On Tue, 03 Jan 2012 12:30:00 -0800, miso wrote:
I suspect Jeff is closer to reality, though I don't doubt your boxes
didn't leak. For electronics, just moisture in the air is enough. You
would probably need a nitrogen purge and pressurization scheme for
electronics. Some camera housings work that way.
Not nitrogen. Too expensive. Just dry air will work.
In a past life, I designed marine radios for Intech Inc. We learned
quite a bit about water proofing. Much of it was learned the hard
way. Most of my radios would work after the boards were soaked with a
bucket of water. There's no magic there, just use low impedances for
literally everything and fairly wide trace spacing. However, I had to
also deal with vendor supplied SCADA hardware, which was full of dense
PCB layout, high impedances, RF sensitive design, and other
nightmares. Kinda like what you find in the typical consumer grade
wireless router. The decision was made to not modify the design, but
rather to protect the sensitive SCADA boards. I won't go into all the
things that didn't work, but I will say that the only thing that
worked every time was a pressurized box, vertically mounted boards (so
that they drain) and dry air. To prevent condensation in case the dry
air went away, there was a small heater to keep the temperature above
the dew point. My idea of dry air was a bicycle pump with an air
compressor dryer filter attached. Add a gauge, a desiccant cartridge,
and the usual warning labels. If you think you can do better with
other technology, you're welcome to try.
Camphor fumes are to reduce rust. I don't know the chemistry behind
this, so I don't know how effective this scheme is for electronics.
Dunno. Most of my stuff was aluminum. There are products that you
can insert inside aluminum tubing (such as for hang gliders and
antennas) that prevent internal corrosion. I don't know the chemistry
offhand.
A weep hole doesn't keep out salt.
Actually, a weep hole has its place. If you can't pressurize with dry
air, then you have to find a place for the water to escape. Water
will eventually evaporate in temperate climates making a weep hole
functional. However, puddling is very bad. That's why I mound the
boards vertically (so they drain).
The absolute worst idea is somewhat sealed box, that's not
pressurized. It works like a water pump. Water collects on the box
seams. The sun comes up, heats up the box, causing some of the inside
air to leak out. The sun goes down, the box cools off, and a partial
vacuum is created inside the box. This sucks the water sitting on the
seams into the box. The next day, the process is repeated.
Eventually, there's quite a bit of water inside the box. It doesn't
just sit on the bottom of the box. It evaporates when warm and
condenses on the electronics. It's MUCH better if the water drain out
the bottom after every cycle instead of being trapped inside.
Fry's has the satellite F connectors. I'm not really impressed with the
satellite coax they sell, though I use their patch cables for temporary
setups.
I make my own CATV cables. The compression type of F connector is
quite waterproof. I've also used it at 2.4Ghz. Works fine. You'll
never notice the 50/75 ohm mismatch as the reflections are all lost in
the high cable losses.
http://www.qsl.net/n9zia/wireless/75_ohm_hardline.html
Quad shielded coax has been at buzzword status for a while. Almost like
drop forged. [Yeah, we dropped that wrench before shipping it to Harbor
Freight.] Yeah, there are four shields. Not the greatest shields.....
Quad shielded exists only because the FCC demanded that CATV leakage
be very very very very low. The only thing that does that is quad
shielded. If you have an ingress problem, use quad. Otherwise,
double shielded (foil + braid) works just fine. My most irritating
problem with RG6a/u is the unplated copper center conductor. It like
to corrode, especially when the mating connector has tin plated
contacts. I've been experimenting with electroless silver plating the
copper, which seems to help.
Archived notes on corrosion by a real expert (not me).
http://yarchive.net/electr/galvanic_corrosion.html
Worth reading methinks.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558