Re: [SLE] rj45 connections
Thanks everyone who contributed. I'm going with the official wiring as we have several runs of around 40 metres. It may also explain some failed NFS stuff we had a while ago. Cheers, Steve. On Saturday 19 April 2003 04:00, you wrote:
It will work fine. I will let you in on a secret the electrons in the wire do not know what color is on the insulation. As long as you always do it the same and put the color striped wire next to the the same solid color one it will work. The only problem is when someone else looks at it as says you are not doing it right.
The electrons still don't know what color the insulation is. The twisted pairs get twisted around each other many times in a long run of cat 5 cable.
pben
On Friday 18 April 2003 02:32 pm, fsanta wrote:
Hi and thanks. So it seems that mine is totally wrong and shouldn't work. So why does it? Or maybe it's not working as fast as it should be? Steve.
On Saturday 19 April 2003 9:09 am, fsanta wrote:
Thanks everyone who contributed. I'm going with the official wiring as we have several runs of around 40 metres. It may also explain some failed NFS stuff we had a while ago. Cheers, Steve.
On Saturday 19 April 2003 04:00, :
It will work fine. I will let you in on a secret the electrons in the wire do not know what color is on the insulation. As long as you always do it the same and put the color striped wire next to the the same solid color one it will work. The only problem is when someone else looks at it as says you are not doing it right.
<pedantic mode=on> The electrons do not know what colour wire they are in, but they certainly know it is harder work if their DC return is not down the respective colour or stripe of the same twisted pair. It is harder work, because they are pushing electrons in other loops which are not twisted pairs and generating cross talk. In part this hinges on the phrase 'next to'. If this means 'go' and 'return' of the same circuit, OK, but it is not automatic that these will be adjacent numberd pins, and even if they are in a patch cable, they might not be in the connectors for premises wiring. <pedantic mode=off> There are 40320 ways to wire an rj45. All of these, if matched consistently at the other end, will provide the correct dc paths, but of these, only 384 will provide the correct ac paths with 'go' and 'return' always in a single twisted pair. Verifying these figures is an exercise for the reader, [and if I got the numbers wrong, remember the pedantic attribute is 'off'...]. But the point remains: It is easier, at the time and more so for later maintenance, to stick to the standard, rather than pick 1 of 384 from 40320 and apply the same 1 of 384 at the other end.
The electrons still don't know what color the insulation is. The twisted pairs get twisted around each other many times in a long run of cat 5 cable.
See my previous post. It is not random, the design is deliberate and done in such a way that all of the crosstalk between pairs cancels out to the maximum practicable extent. regards Vince Littler
participants (2)
-
fsanta
-
Vince Littler