DSAT attenuation - Bill Wright?

Since the change of coding for DSAT BBC Four HD it has suffered pixellation on our Panasonic Freesat TV and Blueray/HDD recorder. Digging around the 'net indicates this is some weird hardware problem for certain Panasonic Freesat kit, that we have...

I've checked the alignment of the dish and have full strength and high quality signals preety much across all Freesat channels. Recently SWMBO'd complained about pixellation on BBC Four *SD*. So checked around the three receivers. Only BBC Four HD on the telly showed pixellation, so not convinced the report was correct but I'm not going to go there.

Anyway half unplugged the cable from the LNB on the telly, signal strength drops one unit, bit error rate drops to zero and pixellation disappears. Properly connect the LNB and the bit error rate increases and pixelation returns. To me this indicates that the signal is "too good" and is overloading something in the telly - the obvious solution is an inline attenuator.

An attenuator would need DC pass through and something above 2100 MHz but what value of attenuation? I have no reference for how much reduction in signal you get from a partial unplugging, or the range that the builtin signal meters operate over. I'm caught between thinking 3dB not being enough and 6dB being too much and messing up the other channels (which are all OK in the partial unplugged condition). erring to 6db?

Actual devices I'm looking at Vision V60-103 or V60-106. At some point there maybe a multiswitch installed, so they'll fit onto the board that'll be mounted on quite nicely.

Reply to
Dave Liquorice
Loading thread data ...

Ideally you need to get a BER reading on the affected muxes. This is the sort of thing where an analyser will usual find the problem.

Why should the signal be too strong? It's extremely unusual. Most receivers will handle anything. There aren't any line amps or active equalisers are there? A multiswitch with a lot of gain?

More likely than too much signal is too much inequality between muxes. Some receivers don't like that. An attenuator can help though, a bit.

Are you using a multiswitch or a multi-output LNB? How long are the cables?

Possibly related, but how is the LNB powered when you half disconnect the feeder?

Are you certain about dish alignment? Sidelobes can receive from adjacent satellites and affect some muxes but not others.

A faulty LNB can add noise that varies greatly across the band. Same with a faulty multiswitch (noise gets into the RF from the PSU).

Damp in the feeder can have a frequency-selective effect, but in general will drop higher frequencies the most. Trapped feeders and dodgy wallplates can have weird effects.

Local interference is occasionally an issue. This can be in the actual transmitted band or in the first IF (the LNB output). If the latter it tends to get in at the receiver, so switch nearby suspects off. Also check connectors for screen continuity.

Better go. I'm being yelled at.

Bill

Reply to
Bill Wright

BBC 4 HD shares the same transponder (11.023 GHz Hor) as:-

BBC 1 HD Scotland BBC 1 HD Wales and BBC News Channel HD

If there's a reception problem with the t/p, it will equally affect those other three channels ?

Reply to
Mark Carver

problem

Thanks Bill.

The telly does have a BER on the "signal condition" page. With the LNB feed half unplugged that drops to zero and stays there, no pixellation. Fully plugged in and it varies quite abit in the 10^-6

10^-7 range and pixellation occurs.

Good question, note that this a "known problem" since BBC Four HD changed coding and only affects some (older) Panasonic equipment and has been determined to be hardware related. Not all vulnerable equipment is affected and there are some hints that the cabling/LNB etc has an affect.

Nope, Quad LNB, 3 outputs in use, one to the telly, two to Blueray/HDD recorder, third has cable attached but is unterminated.

Apart from the odd "minority" channel the built in signal meters of either the telly or HDD are pretty even across all channels and LNB feeds. The weak channels pixellate a little bit occasionally bit but they only reach about 3 or 4 out of ten and the bar changes to orange. BBC Four HD is 10/10.

Quad LNB, about 7 m from dish to where a multiswitch might be fitted at some point. Cables are simply barrel joined at that point with another 15 m or so of cable to the various bits of kit. Cable is foam dielectric rather than air spaced.

Via the commoned, at the LNB, cable screens I expect.

See above about even signal strength across the vast maority of channels.

Reply to
Dave Liquorice

Fairly sure I've checked those. It's the obvious thing to do when you get a reception problem, check that other channels on the same transponder are like. So fairly sure it's only BBC Four HD that suffers.

Reply to
Dave Liquorice

If attenuating seems to fix it, how about (if you have any) adding in a longer coax feed as an attenuator and a cheap way of confirming? Db losses for the cable can then be calculated.

Reply to
Harry Bloomfield

Seems obvious to try an attenuator then. Can't you just buy a 3 and 6 and experiment with them singly and in combination?

The signal meters on tellys etc can be extremely misleading due to scale compression etc.

This isn't about signal strength; it's about s/n ratio.

Bill

Reply to
Bill Wright

Unfortunately we do very occasionally get apparent problems with reception that confound the obvious rule that everything on the same mux should behave the same. I have no idea why this is. My first instinct is to blame faulty reporting, but I'm not 100% about that. I'm wondering if there's something else happening here.

Bill

Reply to
Bill Wright

That will increase any slope across the band, so if the affected channels are near the top they are more likely to be affected by stronger signals on lower frequencies.

Incidentally, some of the very early Sky boxes had to have a minimum of

10m of cable between them and the LNB.

Bill

Reply to
Bill Wright

Monitoring this thread, if it is only one of the many on that transponder then the effect surely has to be down to poor error correction in just one stream on a multiplex. its hard to see how this could be so, considering the whole of the signal has presumably to be decoded to get at the channels.That would imply some kind of secondary processing exists in the system and only applies after the separation of the streams in the set. Therefore, by implication the errors must be being fixed by something in the other channels. Very mysterious.

Brian

Reply to
Brian Gaff

Agreed, Bill, but it could provide a useful check if some spare cable is available:

Establish the IF frequency of the mux in question, then calculate the required loss using the follwing data for the loss in dB per 100m (this data is for WF100 cable)

Frequency dB

1000 Mhz: 19.9 dB 1350 Mhz: 23.4 dB 1750 Mhz: 27.0 dB 2150 Mhz: 30.3 dB 2400 Mhz: 33.5 dB

The slope will be particularly important if the wanted IF is fairlow low in frequency but it doesn't matter for this check.

If the IF is around 1000 MHz, the loss is 1dB per 5m, so 30m will make a 6dB attenuator. Try it and see if it helps. If it does, try again with 15m (3dB).

If that works, check all the other muxes, bearing in mind that a mux at around 2400 MHz will be attenuated by ~5dB. If there are problems at thios end of the band, replace the cable with a length of 9m. (3dB)

If that works, check all the muxes including the problem one. It that also works OK, then job done. Otherwise, you can be confident that an attenuator with a flat loss of 3dB across the band will work.

Time consuming and a lot of messing about, but it could be worth it in the end!

And, of course, this is a DIY group, after all!

Reply to
Terry Casey

I had the unworthy thought that if you put a bit of tape (e.g. duct tape) over part of the front of the horn then you would attenuate the signal.

Depends if your dish is easily accessible of course.

Cheers

Dave R

Reply to
David

affect

I lied.

Those all suffer along with Cbeebies HD (ie daytime BBC Four HD). Again partially unplugging the LNB cures the problem for all those channels...

Wanders off to check all available Freesat HD channels... Only additional HD channel affected from the EPG is ITV HD on Freesat 111 (11053 MHz H). But of course that is a regional (Tyne Tees) variation on ITV. Manually tuning a couple of other transponders to look at other ITV HD variants shows they suffer as well.

These are all coming from Astra 2G (28.2E) with "23000 3/4 8psk" coding. Other HD channels on Astra 2G using "23000 2/3 8 psk" are OK (NHK World, RT, Daystar UK. etc).

Not as simple as a coding issue though. BBC 2 Wales, BBC 2, BBC 1, BBC1 NI, CBBC from Astra 2E (28.5E) also use "23000 3/4 8psk" and are all fine.

Reply to
Dave Liquorice

Yes but it introduces another variable: it alters the slope. So experimentally it doesn't really help. It might solve his problem though.

Bill

Reply to
Bill Wright

That would worsen the signal/noise ratio, which is quite different to reducing the signal level.

Bill

Reply to
Bill Wright

cabling/LNB

Mainly 'cause although the HDD recorder seems OK now, it has "played up" in the past. Indeed it was pixellated recordings that flagged up the issue. So to be sure that it's not going to come back I need to deal with at least 3 feeds and I have an aversion to paying as much in P&P as the goods...

Perzackerly, haven't any idea how no linear (or not) they are. So no real clue as to what a drop of say two "units" means in real terms.

Presumably the "quality" measure is a vague indication of that. Mind you that doesn't vary much (tending to no change) between the fully connected and partially connect LNB states. Dispite the BER staying firmly at zero in the partially connected state and never being zero in the fully connected and pixellating state.

Reply to
Dave Liquorice

1273 MHz for the Astra 2Gg 11023 GHz transponder that carries BBC Four HD etc.

Near enough for jazz, I'll see if I have 25m ish of cable lying about. If nothing else it'll give some indication of the scale of the builtin signal meters and if 3, 6 or even 9 dB is required.

Reply to
Dave Liquorice

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.