Quad shield coax & dielectric?
 

In message , Jerry Stuckle
writes
On 3/18/2014 4:56 AM, Ian Jackson wrote:
In message , Jerry Stuckle
writes
On 3/17/2014 5:03 PM, Ian Jackson wrote:
In message , Jerry Stuckle
writes
On 3/17/2014 3:15 PM, Rob wrote:
Jerry Stuckle wrote:
Most likely the company reduced the transmitted power by a factor
of 10 at the time of the switchover, to put the added link margin
in their own pockets.
(transmitting a megawatt of ERP as was regular in the analog days
puts a serious dent in your electricity bill, even when you have
a lot of antenna gain)


Not at all. If anything, they raised their power.

Here they went from 1 megawatt to about 50 kilowatt (ERP).
And then there are several programmes on one transponder, instead
of one analog programme. This gives significant savings in power.


OK, you mean absolute power. Yes, they can lower the ERP - but that
does not necessarily lower the power for the signal. Remember at 1MW
the power was spread over 4.25 Mhz (assuming video only, of course).
Digital requires much less bandwidth, so they don't need as much
power
to get the same effective signal. However, digital still requires a
stronger signal than analog, in the bandwidth provided. You need
quite
a bit of noise before it becomes visible in analog. Digital, a
single
noise pulse can cause the loss of several bits of information.
Because
of the compression involved, this is more than one or two pixels.

I think not much of that is correct.
The systems differ a bit between US and elsewhere, but over here the
channel spacing of digital and analog is the same, and the bandwidth
is similar (a bit more for digital if anything).

Also there is no discission of "spreading", we are just discussing
peak envelope ERP.
You could argue that a single digital stream sending 5 programmes
means that 1 programme is transmitted at 1/5 the power, but that is
not what I mean. The total ERP for 1 transmitter has been lowered,
and it transmits multiple programmes to boot.

Digital requires less power because it requires less signal-to-noise
ratio at the receiver.


There are major differences between Europe's PAL and the U.S.'s NTSC.

Where do you get these strange ideas? Yes, there are differences (the
major one being the subcarrier phase alternation of PAL), but otherwise
the systems are very similar - and none of the differences really affect
the basics of RF system measurements.


From facts.

Care to share a few of your 'facts'? How do the relatively minor
differences between analogue PAL and NTSC affect how you do RF (and most
video) performance measurements?

But the digital signal has much LESS bandwidth than the old analog one.

No it doesn't. It's the same. In the UK the 8MHz wide analogue channels
have been replaced with an 8MHz wide digital QAM signal, and this
occupies the same channel frequency. Unless I'm very much mistaken, the
US has done the same (except that the channels are 6MHz wide).


Wrong again.

I assure you that what I said about the UK situation is 100% correct.
However, I've lost touch with what you do in the USA - so as I'm wrong,
how do you transmit digital TV. In particular, how do cable TV systems
organise the RF spectrum?

That was the major impetus over here to switch to digital - to free
up major bandspace in the VHF and UHF spectrums.

The 'freeing-up' is because each QAM multiplexed transport stream
carries typically up to 6 TV channels and 6 high-quality stereo radio
channels.


Ah, so the channel is effectively only 1Mhz wide then.

No. In the UK, the 'channel' is 8MHz wide, and occupies the same
frequency slot as the old single analogue channel. It is a QAM (or
sometimes QPSK) multiplex of all the programmes it contains. Yes, if you
have six TV channels in a 6MHz wide QAM mux you could consider that each
channel occupies 1MHz, but they are not six adjacent individual digital
signals. They are all jumbled up in the single 6MHz digital signal. The
decoder sorts them out automagically, and puts the required picture on
your TV screen.

We now have as many (or, in some areas, more) stations in a much
smaller band than before.

Digital requires less power because the bandwidth is much lower.

Only insofar as the digital decoder can successfully extract accurate
data from the 6 or 8MHz wide multiplexed transport stream at
signal-to-noise ratios which, for analogue, would produce barely
watchable pictures.

You really don't understand digital, do you?

I have to admit that I used to be intimately familiar with many aspects
of analogue, but there's a lot about digital that I don't know.
Fortunately, these days I don't really have to!

I hate it when someone tries to tell me my job which I've been doing
for years... But some people just think they know it all...

Se above.

You've lost track of a lot. And I'm tired of teaching a pig to sing.

You know nothing about what I do, who I work with, or what's going on
in the United States. Yet you think you know more than I do about my
job, the people in my industry, and the equipment I use.

I tell you what. You come over and do my job for about ten years.
Then, MAYBE, you'll be qualified enough to know something about it.

Until then, you're just a troll.

I don't know which thread you've been reading, but it's certainly not
this one.

Yes, I don't really know what you do (apart from, I believe you've said,
working for a company that installs lots of cable) - but does that
really matter? Regardless of what you do for a living, it hasn't
prevented some of your statements and assertions being totally and
outrageously incorrect. However, unlike you, I'm not going trying to
prove I'm right by playing an "I've been an XXX for YY years, so I
should know what I'm talking about" card (although you can have a guess,
if you like).

As for coming over and doing your job - thanks for the kind offer, but
no thanks. I've had enough of work to last a lifetime, and I'm now well
into my retirement.
--
Ian

Quad shield coax & dielectric?
 

Jerry Stuckle wrote:
On 3/18/2014 5:11 AM, Rob wrote:
Jerry Stuckle wrote:
That was the major impetus over here to switch to digital - to free
up major bandspace in the VHF and UHF spectrums.

The 'freeing-up' is because each QAM multiplexed transport stream
carries typically up to 6 TV channels and 6 high-quality stereo radio
channels.


Ah, so the channel is effectively only 1Mhz wide then.

That is only an "average" that you get when you divide the total bandwidth
by the number of TV programmes transmitted.
As multiple TV programmes are transmitted on a single channel,
that occupies the width of a classical analog channel that could
transmit only a single programme, more programmes can be put in
the same bandwidth.

However, to receive one of them you really need to receive the
entire channel (several MHz wide), so all link budget calculations
should be based on the full bandwidth. Later, the receiver throws
away most of the bits it has received and decodes only the information
for one of the TV programmes.


You need to check on how the signals are transmitted. You seem claim to
know a lot about how U.S. TV works, even though you're thousands of
miles away. But then you know a lot more about my job, the people I
work with, and the tools I use than I do.

Come on over and do my job for about ten years. Then maybe, just maybe,
you'll be qualified to comment on it.

Otherwise, I just consider you a troll.

I know that the system used in the USA is different from what it is
here, but this mostly concerns the modulation method used for terrestrial
transmissions. Both methods (8VSB and COFDM) have their merit, and this
was discussed a lot in the past.
The same-frequency network we have in operation here would not be possible
with 8VSB, but in long reach operation 8VSB is claimed to be better.

However, the general principle of using a full classical analog channel
(8 MHz here, 6 MHz for you) in its entirety to transmit a multiplexed
transport stream conveying several TV programmes is the same.

SCPC systems (where a transport rate is chosen to just fit the bitrate
required for a single programme, and the resulting channel bandwith
is correspondingly reduced) is used on some satellites, mainly for
satellite newsgathering and other ad-hoc links, and sometimes for DTH
transmission from really small stations. It is not popular because it
wastes bandwidth and transponder output power headroom.

Quad shield coax & dielectric?
 

In article ,
Jerry Stuckle wrote:

You need to check on how the signals are transmitted. You seem claim to
know a lot about how U.S. TV works, even though you're thousands of
miles away. But then you know a lot more about my job, the people I
work with, and the tools I use than I do.

Jerry, I'm intrigued by what you say here.

I just skimmed through the first few parts of the ATSC standard
document (ATSC A/53 parts 1-3). What I see there, indicates that
there's a single MPEG-2 transport stream, carrying several interleaved
elementary streams (audio and video). This transport stream is
trellis-encoded, and then used to modulate a single RF carrier
(VSB), with a 6 MHz channel width.

With that modulation, it seems to me that receiving the OTA
transmission does require demodulating the entire 6 MHz signal
bandwidth to recover the transport stream. The individual program
streams may of course use much less than the full effective bandwidth,
after de-interleaving, but I can't see how it would be practical to
demodulate and decode the VSB signal and "pull out" one individual
elementary stream using only a smaller "slice" of the RF signal.

What am I missing here? Is it actually possible to "receive-slice"
the 8VSB signal with a narrower RF passband, and pull out a specific
elementary stream successfully? Are some broadcasters actually
transmitting multiple modulated carriers within their 6 MHz ATSC
spectrum slice?

Now, I realize that cable TV transmissions may not (and often do not)
use VSB. With an OFDM modulation it would be possible in principle to
"slice" the 6 MHz spectrum segment into smaller, independent sets of
subcarriers carrying different programs... I don't know if any cable
systems do this in practice but it does seem possible.

Quad shield coax & dielectric?
 

In article Jerry Stuckle writes:

OK, you mean absolute power. Yes, they can lower the ERP - but that
does not necessarily lower the power for the signal. Remember at 1MW
the power was spread over 4.25 Mhz (assuming video only, of course).
Digital requires much less bandwidth, so they don't need as much power
to get the same effective signal. However, digital still requires a
stronger signal than analog, in the bandwidth provided. You need quite
a bit of noise before it becomes visible in analog. Digital, a single
noise pulse can cause the loss of several bits of information. Because
of the compression involved, this is more than one or two pixels.

Actually, if you look with a spectrum analyzer, the digital signal has
the power spread much more uniformly over the 6 MHz than analog did.

Digital stations here run substantially less power than their analog
versions did, and deliver much better results. The power is spread across
the 6 MHz channel - the signal uses the whole thing. By contrast, analog
had uneven distribution of the power across its spectrum.

The digital system has levels of error correction in the signal - wiping
out a few bits is unlikely to affect the demodulated result. Wiping out
a big burst of them is more likely to cause a problem.

Alan
wa6azp