Can this be used as client? Jeff? Anyone?

I have an alfa usb radio RTL8187 from which I am able to connect to a
public AP with a homemade antenna/reflector, but the throughput is slow
and connection unstable.

I picked up one of these at a yard sale.
http://www.dlink.com/products/?pid=357 (DWL-G710 High Speed 2.4GHz
(802.11g) Wireless Range Extender)
The router for the public hotspot is also a dlink, but I have no direct
control over it.

Can I use this alone strictly as a client transceiver to get a better
signal? Or in conjunction with the Alfa?

I have no home network, just a lone pc, no router. OS is WinXP.

On Thu, 21 Apr 2011 02:09:35 +0200 (CEST), Anonymous <(E-Mail Removed)>
wrote:

>http://www.dlink.com/products/?pid=357 (DWL-G710 High Speed 2.4GHz
>(802.11g) Wireless Range Extender)
>The router for the public hotspot is also a dlink, but I have no direct
>control over it.

Does it use WEP, WPA, WPA2, or no encryption? Although the data sheet
claims that it will do WPA/WPA2, there might be some gotcha where it
won't work as a store and forward repeater. Easiest way to be sure is
to first update the firmware in the DWL-G710, and just try it.

>Can I use this alone strictly as a client transceiver to get a better
>signal? Or in conjunction with the Alfa?

Looking at the data sheet, there's no mention of using it as a client
bridge. It kinda looks like it's going to be only a repeater.

>I have no home network, just a lone pc, no router. OS is WinXP.

I'm not a big fan of range extenders or repeaters. However, there are
places where they will work. One of those conditions is where the
client radio (your laptop) cannot directly see/hear the remote access
point. This way, there's no way the laptop could attempt to directly
connect to the access point. Everything has to go through the
repeater. Depending on your situation, that may not be too easy to
achieve.

If you need some entertainment, try this test. Drag your laptop and
the range extender to somone's house that has an access point. (Don't
do this at a coffee shop hot spot). With the range extender OFF,
connect to the internet and run some kind of speed test, such as:
<http://www.speedtest.net>
or maybe a local server running iPerf or Jperf:
<http://openmaniak.com/iperf.php>
Now, turn on the range extender (which is assumed to be correctly
configured) and try the speed test again. I've seen some rather
gastly deteriorations in performance. Turn the repeater off, and
things work again. That's what happens when the client can hear the
access point.

In my never humble opinion, you're better off with your USB dongle,
and homemade antenna. If it's lacking in performance, get a better
antenna. Incidentally, the uglier the antenna, the better it works.

--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Jeff Liebermann wrote:

> On Thu, 21 Apr 2011 02:09:35 +0200 (CEST), Anonymous <(E-Mail Removed)>
> wrote:
>
> > http://www.dlink.com/products/?pid=357 (DWL-G710 High Speed 2.4GHz
> > (802.11g) Wireless Range Extender)
> > The router for the public hotspot is also a dlink, but I have no
> > direct control over it.
>
> Does it use WEP, WPA, WPA2, or no encryption? Although the data sheet
> claims that it will do WPA/WPA2, there might be some gotcha where it
> won't work as a store and forward repeater. Easiest way to be sure is
> to first update the firmware in the DWL-G710, and just try it.
>
> > Can I use this alone strictly as a client transceiver to get a
> > better signal? Or in conjunction with the Alfa?
>
> Looking at the data sheet, there's no mention of using it as a client
> bridge. It kinda looks like it's going to be only a repeater.
>
> > I have no home network, just a lone pc, no router. OS is WinXP.
>
> I'm not a big fan of range extenders or repeaters. However, there are
> places where they will work. One of those conditions is where the
> client radio (your laptop) cannot directly see/hear the remote access
> point. This way, there's no way the laptop could attempt to directly
> connect to the access point. Everything has to go through the
> repeater. Depending on your situation, that may not be too easy to
> achieve.
>
> If you need some entertainment, try this test. Drag your laptop and
> the range extender to somone's house that has an access point. (Don't
> do this at a coffee shop hot spot). With the range extender OFF,
> connect to the internet and run some kind of speed test, such as:
> <http://www.speedtest.net>
> or maybe a local server running iPerf or Jperf:
> <http://openmaniak.com/iperf.php>
> Now, turn on the range extender (which is assumed to be correctly
> configured) and try the speed test again. I've seen some rather
> gastly deteriorations in performance. Turn the repeater off, and
> things work again. That's what happens when the client can hear the
> access point.
>
> In my never humble opinion, you're better off with your USB dongle,
> and homemade antenna. If it's lacking in performance, get a better
> antenna. Incidentally, the uglier the antenna, the better it works.


I can't see how a repeater would be of any use in your situation.

I use the very same chipset adapter (8187L) with a cheap beam yagi (16
dBi) and this type of antenna should help a lot.

They work better than anything else I've tried for long distance (at
low cost).

Dirt cheap - don't believe the 20+ dBi claims though.

eg.

http://cgi.ebay.com.au/2-4GHz-20dBi-...LAN-WiFi-Anten
na-New-/370495487530?pt=AU_Computers_Networking_Accessorie s&hash=item564
3407e2a
--

On 22 Apr 2011 08:23:22 +1000, "Dr Who"
<(E-Mail Removed)> wrote:

>I can't see how a repeater would be of any use in your situation.

Put it on the roof top where it has a better view of the hot spot?
However, you're right. A better antenna will do wonders.

>I use the very same chipset adapter (8187L) with a cheap beam yagi (16
>dBi) and this type of antenna should help a lot.

I don't like yagi's. Details on request.

>They work better than anything else I've tried for long distance (at
>low cost).
>
>Dirt cheap - don't believe the 20+ dBi claims though.
>
>eg.
>
>http://cgi.ebay.com.au/2-4GHz-20dBi-...LAN-WiFi-Anten
>na-New-/370495487530?pt=AU_Computers_Networking_Accessorie s&hash=item564
>3407e2a

Same as:
<http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220762554050>
Hopefully, this yagi is NOT the one you're currently using. It's not
clear from your message if you have this yagi, or something else.

I recently bought two of these when the price dropped to less than the
cost of the hardware.
<http://802.11junk.com/jeffl/crud/wi-fi-yagi-that-sucks.jpg>
I figured that it might be useful. So did a friend that bought one.
Well, after much tinkering, we decided that it's a piece of junk and
that the stock 2dBi rubber ducky works better. I don't have any
numbers or test data, but when I have time, I'll do necessary testing
and denunciations.

Notice the crude dipole under the plastic radome. No matching
circuit. In the side view, notice how far the driven element is
offset from the axis of the other elements. It can be off a little,
but not almost 1/4 wavelength. This thing doth suck.

While we're on yagi's, I did an analysis of the MFJ-1800 yagi at:
<http://802.11junk.com/jeffl/antennas/mfj1800/>
<http://www.mfjenterprises.com/Product.php?productid=MFJ-1800>
It had a chance, but the designer apparently forgot to include a
matching circuit. If you happen to have a 200 ohm feed, it should
work. Otherwise, it has a minimum VSWR of about 4:1, which also doth
suck.

Incidentally, in a perfect world, the maximum gain of a yagi antenna
would be 1.66 (the gain of a dipole) times the number of elements. So,
a 15 element yagi, would have a gain of:
10 * log (1.66 * 15) = 14 dBi
20dBi is science fiction as it would require 60 elements and a boom
four times as long as the one in the eBay advertisement. Also, note
that this is best case, and does not include any matching losses.

Duh, I just figured out another big screwup with this yagi. The
designer forgot that when you have a big fat boom, the length of each
element has to include the path travelled around the boom. For
example, if the cut length on an element is 0.5 wavelengths, one
starts measuring from the end of the element, AROUND the boom, and
then to the other end of the element. Straight line won't work.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558
# http://802.11junk.com (E-Mail Removed)
# http://www.LearnByDestroying.com AE6KS

Jeff Liebermann wrote:

> On 22 Apr 2011 08:23:22 +1000, "Dr Who"
> <(E-Mail Removed)> wrote:
>
> > I can't see how a repeater would be of any use in your situation.
>
> Put it on the roof top where it has a better view of the hot spot?
> However, you're right. A better antenna will do wonders.
>
> > I use the very same chipset adapter (8187L) with a cheap beam yagi
> > (16 dBi) and this type of antenna should help a lot.
>
> I don't like yagi's. Details on request.
>
> > They work better than anything else I've tried for long distance (at
> > low cost).
> >
> > Dirt cheap - don't believe the 20+ dBi claims though.
> >
> > eg.
> >
> > http://cgi.ebay.com.au/2-4GHz-20dBi-...ss-WLAN-WiFi-A
> > nten
> > na-New-/370495487530?pt=AU_Computers_Networking_Accessorie s&hash=ite
> > m564 3407e2a
>
> Same as:
> <http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220762554050>
> Hopefully, this yagi is NOT the one you're currently using. It's not
> clear from your message if you have this yagi, or something else.
>
> I recently bought two of these when the price dropped to less than the
> cost of the hardware.
> <http://802.11junk.com/jeffl/crud/wi-fi-yagi-that-sucks.jpg>
> I figured that it might be useful. So did a friend that bought one.
> Well, after much tinkering, we decided that it's a piece of junk and
> that the stock 2dBi rubber ducky works better. I don't have any
> numbers or test data, but when I have time, I'll do necessary testing
> and denunciations.
>
> Notice the crude dipole under the plastic radome. No matching
> circuit. In the side view, notice how far the driven element is
> offset from the axis of the other elements. It can be off a little,
> but not almost 1/4 wavelength. This thing doth suck.
>
> While we're on yagi's, I did an analysis of the MFJ-1800 yagi at:
> <http://802.11junk.com/jeffl/antennas/mfj1800/>
> <http://www.mfjenterprises.com/Product.php?productid=MFJ-1800>
> It had a chance, but the designer apparently forgot to include a
> matching circuit. If you happen to have a 200 ohm feed, it should
> work. Otherwise, it has a minimum VSWR of about 4:1, which also doth
> suck.
>
> Incidentally, in a perfect world, the maximum gain of a yagi antenna
> would be 1.66 (the gain of a dipole) times the number of elements. So,
> a 15 element yagi, would have a gain of:
> 10 * log (1.66 * 15) = 14 dBi
> 20dBi is science fiction as it would require 60 elements and a boom
> four times as long as the one in the eBay advertisement. Also, note
> that this is best case, and does not include any matching losses.
>
> Duh, I just figured out another big screwup with this yagi. The
> designer forgot that when you have a big fat boom, the length of each
> element has to include the path travelled around the boom. For
> example, if the cut length on an element is 0.5 wavelengths, one
> starts measuring from the end of the element, AROUND the boom, and
> then to the other end of the element. Straight line won't work.


Wow Jeff, you really hate that yagi :-)

No, that's not the one I bought, but assumed they were all much the
same.

Mine is a 15 element Ebay model and it works well.

I did some basic testing and it certainly did the job for me.

http://users.picknowl.com.au/~gloaming_agnet/ant3.html

I suppose it's a case of suck it and see.

Cheers

Rob

--

On 22/04/2011 2:09 PM, Jeff Liebermann wrote:
> On 22 Apr 2011 08:23:22 +1000, "Dr Who"
> <(E-Mail Removed)> wrote:
>
>> I can't see how a repeater would be of any use in your situation.
>
> Put it on the roof top where it has a better view of the hot spot?
> However, you're right. A better antenna will do wonders.
>
>> I use the very same chipset adapter (8187L) with a cheap beam yagi (16
>> dBi) and this type of antenna should help a lot.
>
> I don't like yagi's. Details on request.
>
Why not
cheap and work very well on hf
work extremely well in stacks of vhf

Meanwhile, at the alt.internet.wireless Job Justification Hearings, Jeff
Liebermann chose the tried and tested strategy of:

> I recently bought two of these when the price dropped to less than the
> cost of the hardware.
> <http://802.11junk.com/jeffl/crud/wi-fi-yagi-that-sucks.jpg>
> I figured that it might be useful. So did a friend that bought one.
> Well, after much tinkering, we decided that it's a piece of junk

The directors don't seem to be insulated from the boom, from that picture.

--
<http://ale.cx/> (AIM:troffasky) ((E-Mail Removed))
09:42:25 up 12 days, 12:56, 7 users, load average: 0.06, 0.08, 0.14
"I am utterly appalled at how I have been treated like a criminal"
-- Andrew Crossley, ACS:Law, 13 August 2010

On Fri, 22 Apr 2011 09:43:39 +0100, alexd <(E-Mail Removed)>
wrote:

>Meanwhile, at the alt.internet.wireless Job Justification Hearings, Jeff
>Liebermann chose the tried and tested strategy of:
>
>> I recently bought two of these when the price dropped to less than the
>> cost of the hardware.
>> <http://802.11junk.com/jeffl/crud/wi-fi-yagi-that-sucks.jpg>
>> I figured that it might be useful. So did a friend that bought one.
>> Well, after much tinkering, we decided that it's a piece of junk

>The directors don't seem to be insulated from the boom, from that picture.

The directors do NOT need to be insulated in order to work. However,
if they are electrically connected to the boom, the circumference of
the boom needs to be included in calculating their electrical length.
--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Fri, 22 Apr 2011 18:11:21 +1000, atec77 <(E-Mail Removed)> wrote:

>On 22/04/2011 2:09 PM, Jeff Liebermann wrote:
>> I don't like yagi's. Details on request.

>Why not
> cheap and work very well on hf
> work extremely well in stacks of vhf

Short version (because I'm busy today).

Why Yagis suck:
1. Far too many side lobes and marginal front to back ratio. This is
important when trying to reduce interference, and to obtain isolation
from other systems on the same tower.
2. Lousy bandwidth for a given gain.
3. Too many parts making them difficult to build. A comparable panel
or reflector type antenna is much easier to build/manufacture.
4. Excessive boom length to get usable gain. At 2.4GHz a 19dBi dish
is about 1x2ft across. A comparable yagi is about 10ft long. A 24dBi
dish is about 2x3ft across. A comparable yagi would be about 20ft
long.
5. I now have 2 examples of yagis that don't work (MFJ-1800 and that
eBay thing).
6. Somewhat excessive beamwidth compared to a dish. See chart below
scraped from various data sheets and NEC2 models.
Type Gain -3dB BW
dBi degrees
Biquad 10 60
Panel 19 18
Dish 24 8
Yagi 15 30
Dish 15 19
Coffeecan 8 60
7. Personal prejudice: In about 1980(?), I built a 1575/1691Mhz GEOS
WEFAX system. Due to temporary insanity, I built a 25 ft long yagi
antenna out of acrylic tubing pieces with fishing line guys for
stabilization. The logic was that a yagi uses less roof space, and
was more easily shipped than a dish. It barely worked, despite
considerable expert advice and assistance. If learning what NOT to do
is considered educational, this monstrosity was a classic example. I
finally gave up and went back to a conventional dish, which worked the
first time.

On the other foot, there are some benefits to using a yagi.
1. They make great foot pegs on a tower.
2. Birds love to sit on them (horizontally polarized or lower
frequency only).
3. Ice sticks like glue to the elements.
4. They look cool.
5. Fits nicely in the trunk of a car.
6. Is easier to ship than a dish (but a panel is even easier).


--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 22 Apr 2011 17:34:47 +1000, "Dr Who"
<(E-Mail Removed)> wrote:

>Wow Jeff, you really hate that yagi :-)

I don't hate all yagis, just the junk that I keep blundering into.

If you're curious, I like Franklin/AMOS sector antennas. For example:
<http://802.11junk.com/jeffl/antennas/AMOS-7/>
It's not suitable for connecting to a remote hot spot, but great for
covering a wide area or sector.

>No, that's not the one I bought, but assumed they were all much the
>same.

Nope. At 2.4Ghz subtle differences can have a HUGE effect. The one
in your photo uses a folded dipole driven element, which is typically
about 200 ohms impedance. That can easily be matched to 50 ohms with
a 4:1 balun or very careful selection of element spacing (with some
loss in gain). The driven element is also in line with the rest of
the elements. Yours looks much better built than the abomination that
I purchased.

>Mine is a 15 element Ebay model and it works well.
>I did some basic testing and it certainly did the job for me.
>http://users.picknowl.com.au/~gloaming_agnet/ant3.html

Nice web page and tolerable testing method. Without my pile of test
equipment, I use a similar method. I take a reference antenna, with
known gain and characteristics (8dBi patch antenna less coax losses),
and compare the results. It's not hugely accurate but is cheap and
easy.

<http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=180657052471>
Looks like exactly the same antenna as the one I bought, but with a
different driven element. Also looks like a much longer and better
coax cable. My guess(tm) is that someone borrowed the design, and
then cost reduced it with a much cheaper driven element. Maybe I'll
buy one with a folded dipole driven element and compare.

Also looks like yours with the added bonus of an even more inflated
gain claim. While an ideal 15 element theoretical yagi might have
about 14dBi gain (at best), this eBay wonder claims 25dBi.

>I suppose it's a case of suck it and see.

Yep. For what little it costs, it was easy enough to buy and try.

--
Jeff Liebermann (E-Mail Removed)
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558