Re: Fwd: Re: [OT] LAN cable tester in Christchurch to borrow please

[Patrick Dunford]

Christopher Sawtell wrote:

On Sat, 06 Mar 2004 23:35, Patrick Dunford wrote:
 

Christopher Sawtell wrote:
   

Got caught by the flippin' Reply-To: again. Sorry

----------  Forwarded Message  ----------

Subject: Re: [OT] LAN cable tester in Christchurch to borrow please
Date: Sat, 06 Mar 2004 21:12
From: Christopher Sawtell <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
On Sat, 06 Mar 2004 18:00, Yuri de Groot wrote:
     

Hi,

Does anyone have an advanced LAN cable tester I can borrow
in Chch?
I have one that tests each of the 8 strands for electrical
conductivity, but it does not test ability to carry ethernet
traffic - if ya know wha I mean.
       

I came across a very similar situation to your's a while ago.
After much confusion with very odd intermittent faults it transpired that
the poorly put together Ethernet wiring would happily carry 10MHz
signals, but not 100MHz. Sometimes the net worked ok as a 10MHz net, at
others it failed as a 100MHz. We were confused for quite a while. After
rewireing it properly, it worked perfectly.
The point of the story is that it might be worth your while putting an old
10MHz card in one of the machines to force your net to run at 10MHz. If it
works at the lower frequency, then that more or less proves that you have
problems with noise on the cable run.
     

What can cause noise on network cables?
   

Cross-talk from the mains would be the most usual. In any situation remotely 
'industrial' the mains wiring can be a real noise generator.

As a previous poster stated running Ethernet close to mains wiring is usually 
the problem. imho, At least 50 cms in practice leave at least one stud space 
between them in a wall.
 

What about patch leads and power leads of computers running close together?