Should a yagi 2.4 GHz antenna appear to be short circuit a..

[Guest]

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

I have a Dlink 520 PC card connected to a multi-element 15 dB yagi element
on my roof.

It connects most unreliably to my ISP - a clear line of sight some 4 km
away, especially failing to send mail containing even moderate attachments.
ping tests fail about 25% of the time. Signal strength 45% and link quality
near 100% but varying.

Tests with a multimeter show <10 ohm resistance between the outer and the
centra of the coax cable to the roof. To me, this suggests a short in the
coax cable up to the antenna. I would expect the antenna to appear to have
infinite DC resistance.

My installer assures me this 'zero' DC resistance is normal (and recommends
a 5 Ghz setup at much higher cost).

Any expert views ?

Thanks


Paul A Bristow
Prizet Farmhouse, Kendal LA8 8AB UK

 

[Guest]

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

On Fri, 6 Aug 2004 12:38:10 +0100, "Paul A Bristow"
<pbristow@hetp.u-net.com> wrote:

>Tests with a multimeter show <10 ohm resistance between the outer and the
>centra of the coax cable to the roof. To me, this suggests a short in the
>coax cable up to the antenna. I would expect the antenna to appear to have
>infinite DC resistance.

Nope. If the yagi antenna has a gamma match or a T-match (balanced
pair of Gamma matches), it will show infinite impedance because the
driven element feed goes through a coaxial capacitor. However, a
Hairpin match (also known as a Beta match), will exhibit a dead short.
If the driven element is a closed loop as in a folded dipole or loop
yagi, you will also see a dead short.

Incidentally, if you're really getting 10 ohms, something else is
wrong. DC resistance should either zero or open, not something in
between. If you are really getting 10 ohms, you might have some
corrosion or poor connections.

>My installer assures me this 'zero' DC resistance is normal (and recommends
>a 5 Ghz setup at much higher cost).

No comment without some clue as to what you are trying to accomplish,
what you have to work with, and why 2.4Ghz is deemed inadequate.

In my never humble opinion, the absolute worst high gain antenna for
2.4GHz is the yagi. The matching assembly results in a bandwidth that
is often less than the 80MHz necessary. The VSWR never seems to be
low enough. They are seriously affected by nearby objects such as the
mounting hardware. The length of a yagi to obtain the same gain as a
24dBi dish is rediculous and impractical. Radiation angle is narrow
and often skewed (boresight error) with an array of sidelobes.
They're also expensive.

In its place, I suggest either panel antennas for up to about 14dBi
gain, or a 19dBi or 24dBi barbeque grill dish antenna.

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

 

[Guest]

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

On Fri, 06 Aug 2004 09:02:01 -0700, Jeff Liebermann
<jeffl@comix.santa-cruz.ca.us> wrote:

>Incidentally, if you're really getting 10 ohms, something else is
>wrong. DC resistance should either zero or open, not something in
>between. If you are really getting 10 ohms, you might have some
>corrosion or poor connections.

I just looked up the specs on LMR-400 coax. DC resistance is 1.4
ohms/1000ft for the center conductor and 1.6 ohms/1000ft for the
shield. Since 10 ohms is far more than the coax cable resistance,
methinks there might be something wrong with you volts-ohms-guesser,
or you have a corrosion or connection problem (somewhere).


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

 

[Guest]

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

In article <sbKQc.276$5_.151@newsr2.u-net.net>, Paul A Bristow
<pbristow@hetp.u-net.com> writes
>I have a Dlink 520 PC card connected to a multi-element 15 dB yagi element
>on my roof.
>
>It connects most unreliably to my ISP - a clear line of sight some 4 km
>away, especially failing to send mail containing even moderate attachments.
>ping tests fail about 25% of the time. Signal strength 45% and link quality
>near 100% but varying.
>
>Tests with a multimeter show <10 ohm resistance between the outer and the
>centra of the coax cable to the roof. To me, this suggests a short in the
>coax cable up to the antenna. I would expect the antenna to appear to have
>infinite DC resistance.

Shouldn't be a problem. One side of the di-pole element may well be
connected to the metalwork of the aerial via a DC short that is formed
by the actual di-pole unit itself.

Suggest that theres poor aerial alignment, or something's under powered.
Any trees in the way, as these can really clobber a 2.4G signal?.

We've got a link 2x16dB gain aerials etc over neigh on 5 Km with no
problems at all

--
Tony Sayer

Bancom Communications Ltd U.K. Tel+44 1223 566577 Fax+44 1223 566588

P.O. Box 280, Cambridge, England, CB2 2DY E-Mail tony@bancom.co.uk

 

[Guest]

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

Paul A Bristow <pbristow@hetp.u-net.com> wrote:
> I have a Dlink 520 PC card connected to a multi-element 15 dB yagi element
> on my roof.
>
> It connects most unreliably to my ISP - a clear line of sight some 4 km
> away, especially failing to send mail containing even moderate attachments.
> ping tests fail about 25% of the time. Signal strength 45% and link quality
> near 100% but varying.
>
> Tests with a multimeter show <10 ohm resistance between the outer and the
> centra of the coax cable to the roof. To me, this suggests a short in the
> coax cable up to the antenna. I would expect the antenna to appear to have
> infinite DC resistance.

Is <10 7 or 8, or you can't measure how low it is.
How long is the cable?

 

[Guest]

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

Can you provide links for how to build a "24dBi barbeque grill dish
antenna"???

--
Bob Alston

"Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
news:us97h0h6athg7h7rdtbaeu50ip2c1kefno@4ax.com...
> On Fri, 6 Aug 2004 12:38:10 +0100, "Paul A Bristow"
> <pbristow@hetp.u-net.com> wrote:
>
> >Tests with a multimeter show <10 ohm resistance between the outer and the
> >centra of the coax cable to the roof. To me, this suggests a short in
the
> >coax cable up to the antenna. I would expect the antenna to appear to
have
> >infinite DC resistance.
>
> Nope. If the yagi antenna has a gamma match or a T-match (balanced
> pair of Gamma matches), it will show infinite impedance because the
> driven element feed goes through a coaxial capacitor. However, a
> Hairpin match (also known as a Beta match), will exhibit a dead short.
> If the driven element is a closed loop as in a folded dipole or loop
> yagi, you will also see a dead short.
>
> Incidentally, if you're really getting 10 ohms, something else is
> wrong. DC resistance should either zero or open, not something in
> between. If you are really getting 10 ohms, you might have some
> corrosion or poor connections.
>
> >My installer assures me this 'zero' DC resistance is normal (and
recommends
> >a 5 Ghz setup at much higher cost).
>
> No comment without some clue as to what you are trying to accomplish,
> what you have to work with, and why 2.4Ghz is deemed inadequate.
>
> In my never humble opinion, the absolute worst high gain antenna for
> 2.4GHz is the yagi. The matching assembly results in a bandwidth that
> is often less than the 80MHz necessary. The VSWR never seems to be
> low enough. They are seriously affected by nearby objects such as the
> mounting hardware. The length of a yagi to obtain the same gain as a
> 24dBi dish is rediculous and impractical. Radiation angle is narrow
> and often skewed (boresight error) with an array of sidelobes.
> They're also expensive.
>
> In its place, I suggest either panel antennas for up to about 14dBi
> gain, or a 19dBi or 24dBi barbeque grill dish antenna.
>
> --
> Jeff Liebermann jeffl@comix.santa-cruz.ca.us
> 150 Felker St #D http://www.LearnByDestroying.com
> Santa Cruz CA 95060 AE6KS 831-336-2558


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.734 / Virus Database: 488 - Release Date: 8/4/2004

 

[Guest]

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

On Fri, 06 Aug 2004 17:32:58 GMT, "Bob Alston" <bobalston9
NOSPAM@aol.com> wrote:

>Can you provide links for how to build a "24dBi barbeque grill dish
>antenna"???

Build? Not exactly. The antenna dish only looks like a barbeque
grill. It's not made from a barbeque grill. You can make a rather
high gain antenna out of a Primestar or large satellite dish antenna.
http://trevormarshall.com/lapierre.htm
or mine at:

http://www.LearnByDestroying.com/pics/wireless/antennas/slides/primestar.html

However, I don't have any easy plans for fabricating the dish portion.
Best to start with a commercial dish and add a proper feed. The feed
is really the major design challenge. To narrow an illumination
pattern and you end up just lighting up a small spot on the dish and
ending up with lousy gain. To wide a pattern and you end up with lots
of overspray, transmit loss, and pickup from the antenna edges. In
other words, you can't just use any old feed to illuminate a dish.

See:
http://www.qsl.net/n1bwt/contents.htm
for the theory and more detailed explanation.

I suggest you buy a properly designed 24dBi dish:
http://www.pacwireless.com/products/directional.shtml
http://www.fab-corp.com/B1.htm
They're not that expensive.


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

 

[Guest]

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

Bob Alston <bobalston9 NOSPAM@aol.com> wrote:
> Can you provide links for how to build a "24dBi barbeque grill dish
> antenna"???

Jeff didn't suggest that it was homemade. 24dBi is a popular commercial
size, which I think you've found at your favorite vendor site.

Commercial 24dBi, $60 at http://www.fab-corp.com/
mini-USB-"wok" dish "> 15dBi" http://www.usbwifi.orcon.net.nz/
Trevor's 27 dBi PrimeStar http://www.olotwireless.net/catala/biquad.htm
Micheal Erskine shows the NEC plot for a 21 dBi mini-dish, but I don't see
a design for it online.
http://www.freeantennas.com/2400/minidish/graphics.JPG

Hmmm. you might be right about the lack of plans online.

I recall someone going to great effort to lay up the proper wire mesh on a
mold built from a "string parabola".
http://makeashorterlink.com/?W25D15DF8

I never saw any dBi expressed for it, though.



---
Clarence A Dold - Hidden Valley (Lake County) CA USA 38.8-122.5

 

[Guest]

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

On Fri, 6 Aug 2004 19:22:59 +0000 (UTC), dold@ShouldXaXy.usenet.us.com
wrote:

>I recall someone going to great effort to lay up the proper wire mesh on a
>mold built from a "string parabola".
>http://makeashorterlink.com/?W25D15DF8

Ugh. There are better ways. In a previous life, I helped develope a
method of explosive cold forming titanium sheet into various shapes
including parabolic dishes. Not recommended for do it thyself home
systems.

I've built "flat" type parabolic dishes out of a piece of plywood and
some aluminium flashing. The parabolic shape is laid out along the
plywood and cut with a jig saw. The flashing is stapled to the edge
of the plywood forming a parabold. It has to be fairly wide to get
decent gain, but it sure it cheap and easy to build. Oh, designing
the feed was a bit of a pain. This was for a Geos weather satellite
receiver that worked on 1691Mhz. It worked well enough, but was
deemed seriously ugly, impossible to ship, high wind load, difficult
to install, mechanically sloppy, crude, aesthetically disgusting, etc.
However, everyone agreed that it certainly was cheap. It was 8ft wide
and about a foot high. Gain followed the standard parabolic aperature
calculation at about 25dBi or the same as a 1 meter dia dish.
Probably a bit more at 2.4GHz. Photos when I find and scan them.

About 2 years ago, there was a picture in Yahoo News showing satellite
dishes for sale in Kabul made from hammered together paint cans. I
wish I had kept a copy. Here's a close second, made from soda cans:
http://www.internews.org/prs/afghan_media/afgan_media_2002_02.htm

>I never saw any dBi expressed for it, though.

No pain, no antenna gain.


--
# 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

 

[Guest]

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

"Ian Stirling" <root@mauve.demon.co.uk> wrote in message
news:4113ae09$0$536$ed2e19e4@ptn-nntp-reader04.plus.net...
> Paul A Bristow <pbristow@hetp.u-net.com> wrote:
> > I have a Dlink 520 PC card connected to a multi-element 15 dB yagi
element
> > on my roof.
> >
> > It connects most unreliably to my ISP - a clear line of sight some 4 km
> > away, especially failing to send mail containing even moderate
attachments.
> > ping tests fail about 25% of the time. Signal strength 45% and link
quality
> > near 100% but varying.
> >
> > Tests with a multimeter show <10 ohm resistance between the outer and
the
> > centra of the coax cable to the roof. To me, this suggests a short in
the
> > coax cable up to the antenna. I would expect the antenna to appear to
have
> > infinite DC resistance.
>
> Is <10 7 or 8, or you can't measure how low it is.
> How long is the cable?

The smaller division on the meter is 10 ohm, but it looks more like zero.

The cable length is about 15 m :-(

Paul

 

[Guest]

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

"Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
news:us97h0h6athg7h7rdtbaeu50ip2c1kefno@4ax.com...
> On Fri, 6 Aug 2004 12:38:10 +0100, "Paul A Bristow"
> <pbristow@hetp.u-net.com> wrote:
>
> >Tests with a multimeter show <10 ohm resistance between the outer and the
> >centra of the coax cable to the roof. To me, this suggests a short in
the
> >coax cable up to the antenna. I would expect the antenna to appear to
have
> >infinite DC resistance.
>
> Nope. If the yagi antenna has a gamma match or a T-match (balanced
> pair of Gamma matches), it will show infinite impedance because the
> driven element feed goes through a coaxial capacitor. However, a
> Hairpin match (also known as a Beta match), will exhibit a dead short.
> If the driven element is a closed loop as in a folded dipole or loop
> yagi, you will also see a dead short.
> >My installer assures me this 'zero' DC resistance is normal (and
recommends
> >a 5 Ghz setup at much higher cost).
>
> No comment without some clue as to what you are trying to accomplish,
> what you have to work with, and why 2.4Ghz is deemed inadequate.

My installer says:
The problem, as we told you at install time, isn't really a question of
signal strength, it's a question of distance. 2.4GHz signals

take a finite time to travel over distance, and the MAC layer of the 802.11b
protocol has quite tight timing restrictions. This is

why most manufacturers quote 100s of metres rather than kilometers for the
operation of their equipment. We have found ways to

increase this tolerance, but only within reason, signals simply take too
long to reach you and return within acceptable limits.

Fitting a higher gain antenae will achieve 2 x things, it'll take you over
the 20db maximum permitted power output in the UK, and

it'll probably bring the pole down in the first storm, what it won't do is
improve your signal timing (and hence the quality of your

service).

The only real solution is to upgrade to 5.8GHz which has half the wavelength
and much more leniant MAC timings, making distances of

xx kilometers quite possible. This is no problem, but 5.8G equipment is
expensive as it is not widely adopted, involving an

additional cost of around £ 250.00. We can however guarantee that it will
work or we won't install it (5.8GHz tends to be all or

nothing, it works or it doesn't, there is no half way house as with
802.11b).

I'm about 4 km away from the TV tower with clear line of site (no trees or
anything in the Fresenel zone).

Views on this please.

Thanks



Paul

 

[Guest]

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

On Sat, 7 Aug 2004 11:25:03 +0100, "Paul A Bristow"
<pbristow@hetp.u-net.com> wrote:

>My installer says:
>The problem, as we told you at install time, isn't really a question of
>signal strength, it's a question of distance. 2.4GHz signals
>
>take a finite time to travel over distance, and the MAC layer of the 802.11b
>protocol has quite tight timing restrictions. This is
>
>why most manufacturers quote 100s of metres rather than kilometers for the
>operation of their equipment. We have found ways to
>
>increase this tolerance, but only within reason, signals simply take too
>long to reach you and return within acceptable limits.

Partially right as it depends upon the manufactory. Please read:
http://peertech.org/LongShotWiFiTiming
I'm late... gotta run.

>The only real solution is to upgrade to 5.8GHz which has half the wavelength
>and much more leniant MAC timings, making distances of

Absolute baloney. RF frequency has no effect on signal timing. If
what you say were true, then the timing of every TV broadcast station
would be different. They're not. The timing of TV video and 802.11
signals stays the same regardless of RF frequency.

Incidentally, there are web pages with very long 802.11b paths. The
wi-if shootout at Defcon every year has some really impressive
numbers.
http://www.wifi-shootout.com

http://www.wired.com/news/culture/0,1284,64440,00.html?tw=wn_story_top5
55 miles this year. The worlds record is 192 miles to a ballon.

>xx kilometers quite possible. This is no problem, but 5.8G equipment is
>expensive as it is not widely adopted, involving an
>
>additional cost of around £ 250.00. We can however guarantee that it will
>work or we won't install it (5.8GHz tends to be all or
>
>nothing, it works or it doesn't, there is no half way house as with
>802.11b).
>
>I'm about 4 km away from the TV tower with clear line of site (no trees or
>anything in the Fresenel zone).
>
>Views on this please.
>
>Thanks
>
>
>
>Paul
>
>
>
>

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

 

[Guest]

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

In article <t32Rc.279$5_.63@newsr2.u-net.net>, Paul A Bristow
<pbristow@hetp.u-net.com> writes
>"Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
>news:us97h0h6athg7h7rdtbaeu50ip2c1kefno@4ax.com...
>> On Fri, 6 Aug 2004 12:38:10 +0100, "Paul A Bristow"
>> <pbristow@hetp.u-net.com> wrote:
>>
>> >Tests with a multimeter show <10 ohm resistance between the outer and the
>> >centra of the coax cable to the roof. To me, this suggests a short in
>the
>> >coax cable up to the antenna. I would expect the antenna to appear to
>have
>> >infinite DC resistance.
>>
>> Nope. If the yagi antenna has a gamma match or a T-match (balanced
>> pair of Gamma matches), it will show infinite impedance because the
>> driven element feed goes through a coaxial capacitor. However, a
>> Hairpin match (also known as a Beta match), will exhibit a dead short.
>> If the driven element is a closed loop as in a folded dipole or loop
>> yagi, you will also see a dead short.
>> >My installer assures me this 'zero' DC resistance is normal (and
>recommends
>> >a 5 Ghz setup at much higher cost).
>>
>> No comment without some clue as to what you are trying to accomplish,
>> what you have to work with, and why 2.4Ghz is deemed inadequate.
>
>My installer says:
>The problem, as we told you at install time, isn't really a question of
>signal strength, it's a question of distance. 2.4GHz signals
>
>take a finite time to travel over distance, and the MAC layer of the 802.11b
>protocol has quite tight timing restrictions. This is
>
>why most manufacturers quote 100s of metres rather than kilometers for the
>operation of their equipment. We have found ways to
>
>increase this tolerance, but only within reason, signals simply take too
>long to reach you and return within acceptable limits.

Really?, Well lets see most manufactures only quote hundreds of yards
because they only intend such equipment to work over short distances
with the built in aerials provided.

Well we've used it at 5 Km and had no problems at all!. With external
aerials of course..
>
>Fitting a higher gain antenae will achieve 2 x things, it'll take you over
>the 20db maximum permitted power output in the UK, and
>

Well that depends on a number of things inc cable and connector losses
etc anyway at that range you shouldn't need such gain. Are you sure the
path is really line of sight?...

>it'll probably bring the pole down in the first storm, what it won't do is
>improve your signal timing (and hence the quality of your
>
>service).
>

Well put up a stronger pole then, silly argument!..

>The only real solution is to upgrade to 5.8GHz which has half the wavelength
>and much more leniant MAC timings, making distances of
>
>xx kilometers quite possible. This is no problem, but 5.8G equipment is
>expensive as it is not widely adopted, involving an
>
>additional cost of around £ 250.00. We can however guarantee that it will
>work or we won't install it (5.8GHz tends to be all or
>
>nothing, it works or it doesn't, there is no half way house as with
>802.11b).
>
>I'm about 4 km away from the TV tower with clear line of site (no trees or
>anything in the Fresenel zone).

It seems to me that theres another reason behind this somewhere perhaps
their system is congested at 2.4 or some other reason.....
>
>Views on this please.
>
>Thanks
>
>
>
>Paul
>
>
>
>
>

--
Tony Sayer

 

[swich]

My post it`s a little bit late, but ...one thing ...the problem you claimed is to be with the emails that have some kind of attachment ..well i did have the same problem and the email server was in the basement. So you can not say that is a connection problem between a PC -> Switch -> Server ...no more than 15 meters of CAT5e cable. The problem was with the service provider ...their server seems not to be so happy handling attachments on POP and SMTP ports. Maybe i`m wrong but i think you do not have a connection problem.

If anyone has this kind of problem, before jumping to conclusions and look into the connection quality try to test the connection using a ftp direct download, making a lot of traffic in a short time, it would be best if you know the isp`s ftp site and download smtg above 1 MB (1 to 10 MB) ...look for transfer rate jumps ..if all goes smooth your connection it`s fine ...

NOTE :
THIS IS NOT A SPECIALIST OPINION !!!
Constantin Catalin Buica
IT&C Service Technician
Romania