Snow Go Area

On Jan 8, 7:52*pm, John Legon wrote:
At 07:48:23 Fri, 8 Jan 2010,
wrote in article 523ed8b5-cdc9-42e3-9d81-2a5a664
:

On Jan 8, 1:03*pm, John Legon wrote:
Conclusion: *Reception of Freeview is fine here with sub-zero temperatures
outside, provided there is enough snow on the roof to insulate the loft :-)

I've never heard of a faulty masthead amp where the fault only showed
up when it was cold.

That's very interesting. *The next time the temperature's below freezing
and I start losing Freeview muxes, I'll go up into the loft and give the
amplifier a quick blast with my hot air gun ! *If that doesn't fix it then
I'll think again.

Of course I really should have said, "I haven't heard of a faulty
masthead amp where the fault only showed
up when it was cold SO FAR!" I was tempting fate really, and I'll
probably come across one in the next few days!

Bill

In article , John Legon
wrote:
At 15:01:23 Fri, 8 Jan 2010, Jim Lesurf wrote in
article :
In article , John Legon
wrote:

Shortly before Christmas I reported the complete loss of three muxes,
and attributed this to the layer of snow on the roof at that time. I
also suggested that the sub-zero temperature might be affecting the
masthead- type amplifier.

Alternatively, the cold is accompanied by a vertical profile in the air
density, etc, which is reducing the signal level at your location. So
may have nothing to do with your amplifier.

This would have to be a very localised effect.

Depends on the details of the atmospheric distribution. And of course any
individual antenna *is* 'localised'. :-)

Recently, up here is the 'frozen north' we have had a comment on the news
that the inversion has been quite noticable in places and that on some of
the higher more isolated hills it was warmer than lower down. So given the
clear air, snow, etc, I suspect inversions have been quite widespread and
persistent.

However I've never done any measurements on propagation below 30GHz, so
can't be sure of the effects down at c1GHz! Just that the laws of physics
will be similar, I guess. :-)

There is no noticeable
loss of signal at the other end of the house where I have an external
aerial on a mast, again with masthead amplifier...

Similar here. I also have another (larger) antenna in the loft with a
following distribution amp. The signals from that are fine on the RXs it
feeds. But since DTTV will work fine if the signal is strong enough that
only tells me that this antenna picks up a bigger signal so have a larger
fade margin before I'd observe any problems.

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/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

In message
,
" writes
On Jan 8, 7:52*pm, John Legon wrote:
At 07:48:23 Fri, 8 Jan 2010,
wrote in article 523ed8b5-cdc9-42e3-9d81-2a5a664
:

On Jan 8, 1:03*pm, John Legon wrote:
Conclusion: *Reception of Freeview is fine here with sub-zero temperatures
outside, provided there is enough snow on the roof to insulate the
loft :-)

I've never heard of a faulty masthead amp where the fault only showed
up when it was cold.

That's very interesting. *The next time the temperature's below freezing
and I start losing Freeview muxes, I'll go up into the loft and give the
amplifier a quick blast with my hot air gun ! *If that doesn't fix it then
I'll think again.

Of course I really should have said, "I haven't heard of a faulty
masthead amp where the fault only showed
up when it was cold SO FAR!" I was tempting fate really, and I'll
probably come across one in the next few days!

I was once asked to have a look at a radar IF amplifier (something like
50dB gain, 10MHz wide bandpass, at 70MHz) which 'took off' as it warmed
up -and then became stable again.

Each stage of amplification was in a screened compartment, and a 12V HT
rail ran from stage to stage via ~1000pf feedthrough capacitors in the
compartment walls. There were no other decoupling components so,
effectively, there were several capacitors in parallel across the 12V
rail, each joined to the next via 1/2" of wire..

By chance, I had previously seen what can happen at RF when you directly
parallel capacitors, and it didn't take long to establish that the
capacitors were forming a paralleled tuned circuit(s). Instead of
providing the intended excellent decoupling of the 12V rail, at around
50MHz, they were doing exactly the opposite - leaving it totally
undecoupled.

As the feedthrough capacitors were of the 'High-K' type, their value
changed considerably with temperature (reducing as it increases), and a
short blast from a hair drier soon showed that the 50MHz resonance
quickly moved up to 70MHz - which was the obvious reason why the unit
went into oscillation. This was cured by the minimal modification of the
addition of a ferrite bead onto the wires linking each capacitor.

Now...
It is just possible that the masthead amplifier has a couple of
paralleled capacitors, their resonance hits a sensitive frequency at low
temperatures, and the amplifier starts to oscillate - and this is the
cause of the lack of gain. This may be aggravated by the fact transistor
gain usually increases at lower temperatures.

[Note that, despite what many textbooks and pundits may tell you,
paralleling capacitors is not a universal 'good practice' in VHF and UHF
circuitry. The technique is intended to minimise the inherent parasitic
inductance of a single capacitor. However, what happens is that even
with the shortest of leads, you can produce an unintentional
parallel-tuned circuit (think of it as the capacitance of one capacitor
resonating with the inductance of the other). If you're not careful,
such unwanted (and unexpected) resonances can have unpredictable
results.]
--
Ian

In article , Stephen
Wolstenholme wrote:
On Fri, 08 Jan 2010 15:29:39 +0100, J G Miller wrote:

On Fri, 08 Jan 2010 13:41:26 +0000, Charles wrote:

The temperature ratings of most capacitors do not include negative
temperatures.

Which raises the point that presumably they have to use a special grade
of capacitors for satellites and space craft??

Years ago I worked in a factory producing TV signal distribution line
devices. As they were for outdoor use they had to work in any weather
conditions. Testing involved a range of temperatures to simulate the
extremes of weather. Some devices failed testing at the high temperature
but nothing ever failed because it was too cold. I don't recall any
special components being used.

Bear in mind that components used for space (and in high altitude aircraft)
may need to cope with even lower temperatures than overnight in Altnaharra.
Making repairs can also be awkward if you don't have a ladder high enough.
:-)

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/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

Actually, I'd suspect it is more likely you have moisture in the system at
some point, than an amp that does not like low temps. I have used many head
amps in the past with more primitive components and never seen such a
problem. However, if condensation forms inside the box or in the coax you
get a terrible mess.

Brian

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


"John Legon" wrote in message
...

I have been surprised to find that my reception of Freeview has not been
affected by the thick layer of snow on the roof, even though I'm using a
loft aerial in a weak signal area. Reception is fine for the usual four
muxes with no evident loss of signal on the STB indicator.

Shortly before Christmas I reported the complete loss of three muxes, and
attributed this to the layer of snow on the roof at that time. I also
suggested that the sub-zero temperature might be affecting the masthead-
type amplifier.

Having lost the same muxes a few days ago, when there was no snow on the
roof but the temperature outside was four degrees below zero, it is now
clear that the amp doesn't work well at a temperature much below freezing.
Reception returned to normal during the day as the loft space warmed up.
The signal here being not far above the "digital cliff", the loss of a few
dB in amplifier gain is enough to throw Freeview over the edge...

Ironically, even though the outside temperature is now minus 5, reception
is still fine. I'm sure this is because the snow on the roof is acting as
an insulator and keeping the loft warm. I've made a remote temperature
sensor so I can monitor the loft temp without going up there. Plugs into
the game port on my PC...

Conclusion: Reception of Freeview is fine here with sub-zero temperatures
outside, provided there is enough snow on the roof to insulate the loft
:-)

They do indeed use these but increasingly, 'normal' ones are very good in
this respect, so I'd have my doubts if this is the problem here.

Brian

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


"J G Miller" wrote in message ...
On Fri, 08 Jan 2010 13:41:26 +0000, Charles wrote:

The temperature ratings of most capacitors do not include negative
temperatures.

Which raises the point that presumably they have to use a special grade
of capacitors for satellites and space craft??

No, its damp getting in or condensation. Its either in the aerial connection
box, the coax or amp.

Brian

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


"John Legon" wrote in message
...
At 13:41:26 Fri, 8 Jan 2010, charles
wrote:
In article ,
John Legon wrote:
Having lost the same muxes a few days ago, when there was no snow on the
roof but the temperature outside was four degrees below zero, it is now
clear that the amp doesn't work well at a temperature much below
freezing.

The temperature ratings of most capacitors do not include negative
temperatures.

Well, in a previous thread I mentioned that I thought the problem might
be an electrolytic capacitor, though I doubt that a drift in value would
be critical in this application. There is also the fact that masthead
amps in general don't stop working because of negative temperatures.

Perhaps the age of the amp is significant - it's been running 24/7 for
at least 15 years - though it still works fine at normal temperatures.
An electrolytic at the end of its lifetime?

--
John Legon

Ian Jackson wrote:
In message
,
" writes
On Jan 8, 7:52 pm, John Legon wrote:
At 07:48:23 Fri, 8 Jan 2010,
wrote in article
523ed8b5-cdc9-42e3-9d81-2a5a664
:

On Jan 8, 1:03 pm, John Legon wrote:
Conclusion: Reception of Freeview is fine here with sub-zero
temperatures
outside, provided there is enough snow on the roof to insulate the
loft :-)

I've never heard of a faulty masthead amp where the fault only showed
up when it was cold.

That's very interesting. The next time the temperature's below freezing
and I start losing Freeview muxes, I'll go up into the loft and give the
amplifier a quick blast with my hot air gun ! If that doesn't fix it
then
I'll think again.

Of course I really should have said, "I haven't heard of a faulty
masthead amp where the fault only showed
up when it was cold SO FAR!" I was tempting fate really, and I'll
probably come across one in the next few days!

I was once asked to have a look at a radar IF amplifier (something like
50dB gain, 10MHz wide bandpass, at 70MHz) which 'took off' as it warmed
up -and then became stable again.

Each stage of amplification was in a screened compartment, and a 12V HT
rail ran from stage to stage via ~1000pf feedthrough capacitors in the
compartment walls. There were no other decoupling components so,
effectively, there were several capacitors in parallel across the 12V
rail, each joined to the next via 1/2" of wire..

By chance, I had previously seen what can happen at RF when you directly
parallel capacitors, and it didn't take long to establish that the
capacitors were forming a paralleled tuned circuit(s). Instead of
providing the intended excellent decoupling of the 12V rail, at around
50MHz, they were doing exactly the opposite - leaving it totally
undecoupled.

As the feedthrough capacitors were of the 'High-K' type, their value
changed considerably with temperature (reducing as it increases), and a
short blast from a hair drier soon showed that the 50MHz resonance
quickly moved up to 70MHz - which was the obvious reason why the unit
went into oscillation. This was cured by the minimal modification of the
addition of a ferrite bead onto the wires linking each capacitor.

Now...
It is just possible that the masthead amplifier has a couple of
paralleled capacitors, their resonance hits a sensitive frequency at low
temperatures, and the amplifier starts to oscillate - and this is the
cause of the lack of gain. This may be aggravated by the fact transistor
gain usually increases at lower temperatures.

[Note that, despite what many textbooks and pundits may tell you,
paralleling capacitors is not a universal 'good practice' in VHF and UHF
circuitry. The technique is intended to minimise the inherent parasitic
inductance of a single capacitor. However, what happens is that even
with the shortest of leads, you can produce an unintentional
parallel-tuned circuit (think of it as the capacitance of one capacitor
resonating with the inductance of the other). If you're not careful,
such unwanted (and unexpected) resonances can have unpredictable results.]
Indeed, if something like this is happening it will be down to the cheap
"Z" tolerance (-20% +80%) capacitors, typically used for power supply
decoupling, that change their value with temperature, in some cases
quite dramatically.
See he
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/ADB5E2666EC6BA90CA2570A50016092E/$file/F3102_Y5V.pdf
or here
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/B55A869589CF0BB2852571E900692CB4/$file/F3296_ProductSelection.pdf#page=3
and more info he
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/9B4CE7C36CF87BBCCA2570A50016092B/$file/F3102_CE.pdf
Dave

--
Blow my nose to email me

Kellerman kellerman snot wrote:
Ian Jackson wrote:
In message
,
" writes
On Jan 8, 7:52 pm, John Legon wrote:
At 07:48:23 Fri, 8 Jan 2010,
wrote in article
523ed8b5-cdc9-42e3-9d81-2a5a664
:

On Jan 8, 1:03 pm, John Legon wrote:
Conclusion: Reception of Freeview is fine here with sub-zero
temperatures
outside, provided there is enough snow on the roof to insulate
the loft :-)

I've never heard of a faulty masthead amp where the fault only showed
up when it was cold.

That's very interesting. The next time the temperature's below
freezing
and I start losing Freeview muxes, I'll go up into the loft and give
the
amplifier a quick blast with my hot air gun ! If that doesn't fix
it then
I'll think again.

Of course I really should have said, "I haven't heard of a faulty
masthead amp where the fault only showed
up when it was cold SO FAR!" I was tempting fate really, and I'll
probably come across one in the next few days!

I was once asked to have a look at a radar IF amplifier (something
like 50dB gain, 10MHz wide bandpass, at 70MHz) which 'took off' as it
warmed up -and then became stable again.

Each stage of amplification was in a screened compartment, and a 12V
HT rail ran from stage to stage via ~1000pf feedthrough capacitors in
the compartment walls. There were no other decoupling components so,
effectively, there were several capacitors in parallel across the 12V
rail, each joined to the next via 1/2" of wire..

By chance, I had previously seen what can happen at RF when you
directly parallel capacitors, and it didn't take long to establish
that the capacitors were forming a paralleled tuned circuit(s).
Instead of providing the intended excellent decoupling of the 12V
rail, at around 50MHz, they were doing exactly the opposite - leaving
it totally undecoupled.

As the feedthrough capacitors were of the 'High-K' type, their value
changed considerably with temperature (reducing as it increases), and
a short blast from a hair drier soon showed that the 50MHz resonance
quickly moved up to 70MHz - which was the obvious reason why the unit
went into oscillation. This was cured by the minimal modification of
the addition of a ferrite bead onto the wires linking each capacitor.

Now...
It is just possible that the masthead amplifier has a couple of
paralleled capacitors, their resonance hits a sensitive frequency at
low temperatures, and the amplifier starts to oscillate - and this is
the cause of the lack of gain. This may be aggravated by the fact
transistor gain usually increases at lower temperatures.

[Note that, despite what many textbooks and pundits may tell you,
paralleling capacitors is not a universal 'good practice' in VHF and
UHF circuitry. The technique is intended to minimise the inherent
parasitic inductance of a single capacitor. However, what happens is
that even with the shortest of leads, you can produce an unintentional
parallel-tuned circuit (think of it as the capacitance of one
capacitor resonating with the inductance of the other). If you're not
careful, such unwanted (and unexpected) resonances can have
unpredictable results.]
Indeed, if something like this is happening it will be down to the cheap
"Z" tolerance (-20% +80%) capacitors, typically used for power supply
decoupling, that change their value with temperature, in some cases
quite dramatically.
See he
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/ADB5E2666EC6BA90CA2570A50016092E/$file/F3102_Y5V.pdf

or here
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/B55A869589CF0BB2852571E900692CB4/$file/F3296_ProductSelection.pdf#page=3

and more info he
http://www.kemet.com/kemet/web/homepage/kechome.nsf/weben/9B4CE7C36CF87BBCCA2570A50016092B/$file/F3102_CE.pdf

Dave

I eventually found a nice simple datasheet

http://www.niccomp.com/catalog/nmc3.pdf
A change in value of up to -40% is possible with a Y5V dielectric for a
temperature change from +25C to -10C.
Goodness knows what havoc this might cause in a marginal amplifier
design that's been costed by an accountant.
Dave


--
Blow my nose to email me

At 09:23:56 Sat, 9 Jan 2010, Jim Lesurf wrote in
article :
In article , John Legon
wrote:

There is no noticeable
loss of signal at the other end of the house where I have an external
aerial on a mast, again with masthead amplifier...

Similar here. I also have another (larger) antenna in the loft with a
following distribution amp. The signals from that are fine on the RXs it
feeds. But since DTTV will work fine if the signal is strong enough that
only tells me that this antenna picks up a bigger signal so have a larger
fade margin before I'd observe any problems.

When I say there is no noticeable loss of signal, I'm referring to the
signal level as displayed by the STB signal indicator bar. For the
external aerial, the indicated signal level has been consistently at
around 90% of the bar length, regardless of snow or outside temperature.
Hence AFAICS there has been no localised loss of signal reception caused
by temperature inversion in this area.

For the loft aerial and amplifier, the signal level as displayed by the
same type of STB is typically around 60-70% of the bar length, depending
on the mux. The signal appears to be well above the threshold for DTTV
and the picture is essentially "perfect".

When the loft temperature falls sufficiently, however, the signal level
for this aerial drops to little or nothing for three of the muxes, and
there is no picture. The "BBC" mux comes in at something like 30% of the
bar-length but still gives a "perfect" picture.

I hope this explains why I ascribe the reduction in signal level at low
temperatures to a loss of gain in the loft amplifier, presumably caused
by a change in the component characteristics. Frustratingly, whilst it
is very easy to avoid the problem by keeping the loft warm, it won't be
so easy to cool the amplifier sufficiently to replicate the problem....