An astonishing solution to a Noise Margin problem!

I'm posting this in the hope that it may help someone.

I have been doing my nut for weeks trying to sort out a poor Noise
Margin on my Netgear DG834 router. I did all sorts of experiments. I
investigated every suggestion I saw on this newsgroup and on various
bulletin boards.
I asked all my friends, some of them very experienced, and couldn't find
the cause of the problem - until this morning - when we found the cause
was a power supply on one of the computers.

If that machine was plugged into the mains, even if powered off, it
caused a severe reduction in router SNR.

Amazing! Never saw that suggested before!
I am so delighted to have tracked it down at last!

Add it to your checklist.
--
Chris

In an earlier contribution to this discussion,
Chris <nospam@[127.0.0.1]> wrote:

> I'm posting this in the hope that it may help someone.
>
> I have been doing my nut for weeks trying to sort out a poor Noise
> Margin on my Netgear DG834 router. I did all sorts of experiments. I
> investigated every suggestion I saw on this newsgroup and on various
> bulletin boards.
> I asked all my friends, some of them very experienced, and couldn't
> find the cause of the problem - until this morning - when we found
> the cause was a power supply on one of the computers.
>
> If that machine was plugged into the mains, even if powered off, it
> caused a severe reduction in router SNR.
>
> Amazing! Never saw that suggested before!

Well it has been - many times!

There has been much discussion here about noisy power supplies which come
with some 3Com routers - which prevent BB from working properly. And others
have mentioned that noise from other nearby equipment - sometimes not even
computer equipment - can have a detrimental effect.
--
Cheers,
Roger
______
Please reply to newsgroup.
Reply address IS valid, but is disposable in the event of excessive
spam.

On 16 Mar 2006 11:49, Chris <nospam@[127.0.0.1]> wrote:

>we found the cause was a power supply on one of the computers.

Hmmm, major pain if you only had that one PC... I'd seen comments
about other household kit, and sometimes router PSU, but never of
a PC PSU being the problem.

>If that machine was plugged into the mains, even if powered off,
>it caused a severe reduction in router SNR.

Wonder why/how it causes a problem when plugged in - unless (like
a Dell I have here) it isn't truly "powered off" - the Dell shows
an ethernet connection unless completely unplugged... could have
some noisy circuitry in yours, I suppose). Glad you found it!


--
USENET news service ? <http://tinyurl.com/3rjw4> (plans from < US $5 )

poster <us-(E-Mail Removed)> wrote:
>
> >If that machine was plugged into the mains, even if powered off,
> >it caused a severe reduction in router SNR.
>
> Wonder why/how it causes a problem when plugged in - unless (like
> a Dell I have here) it isn't truly "powered off" - the Dell shows
> an ethernet connection unless completely unplugged... could have
> some noisy circuitry in yours, I suppose). Glad you found it!
>
It's pretty common, I had a PC PSU that did exactly that, it destroyed
radio reception pretty effectively when plugged in even though it
wasn't powering anything. I cured the worst of it by fitting a
filtered IEC socket in place of the standard one, the noise was mostly
being radiated from the mains lead.

--
Chris Green

poster <us-(E-Mail Removed)> wrote in
news:(E-Mail Removed):

> On 16 Mar 2006 11:49, Chris <nospam@[127.0.0.1]> wrote:
>
>>we found the cause was a power supply on one of the computers.

>>If that machine was plugged into the mains, even if powered off, it
>>caused a severe reduction in router SNR.

> Wonder why/how it causes a problem when plugged in - unless (like
> a Dell I have here) it isn't truly "powered off" - the Dell shows
> an ethernet connection unless completely unplugged...

.... so that power-on by LAN/modem/keyboard/mouse will work!

In message <(E-Mail Removed)>, poster
<us-(E-Mail Removed)> writes
>Wonder why/how it causes a problem when plugged in - unless (like
>a Dell I have here) it isn't truly "powered off"
Most PCs aren't ever truly switched off unless you pull the plug or are
lucky enough to have a mains switch on the back you can turn off, all
ATX supplies have a standby mode that produces 5v at anything up to a
couple of amps.

--
Clint Sharp

On 16 Mar 2006 19:44, Clint Sharp <(E-Mail Removed)> wrote:

>all ATX supplies have a standby mode that produces 5v at anything up
>to a couple of amps.

Thanks for that - yes, some of my PCs do have a rocker on back :-)


--
USENET news service ? <http://tinyurl.com/3rjw4> (plans from < US $5 )

On 16 Mar 2006 14:57, Chris Cheney wrote:

>... so that power-on by LAN/modem/keyboard/mouse will work!

In which case, I'd expect some indicator on keyboard to still
be on. It might be a BIOS setting, of course, which makes my
keyboard and mouse "dead" when the power is "off". Peter M.


--
USENET news service ? <http://tinyurl.com/3rjw4> (plans from < US $5 )

poster <us-(E-Mail Removed)> wrote in
news:(E-Mail Removed):

> On 16 Mar 2006 14:57, Chris Cheney wrote:
>
>>... so that power-on by LAN/modem/keyboard/mouse will work!
>
> In which case, I'd expect some indicator on keyboard to still
> be on. It might be a BIOS setting, of course, which makes my
> keyboard and mouse "dead" when the power is "off". Peter M.

On one of my computers, you can just see the keyboard lights are on very
dimly, but only if the room is dark.

"poster" <us-(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> On 16 Mar 2006 11:49, Chris <nospam@[127.0.0.1]> wrote:

> >If that machine was plugged into the mains, even if powered off,
> >it caused a severe reduction in router SNR.
>
> Wonder why/how it causes a problem when plugged in...

Another possible cause (in addition to the ones already mentioned in this
thread) is an "earth loop".

If you have two earthed bits of kit plugged into different sockets, and
there is some other earthed connection between them, such as a shielded
signal cable, the earth connections form a big loop. Any varying magnetic
field (e.g. from mains or cordless phones) in the vicinity will induce a
current around that loop, and a voltage where the resistance is highest.

You can reduce this problem by reducing the size of the loop (plug things
into the same mains socket, twist cables) or disconnecting any shielding or
earth connections (dangerous!!).

--
Phil McKerracher
www.mckerracher.org