Re: Another ZigBee convert

[Joerg <notthisjoergsch@xxxxxxxxxxxxxxxxxxxxx>]

Hello Dave,

>>I wonder how well that stuff holds up in practice. Many houses have
>>in-wall insulation using fibers on an aluminum foil backer and this can
>>include some inside walls. When I walk into our garage with anything RF
>>such as a cordless phone or LAN it often quits.
>
> That type of insulation is usually only in exterior walls. Interior walls,
> when insulated, tend to use batts with either no backing kraft paper
> backing. I heven't been on a new construction site in many, many years but I
> suspect most exterior flashing which once was foil based is now Tyvek based.

Right, they do it differently on many houses today. The latest I have
seen is what looks like "Concrete-Styrofoam mix" blocks. I hope they
have investigated the toxicity in a fire before releasing that stuff to
market.

However, lots of the existing houses, especially upscale ones with
people in there who'd have the dough for home automation, have aluminum
backed fiber. It seems they kept huge rolls at the site and simply used
it everywhere insulation was needed. Garage walls are one area because
the garage is vented and very cold in winter, hot in summer. I have even
seen foil-backed fiber in ceilings.

Then there is the stucco craze to create that mediterranean feel.
Decorating interior walls typically requires a metal mesh be tacked down
first so the mortar has something to grab onto, just like outside
stucco. When we move an FM radio to the north side of a lava rock wall
some stations fade away. It can be the same behind tile walls that were
done in the classical mud bed technique. In our case it's actually all
floor areas because they floated a 2" mud bed onto mesh throughout.

> That said, I have seen more reports of RF problems from Canada than from the
> US and have wondered whether their insulation practices might be a factor.

They probably use more aluminum because it reflects heat. In Germany
pretty much any insulation was aluminum backed. When I did an attic
there it was the only thing they had. No Kraft paper, too flammable.

> Crestron is very high-end so I doubt they would introduce something that
> isn't reliable or that requires excessive service calls. Their customers,
> dealers and installers would not be happy.

Hopefully. Testing for RF blinding requires lots of miles through all
kinds of different types of houses, analyzer in hand.

> Most of the RF mesh networks require devices to be rather closely spaced
> (20-25'). The devices act as repeaters so overall range is not a major
> factor. It does mean that the number of devices tends to grow geometrically
> with the volume needing coverage (which is a plus for the manufacturer). If
> the protocol is a routed one, it also means that it may take several hops to
> traverse the network which may mean delays and even missed signals if the
> max hops is too low to traverse the network.

This "mote approach" can work nicely. But it does get old if there is a
spot where you have to place a mote just for repeater purposes and then
there is no power in that area.

Regards, Joerg

http://www.analogconsultants.com