Well, glass, silicon dioxide is a good dielectric. And dielectric materials can make RF lenses. So if it is flat, it will not refract the signal and should faithfully transmit it with low loss.
The amount of loss, assuming you are out of the reactive near field range, is related to a factor called the loss tangent or dissipation factor. It is dependent on frequency. Air =0 (depends on weather and atmospheric parameters) Glass = .02 (decreases with higher frequency) ABS plastic I use for radomes = .01 Wood = as much as .4 Commonly in the .02 range @ 3 GHz if dry. Walnut wood = 1.4 @ 10 MHz Water = .157 Bacon (smoked) = .05 From: Christopher Gray Sent: Saturday, February 20, 2016 4:00 PM To: [email protected] Subject: Re: [AFMUG] Antennas Behind Wood or Glass in Old Building? The building is 200+ years old, so most of the glass is old enough. I've requested to replace some panes with acrylic sheets, but I don't think they will let me. Thanks for the feedback. It sounds like mounting behind glass is much preferred over wood. I have not found good loss estimates yet, but I haven't dug into it too far. On Sat, Feb 20, 2016 at 4:59 PM, Eric Kuhnke <[email protected]> wrote: Old windows are not so bad RF transparency wise. It's even possible to use 80 GHz through glass in high rise office buildings that predate 1982 or so, when metallic coatings and special IR/UV coatings on windows started to become possible. It's the *new* windows you have to worry about. On Fri, Feb 19, 2016 at 7:37 PM, Christopher Gray <[email protected]> wrote: I may have an opportunity to install some radios inside a steeple with some very specific requirements. I'm currently considering 5 GHz and 3.65 GHz radios for this location. I'd like to do some PTP and PMP links, but I cannot afford to lose too much. I have the option between mounting behind 1" thick solid boards, 2x 1" thick solid boards, or behind original windows. Are locations with such barriers even worth entertaining? If so, would it be best to ask for locations behind wood or glass? Thanks you, Chris
