wireless woes

Archived from groups: alt.internet.wireless (More info?)

i thought that you might want to know that the antenna is omnidirectional...

thanks!!

Archived from groups: alt.internet.wireless (More info?)

Calvin <paxsonc001@hawaii.rr.com> wrote:
> I have scanned throught this group and if you all have previously
> answered my question, then i must not understand the answer or my
> question well enought. I was wondering if anyone could tell me why we
> keep "frying" access points? Here is our setup: We have tried 2
> different access points, a linksys wap11 v. 2.8 and a hp access point
> (made by?) and both stopped working after about 3 days to 1 week. we

What's the mains power like?
Might it be spikes killing the APs?

Archived from groups: alt.internet.wireless (More info?)

On 23 Jul 2004 15:47:33 -0700, paxsonc001@hawaii.rr.com (Calvin)
wrote:

>I have scanned throught this group and if you all have previously
>answered my question, then i must not understand the answer or my
>question well enought. I was wondering if anyone could tell me why we
>keep "frying" access points?

Have you done a post mortem? Is the xmitter dead, receiver dead, the
router dead, the power supply dead, or what? Any lights on? Any
visible damage on the boards? Fried is not exactly a technical term.

>Here is our setup: We have tried 2
>different access points, a linksys wap11 v. 2.8 and a hp access point
>(made by?) and both stopped working after about 3 days to 1 week. we
>have a 3.5 acre campus which we are trying to cover with a 26 db gain
>antenna with no amp. the antenna is about 9 feet from the access
>point connected with a thick cable that we had made at a local radio
>shop. the calbe test okay with a meter (no short)and is about 8 ft.
>long. we can rule out lightning strikes as we are in hawaii and we
>only have lightning about once a year.

Well, if you pointed your high gain antenna (dish?) at a military
radar, it's highly likely that something will get overloaded. I don't
think it will blow anything up unless you're right in the beam path.
Have you done a site survey with a spectrum analyzer?

More likely is some kind of AC power ground loop between the antenna
and the radio. Grab a volts-ohms guesser and look for AC voltage
between the antenna mounting pipe and the AC power ground (green
wire). Be careful here as anything capable of blowing up the radio,
can also deliver a jolt.

Another possibility are AC power line spikes. There are some cheapo
power line monitoring contraptions available. I helped troubleshoot
one rooftop WISP installation that was constantly hanging and
eventually just died. Every time the elevator went to the top floor,
the load on the motor was sufficient to deliver a large voltage spike
on the AC line. The clue was when we discovered that numerous other
AC powered devices in the top floors were also blowing up.


--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# jeffl@comix.santa-cruz.ca.us
# 831.421.6491 digital_pager jeffl@cruzio.com AE6KS

Archived from groups: alt.internet.wireless (More info?)

On 23 Jul 2004 15:47:33 -0700, paxsonc001@hawaii.rr.com (Calvin)
wrote:

>I have scanned throught this group and if you all have previously
>answered my question, then i must not understand the answer or my
>question well enought. I was wondering if anyone could tell me why we
>keep "frying" access points? Here is our setup: We have tried 2
>different access points, a linksys wap11 v. 2.8 and a hp access point
>(made by?) and both stopped working after about 3 days to 1 week. we
>have a 3.5 acre campus which we are trying to cover with a 26 db gain
>antenna with no amp. the antenna is about 9 feet from the access
>point connected with a thick cable that we had made at a local radio
>shop. the calbe test okay with a meter (no short)and is about 8 ft.
>long. we can rule out lightning strikes as we are in hawaii and we
>only have lightning about once a year.
>
>Mahalo nui loa!!!
>
>Calvin Paxson

Impedance mismatch between the wap, cable and antenna frying the
transmitter final amps?

Archived from groups: alt.internet.wireless (More info?)

> Have you done a post mortem? Is the xmitter dead, receiver dead, the
> router dead, the powers supply dead, or what? Any lights on? Any
> visible damage on the boards? Fried is not exactly a technical term.

I did not open the wap as I am trying* to get it replaced under
warranty. But it is only a $30 piece of plastic so I will crack it
open and let you know what I see. This is what I know about it right
now... the power light comes on, link light which is supposed to flash
quickly or stay solid (solid = connected flash = traffic) blinks
slowly on and off.... and the link light on the hub that I plugged it
into blinked simultaneously. I crack it open later today and tell you
what is *fried*.

> Well, if you pointed your high gain antenna (dish?) at a military
> radar, it's highly likely that something will get overloaded. I don't
> think it will blow anything up unless you're right in the beam path.
> Have you done a site survey with a spectrum analyzer?

As I said in my first follow-up post it is an omni antenna. I don't
have the $$$ for a spectrum analyzer. I have 2 other AP's setup on
the campus that don't have external antenna's just the factory ones
and they work just fine.

>
> More likely is some kind of AC power ground loop between the antenna
> and the radio. Grab a volts-ohms guesser and look for AC voltage
> between the antenna mounting pipe and the AC power ground (green
> wire). Be careful here as anything capable of blowing up the radio,
> can also deliver a jolt.
>
I'll test the pipe… I have a long enough lead somewhere!

> Another possibility are AC power line spikes. There are some cheapo
> power line monitoring contraptions available. I helped troubleshoot
> one rooftop WISP installation that was constantly hanging and
> eventually just died. Every time the elevator went to the top floor,
> the load on the motor was sufficient to deliver a large voltage spike
> on the AC line. The clue was when we discovered that numerous other
> AC powered devices in the top floors were also blowing up.

I have the wap's wall wart plugged into the back of a nice ups. Same
one that our server is plugged into.

Thanks again! I'll post the results of my surgery into the wap.
Learn by destroying right?

Archived from groups: alt.internet.wireless (More info?)

On 24 Jul 2004 12:22:06 -0700, paxsonc001@hawaii.rr.com (Calvin)
wrote:

>This is what I know about it right
>now... the power light comes on, link light which is supposed to flash
>quickly or stay solid (solid = connected flash = traffic) blinks
>slowly on and off.... and the link light on the hub that I plugged it
>into blinked simultaneously.

I'm not familiar with the WAP11 v2.8. However, what you're seeing
sounds almost normal. I have another guess. Has someone hacked this
router and trashed the setting or trashed the flash? If the
previously failed units had the exact same configuration and password,
it's highly probable. Try a grand reset (stomp on the reset button
and hold it for 60 seconds), re-flash the firmware, reset to defaults,
and reload the settings. Be sure to change the SNMP read and write
community names as that's the most likely point of entry.

>I crack it open later today and tell you
>what is *fried*.

I don't think you'll find anything. I was hoping for some clue as to
whether it belches RF, delivers traffic (apparently yes judging by the
flashing hub lights), and can be configured.

>As I said in my first follow-up post it is an omni antenna.

Is it at a crowded site with lots of other antennas? Some paging
xmitters belch 300 watts at 930MHz and can do lots of damage. Is the
WAP11 antenna in the pattern of any type of other xmitter?

>I don't
>have the $$$ for a spectrum analyzer.

I bought mine for peanuts. Nobody wanted it because it was too big an
not portable. You might wanna invest in one of these.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5711...
It's not terribly sensitive (thanks to the loss of not having any
processing gain), but it does work well for strong signals. I don't
have one but borrow a friends erratically.

>I have 2 other AP's setup on
>the campus that don't have external antenna's just the factory ones
>and they work just fine.

Well, then rotate the radios. Put one of those 2 AP's at the location
that destroys radios. That will determine if it's the radio or the
loation.

>I'll test the pipe. I have a long enough lead somewhere!

Ummm, just how long is this coax and what flavour? Is it grounded
anywhere?

>I have the wap's wall wart plugged into the back of a nice ups. Same
>one that our server is plugged into.

How nice? I've measured rediculously high spikes on some cheapo UPS's
when feeding a pure inductive load. Anything else that's plugged into
the UPS that resembles an inductive load (fluorescent lamps, antenna
rotator control box, motors of any sort, etc)?

Here's a cheap fun test. Find a 180v MOV (metal oxide varistor).
Solder a lower current fuse (about 0.5A) in series. Shove it into the
UPS power output. Insulate it so nobody gets fried when they touch
it. If the fuse is blow when the WAP11 blows up again, you've
experienced a high power glitch from the UPS or from the electric
utility.

>Thanks again! I'll post the results of my surgery into the wap.
>Learn by destroying right?

Right. However, I don't think you'll find any damage. Look for
electrolytic capacitors with bulging tops, charred components, fused
traces on the board, and lumps on IC's where they don't belong. Also
look for excessive heat coming from any components.


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558

Archived from groups: alt.internet.wireless (More info?)

I said that i would do a post mortem on the ap and here it is...
Nothing is burnt inside. I did find that the lan port is not working
correctly, but that is about all. I am going to replace the ground on
the antenna as a couple of students installed it for us and the ground
doesn't seem trust worthy. I am also waiting for my proxim harmony to
arrive in the mail *smile* thanks for the tip on that one it will be
a nice tool.

Thanks a ton,

Calvin Paxson

Archived from groups: alt.internet.wireless (More info?)

On 28 Jul 2004 16:34:22 -0700, paxsonc001@hawaii.rr.com (Calvin)
wrote:

>I said that i would do a post mortem on the ap and here it is...
>Nothing is burnt inside. I did find that the lan port is not working
>correctly, but that is about all.

Are you using shielded CAT5 cable with a ground sleeve around the
RJ-45 connector? If so, look for AC power ground loops.

Also, if your radio does NOT have a transformer, and uses an active
instrumentation amplifier type of differential transceiver, you could
easily blow it up with ground loops or induced voltages from nearby
motors or transformers. I've had this happen when I mounted a radio
on a rooftop HVAC box. Every time the HVAC unit started, there would
be a 50V glitch between the shield and the antenna ground. I blew up
two DWL-900AP+ radios before I realized what was happening. I
"solved" the problem by installing a cheapo 10baseT hub at the radio.
It had a transformer and was therefore unaffected by the glitches.

>I am going to replace the ground on
>the antenna as a couple of students installed it for us and the ground
>doesn't seem trust worthy.

Are the students trustworthy? Never mind, that's a rhetorical
question.

>I am also waiting for my proxim harmony to
>arrive in the mail *smile* thanks for the tip on that one it will be
>a nice tool.

Yep. They're nice. Only one warning. They're not as sensitive as
you would expect. That's because you lose -10dB because of the lack
of spread spectrum processing gain. Good luck.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831.336.2558 voice http://www.LearnByDestroying.com
# jeffl@comix.santa-cruz.ca.us
# 831.421.6491 digital_pager jeffl@cruzio.com AE6KS