Re: Digital TV compatable aerial


"Jim Lesurf" <jcgl@xxxxxxxxxxxxxxxxxx> wrote in message
news:4dd92b9ccejcgl@xxxxxxxxxxxxxxxxxxxxx
> In article <dnmp8l$rlv$1@xxxxxxxxxxxxxxxxxxxxx>, Bill Wright
> <insertmybusinessname@xxxxxxx> wrote:
>
>> In the vast majority of cases the transmitted power of the digital
>> multiplexes is about a hundredth of that for the analogue channels. This
>> equates to a received signal differential of 20dB. Since the carrier to
>> noise ratio need by digital is something like 20dB less than that
>> required for analogue, the coverage is much the same.
>
> I am wary of the above
Yes, it is rather simplistic maybe.

for two reasons:
>
> One is that we have to be carefuly not to unintentionally accept what I
> now
> tend to call "Spiney's Flaw". :-) This is to base our assumptions mainly
> on 'gaussian noise/statistics' and forget - for example - the effects of
> impulse Interference.
But impulse interference doesn't seem to that much of a problem, these days.
I wouold go so far as to say that it is a very rare problem. The DTT
reception would be falling over due to poor signal strength per se before
passing motorbikes became a big factor. I agree it is a problem in places
where there is an exceptional amount of impulse noise, but this comes back
in a way to my point that people will tolerate poor analogue reception but
they will not tolerate the digital equivalent. I suppose that's fair enough
though. I mean, what matters really is the viewers' perception. People will
tolerate an occasional white flash but they are less happy about the
occasional freeze and lock up. But receivers seem to have got much better.
They don't seem to crash as much as they used to.

>
>> Practical experience shows that analogue and digital coverage is much
>> the same where the transmitted power differential is 100:1. Of course
>> there are some transmission sites where this is not the case.
>
> The papers, etc, I've seen from the BBC RD and others seem to present a
> slightly different situation.
>
> For example in "All-digital planning and Digital Switchover" by Nokes, et,
> al, they show relative service maps for analogue and digital from Crystal
> palace with a 100:1 TX power ratio and this indicates that 20dB less for
> DTTV would not provide the same level of coverage. They state that more
> like 10dB down for DTTV would be required for the same coverage as
> analogue.
I'm sure this is pessimistic. Absolutely certain. Maybe these figures are
based on early receivers, or maybe they are just very cautious. My
overwhelming impression of DTT is that it works significantly better than it
should! Take as an example the village of Tickhill. Most of the village is
badly screened from Emley Moor. Ground level in the village is 50ft ASL. The
high ground at Clifton that's in the way is 450ft ASL. In days of yore we
would use Belmont, Bilsdale and even Waltham because Emley was snowy and
ghosty. We now almost always use Emley, normally installing a good 18
element aerial on a fairly tall mast and a masthead amp. This generally
gives four reasonable (but not perfect) analogue channels, often with
visible ghosting, one snowy ghosty analogue channel (C5 is transmitted from
lower down the mast) and perfect DTT. People do not ring up during the
following weeks and months to say that the DTT reception is defective.
Impulse interference is not an issue, even on the main road.

However (and it's a big however!) in a place where Emley Moor analogue is
the best available, but is really poor with a decent aerial producing
unmistakably snowy pictures, DTT will in general not work at all. Not a
dickybird!

Incidentally we always plan our distribution systems so that the levels of
the muxes are only 12 to 15dB below the analogue channels. This is because
there's no reason why not, I suppose, and because it allows to some extent
for poor connections in the living rooms.
>
> I'd need to re-read other references to be sure, but my recollection from
> them, including those which deal with II, is that the required powers will
> vary with the situation, and the choice of modulation (16/64QAM), but that
> -20dB is likely to be too low if we want a similar coverage areas with
> minimal II/fade problems. My recollection is that that between -15dB and
> -10dB would be more realistic *if we wanted the same levels of reliable
> coverage from the same TX sites*.
I'd agree with 15dB as a very rough practical rule of thumb for rock solid
DTT because digital deficiencies are subjectively less tolerable. But having
said that I've done many many installations like the Tickhill examples
above, where people are really pleased to have DTT after enduring dodgy
analogue.

Another pointer is to consider what happens when Belmont viewers in this
area come home with a DTT box. They ring us and say "Reception of the BBC
channels is much better than the analogue -- it's definitely the best
reception we've ever had -- but we can't get ITV." Of course, only the BBC
mux in in band. But the point is, they go from indifferent analogue to (what
they perceive as perfect) DTT.

You certainly don't need a differential of only 10dB. As I said, we keep the
muxes 12 to 15dB below the analogue channels on distribution systems and we
have no problems with that. If we did we'd simply lift up the DTT as much as
necessary, and we don't because there isn't a problem.

Here are a few totally unscientific figures, derived from a wide variety of
installations. They assume the use of a good masthead amp where appropriate,
no significant CCI and a normal urban environment. Signal level of a mux at
the aerial terminals:
-30dBmV: No way will this work
-25dBmV: 64QAM no chance; 16QAM very dodgy
-20dBmV: 64QAM won't be reliable; 16QAM generally OK but could break up
-15dBmV: Normally reception is perfect
-10dBmV: Rock solid
Analogue (measured at the aerial, but again the comment about picture
quality assumes the use of a good masthead amp):
-30dBmV: very snowy, prone to impulse interference
-20dBmV: snowy, prone to impulse interference
-10dBmV: Can be quite reasonable, but still some impulse interference and
snow
0dBmV: Usually just about OK, but needs a masthead amp of course
10dBmV: Rock solid.


> I would not say that DTTV is the 'poor relation'. However I think we need
> to be wary of assuming that ERPs 20dB below analogue would be sufficient.
> I'd have said more than that would be desirable if we want to cover the
> same areas with minimal problems.

I agree. But it's because we don't have 'minimal problems' with analogue
either at the edges of the service areas. We have every kind of damned
problem! I'll tell you what Jim, suppose you said to me "Here's 100
addresses. They're in all sorts of locations but they're all in places where
reception is reputedly 'poor' to a greater or lesser degree The
digital/analogue differential is nominally 20dB though, at every location.
See what you can do to get the people some decent telly." I can guarantee
that I'd have more happy viewers with digital than with analogue. You'll
find that astonishing, but it's true. Quite frankly, it's a bit of a bugger
from the point of view of domestic aerial riggers. And when the broadcasters
up the DTT power by 10dB I think that will be the end of the road for aerial
rigging in many parts of the country.

Bill

PS Written in haste and when very tired, so E&OE.


.

Loading