On Sat, 24 Nov 2012 14:47:25 GMT, (Don Pearce) wrote:

On Sat, 24 Nov 2012 14:30:18 +0000 (GMT), Jim Lesurf
wrote:


That may be so, but 1800 goes through windows easier than 800.

Curious. Why is that? Are the thicknesses/layers of double glazing a better
reflector at 800 than 1800?

You just reminded me - I was working for Marconi Electronics, and I
developed a 43GHz TV transmitter for MVDS (always too expensive). Our
windows were metallized inside and out for EMC screening as part of a
security measure. I was testing a link one day from the lab roof to my
office when I realised the window was closed, and it was still working
perfectly. Turned out the glass thickness was just about perfect for
constructive interference.


I'd be interested in any test results on 800 v 1800 MHz loss through
glass. Intuition suggests there would be little difference.

On Sat, 24 Nov 2012 15:25:23 +0000, lid wrote:

On Sat, 24 Nov 2012 14:47:25 GMT, (Don Pearce) wrote:

On Sat, 24 Nov 2012 14:30:18 +0000 (GMT), Jim Lesurf
wrote:


That may be so, but 1800 goes through windows easier than 800.

Curious. Why is that? Are the thicknesses/layers of double glazing a better
reflector at 800 than 1800?

You just reminded me - I was working for Marconi Electronics, and I
developed a 43GHz TV transmitter for MVDS (always too expensive). Our
windows were metallized inside and out for EMC screening as part of a
security measure. I was testing a link one day from the lab roof to my
office when I realised the window was closed, and it was still working
perfectly. Turned out the glass thickness was just about perfect for
constructive interference.


I'd be interested in any test results on 800 v 1800 MHz loss through
glass. Intuition suggests there would be little difference.

It isn't a question of glass loss so much as penetration of the
opening. Think of a car radio. When you drive into a tunnel, the long
wavelength of medium wave can't penetrate, and it stops dead. But the
short VHF wave of FM goes probably 100 yards into the tunnel.

d

On Sat, 24 Nov 2012 15:23:37 +0000, Andy Burns
wrote:

Don Pearce wrote:

Most 3G is now at 1800MHz isn't it?

No, most 3G is 2100MHz, O2 have re-purposed some of their 900MHz
spectrum for 3G.

1800MHz was mostly Orange and T-Mobile (now EE) 2G (DCS), I think O2 and
Vodafone had a small slice too, though EE have re-purposed some as 4G
and sold some to Three for the same.

Sorry, I meant to include that - near enough the same frequency. I was
trying to distinguish it from the 800MHz band.

d

In article , Don Pearce
wrote:
On Sat, 24 Nov 2012 15:25:23 +0000, lid wrote:



I'd be interested in any test results on 800 v 1800 MHz loss through
glass. Intuition suggests there would be little difference.

It isn't a question of glass loss so much as penetration of the opening.
Think of a car radio. When you drive into a tunnel, the long wavelength
of medium wave can't penetrate, and it stops dead. But the short VHF
wave of FM goes probably 100 yards into the tunnel.

However a glass window in a brick wall isn't really the same situation as a
tunnel surrounded by earth.

Ignoring surface reflection effects I'd expect the glass to have an
absorbtion that rises with frequency in this region as it is (I think) well
below any resonant lines. So it is being absorbed by the 'wings' of all the
higher frequency resonances. (I'm assuming we can ignore d.c. conductivity
for glass.)

So the glass itself should be far more lossy at 1800MHz than 800MHz.

I'd think the window is larger than a wavelength or two in aperture. But
the thicknesses of the glass and the surrounding walls *won't* be many
wavelengths.

It would be more like an aperture than a tunnel. And in this case the
material in which the aperture is places is, itself, not totally absorbing
or reflecting. You can certainly get an adequate signal for UHF TV though a
brick wall, although it obviously isn't the ideal arrangement.

So were you just assuming what you said would be the case? Or do you have
some practical data or loss values, etc?

Above all said, AIUI one of the assumptions behind the policy is that
800MHz is preferrable to 1800MHz because it will travel though building
walls, etc, more efficiently.

I can probably find loss values for glass somewhere. But the data I have
may not go down below a few tens of GHz. :-)

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 article , Don Pearce
wrote:
On Sat, 24 Nov 2012 14:30:18 +0000 (GMT), Jim Lesurf
wrote:


That may be so, but 1800 goes through windows easier than 800.

Curious. Why is that? Are the thicknesses/layers of double glazing a
better reflector at 800 than 1800?

You just reminded me - I was working for Marconi Electronics, and I
developed a 43GHz TV transmitter for MVDS (always too expensive). Our
windows were metallized inside and out for EMC screening as part of a
security measure. I was testing a link one day from the lab roof to my
office when I realised the window was closed, and it was still working
perfectly. Turned out the glass thickness was just about perfect for
constructive interference.

Shows you have to be careful about assuming the behaviour of 'thin'
barriers. :-)

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 article ,
wrote:
On Sat, 24 Nov 2012 14:47:25 GMT, (Don Pearce) wrote:

On Sat, 24 Nov 2012 14:30:18 +0000 (GMT), Jim Lesurf
wrote:


That may be so, but 1800 goes through windows easier than 800.

Curious. Why is that? Are the thicknesses/layers of double glazing a
better reflector at 800 than 1800?

You just reminded me - I was working for Marconi Electronics, and I
developed a 43GHz TV transmitter for MVDS (always too expensive). Our
windows were metallized inside and out for EMC screening as part of a
security measure. I was testing a link one day from the lab roof to my
office when I realised the window was closed, and it was still working
perfectly. Turned out the glass thickness was just about perfect for
constructive interference.


I'd be interested in any test results on 800 v 1800 MHz loss through
glass. Intuition suggests there would be little difference.


I don't know the values here. But in general at 'low' frequencies (i.e.
well below line resonances) it approximates to a given 'loss tangent'
value. So tends to be a loss proportional to frequency. The main
modifications tend to be due to the shape and size setting up reflective
and interference effects. As you go to higher frequencies closer to the
lines the loss starts rising more swiftly than proportional to frequency.

Slainte,

So for bulk glass I'd expect the loss factor at 1800 to be a bit more than
double that at 800. But I have no idea of the values. For all I know, the
absorbtion losses for both though a window are tiny if it isn't increased
too much by resonances.

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

Andy Burns wrote:
Don Pearce wrote:

Most 3G is now at 1800MHz isn't it?

No, most 3G is 2100MHz, O2 have re-purposed some of their 900MHz
spectrum for 3G.

1800MHz was mostly Orange and T-Mobile (now EE) 2G (DCS), I think O2 and
Vodafone had a small slice too, though EE have re-purposed some as 4G
and sold some to Three for the same.

SWMBO gets a 3G carrier far more often, and in far more places on her
T-Mobile phone (now EE of course), than I do on Voodofone.

I read somewhere that they were able to horse the 2100 MHz transmitters
into their existing 1800 MHz antenna systems, and therefore have rolled out 3G
much faster than VF and 02 who have had to duplicate 2100 MHz systems
alongside 900 MHz ?


--
Mark
Please replace invalid and invalid with gmx and net to reply.

www.paras.org.uk

Don Pearce wrote:


That may be so, but 1800 goes through windows easier than 800.

Before or after the latest building regulations?

The glass required for new double glazing is relatively radio opaque
because of the way it is treated to be heat reflective.

Mark Carver wrote:

I read somewhere that they were able to horse the 2100 MHz transmitters
into their existing 1800 MHz antenna systems, and therefore have rolled out 3G
much faster than VF and 02 who have had to duplicate 2100 MHz systems
alongside 900 MHz ?

Vaguely related, I found this blog post showing the inside of a GSM base
station cavity duplexer quite interesting ...

http://laforge.gnumonks.org/weblog/2012/11/22/#20121122-inside_cavity_duplexer

On Sat, 24 Nov 2012 17:36:06 +0000, Andy Burns
wrote:

Mark Carver wrote:

I read somewhere that they were able to horse the 2100 MHz transmitters
into their existing 1800 MHz antenna systems, and therefore have rolled out 3G
much faster than VF and 02 who have had to duplicate 2100 MHz systems
alongside 900 MHz ?

Vaguely related, I found this blog post showing the inside of a GSM base
station cavity duplexer quite interesting ...

http://laforge.gnumonks.org/weblog/2012/11/22/#20121122-inside_cavity_duplexer

That's quite a filter pair. Nine pole elliptical. Very steep sides
with nulls in the adjacent bands. Easy enough to do the electrical
design, but implementing it mechanically is not so simple.

d