere - shock horror

J

[backintime]

Thanks for the replies everyone - the modulators are in the post and I'll take pics when they arrive.

[backintime]

Hi everyone, the modulators arriveth!

The three channels will come from two PROception Mk2 units and one Panasonic NV-J45. Any suggestions are appreciated!

[Nuvistor]

The Proception as a RF output of 75dbuv, that should be enough to drive 3 sets especially if you choose band 4 not band 5 for the channels and cable runs are short. Don’t know about the Panasonic.
75dbuv will be about 5mv if my calculations are correct, if not someone will correct me.

Just realised you want to combine them into one coax and then spilt them again, that could be a bit too much attenuation, still worth trying before buying amplifiers.

[terrykc]

It is much easier to work in dB - makes losses easy to calculate. 75dBuV = +15dBmv (easier to think in mV!

Let us know the output level of the other modulators and if there is a loop through facility.

Also which frequencies you are going to use.

I will then be able to offer sound advice (as I'm sure will others)

On the face of it, if cable lengths are reasonably short, it should be possible to do it with two 3-way splitters.

Do you have any splitters or access to a source of them? If so, what are they?

[G6Tanuki]

If you've got plenty of RF drive available, and you're only working over short distance [as your 'domestic' situation implies] you can easily get away with simple resistive 'tap-off' splitters, as were generally found used in low-rise council-flats etc back in the 50s/60s/70s.

Over such short distances passive resistive combiner/splitters will work fine - provided you're using properly-shielded coax and select the head-end frequencies so they're not clashing with any of the local/semi-local UHF telly-transmitters.

[OscarFoxtrot]

Have a look at the articles on
http://www.wrightsaerials.co.uk/articles/index.shtml
especially Domestic RF distribution systems for television and radio
and Domestic Multi-Channel TV Distribution Systems
and
http://www.wrightsaerials.co.uk/reference/index.shtml
UHF channel planner
This is one of Bill's installations with 11 modulators for CCTV
http://www.wrightsaerials.co.uk/ourwork/022.shtml

Bill can be found in news:uk.d-i-y and news:uk.tech.digital-tv (also carried on Google Groups if you don't have NNTP access)

[kan_turk]

Hi
Both the proception and panasonic modulators have RF loop through so you do not need combiners - just connect rf output of first modulator to rf input of 2nd modulator and rf output of 2nd modulator to rf input of 3rd modulator - output of 3rd modulator will have all 3 channels
Re selection of channels (frequencies) the main issue is that you must leave at least one unused channel between modulator channels as these are double sideband units and adjacent channel operation is not poossible - hope this makes sense - if not post query on here and I'm sure someone will know the answer

Rgds

J

[backintime]

Wow - thanks for the replies everyone!

Frank and Terry - I believe the modulators do have a loop through facility (presuming that's when the modulator has an RF in socket). I don't know the dBmV of the VCR. Can you tell from the picture I've attached? By band 4 and 5, do you mean what the double digit number on the modulator ends with?

The distance from the combiner will be 0, 6 and 10 metres away.

Terry - Two modulators are the same; ProCeption MK2. The third signal is from the VCR. I don't know what frequencies I should use. Recommendations are appreciated.

G6 - Looking into the taps now - thank you!

OscarFoxTrot - Terrific articles, thank you! I fear I'm going to spend the rest of the evening on Bill's website. It's particularly helpful as I don't know anything about dBmv's yet.

[Nuvistor]

Band 4 are channels from 21 to 38, the group 35-38 were not in the original band plan, cable losses in these channels will be less than the higher channels in Band 5.

Take note of keeping clear of broadcast transmissions and adjacent channels as per previous posts.

[backintime]

Originally Posted by kan_turk

Both the proception and panasonic modulators have RF loop through so you do not need combiners - just connect rf output of first modulator to rf input of 2nd modulator and rf output of 2nd modulator to rf input of 3rd modulator - output of 3rd modulator will have all 3 channels
Re selection of channels (frequencies) the main issue is that you must leave at least one unused channel between modulator channels as these are double sideband units and adjacent channel operation is not poossible - hope this makes sense - if not post query on here and I'm sure someone will know the answer

Terrific - thanks J. I'll space the frequencies accordingly. Can I get away with then using a passive splitter after all the loop throughs?

[kan_turk]

Absolutely - there should be adequate levels if you are only splitting 3 ways
Re choice of channels you should indeed avoid local dtt broadcast channels - am I correct in supposing that you are/will be only carrying your modulator analogue signals on your distribution system and not broadcast digital signals as well? - this considerably relaxes the requirements

J

[Scimitar]

Analogue DSB modulators are generally very dirty. Avoid +- 1 and +-5. A mini spreadsheet or manual table is handy for doing the calculations.
I often had problems trying to loopthrough these cheap products, so my favoured approach is a 3 or 4 way combiner. That also allows individual modulators to be attenuated if required. Through losses are covered by the 75dB ish outputs.

[terrykc]

Originally Posted by backintime

It's particularly helpful as I don't know anything about dBmv's yet.

Don't worry, you will soon!

First of all, you need to establish what the end result will be - the level of the signals at each outlet.

This should be a minimum of 1mV - any lower and you might run into noise problems - and a maximum of 3mV (to avoid overloading).

These levels correspond to 0dBmV (1mv) and +10dBmV (3mV). The ideal aiming point is, of course, 2mV which is +6dBmV.

Cable loss won't be a problem to you. A good quality well screened cable should be used - CT100 type preferably - if you want to argue about it, Bill Wright is your man (you'll lose, by the way!).

The later foam filled dielectric versions such as WF100 or PF100 are better as they are resistant to kinking. Losses per 100m at UHF range from 19.8dB at 862MHz down to 14.6dB at 470MHz. As your longest run is 10m, this boils down to 2dB and 1.5dB. You should always use the worst case figure, so 2dB it is. (Always round fractions of a dB up to the next round figure. Apart from anything else, it helps to compensate for things like connector loss, etc. - and, as I said before, working in dB makes things much simpler, so why complicate the process?)

The awkward bit here is the VCR. I assumed from your first post that you were going to use three clearly specified modulators, not two plus a VCR. A third modulator fed from the VCR would make things much simpler!

VCRs aren't usually intended to feed distribution systems, so the output can be expected to be low. (Feeding it with a high level from the modulators may also cause problems as any active devices will be expecting much lower levels from a domestic aerial and putting the VCR first in the queue will introduce further losses which you can ill afford).

I'm not in favour of resistive splitters at the best of times and wonder if stray capacity effects would have unpredictable effects at UHF.

Also, resistive splitters are voltage splitters whereas inductive splitters are power splitters. Taking a 2-way splitter as an example, the resistive version will have a loss of (at least) 6dB whereas a good quality inductive splitter will typically have a loss of 3.5 to 3.7dB, so call it 4dB. Moving on to a three way splitter, the losses will be about 10dB and 6dB.

So, ignoring the VCR for the moment, let's see what we've got: 10dB resistive splitter loss plus 2dB cable loss = -12dB.

The Proception spec says 70 – 75 dBμV, not 75dBμV, so, worst case, the output might only be +10dBmV. 10-12 = -2dBmV, so we've missed the 0dBmv minimum limit we started with!

Use an inductive splitter and we should have +2dBmV, short of our +6dBmV target but still within range.

We definitely need to know the output of that VCR if you aren't planning to get a third modulator. Do you have a UHF LMS - or know someone who has?

Then you could measure it and know for sure.

[kan_turk]

Hi
All of the above advice/comment is absolutely valid - its just that it is difficult to weigh up cost versus performance - you generally get what you pay for
Comments about the VCR being the imponderable are spot on - it would be preferable to have 3 identical modulators
I would not consider anything other than an inductive splitter for the reasons stated
Personally I would consider using a selective combiner to combine outputs of modulators as this would address a number of the issues - dirty modulator output and it attenuates the unwanted sideband etc - a search for 'selective UHF combiner' will throw up a supplier in Slovakia which I would highly recommend (no connection other than being a satisfied customer) - you just specify the channels you want to use and the unit arrives pre tuned (requires 12v power supply) - re choice of channels I would suggest 21, 23 & 25 unless these are used by local dtt mux - it all depends on whether you want to spend another £30

Hope this helps and does not confuse further

Rgds

J

[Nuvistor]

UHF LMS? I have searched but they come back with various information?

Although you require a 3 way, just a word of caution about some 2 way splitters on the market, I bought one and was suspicious of the performance, on opening it the three outers were connected together and the 3 centres were connected together, not really good enough. This was from a big brand supermarket, perhaps I should have known better.

[backintime]

Thanks for the replies J, Scimitar and Terry - really appreciated!

Terry - I've just ordered 20 metres of CT100 coax!

I found a manual online (Though I think we're not allowed to link to manuals?) - it states the RF Output is 75 OHM Unbalanced.

[backintime]

J, do you think this would be a suitable selective combiner?