Weatherproof Enclosure?

Have you seen this from Linksys?

Geoff,

Never, ever, put a linksys router outside. Tried it once and when it got cold it stopped working.

BTW, Hyperlink should sell direct, they did to me before.... but that was B2B sales.

Chris

Thanks for the suggestion - It's my 'fallback' position.

Cheers, Geoff Glave Vancouver, Canada

Yeah, but I'm in Canada, and therfore 'international', which means I have to order $250 USD worth of gear. I'm not sure the box will come to that much.

Cheers, Geoff Glave Vancouver, Canada

"rieker" hath wroth:

WAP54GPE is about $380.

I've had a Linksys WAP54G in a regular old NEMA 4 box for 2 years now and haven't had a problem. They are available at home depot for about $30. I put in a small fan to keep the unit cool in summer (120F temps here) and it has worked fine, even when the temp has dipped to 13F during the winter. It's probably not the best solution, but it was quick and easy and it has worked. I just took the extra precaution of sealing everything with a bead of silicone, just to be extra safe.

I flipped a tupaware cookie jar upside down , painted it white and stuck the router inside , so far over 12 months in heat up to 40 deg heat and not a problem.

Don't know how cold you're talking about, but have a Linksys WRT54G tucked up under an eaves and inside a simple Walmart cover plastic bin to keep the dust out and its been working fine for two weekes with daytime temps below freezing.

snipped-for-privacy@softtracks.com hath wroth:

Lots of choices. Actually, too many choices. All are either PVC, PolyCarbonate, or fiberglass. No metal. Try the "CL" series or "classic series" which are cheap.

Plastic is kinda nifty because you can mount the antenna inside. Try to avoid building an oven. I posted some rubbish on how to deal with heat issues in the past. For example:

etc...

You don't need a PoE adapter with a WRT54G. The internal swithcing regulator has a very wide range of applied voltage. It will run on anything between about 4VDC and 18VDC. Here's a photo of a similar unit (BEFW11S4) with the same switcher running on 3.6VDC:

Basically, just hot wire the extra 4 wires in the CAT5 cable to the circular power connector on the WRT54G. Connect the other end to the

I don't know much about dealing with cold weather. The People's Republic of Santa Cruz has very temperate weather. If you can manage it, install a bicycle valve in the case, seal the box, and pressurize it with a bicycle pump or small compressor to a few PSI above ambient pressure. That should keep out any water.

Bah. Drill a few holes and do it thyself.

More:

This is 7 pages long. Hit "next" in lower right corner. Ignore the PoE stuff.

There are antenna builders that claim that plastic that isn't spec'ed for RF may not be as "transparent" as we would like.

FWIW.

It wouldn't be hard to test a box that you'd like to use for attenuation. netstumbler and other WiFi sniffer software report DB signal strength.

google for "diy poe" and you'll come up with some diagrams.

decaturtxcowboy hath wroth:

We just had a few days below freezing in the area. Of the 8 assorted outdoor routers of various types, none even showed any sign of a problem. Also, no water in the coax cables. However, I did have a PVC mounting pipe that had some water in the bottom, which froze, and then split open.

About a year ago, Floyd L. Davidson mentioned that he had an outdoor WRT54G router inside some kind of enclosure running in cold weather. He's in Ukpeagvik (Barrow), Alaska which qualifies as very cold. Unfortunately, I couldn't find the specific article in the Google Groups archives.

The reason it works is simple. There's about 6 watts of heat coming out of the board. In a closed small enclosure, that's quite a bit of heat. It will keep the board fairly warm and generally dry. As long as no moisture condenses on the board, it will work just fine. Meltdown at high summer temperatures is a more serious problem. It's quite easy to turn such a box into an oven.

snipped-for-privacy@panix.com (Al Dykes) hath wroth:

True. However, we have a very easy test to see how much RF the plastic will absorb. Shove a sample of the plastic into a microwave oven. Insert a glass of water to keep from arcing the magnetron. If the plastic gets hot, melts, or catches fire, it's unsuitable for RF at 2.4Ghz.

It's worth quite a bit. PVC tends to get doped with all manner of disgusting fillers some of which will dissipate RF. For example, black plastic PVC pipe is mostly graphite, which is the last thing you would want in the antenna path. However, if you don't want the antenna inside, just mount it outside. End of problem.

Actually, it's VERY hard to do that with those tools. It's very difficult to do comparative measurements due to the sensitivity of the antenna test range to reflections, movement, small changes in position, and even temperature stability of the wireless devices. I consider myself quite fortunately if I can repeat an antenna test measurement to within 3dB.

What's so hard. Put the AP in a table, put a laptop with netstumbler on a table far enough away to get a middle reading and then go back and put the plastic box over the AP. Don't move anything. walk back to the laptop. look at the graph.

I agree that, in general antenna measurements are a PITA.

snipped-for-privacy@panix.com (Al Dykes) hath wroth:

Have you actually tried it? I have. Let's try something simple. Setup your laptop on the table with Netstumbler and have it probe your own access point. Never mind the pretty graphs, just look at the signal strength and SNR numbers. I'll assume you have the typical rubber ducky antennas on your access point.

Now, take a 6" or so piece of PVC or ABS pipe and slide it over one or both antennas on the access point. Record the measurements before and after inserting the pipes. Go away and get some coffee or tea. Record the before and after measurements again. Go away and watch a TV show. Record the measurements before and after. Open or close a door or window. Do it again. If you're really lucky, you'll have a +/- 3dB spread in values of signal strength, and about twice that for SNR.

Yep, especially indoors where uncontrolled measurements are the norm.

Incidentally, if you have access to a spectrum analyzer that does NOT have a digital readout, put a spread spectrum signal on the screen from a controlled source (signal generator or directly connected access point), and ask your available friends and accomplices to read the indicated level off the screen. My guess is about +/- 2dB variation depending on the focus. Incidentally, tall people tend to produce low readings, while short people produce high readings, both due to parallax.

So, have you figured out the spectrum analyzer quiz question?

Jeff Liebermann hath wroth:

Never mind. Wrong thread. Sorry.

Hi,

I have an old 802.11b WAP outside in a little birdhouse near my house. I put the WAP inside tupperware. The antenna sticks out of a hole in the lid and also out the roof the birdhouse. I simply just epoxied the hell out of the seam of the tupperware lid/base, hole in side of the base for power and CAT5, and the along the antenna hole in the lid. Power (DC, spliced the power adapter cord to run it longer) and CAT5 simply runs down a buried trench and up through the hollowed pole. The power adapter is in a closed in area (porch).

Later, after reading one of Jeff's posts, punched a couple small holes in the base to allow it to breath against condensation.

Is this the best way to do it? Probably not! I just figured I'd run the old thing until it died, but its coming up on year two of cooking...

You do understand that your Linksys is not operationaly rated for that low a temp, right? Its not just moisture that you need to be concerned about.

That was the problem I ran into for a setup in North Dakota. We ended up building a box and installed a thermostatic heater. The supplier you cited also sells boxes with heaters.

I'm not sure what the quiz is, but it's an interesting effect.

I agree that the table-top test is crude but for controlled environment (which my big room is if I don't change anything) I'm not sure why the reading would wonder around by 6dB, all on it's own.