Satellite signal strength shows 70% with LNB covered up

[Dave W]

I have just acquired a Technosat T888 receiver and dish. It was
working at the previous location, but now when I am trying to set up
the dish, the displayed signal level shows about 70% irrespective of
where the dish is pointing, or me putting my hand over the LNB.
'Quality' displays as zero.

I've found that if I hold the LNB about 1cm from my metal fridge door,
the signal level goes to zero, but goes to 70% again when in contact.
This looks to me like the LNB is oscillating, and forming a standing
wave facing the door.

Is this a fault, and if so is it likely to be the LNB or the receiver?
I was hoping to use the displayed signal strength as an aid to
pointing the dish.

Dave W

[Bill Wright]

Are you testing the equipment with a very short cable? The meter
probably expects a 15m or so cable. I shouldn't worry about what happens
when you point the LNB at the fridge. When you point a spaniel at a
fridge its tail will oscillate, but this is not a fault.

The meter is probably a bit hyperactive, like many spaniels.

Just put the LNB on the dish and point the thing at the sky and wiggle
it about and I'm sure everything will be all right. If you get some
dusky bloke with a tea towel on his head reading the news, then you've
got the wrong satellite.

Bill

[Graham.]

On Tue, 22 May 2012 00:39:15 +0100, Bill Wright <bi...@invalid.com>
wrote:

Jim could probably think up an algorithm for direct reading of the
inside of the fridge in degrees Kelvin, which will be a damn sight
more use than the signal strength indication is for measuring signal
strength.



--
Graham.
%Profound_observation%

[Andy Burns]

Dave W wrote:

> I've found that if I hold the LNB about 1cm from my metal fridge door,
> the signal level goes to zero, but goes to 70% again when in contact.
> This looks to me like the LNB is oscillating, and forming a standing
> wave facing the door.

Check John Legon's past threads on pointing LNBs at hot cups of tea!

For initial location a compass and a 'satfinder' meter will help, when
all's said and done it's the quality indicator you're most interested in
for fine adjustments, not the strength indicator.

[R. Mark Clayton]

This will be background noise.

It will go down if you point it at something very cold (fridge / night sky),
but up if you point it as something hot (oven, sun*)




* don't actually point it at the sun you will fry the LNB.


"Dave W" <dave...@yahoo.co.uk> wrote in message
news:jpedev$c3d$1...@speranza.aioe.org...

[Bill Wright]

R. Mark Clayton wrote:
> This will be background noise.
>
> It will go down if you point it at something very cold (fridge / night sky),
> but up if you point it as something hot (oven, sun*)

It's almost all noise generated inside the LNB. Only a small fraction is
environmental.

> * don't actually point it at the sun you will fry the LNB.

I beg to differ. The phenomenon you describe is unknown to me.

Bill

[John Legon]

Andy Burns wrote:
> Dave W wrote:
>
>> I've found that if I hold the LNB about 1cm from my metal fridge door,
>> the signal level goes to zero, but goes to 70% again when in contact.
>> This looks to me like the LNB is oscillating, and forming a standing
>> wave facing the door.
>
> Check John Legon's past threads on pointing LNBs at hot cups of tea!

As it happens, there's a summary (with videos) here:

http://www.john-legon.co.uk/thermal.htm

[Ian Jackson]

In message <jpfpc4$buo$1...@speranza.aioe.org>, Bill Wright
<bi...@invalid.com> writes

I have virtually no experience of satellite installations. However, when
I swapped a single output LNB for a double, and re-checked the position
of the dish, I found that the satellite finder meter needle was
permanently pinned on the endstop, regardless of where the dish - or
just the LNB itself - were pointing (unless I turned the sensitivity
down to minimum, when it read zero). I guessed that the system was
oscillating. To tame it, I needed 20dB of power-passing attenuation (two
8-way splitters in series) between the meter and the LNB.
--
Ian

[Dave W]

It's very encouraging that others have found the reading high whatever
the dish is pointing at. Does this mean that the reading may go down
when the dish is pointing at a satellite? If so, will it still show a
greater reading when the direction is spot on than when it isn't
quite? I haven't found any noticeable peaks so far.

Would an independent finder meter act any differently? I can't put an
attenuator between the LNB and the receiver because that would short
out the DC power to the LNB, and anyway the receiver meter is not at
full scale.

The fridge door I mentioned is not cold, it was just the nearest metal
for my experiment.

Dave W

[Ian Jackson]

In message <jpedev$c3d$1...@speranza.aioe.org>, Dave W
<dave...@yahoo.co.uk> writes

I've just re-read your original post, and I now realise that you are
getting the constant 70% on the bar-graph of the satellite receiver
(presumably when it is tuned to any of the TV channels). This is a
completely different situation from getting a constant high reading on a
£10 satellite finder - which is essentially just a wideband amplifier
(tuned broadly to 11GHz), plus a diode detector feeding a DC amplifier
driving a meter. It feel that it's unlikely that an oscillating LNB
would produce this effect, because the receiver simply wouldn't make any
attempt to decode the incoming signal. However, I really just don't
know!
--
Ian

[John Legon]

Dave W wrote:
> It's very encouraging that others have found the reading high whatever
> the dish is pointing at. Does this mean that the reading may go down
> when the dish is pointing at a satellite? If so, will it still show a
> greater reading when the direction is spot on than when it isn't
> quite? I haven't found any noticeable peaks so far.

A satellite receiver should give two readings, one for signal quality
and the other for signal strength. For practical purposes you can
ignore the signal strength - which is mostly noise and essentially only
shows that the LNB is working. It's the signal quality that matters,
and this is likely to be zero until the dish is accurately aligned on a
satellite.

>
> Would an independent finder meter act any differently?

Yes, a simple satellite finder won't show signal quality, but it will
have a sensitivity control which you can adjust so that the reading is
in the middle of the range when the dish is pointing anywhere in the
sky. As the dish is moved close to the position of a satellite, the
reading should increase and probably go off the scale. The control is
then turned down and the process repeated until no further improvement
in the reading can be obtained. You should then find that the receiver
is giving a reading for signal quality, which you can use to tweak the
alignment of the dish and to set the skew of the LNB.

I can't put an
> attenuator between the LNB and the receiver because that would short
> out the DC power to the LNB, and anyway the receiver meter is not at
> full scale.

You shouldn't need to use an attenuator, although it is true that some
meters are so sensitive that they don't work well with high-gain LNBs,
and may show signs of instability. I suggest you get an SLX or Konig
satellite finder with moving coil meter, such as can be bought for about
15 quid from B&Q. It has a varying audible tone which is very useful
when setting the dish alignment.

[R. Mark Clayton]

"Martin" <m...@address.invalid> wrote in message
news:5utmr7130pgiab6f9...@4ax.com...

> Doncaster/Rotherham is in a place where the sun doesn't shine?

Close to the ****hole of the universe - so likely...

> --
>
> Martin
>

[Andy Wade]

On 22/05/2012 15:25, Dave W wrote:

> It's very encouraging that others have found the reading high whatever
> the dish is pointing at. Does this mean that the reading may go down
> when the dish is pointing at a satellite?

Possibly, depending on what it was pointing at before, but what matters
is the increase when it's pointing at the satellite, relative to the
reading when it's pointing at 'cold sky' - that being an indication of
the signal to noise ratio.

> If so, will it still show a greater reading when the direction is
> spot on than when it isn't quite? I haven't found any noticeable
> peaks so far.

It should, but - as has already been said - it's better to forget the
strength reading altogether and concentrate on peaking the quality bar.
Also check the Network ID (0002 for Astra 2 at 28.2 E) to make sure
you're on the right bird.

> The fridge door I mentioned is not cold, it was just the nearest metal
> for my experiment.

Its temperature is largely irrelevant. The door will just act as a
reflector and the LNB will just see the temperature of the surroundings.
If the feedhorn is very close to the door its front end will be
detuned and any 'results' will be quite meaningless.

--
Andy

[Andy Wade]

On 22/05/2012 11:21, Bill Wright wrote:

> It's almost all noise generated inside the LNB. Only a small fraction is
> environmental.

Hmm, that needs qualifying a bit. In a literal sense it's true, since
most of the output noise power is coming from the high-gain amplifier(s)
in the LNB (and thus ultimately from its power supply).

That view's a bit misleading though: the amplifier, being just a linear
'power multiplier', is just scaling the weak input noise sources. As an
example let's take an LNB with a noise figure of 1 dB - i.e. an
equivalent input noise temperature of very close to 75 K. If it's
looking at warm terrestrial surroundings at 300 K, say, the total input
noise temp is 375 K, so 300/375, i.e. 80%, of the input or output noise
is attributable to the environment. If the surroundings could be cooled
to absolute zero (without affecting the LNB internally) then the output
noise would drop by 80% (7 dB).

A small dish looking through a clear sky might have an antenna noise
temp of around 40 K. This time about about 35% (40/115) of the noise is
environmental (including the cosmos).

--
Andy

[Bill Wright]

Andy Wade wrote:
> On 22/05/2012 11:21, Bill Wright wrote:
>
>> It's almost all noise generated inside the LNB. Only a small fraction is
>> environmental.
>
> Hmm, that needs qualifying a bit. In a literal sense it's true, since
> most of the output noise power is coming from the high-gain amplifier(s)
> in the LNB (and thus ultimately from its power supply).
>
> That view's a bit misleading though: the amplifier, being just a linear
> 'power multiplier', is just scaling the weak input noise sources. As an
> example let's take an LNB with a noise figure of 1 dB - i.e. an
> equivalent input noise temperature of very close to 75 K. If it's
> looking at warm terrestrial surroundings at 300 K, say, the total input
> noise temp is 375 K, so 300/375, i.e. 80%, of the input or output noise
> is attributable to the environment. If the surroundings could be cooled
> to absolute zero (without affecting the LNB internally) then the output
> noise would drop by 80% (7 dB).

So what would you expect to happen if the LNB was in a sealed box with
absolutely no RF input?

Bill

[Dave W]

At ease everybody - I've now got programmes.

I decided not to buy a 'finder meter' because it looked too simple
(like Ian described) and did not have a 'Quality' display. Instead I
decided to look very closely at the 'Power' bargraph while moving the
dish, and found that I could increase it from its normal 70%, to 72%.
At this point the 'Quality' bargraph instantly went from zero to 50%,
and further slight tweaking could maximise this to 70%. Now when I
selected a satellite on the receiver's list that showed a 'Quality'
reading, and scanned for programmes, they appeared.

This is a 2006 vintage receiver. On its list the satellites that
worked were Astra 19.5E, Astra 35E, and Hotbird1. It is my belief that
the angles stated are no longer true - Wikepedia says they should be
at 28 degrees East. My programme listing just shows these titles
rather than what they are now.

Dave W

[John Legon]

That's not a realistic scenario, unless the box has been cooled to
absolute zero. Otherwise, it will depend on the material of the box.
Some types of plastic and wood are largely transparent to microwaves, so
the LNB will "see" the thermal background radiation of the surroundings,
depending on the ambient temperature.

I think you devised a wide-band noise generator using an LNB as the
source - I may have the pdf for it somewhere - based on the assumption
that the noise was generated internally by the LNB...

[John Legon]

Dave W wrote:
> At ease everybody - I've now got programmes.
>
> I decided not to buy a 'finder meter' because it looked too simple
> (like Ian described) and did not have a 'Quality' display. Instead I
> decided to look very closely at the 'Power' bargraph while moving the
> dish, and found that I could increase it from its normal 70%, to 72%.
> At this point the 'Quality' bargraph instantly went from zero to 50%,
> and further slight tweaking could maximise this to 70%. Now when I
> selected a satellite on the receiver's list that showed a 'Quality'
> reading, and scanned for programmes, they appeared.

Well done! You were lucky to notice the change in signal strength but
it shows that a little patience pays off. In general it is much easier
to align a dish using a finder even though they don't show quality - the
reason being that they give a much more visible (and audible) response
when the dish is further away from the target, and they also respond
immediately to movements of the dish whereas a receiver will take a
second or two to update the bargraph display.

>
> This is a 2006 vintage receiver. On its list the satellites that
> worked were Astra 19.5E, Astra 35E, and Hotbird1. It is my belief that
> the angles stated are no longer true - Wikepedia says they should be
> at 28 degrees East. My programme listing just shows these titles
> rather than what they are now.

You should now go a site such as www.lyngsat.com to check that the
transponder details are correct and make sure that you are getting all
the available programmes.

[Andy Wade]

On 23/05/2012 01:22, Bill Wright wrote:

> So what would you expect to happen if the LNB was in a sealed box ...

Ah, that old question :~).

Fairly obviously if the box is lined with radar absorbent material, or
similar, the LNB will see the physical temperature of that material.

If it's just a plain metal box I'm fairly sure that the answer is the
same (LNB sees the temperature of the box walls). Think of it as a
cavity resonator and realise that there is always some resistive loss
(the Q-factor cannot be infinite) and the temperature of the losses is
again the wall temperature.

A big assumption here, especially in the latter case, is that the noise
figure and gain of the LNB don't change when it's put in the box. In
practice this may well not be true, and only a small change in gain
could completely bugger-up your attempts to use the LNB as a
thermometer. There might be other effects too, such as changes in the
level of translated LO noise sidebands reaching the IF output, as the
reflection coefficient seen at the input changes. LNBs are not
high-performance receiver front-ends; they're just designed to be cheap.

> ... with absolutely no RF input?

No such thing in practice. There must always be some leakage, even if
the walls are hundreds or thousands of skin-depths thick, and any
practical box will have imperfect joints.

I'm not sure what would happen if the box (and its contents) were
superconducting. Could you really have an infinite-Q resonator? Could
such a box retain the radiation temperature present when it was sealed -
like an object placed in a perfectly thermally-insulated box? We're
into the realm of thought-experiments now. Is Jim around...?

--
Andy

[Dave W]

John Legon <jo...@nospam.demon.co.uk> wrote:

>Dave W wrote:
>> At ease everybody - I've now got programmes.

....

>> This is a 2006 vintage receiver. On its list the satellites that
>> worked were Astra 19.5E, Astra 35E, and Hotbird1. It is my belief that
>> the angles stated are no longer true - Wikepedia says they should be
>> at 28 degrees East. My programme listing just shows these titles
>> rather than what they are now.

>You should now go a site such as www.lyngsat.com to check that the
>transponder details are correct and make sure that you are getting all
>the available programmes.

Thanks. I did that before I read your comment, and have now used a
programmable "User" satellite in my receiver, renaming it "28deg East"
to encompass the four actual satellites involved. As soon as I entered
the 22000 Symbol Rate the Quality shot up from zero to 75%, allowing
me to do an automatic search for all frequencies within it (and other
symbol rates appeared).

Out of the 350 free channels that appeared, only about a dozen that
are not already on terrestrial Freeview were sufficiently worthy of
occasional use for me!

Dave W

[Jim Lesurf]

In article <a242b1...@mid.individual.net>, Andy Wade

<spamb...@maxwell.myzen.co.uk> wrote:
> On 23/05/2012 01:22, Bill Wright wrote:

> > So what would you expect to happen if the LNB was in a sealed box ...

> Ah, that old question :~).

> Fairly obviously if the box is lined with radar absorbent material, or
> similar, the LNB will see the physical temperature of that material.

> If it's just a plain metal box I'm fairly sure that the answer is the
> same (LNB sees the temperature of the box walls). Think of it as a
> cavity resonator and realise that there is always some resistive loss
> (the Q-factor cannot be infinite) and the temperature of the losses is
> again the wall temperature.

Yes. If the 'box' is 'closed' at RF then the field of view of the antenna
is the inside of the box. The possible qualifier is that the antenna may
also see itself via reflections in the box. So the overall power level it
sees will be some combination of box temperature and signals emitted by
what is connected to the antenna, that bounce around the box and get back
to into the antenna. (i.e. any amps/mixers.etc that may be sending any
power back up the antenna and which in normal use would be lost to the
sky.) Hence the points that Andy then added that said LNB may also have
behaviour that reacts to having a 'mirror' over the antenna.

Even when not 'oscillating' a mixer or amp may be producing wideband noise
that squirts out of the RX antenna. And its gain, etc, may alter if you
reflect that back. Common problem for radio telescopes where an RX can see
a slight reflection of itself in something like a secondary mirror. A
cassgrain telescope can easily act as a lossy resonator.

> I'm not sure what would happen if the box (and its contents) were
> superconducting. Could you really have an infinite-Q resonator? Could
> such a box retain the radiation temperature present when it was sealed -
> like an object placed in a perfectly thermally-insulated box? We're
> into the realm of thought-experiments now. Is Jim around...?

In theory you can have a perfectly reflecting box. People have done the
equivalent experiment of having a resonator made of superconductor and
check that the signal persists after an arbitrarily long time. A close coil
of wire also behaves a bit like this at 'dc', although you would not
normally think of a 0Hz resonator.

Of course, as soon as you checked you added a loss or escape route to sniff
out the enclosed energy... :-)

That said, as you increase the frequency there is a tendency for loss
mechanisms to appear and the system isn't quite as 'superconducting'. But
I've forgotten the details, I'm afraid.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html