DTT signal levels: a problem and the solution (dull)

In 2004 we installed a complete new system at a development that had
six blocks, all well spaced in large grounds. The place was listed so
we were only allowed one aerial and dish, so the system has quite long
underground links, which are CT167 Direct Burial. At the time of
installation I made myself unpopular with the builder by insisting on
separate cables to each block.

Recently the on-site manager reported problems with DTT. A lot of
people were complaining about pixilation on the channels carried by
mux 2.

I started at the beginning. Signals from the aerial were fine. The
head-end output was fine. Nothing had changed since installation. At
the nearest repeater I found that losses on the underground link had
increased a little, which is what you’d expect. Signal levels were
still more than adequate though, since I plan my systems with a lot of
signal in reserve, as it were. Well, I thought they were adequate. The
output of the repeater on mux 2 was
-6dbmV. That, I thought, must surely be adequate. But in the flats I
found that the downlead losses, which should be between 3 and 7dB, had
crept up and in some cases were 10 or 11dB. So? That’s still enough
isn’t it? Well no. -17dBmV is borderline for a 64QAM mux, and of
course after the system amplification the noise floor was somewhat
elevated. Hence the pixilation, from what at first seemed like a
healthy system. The combination of extra loss on the trunks and the
downleads, although not being much when expressed as a percentage, was
enough to bugger things up comprehensively.

At the repeaters further from the head end, the problem was more
obvious. When the system was new the analogue signal on ch67 arrived
at the farthest repeaters at about the same level as the one on ch21.
After equalisation at the head end to make up for transmitter changes
it was now arriving 3dB down. The higher channel muxes were similarly
afflicted. This is simply the effect of five and a half years’ aging
on top quality trunk cable. Interestingly the satellite signals had
dropped by 4 to 8dB, but this had had no effect on reception and there
had been no complaints about ‘Sky’ – although the vast majority of
them have got it. Presumably the downlead losses are in some cases
astronomic, but, well, there are no complaints. I’m guessing that the
levels at the outlets are in some cases 20dB below what we planned in
2004, but they are still adequate. When the problem arises I will fix
it by brute amplification. This should get us through to the time when
either (a) I am dead or (b) we are ripping it all out and putting
fibre in.

Each repeater had a terrestrial amplifier with 20dB gain. I replaced
them all with amps having 30dB gain and the capability for 18dB
equalisation across Bands I to IV, which I set at or near maximum.
This gave a gain of 22dB on ch 21 and 30dB on ch67. It was about
right. A rather modest increase in signal levels at the outlet took
the difficult mux from severe pixilation to perfection.

Comments anyone?

Bill

" wrote in message
...
In 2004 we installed a complete new system at a development that had
six blocks, all well spaced in large grounds. The place was listed so
we were only allowed one aerial and dish, so the system has quite long
underground links, which are CT167 Direct Burial. At the time of
installation I made myself unpopular with the builder by insisting on
separate cables to each block.

Recently the on-site manager reported problems with DTT. A lot of
people were complaining about pixilation on the channels carried by
mux 2.

I started at the beginning. Signals from the aerial were fine. The
head-end output was fine. Nothing had changed since installation. At
the nearest repeater I found that losses on the underground link had
increased a little, which is what you’d expect. Signal levels were
still more than adequate though, since I plan my systems with a lot of
signal in reserve, as it were. Well, I thought they were adequate. The
output of the repeater on mux 2 was
-6dbmV. That, I thought, must surely be adequate. But in the flats I
found that the downlead losses, which should be between 3 and 7dB, had
crept up and in some cases were 10 or 11dB. So? That’s still enough
isn’t it? Well no. -17dBmV is borderline for a 64QAM mux, and of
course after the system amplification the noise floor was somewhat
elevated. Hence the pixilation, from what at first seemed like a
healthy system. The combination of extra loss on the trunks and the
downleads, although not being much when expressed as a percentage, was
enough to bugger things up comprehensively.

At the repeaters further from the head end, the problem was more
obvious. When the system was new the analogue signal on ch67 arrived
at the farthest repeaters at about the same level as the one on ch21.
After equalisation at the head end to make up for transmitter changes
it was now arriving 3dB down. The higher channel muxes were similarly
afflicted. This is simply the effect of five and a half years’ aging
on top quality trunk cable. Interestingly the satellite signals had
dropped by 4 to 8dB, but this had had no effect on reception and there
had been no complaints about ‘Sky’ – although the vast majority of
them have got it. Presumably the downlead losses are in some cases
astronomic, but, well, there are no complaints. I’m guessing that the
levels at the outlets are in some cases 20dB below what we planned in
2004, but they are still adequate. When the problem arises I will fix
it by brute amplification. This should get us through to the time when
either (a) I am dead or (b) we are ripping it all out and putting
fibre in.

Each repeater had a terrestrial amplifier with 20dB gain. I replaced
them all with amps having 30dB gain and the capability for 18dB
equalisation across Bands I to IV, which I set at or near maximum.
This gave a gain of 22dB on ch 21 and 30dB on ch67. It was about
right. A rather modest increase in signal levels at the outlet took
the difficult mux from severe pixilation to perfection.

Comments anyone?

Bill

yes, yyyyaaaaaaaaaaaaaaaawwwwwwwwwwwwnnnnnnnnnnnnnnnnn

Well its obvious, and was to many when the terrestrial standard was
brought in, that the robustness of the system is poor compared to analogue,
which gracefully fails into noise.

Still we are stuck with it, I suppose.

I hate to think what it must be like in areas with multipath reflections or
hills in the way like in parts of Wales!

Brian

--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email:
__________________________________________________ __________________________________________________ __________


" wrote in message
...
In 2004 we installed a complete new system at a development that had
six blocks, all well spaced in large grounds. The place was listed so
we were only allowed one aerial and dish, so the system has quite long
underground links, which are CT167 Direct Burial. At the time of
installation I made myself unpopular with the builder by insisting on
separate cables to each block.

Recently the on-site manager reported problems with DTT. A lot of
people were complaining about pixilation on the channels carried by
mux 2.

I started at the beginning. Signals from the aerial were fine. The
head-end output was fine. Nothing had changed since installation. At
the nearest repeater I found that losses on the underground link had
increased a little, which is what you’d expect. Signal levels were
still more than adequate though, since I plan my systems with a lot of
signal in reserve, as it were. Well, I thought they were adequate. The
output of the repeater on mux 2 was
-6dbmV. That, I thought, must surely be adequate. But in the flats I
found that the downlead losses, which should be between 3 and 7dB, had
crept up and in some cases were 10 or 11dB. So? That’s still enough
isn’t it? Well no. -17dBmV is borderline for a 64QAM mux, and of
course after the system amplification the noise floor was somewhat
elevated. Hence the pixilation, from what at first seemed like a
healthy system. The combination of extra loss on the trunks and the
downleads, although not being much when expressed as a percentage, was
enough to bugger things up comprehensively.

At the repeaters further from the head end, the problem was more
obvious. When the system was new the analogue signal on ch67 arrived
at the farthest repeaters at about the same level as the one on ch21.
After equalisation at the head end to make up for transmitter changes
it was now arriving 3dB down. The higher channel muxes were similarly
afflicted. This is simply the effect of five and a half years’ aging
on top quality trunk cable. Interestingly the satellite signals had
dropped by 4 to 8dB, but this had had no effect on reception and there
had been no complaints about ‘Sky’ – although the vast majority of
them have got it. Presumably the downlead losses are in some cases
astronomic, but, well, there are no complaints. I’m guessing that the
levels at the outlets are in some cases 20dB below what we planned in
2004, but they are still adequate. When the problem arises I will fix
it by brute amplification. This should get us through to the time when
either (a) I am dead or (b) we are ripping it all out and putting
fibre in.

Each repeater had a terrestrial amplifier with 20dB gain. I replaced
them all with amps having 30dB gain and the capability for 18dB
equalisation across Bands I to IV, which I set at or near maximum.
This gave a gain of 22dB on ch 21 and 30dB on ch67. It was about
right. A rather modest increase in signal levels at the outlet took
the difficult mux from severe pixilation to perfection.

Comments anyone?

Bill

On 17/02/2010 06:02, Brian Gaff wrote:
Well its obvious, and was to many when the terrestrial standard was
brought in, that the robustness of the system is poor compared to analogue,
which gracefully fails into noise.

Still we are stuck with it, I suppose.

I hate to think what it must be like in areas with multipath reflections or
hills in the way like in parts of Wales!

Brian

Most of us here have given DVB-T the finger and have gone with S
flavoured DVB...

Rob.

Brian Gaff wrote:
Well its obvious, and was to many when the terrestrial standard was
brought in, that the robustness of the system is poor compared to analogue,
which gracefully fails into noise.

I don't really think that is fair if you consider the signal level. The
digital system is much more robust than analogue and will get perfect
pictures when analogue has no picture atall, and in fact when most test
equipment cannot even measure the signal (even if it says it does it is
lying). The problem is of course like any newer system (eg ADSL) that
robustness has been used to reduce signal levels or use spare
characteristics that were not relied on before. So now when stuff goes
out of spec on your system it doesn't work because there isn't enough
signal level. Of course all that will change with DSO and TV should
work perfectly with significantly degraded equipment.

As we engineer things to be more efficient they will require more
maintenance or better design for reliability. For example POTS is very
robust, but ADSL is not so phone lines now are more sensitive to
problems, increased maintenance was required, however fibre will be a
longer lasting solution.

My view is that analogue is very wasteful, and it is amazing that we
have managed to squeeze in 30+ more channels when no more could be added
with the existing analogue. It is a testament to digital that we have a
reasonably workable system in tandem with analogue. With DSO we will
get even more benefit.


Still we are stuck with it, I suppose.


Not with the lack of system margin, which will increase with DSO.

I hate to think what it must be like in areas with multipath reflections or
hills in the way like in parts of Wales!


DVB-T is very good a dealing with multipath up to the guard band limit,
aerial systems simply have to keep within it.

--
Tony

"Tony" wrote in message
. ..

As we engineer things to be more efficient they will require more
maintenance or better design for reliability. For example POTS is very
robust, but ADSL is not so phone lines now are more sensitive to
problems......

Well, just to nitpick a little: research conducted at BT Labs (no longer
called that) at Martlesham (my workplace for many years) showed that ADSL
was actually MORE robust than telephony to all the usual line faults, such
as battery or earth contact, high resistance joints, noisy joints, and so
on.

It's one weakness arises from differential mode noise on the line in the
frequency range used by ADSL, to which telephony is completely immune. This
noise is usually induced at the customer premises from the bell wire, which
is the third wire (terminal 3) running around your extensions in addition to
the two legs (A leg and B leg, terminals 5 and 2 respectively) of the phone
line itself. Its coupling to the A and B legs is unequal (to the B leg via
a capacitor, to the A leg via a resistor), so any noise picked up by the
bell wire (in effect, and end-fed long wire antenna) appears as a
differential mode signal on the A and B legs, buggering up the ADSL.

The easiest solution is simply to disconnect the bell wire (terminal 3) at
the master socket, because the plug-in filters used for each extension phone
regenerate that third wire for the phone anyway (furthermore, some phones
don't even require the bell wire). BT make available a special master
socket with a filter for the bell wire, which has a similar effect.

Of course, I'm not talking about people suffering poor ADSL performance due
to sheer line length - that is a systematic limitation in performance,
rather than a resilience or reliability issue.

SteveT

"Tony" wrote in message
. ..
Brian Gaff wrote:
Well its obvious, and was to many when the terrestrial standard was
brought in, that the robustness of the system is poor compared to
analogue, which gracefully fails into noise.

I don't really think that is fair if you consider the signal level. The
digital system is much more robust than analogue and will get perfect
pictures when analogue has no picture atall, and in fact when most test
equipment cannot even measure the signal (even if it says it does it is
lying). The problem is of course like any newer system (eg ADSL) that
robustness has been used to reduce signal levels or use spare
characteristics that were not relied on before. So now when stuff goes
out of spec on your system it doesn't work because there isn't enough
signal level. Of course all that will change with DSO and TV should work
perfectly with significantly degraded equipment.

As we engineer things to be more efficient they will require more
maintenance or better design for reliability. For example POTS is very
robust, but ADSL is not so phone lines now are more sensitive to problems,
increased maintenance was required, however fibre will be a longer lasting
solution.

My view is that analogue is very wasteful, and it is amazing that we have
managed to squeeze in 30+ more channels when no more could be added with
the existing analogue. It is a testament to digital that we have a
reasonably workable system in tandem with analogue. With DSO we will get
even more benefit.


Still we are stuck with it, I suppose.


Not with the lack of system margin, which will increase with DSO.

I hate to think what it must be like in areas with multipath reflections
or hills in the way like in parts of Wales!


DVB-T is very good a dealing with multipath up to the guard band limit,
aerial systems simply have to keep within it.


My own experience bears that out, my analogue Mendip reception was was
always iffy WRT multipath, due to foliage, especially in windy and rainy
conditions, which was on occasions so bad that the ghost images would
pulsate in and out, I switched to digital Freeview almost as soon as it
started and have never bothered with analogue since, because it works so
well, despite the transmission power still only being at one tenth of its
intended power after DSO.
The only problem I've ever experienced was occasional impulse noise (due to
a young neighbours motorcycle) this was completely eliminated with a new
aerial and a CT 100 rewire from aerial to amplifier and all six outlets, and
maybe even that would have disappeared of its own accord had I waited for
Mendip to put out 100 kW and switch transmissions to 8K after DSO.

My own experience bears that out, my analogue Mendip reception was was
always iffy WRT multipath, due to foliage, especially in windy and rainy
conditions, which was on occasions so bad that the ghost images would
pulsate in and out, I switched to digital Freeview almost as soon as it
started and have never bothered with analogue since, because it works so
well, despite the transmission power still only being at one tenth of its
intended power after DSO.
The only problem I've ever experienced was occasional impulse noise (due
to a young neighbours motorcycle) this was completely eliminated with a
new aerial and a CT 100 rewire from aerial to amplifier and all six
outlets, and maybe even that would have disappeared of its own accord had
I waited for Mendip to put out 100 kW and switch transmissions to 8K after
DSO.

That bears out my experience on holiday below Caradon Hill. An indoor Silver
Sensor aerial was useless on analogue, requiring tweaking between each
channel change and giving at best really ghosty results. DTTV however was
rock solid and I was very happy to have packed the set top box!

On 17 Feb, 19:08, "Doctor D" wrote:
My own experience bears that out, my analogue Mendip reception was was
always iffy WRT multipath, due to foliage, especially in windy and rainy
conditions, which was on occasions so bad that the ghost images would
pulsate in and out, I switched to digital Freeview almost as soon as it
started and have never bothered with analogue since, because it works so
well, despite the transmission power still only being at one tenth of its
intended power after DSO.
The only problem I've ever experienced was occasional impulse noise (due
to a young neighbours motorcycle) this was completely eliminated with a
new aerial and a CT 100 rewire from aerial to amplifier and all six
outlets, and maybe even that would have disappeared of its own accord had
I waited for Mendip to put out 100 kW and switch transmissions to 8K after
DSO.

That bears out my experience on holiday below Caradon Hill. An indoor Silver
Sensor aerial was useless on analogue, requiring tweaking between each
channel change and giving at best really ghosty results. DTTV however was
rock solid and I was very happy to have packed the set top box!

Quite! Based on my experience with the Wales/SW England DSO, I think
the sheer increase in signal levels post switchover will immediately
remedy any problems caused by normal equipment wear and tear.

Post DSO DTT is a real revelation compared to the flawed transitional
system we've struggled with since 1998.

"Steve Thackery" wrote in message
...

"Tony" wrote in message
. ..

As we engineer things to be more efficient they will require more
maintenance or better design for reliability. For example POTS is very
robust, but ADSL is not so phone lines now are more sensitive to
problems......

Well, just to nitpick a little: research conducted at BT Labs (no longer
called that) at Martlesham (my workplace for many years) showed that ADSL
was actually MORE robust than telephony to all the usual line faults, such
as battery or earth contact, high resistance joints, noisy joints, and so
on.

Someone should tell the Openreach engineers.
I really had to fight to get them to accept that there was a fault in the
ADSL equipment at the exchange, they kept whining about minute problems
they'd detected on the line and tiny imbalances caused by one of my surge
filters.

--

Brian Gregory. (In the UK)

To email me remove the letter vee.