 RickezGoinginsane
join:2000-09-02
Three Rivers, MA
 ·Comcast
| cat 5e slow, how to test I was working on a project with a buddy. We ran cat5 e up a floor made crimped the ends as described on the package and tested with a cat 5 network tester. the test ran fine. but when we tested the router with his laptop and it was extremely slow...
after checking all kinds of situations we thought of that could cause this, such as a wireless phone base near by, it just appears to be something with our wire.
our thoughts are perhaps the crimp is weak... maybe the wire is bad... ( it was not stretched as the run was loose...)
Is there a way to check the quality of the signal in run rather than just checking it for connection like the tester does?
thanks, rick
--
Got a one way ticket, going the wrong way.... |
|
 More FiberPremium,MVM
join:2005-09-26
West Chester, PA
kudos:28 | said by Rickez:Is there a way to check the quality of the signal in run rather than just checking it for connection like the tester does? There are testers for that, but they are expensive. » www.flukenetworks.com/datacom-ca···r-series•What were you testing against at the other end? If you were testing against a speedtest site, you're going to be constrained by the link to your ISP. •What speeds did you get doing the same when you connected the laptop directly to the router with a known good cable? •Are you sure the laptop was using the wired link and not it's wireless link? --
There are 10 kinds of people in the world; those who understand binary and those who don't.
|
|
bdnhsv
join:2012-01-20
Huntsville, AL | reply to Rickez
Rick - Fluke makes great test gear as had been recommended. If you intend to do this type of work I'd definitely recommend them. If this is a 1 off type deal maybe you could use something like iperf/jperf just to test this one drop. It won't tell you what is incorrect with the run but it will show you the performance. |
|
|
|
2 edits | reply to Rickez
it you tested it with a cable tester and got the pins right, then that is a good start.
how long is the run? you are limited to just over 100m with data on cat5e. |
|
| reply to Rickez
Second Fluke testers, but as said already, the high end models are EXPENSIVE, and unless you're a cabling company
that does multiple installs daily, kind of an expensive investment to check just one cable.
Regards |
|
RickezGoinginsane
join:2000-09-02
Three Rivers, MA Reviews:
·Comcast
| reply to Rickez
Our run fairly short, 25 feet or so.
The router works fine with older premade cable.
There seems to be nothing to interfere, no florescent lights minimal electrical wiring.
After coming home from the job i decided to look on the web for T568B, the order of pins on the back of the cat 5 plugs is different than all the web sites I searched. The order on the package was
os o bs b gs g bs b
I found on the web it should be:
os o gs b bs g bs b
Could this error cause the problem???
I will get t try a new connector Saturday when i return to hopefully rap this up...
as for the laptop, it was connected wireless, but the wire from the modem to the router is where we have the run with the new wire....
--
Got a one way ticket, going the wrong way.... |
|
 CabalPremium
join:2007-01-21
Austin, TX
·Suddenlink
| Yes, using the wrong wire order can cause interference issues, especially on longer runs (not that 25 feet is that long). I remember having exactly that issue when doing an attic-to-basement run many years ago (in my youth  ). Definitely start there.
--
If you can't open it, you don't own it. |
|
| the wrong wire order wont matter if he did the same thing on both ends.
i would re-crimp and follow the t568b standard.
»en.wikipedia.org/wiki/TIA/EIA-568 |
|
 shdesignsPowered By Infinite Improbabilty DrivePremium
join:2000-12-01
Stone Mountain, GA | said by tomdlgns:the wrong wire order wont matter if he did the same thing on both ends.
Yes it will matter. The impedance will be wrong as you have signals split between pairs instead of running as a pair.
10mbit will work, 100/1000 will not in most cases.
--
Scott Henion
Embedded Systems Consultant,
SHDesigns home - DIY Welder |
|
| as i stated he should follow t568b, i even linked it, but i have been on network (not that i own, just worked on) where the cables were terminated with the wrong color code, but the same on both sides and everything worked.
there is no sense in going back and forth, what works in one place might not work in another place with different/better equipment....etc. |
|
LLigetfa
join:2006-05-15
Fort Frances, ON
kudos:1 | said by tomdlgns:there is no sense in going back and forth...
...unless you are giving bad advice. The colours of the pairs may not matter but one still needs to use pairs, not 1 wire from one pair and another wire from another pair.
So, pins 1 & 2 form a pair and pins 3 & 6 form the other pair (for 10/100) while 4 & 5 form the third pair and 7 & 8 form the last pair.
--
Strange as it seems, no amount of learning can cure stupidity, and formal education positively fortifies it. -- Stephen Vizinczey |
|
| said by LLigetfa:said by tomdlgns:there is no sense in going back and forth...
...unless you are giving bad advice. The colours of the pairs may not matter but one still needs to use pairs, not 1 wire from one pair and another wire from another pair. So, pins 1 & 2 form a pair and pins 3 & 6 form the other pair (for 10/100) while 4 & 5 form the third pair and 7 & 8 form the last pair. sorry, you are right, this is what i meant. i wanted to say that if a pair is reversed on one end and it is reversed the same on the other end, it should still work.
i can see how leaving that out matters. |
|
RickezGoinginsane
join:2000-09-02
Three Rivers, MA | reply to Rickez
I am going to use the proper t568b method. And inform ace hardware of their error on the package I guess....
--
Got a one way ticket, going the wrong way.... |
|
| said by Rickez:I am going to use the proper t568b method. And inform ace hardware of their error on the package I guess....
take a pic and upload for us. |
|
 stormbowFreedom isn't FREEPremium
join:2002-07-31
Simi Valley, CA | reply to Rickez
Rick,
Tom is partially right. The reason for using pairs of wires for transmit/receive pair is that it suppresses interference. If there were no interference, the wire would work fine. If there is a power line or fluorescent light along the path, it could do all kinds of bad things to the signal. |
|
LLigetfa
join:2006-05-15
Fort Frances, ON
kudos:1 | I had a former admin use 8 cond silver satin phone cord for jumpers at the PC. The premise wiring though was all CAT5 as were the jumpers at the hub. It worked with the original 10 meg Synoptics hubs but failed miserably when I replaced the hubs with 100 meg Nortel switches.
--
Strange as it seems, no amount of learning can cure stupidity, and formal education positively fortifies it. -- Stephen Vizinczey |
|
shdesignsPowered By Infinite Improbabilty DrivePremium
join:2000-12-01
Stone Mountain, GA | reply to stormbow
said by stormbow:Rick,
Tom is partially right. The reason for using pairs of wires for transmit/receive pair is that it suppresses interference. If there were no interference, the wire would work fine. If there is a power line or fluorescent light along the path, it could do all kinds of bad things to the signal.
There is also impedance of the signal. Using mixed pair wires will cause the signal to bounce back and forth and cause wrong levels. This will cause significant errors regardless if there is noise present.
--
Scott Henion
Embedded Systems Consultant,
SHDesigns home - DIY Welder |
|
| said by shdesigns:said by stormbow:Rick,
Tom is partially right. The reason for using pairs of wires for transmit/receive pair is that it suppresses interference. If there were no interference, the wire would work fine. If there is a power line or fluorescent light along the path, it could do all kinds of bad things to the signal.
There is also impedance of the signal. Using mixed pair wires will cause the signal to bounce back and forth and cause wrong levels. This will cause significant errors regardless if there is noise present. we quickly dismissed the mixed pair scenario. |
|
stormbowFreedom isn't FREEPremium
join:2002-07-31
Simi Valley, CA | said by tomdlgns:said by shdesigns:said by stormbow:Rick,
Tom is partially right. The reason for using pairs of wires for transmit/receive pair is that it suppresses interference. If there were no interference, the wire would work fine. If there is a power line or fluorescent light along the path, it could do all kinds of bad things to the signal.
There is also impedance of the signal. Using mixed pair wires will cause the signal to bounce back and forth and cause wrong levels. This will cause significant errors regardless if there is noise present. we quickly dismissed the mixed pair scenario. Here I try to play peacemaker and both sides jump on me! A straight-through cable will work, REALLY poorly, but you can get link. It will pass traffic, but I would never run that way, and that was all I was trying to say, SHEESH.
:P |
|
| i agree, i shouldn't have said anything other than to use t568b. |
|
