USB mouse cable has an effect on signal strength

How does this work?

I just moved my old test PC (1ghz socket 370 M754 PCchips with Belkin
54g PCI card) to the back room of the house.

Its now 4 rooms away from the access point in a double brick wall house.

I expected signal strength to suffer, and it did.

Even so the Belkin 54g setup didn't disgrace itself too badly, and I
recorded 43% signal strength at 48 mbs.

As the arial at the back of the PC was facing away from the signal (ie
the ATX case is between it and the access point) I placed the metal
side panel off of an old PC case about 5" away from it to reflect the
signal to it and past the ATX case towards the access point.

This brought the strength up to 56% at 54mbs. Not too bad.

Then as I moved cables around to get a good route for the mouse cable I
noticed that the signal strength altered about 10% depending how I
routed it.

Interesting.

After playing around with various positions I finally got a stable 63%
strength.

How does this come about?

Is the USB mouse cable acting as an arial or primitive repeater in
combination with the reflector panel?

Overall I thought the Belkin did a good job.

Are there better mediums than sheet steel to act as a reflector...eg.
alfoil sheet?

Cheers

Rob

"me here" <(E-Mail Removed)> hath wroth:

>Is the USB mouse cable acting as an arial or primitive repeater in
>combination with the reflector panel?

Yes. The USB dongle has a very short antenna inside which belches RF
in almost all directions including backwards towards the connector end
of the dongle. The cable just reflects that signal towards the access
point. It's not a great reflector, but does add a little gain. I
would not count on it being a stable configuration because such
arrangement are affected by literally everything in the room. If you
want to experiment, punch a hole in the middle of a pie tin, pass the
USB dongle through, and see if works better.

>Are there better mediums than sheet steel to act as a reflector...eg.
>alfoil sheet?

Yes. See:
http://www.freeantennas.com
for some clues on building reflectors. You can pickup quite a bit of
gain with a reflector.

Sheet mild steel is kinda marginal for RF reflectors. Aluminum,
stainless steel, and copper flashing, work best.

--
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:

> "me here" <(E-Mail Removed)> hath wroth:
>
> > Is the USB mouse cable acting as an arial or primitive repeater in
> > combination with the reflector panel?
>
> Yes. The USB dongle has a very short antenna inside which belches RF
> in almost all directions including backwards towards the connector end
> of the dongle. The cable just reflects that signal towards the access
> point. It's not a great reflector, but does add a little gain. I
> would not count on it being a stable configuration because such
> arrangement are affected by literally everything in the room. If you
> want to experiment, punch a hole in the middle of a pie tin, pass the
> USB dongle through, and see if works better.
>
> > Are there better mediums than sheet steel to act as a
> > reflector...eg. alfoil sheet?
>
> Yes. See:
> http://www.freeantennas.com
> for some clues on building reflectors. You can pickup quite a bit of
> gain with a reflector.
>
> Sheet mild steel is kinda marginal for RF reflectors. Aluminum,
> stainless steel, and copper flashing, work best.

Thanks Jeff.

So the USB cable (this is not a wireless mouse) is just acting as a
reflector?

I suppose this could apply to any other cable in the near vicinity?

Cheers

Rob

On 25 May 2006 10:07:07 +0950, "me here" <(E-Mail Removed)>
wrote:

>So the USB cable (this is not a wireless mouse) is just acting as a
>reflector?
>
>I suppose this could apply to any other cable in the near vicinity?

Yes. Any metal in the general vicinity of the USB dongle will have an
effect on the radiation pattern and gain. My rule of thumb is that
metal has to be at least 6 wavelengths (or more) away from the RF
source to not have a major effect. That's about 75cm at 2.4GHz. The
tiny 2.4GHz antenna found in most USB dongles probably has a gain of
about -3dBi. Put a pie plate refector behind the USB dongle and
you'll perhaps get up to +8dBi gain. That's a huge improvement. I
don't have a clue what tinkering with the USB cable will do, but it
probably won't be any worse than the original -3dBi gain.

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

Jeff Liebermann <(E-Mail Removed)> wrote in
news:(E-Mail Removed):

> On 25 May 2006 10:07:07 +0950, "me here" <(E-Mail Removed)>
> wrote:
>
>>So the USB cable (this is not a wireless mouse) is just acting as a
>>reflector?
>>
>>I suppose this could apply to any other cable in the near vicinity?
>
> Yes. Any metal in the general vicinity of the USB dongle will have an
> effect on the radiation pattern and gain. My rule of thumb is that
> metal has to be at least 6 wavelengths (or more) away from the RF
> source to not have a major effect. That's about 75cm at 2.4GHz. The
> tiny 2.4GHz antenna found in most USB dongles probably has a gain of
> about -3dBi. Put a pie plate refector behind the USB dongle and
> you'll perhaps get up to +8dBi gain. That's a huge improvement. I
> don't have a clue what tinkering with the USB cable will do, but it
> probably won't be any worse than the original -3dBi gain.
>

Jeff,

Why do you keep saying USB dongle ?

From the OP.....

'I just moved my old test PC (1ghz socket 370 M754 PCchips with Belkin
54g PCI card) to the back room of the house.'

'As the arial at the back of the PC was facing away from the signal'

On Thu, 25 May 2006 16:36:18 -0500, DanS
<(E-Mail Removed)> wrote:

>Why do you keep saying USB dongle ?

Too much RF exposure, brain damage, too many windows open on the
monitor, trying to do too much at the same time, talking on the phone
with customers while typing, watching TV on the side, lack of sleep,
irate at moderators playing censor, eating my own cooking for dinner,
limited attention span, trying to remember what was posted days later
instead of re-reading the entire thread, etc. Lots of reasons.

>From the OP.....
>
>'I just moved my old test PC (1ghz socket 370 M754 PCchips with Belkin
>54g PCI card) to the back room of the house.'
>
>'As the arial at the back of the PC was facing away from the signal'

Ok, I missed it. I saw:
"Is the USB mouse cable acting as an arial or primitive repeater
in combination with the reflector panel?"
and somehow turned it into a USB wireless dongle. Sorry.

Anyway, expept for the pie plate suggestion, my comments still apply.
All the neaby metal wires, cables, and case parts have an affect on
the antenna pattern. The exact effect is difficult to estimate.
Moving the wires around randomly is not a great idea because even if a
combination can be found that works, it will not be stable if anything
moves. The best plan is to get the antenna out of the tangle of wires
with an external antenna.

Like all bad ideas, there's always a way to make it worse:
ftp://ftp10.dlink.com/images/product...47/DWA-547.jpg
Sigh.

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

Jeff Liebermann wrote:

> On Thu, 25 May 2006 16:36:18 -0500, DanS
> <(E-Mail Removed)> wrote:
>
> > Why do you keep saying USB dongle ?
>
> Too much RF exposure, brain damage, too many windows open on the
> monitor, trying to do too much at the same time, talking on the phone
> with customers while typing, watching TV on the side, lack of sleep,
> irate at moderators playing censor, eating my own cooking for dinner,
> limited attention span, trying to remember what was posted days later
> instead of re-reading the entire thread, etc. Lots of reasons.
>
> > From the OP.....
> >
> > 'I just moved my old test PC (1ghz socket 370 M754 PCchips with
> > Belkin 54g PCI card) to the back room of the house.'
> >
> > 'As the arial at the back of the PC was facing away from the signal'
>
> Ok, I missed it. I saw:
> "Is the USB mouse cable acting as an arial or primitive repeater
> in combination with the reflector panel?"
> and somehow turned it into a USB wireless dongle. Sorry.
>
> Anyway, expept for the pie plate suggestion, my comments still apply.
> All the neaby metal wires, cables, and case parts have an affect on
> the antenna pattern. The exact effect is difficult to estimate.
> Moving the wires around randomly is not a great idea because even if a
> combination can be found that works, it will not be stable if anything
> moves. The best plan is to get the antenna out of the tangle of wires
> with an external antenna.
>
> Like all bad ideas, there's always a way to make it worse:
> ftp://ftp10.dlink.com/images/product...47/DWA-547.jpg
> Sigh.

Sorry to say it, but I was a bit confused on the dongle thing also.

I didn't push the point - as Jeff has been extremely helpful in the
past regarding setting up a 3Com888 for my dialup, and just appeared to
have assumed the arial was a dongle type.

In the grand scheme of things it was a non-issue.

Moving right along, I suppose I have two options with this.

1) Continue to use a reflector panel of some sort to bounce the signal
back towards the rear of the PC to the PCI arial.

or

2) see if I can get a remote arial with a cable connector to the 54g
Belkin PCI card.

Would I have to use a Belkin one, or can I use aftermarket ones - are
these connectors standard?

Directional or with a reflector?

If I continue to use a reflector attached to the wall behing the PC as
I am now, would I be better off with a flat pannel (as is now) or would
a curved panel be better?

Flat is easier to fit as it hides behind the curtain :-)

Cheers

Rob

On 26 May 2006 10:25:25 +0950, "me here" <(E-Mail Removed)>
wrote:

>Jeff Liebermann wrote:
>> Like all bad ideas, there's always a way to make it worse:
>> ftp://ftp10.dlink.com/images/product...47/DWA-547.jpg
>> Sigh.

>Moving right along, I suppose I have two options with this.
>
>1) Continue to use a reflector panel of some sort to bounce the signal
>back towards the rear of the PC to the PCI arial.

Yech. Don't bother.

>2) see if I can get a remote arial with a cable connector to the 54g
>Belkin PCI card.

Yep.

>Would I have to use a Belkin one, or can I use aftermarket ones - are
>these connectors standard?

Anything with a Reverse-SMA connector or adapter from something else.

>Directional or with a reflector?

You have quite a few options. Since this isn't going to be moving (as
in a vehicle) I suggest a directional antenna. I don't think it has
to have much gain. An 8dBi patch antenna will work. The only problem
is that almost all of the aftermarket antennas have excessively small
cables that are also excessively long. One omni antenna I found has a
gain of 5dBi with a cable loss of 5dBi resulting in no net gain.
Therefore look for something with decent directional gain, but a
fairly short (1-2 meter) cable.

I usually build my own such antennas. I'm partial to a biquad but a
coffee can antenna will also work. If these are aesthetically
disgusting, then something smaller will work. I hate to recommend
products I haven't tried, so beware of suprises here. Some typical
products are:

| http://www.hawkingtech.com/products/...=58&ProdID=122
| http://www.hawkingtech.com/products/...=58&ProdID=221
| http://www.buffalotech.com/products/...2&categoryid=8
| http://www.buffalotech.com/products/...4&categoryid=8
There are probably more companies making similar antennas. I'm too
lazy to search for vendors. Make sure it has a RP-SMA connector or
that an adapter is available.

>If I continue to use a reflector attached to the wall behing the PC as
>I am now, would I be better off with a flat pannel (as is now) or would
>a curved panel be better?
>
>Flat is easier to fit as it hides behind the curtain :-)

Flat plate is fine. I did a model comparison of the gains of a small
Hawking "dish" antenna with the equivalent diameter flat plate
reflector.
Flat plate:
http://802.11junk.com/jeffl/antennas...lateReflector/
Dish antenna:
http://802.11junk.com/jeffl/antennas/HawkDish08/
The dish showed a gain of 7.8dBi, while the flat plate came in at
8.3dBi. Looks like the flat plat has 0.5dB more gain.

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

Jeff Liebermann wrote:

> On 26 May 2006 10:25:25 +0950, "me here" <(E-Mail Removed)>
> wrote:
>
> > Jeff Liebermann wrote:
> >> Like all bad ideas, there's always a way to make it worse:
> >> ftp://ftp10.dlink.com/images/product...47/DWA-547.jpg
> >> Sigh.
>
> > Moving right along, I suppose I have two options with this.
> >
> > 1) Continue to use a reflector panel of some sort to bounce the
> > signal back towards the rear of the PC to the PCI arial.
>
> Yech. Don't bother.
>
> > 2) see if I can get a remote arial with a cable connector to the 54g
> > Belkin PCI card.
>
> Yep.
>
> > Would I have to use a Belkin one, or can I use aftermarket ones -
> > are these connectors standard?
>
> Anything with a Reverse-SMA connector or adapter from something else.
>
> > Directional or with a reflector?
>
> You have quite a few options. Since this isn't going to be moving (as
> in a vehicle) I suggest a directional antenna. I don't think it has
> to have much gain. An 8dBi patch antenna will work. The only problem
> is that almost all of the aftermarket antennas have excessively small
> cables that are also excessively long. One omni antenna I found has a
> gain of 5dBi with a cable loss of 5dBi resulting in no net gain.
> Therefore look for something with decent directional gain, but a
> fairly short (1-2 meter) cable.
>
> I usually build my own such antennas. I'm partial to a biquad but a
> coffee can antenna will also work. If these are aesthetically
> disgusting, then something smaller will work. I hate to recommend
> products I haven't tried, so beware of suprises here. Some typical
> products are:
>
> >
> > http://www.hawkingtech.com/products/...tID=32&FamID=5
> > 8&ProdID=122
> > http://www.hawkingtech.com/products/...tID=32&FamID=5
> > 8&ProdID=221
> > http://www.buffalotech.com/products/...?productid=82&
> > categoryid=8
> > http://www.buffalotech.com/products/...?productid=84&
> > categoryid=8
> There are probably more companies making similar antennas. I'm too
> lazy to search for vendors. Make sure it has a RP-SMA connector or
> that an adapter is available.
>
> > If I continue to use a reflector attached to the wall behing the PC
> > as I am now, would I be better off with a flat pannel (as is now)
> > or would a curved panel be better?
> >
> > Flat is easier to fit as it hides behind the curtain :-)
>
> Flat plate is fine. I did a model comparison of the gains of a small
> Hawking "dish" antenna with the equivalent diameter flat plate
> reflector.
> Flat plate:
> http://802.11junk.com/jeffl/antennas...lateReflector/
> Dish antenna:
> http://802.11junk.com/jeffl/antennas/HawkDish08/
> The dish showed a gain of 7.8dBi, while the flat plate came in at
> 8.3dBi. Looks like the flat plat has 0.5dB more gain.

Thanks Jeff.

Much appreciated.

Cheers

Rob