loss of signal through Glass

Hi,

Interesting question!

According to this:

for untinted glass, attenuation would be "low". I guess that the actual loss would probably depend on what frequency you're talking about, and the thickness of the glass itself.

: >

: : : Hi, : : Interesting question! : : According to this: : :

: for untinted glass, attenuation would be "low". I guess that the actual : loss would probably depend on what frequency you're talking about, and : the thickness of the glass itself.

Would likely be dependant on the impurities in the glass. Lead and iron comne to mind.

:

How thick? Any doping? Tinted? Metalized?

My guess(tm) is that for ordinary 1/8" window glass, you'll lose about

-1 to -2dB at 2.4GHz. However, if the window is IR reflective, or is lead doped for clarity, the attenuation will be much higher. Recent building construction at the local university uses such glass. Operating a cell phone in the building is ummm... tricky. Another is architectural glass "brick" walls, commonly found in offices. They're good for about -6dB loss because of the thickness and doping. There are also some glass compositions that are essentially RF barriers, but you probably won't run into those.

They just had to install microcells inside my company's new office building due to this effect. The company-supplied Blackberries didn't even work... genius.

In strong signal areas, I successfully installed a passive repeater. Two panel antennas on the roof, diplexer, coax to the basement, wall mounted panel for inside distribution. It wasn't a 100% fix, but it sure was cheap.

Same problem at UCSC using NextG hardware:

I once calculated that the energy consumption and costs of the micro-cellular system more than negated any energy savings from the RF blocking IR reflective windows. Of course, I was ignored. They also could have used clay or ceramic coated windows, which offer the same Low-E performance as the metalized flavor, but with less RF blocking. However, these are admittedly more expensive and less durable than metalized coatings. The system has been up since 2006. Opinions vary as to how well it works.

Incidentally, UCSC put the new microcellular up/down converters and antennas on the roof tops and in the basement. Fiber optic cables everywhere. The result is that the local microcellular site works well on the top floors, and in the basement, but is not so great in between. Oh well.

Hi,

The OP indicated the window (singular?) was 50 years old.

Jim

On Sun, 27 Sep 2009 22:29:52 -0400, ohaya wrote in :

Very old glass often had metal (e.g., lead) added.

Some of us pay good money for such glass.

From what I remember from school (about as long ago as your window), glass is a liquid. So it would tend to be much thicker at the bottom of the pane than at the top, after all these years.. So kind of wiggle your antenna higher up. :P

Glass is not a liquid:

The anecdote about the glass being thicker at the bottom was the result of misinterpreting the thickening of the glass in mideaval stained glass windows. The thickening was an artifact of the way they made the glass and not from any cold flow effects.

Ummm... none if this has anything to do with glass attenuating RF.

I've been trying to find time to hunt down one of my physics teachers to rub this one in his face...

If glass did interfere with RF signal based on the thickness of the glass, this would be very significant.

Don't feel too bad about it. I was taught the same thing in early college fizzix. I think that was about 1967. Enlightenment came much later. Today, such things provide considerable entertainment on the Mythbusters TV show and various urban legends web sites. The supply of erronious fizzix seems endless. It may take some time to convince everyone that glass is not a liquid. I eventually migth be convinced that the planet is round, but logic still tells me that it's flat.

Sure, but the change in thickness isn't all that much. Glass does not attenuated 2.4GHz very much. (Hint: Most outdoor antenna radomes are made from fiberglass). It's more dependent on the type of glass, the doping, aperture size (window panes), and any coatings, than on small changes in thickness. I'm not even sure that the difference in attenuation can be accurately measured as it's probably less than the measurement error.

The way I previously measured material attenuation was to stack up as much of the stuff as possible between the emitter and detector. That gave me large number to work with, which tended to remove small variations in measurements. So, for glass, I would take my two calibrated horn antennas, point them at each other in a handy RF anechoic torture chamber, and pile about a foot of window glass plates in between the horns. Divide the measured attenuation by the number of glass plates and we have a reasonable average for the attenuation through a window glass pane.

Incidentally, moving the antenna closer to the suggested thicker bottom edge of the window pane is a bad idea as the the window frame might interfere with the antenna pattern and possibly block some of the signal.