Analog TV v's Digital

General Discussion Forums Life, The Universe and (almost) Everything!
2

Ever since TV went digital (apart from the initial period - which was ‘EEEK’) I now find digital annoying, for odd video breaks, and far worse, breaks in speech, or missing words altogether. This I find the most annoying. So - I’m still using an analog antenna, backed up by an analog amplifier - wide band, Philips, which gives me around 26 DBm of gain.

Do I blame the broadcasters, the antenna/amplifier, or does anyone else notice this, even if using a digital front end, or if not - is that my problem?

Worse case seems to be reports from far away places, where they too are using digital formats to report - often from mobile phones? ‘Comedy of Links?’

3

its your kit I reckon.  Freeview reception should be 99.99% flawless, (but in my experience only in normal quality, the HD version can be a bit less stable)  assuming that’s what you are viewing and not using satellite. But I suspect it works better with a modern aerial designed for it with decent cable between that and the tv/receiver box. That has to be your starting point… 

 

4

According to the comments below I’m talking rubbish so post removed. Sorry.

5

The rf signal for ‘digital TV’ is still an analogue waveform.  It is extremely important that any rf amplifier does not distort it with clipping or non-linearity.

I was involved with the very early BBC tests on digital TV (leading up to OnDigital), being in a fringe area with lots of ghosts my location was ideal. Digital TV actually benefits from what would have been ghosts on analogue TV because it can take several bites of the cherry to grab the signal (‘diversity’ is the key word), but if there are too many small ghosts too close together then it becomes confused.

I was having dreadful problems to begin with and eventually we found that my recently installed “lo-loss” cable was totally useless because the open weave braid outer let in too much mini ghosting.  As soon as I changed my 10 metre long down lead to 100% cover CT100 (satellite cable) all the mini-ghost problems went away, and the quality reading jumped from 20-40% to 95-100%, signal strength improved a little bit. 

The two other big improvements I added a year later were changing to a big Triax brand aerial with a built-in Bal-Un type connector card (very important, don’t buy a TV aerial without one) to correctly match the balanced aerial to the unbalanced coax, and a new ultra-linear pre-amplifier to preserve the shape of the rf signal with no analogue distortion of the digital signal ‘constellation’.  Both the aerial and pre-amp are no longer on the market, but they are still working well almost twenty years later.  I expect the more modern units will be a further improvement as the technology progresses.

 

6

Terrestrial Digital TV uses the same bands of frequencies as the old analogue channels 21 to 69.  Each of the original 8MHz wide carriers is now filled with a ‘multiplex’ carrying a varied number of digital channels. The nature of the digital multiplex means there is no co-channel interference so all the adjacent channels can be used. There might also be some 4G stealing the odd old channel number, but it should not interfere if the TV aerial is correctly terminated to a fully screened down-lead with an internal  Bal-Un.

Allocation of multiplexes still tends to follow the old analogue channel groupings, eg in London a Group A aerial is still relevant.  However the opportunity to increase the numbers of program channels in other regions might have upset the previous Group arrangements, hence why in some places a ‘Wideband’ antenna is used.  However a cheap Wideband aerial is not suitable for London fringe areas because its gain is often inadequate for the bottom two multiplexes (22, 23) and the old Group A aerials are still better.

Also make sure any patch leads are also using good thick cable with 100% foil screening and not the thin cheap and nasties given away with set-top boxes and TVs.

Check the Freeview website for local reception info. and how many digital channels you can expect, but bear in mind the technical advice is way out of date - who still uses a set-top box with a SCART connector for watching Freeview?

 

7

Sorry for the poor information above some of it just nonsense.

Digital TV as we have it via Freeview is a number of channels say 30 to 50 but not all on at the same time mixed or muliplexed into around up to 7 to 8 broadcast streams or muxes depending on what transmitter you receive your signal from.

Assuming you have good reception and that may not be the case for all the muxes, an adequate aerial and a good quality down lead that has not got water trapped in it in most cases you will receive a reasonable picture quality.

In saying that some muxes are more compressed than others to get more viewing channels carried on it, therefore running nearer their limit or providing slightly less clear pictures some times more noticeable than other times.

While from time to time you may get some break in signal, there are from time to time problems with the multiplexing and coding equipment that are usually the cause of what you are complaining about. In that case as an industry it is a case of the amount of data through the system is just to near the limit of existing equipment. I notice it from time to time as you say.

With regards to the problems with outside broadcasts, that is a problem with the outside broadcast equipment that again is right at the edge of data transmission technology.

8

An excellent modern terrestrial aerial that will work anywhere in UK, uses proper F-connectors and will reject 4G is this one, it is well worth the extra over the cheap ones.

https://www.amazon.co.uk/AERIAL-LTE-UNIX-108451-TRIAX/dp/B0118KAKIM/

9

Well I was going to say no-one commented about the basic problem I mentioned, but thanks to Eddie- someone has.

I’ve had eleven years of experience in TV distribution, having worked for two years with an outfit in Canada on a ‘pay as you go’ local system, as system, cable and route planner and another nine back here with Telefusion (contracts division) where I was combined planner, engineer and supervisor and did quotes too. That was on what was called communal television. CATV. However, having had to learn and relearn different disciplines several times during my working life, last TV job is now over 40 years ago. That means I’m a bit rusty, as you get when you move on from one job to the next. “Forget that - learn this instead.” (Ten times over a working life!)

My home system was planned and installed by me, on CATV principals, so cable is RG59U with all joints either terminated with co-ax plugs (good ones) or matched Tee’s and splitters, finishing at 75 ohm outlets for 4 television points (and four TV’s) around the house. There were more, but I’ve diss’d a couple. I didn’t install the original antenna, one of our crews did, I hate working above 10’ off the ground! The whole system is 75 ohm matched, and the cable is close wound screening (cris-cross weave pattern) so while not the PD 375 cable we used for main distribution - which did have solid screening, we never had problems with it.

I’ve not seen - anywhere, what digital matching is, but I agree that a balun on the antenna is essential- as older antennas were 300 ohm inherently. The whole system covered VHF and UHF, up to channel 69, as BBC 2 started off at UHF, and was way up there from square 1. We got good cover from Sutton Coldfield originally and probably still do, but when the first antenna mount went, a fitter insisted we covered Waltham instead.

So thanks for the equipment tips, but I’m not convinced the problem lies with it. On the other hand, not convinced it isn’t either! Qualified CATV system engineers are thin on the ground these days, and many of our old systems now have individual antennas - whereas under old rules, council estate tenants weren’t allowed. Telefusion ceased to exist many moons ago, having made diabolical mistakes under bad management. (Give a bad accountant free reign - and chaos rules.)

10

 

Hmm, RG59 is OK for short runs and lower frequency stuff such as baseband video and CCTV and it has a good voltage rating, but it is quite lossy at UHF, especially when compared with a modern satellite cable. At work we always used expensive PSF1/2 and PSF1/3 for baseband video and SDI; this was for predictability and in preference to the more generic assorted RG59 types because those had such great variations in performance from different manufacturers.

Another Hmm, is it the same with all the TVs?  With your matched Tees and Splitters, (directional couplers as well?), do you leave the unused outlets unterminated?  If so there will be a reflection back up the cable.  Always leave a split line fully terminated using dummy load 75ohm plugs where necessary, or replace the split with a barrel.  To illustrate the point, when we needed to put a colour-sub-carrier notch filter into a lead, the simplest minimum delay filter was a simple stub of coax on a Tee-piece (I forget the exact length it varied slightly depending on the type of cable and was tuned down to length with a wire-cutter and watching the destination burst amplitude on a vector scope).

If you want to be dazzled into brain numbness the wiki article on OFDM is very good indeed, they seem to have got it right (as far as can I remember from two decades ago). Amongst many other points it highlights the requirements for managing the analogue portion of transmission, ie the rf, very carefully.

https://en.wikipedia.org/wiki/Orthogonal_frequency-division_multiplexing

A simple test is to put a single piece (no joins) of CT100 satellite cable from a TV to a good bal-un equipped aerial and point that at the transmitter from a window in the house.  If the problems vanish, then you have the answer.  If they are still there, the plot thickens.

And no, I don’t know what ‘digital matching’ is either…

11

Richard - will reply tomorrow (which it is already - 12.30 am) so got half way down that Wiki article, and to both our surprise, I understood most of it. So going to bed - early for a change!

12

O/K - Further thoughts on the issue.

As I have no protection against 4G, other than the set itself (which apparently seems unaffected) I may get round to swapping the antenna, for the one you (Richard) suggested. However, deeper thought has invoked some distant memories, one is research we did in the workshop when some CATV users complained of low - or very low signals from BB2, while the other channels were still on VHF. (that’s a long time ago!) As tenants further down the main feed were fine, it seemed odd that one earlier on the same feed was down.
We had one brilliant engineer, a former TV service bod, who fed up with that routine had joined us instead.He suggested that frequency was the main culprit, with length of UHF cycle much shorter than VHF, and perhaps the problem lay in length of tap cable, rather than discrimination elsewhere. So, we chopped a short piece of the co-ax feed off, and lo and behold - bingo, the signal shot up. From this it seemed obvious that wave length was the problem, so we directed our crews to always leave a slight amount of spare on ‘drops’ so that adjustment could be made. That applied to both feeder and drop, as either could be affected.

Later on, one of our TV service engineers rang us, and said they had the same problem in their test workshop, with benches fed from a small system we had supplied them with. I told them what to do, for any outlet that suffered low UHF, and he rang me back a day later and said “That worked great - how did you know?”. I told him the story, and passed his thanks onto our ‘withit engineer’.

Taking this further - is this possibly also my problem? - The signal seems fine, but breaks in speech seem to occur mainly with female voices (though not always) whose voices are obviously higher pitched, and with intonation from individuals, plus shortness of wavelength at those frequencies - may this be the root cause?

Broadcasters generally often apologise for loss or break in video or audio content, but not always, so that still leaves me in doubt that the above is the real cause of my problem, unless it is either my equipment or interference from 4G or multiplexing, as Eddie suggests.

Broadcasts over the weekend from Glastonbury were often poor, with breaks in both mediums, so I still put that down as outside broadcast failings, which often occur anyway, as I said above. Anyone with family present, even the dog or cat, may not notice these, but I do, as nothing much bothers me most of the time - till the door bursts open and a voice interrupts - fairly often - -

Not related to the above, but as a former audio engineer (did that too) the quality of many bands when broadcasting is often pathetic, as the sound levels are extremely poor, with bands drowning out vocals so much that in some cases they may as well not bother singing at all - very evident from the Manchester event, but NEVER from orchestral concerts. Are there no good sound engineers out there these days? Glad I got that off my chest, though it’s in the wrong place.
Ho Hum.

13

Hmm, the stub problem is only relevant for an unterminated cable - common with distribution amps using splitters… 

If the cable is terminated the signal goes on (apparently) for ever and never reflects back because the cable is infinitely long so far as the signal is concerned, in other words a classical ‘transmission line’.  But the unterminated stub is a bit like giving a tennis ball a brick wall to bounce off, and with the example of the PAL analogue colour-sub-carrier a quarter wavelength of cable (approx 13.5 metres if velocity factor is about 0.8) allowed the signal to bounce off the open end and arrive back at the Tee exactly 180 degrees out of phase to cancel itself out.

Modern digital audio does not suffer from any of the old analogue problems, and in general it either works or breaks totally.  However this does not stop the mix from possibly being poor and optimised for a much higher listening level than domestic and on better fidelity speakers than a TV; or quality might be compromised at source by the broadcaster being mean on the bit budget and then it can sound like a cheap mobile or as if it is coming from underwater.  Or both!  Remember that music suffers sooner than speech because it is a more complex source and needs at least double the bit budget, and if there is both music and speech/vocals combined then the bit  budget should be that for good music because we notice failures in speech much more readily than in music. 

Digital video again is compromised on bit budget and it is one of the several reasons I don’t have Sky; they tend to charge for the bits and the channels using Sky pay as little as they can get away with.  Modern good TV needs about 1.5Mbits/s for OK pictures, but some of the Sky channels we measured at about 200-400Kbits/s (back in 2000-2005).  A typical symptom is a football match where the players and pitch look OK until the camera moves and the pitch turns to water and all the detail vanishes into blocks.  With the latest coding scheme good Standard Definition 625 digital TV needs about 4-6Mbits/s before I can’t see the artefacts (unfortunately, I know what to look for), and High Definition 1080 about 8-13Mbits/s, typically the better broadcasts give us 1.5-3Mbits/s for SD and 3-8Mbits/s for HD, and vary it according to what is going on in the picture - the more movement and change the more bits required.

Satellite digital can be upset by the the dish seeing the sun in line with the satellite (but this is rare), and heavy rain and lightning can upset both satellite and terrestrial broadcasts. 

Ignition interference can upset both, as can the high power sparking from a bad mains electrical connection in the house (eg a couple of years ago the switch in the brand new MK wall socket for my washing machine), or a badly shielded microwave oven.

One last and totally non-obvious oddity we used to find with the old Belling TV connectors was that the really cheap aluminium ones with a tinned centre pin were OK, but the expensive shiny plated ones used to cause problems and needed to be regularly exercised to prevent a ‘dissimilar metals contact diode’ damaging the signal (a bit like the iffy battery contacts in some remote controls with certain makes of battery).  The F connectors are a vast improvement (in general but avoid the really cheap ones that fall apart) and should be used whenever possible.