Building cat5e Cables

[brownjm74]

If this has been posted elsewhere I apologize, but I couldn't find anything. I need to make quite a few cables and I didn't know if there is a specific way these need to be built. Such as the wires need to be reversed on the other end or what not. Thanks in advance for your help.

[Don]

Jeremy,

Check out this link: http://en.wikipedia.org/wiki/TIA/EIA-568-B

It some good information on how a CAT5 cable should be wired.

[bwaldrep]

Straight through (same on both ends) is what you want to do. the T568B standard would be my recommendation to follow.

[Frank A.]

Good evening

T568B is what you want on both ends for

LOR network cables

Frank A.


Attached files

[WhitePlainsNY]

Do yourself a favor and buy a cable tester too. Its very easy to crimp the ends and then find out that all the cables aren't engaged. Better to test it while you've got the cable in your hands versus after you've run the cable and connected it to your gear.

[DrHudd]

Frank A. wrote:

Good evening

T568B is what you want on both ends for

LOR network cables

Frank A.

Just to make sure everyone understands...T568A and T568B are BOTH straight through wiring standards. I completely agree that all of your cables should be the same either 568A or 568B, but it really doesn't matter as long as the cables are terminated the same on each end.

If you had three controllers connected with two seperate cables and one was wired (on both ends) as 568A and the other was wired (on both ends) as 568B there would be no difference in the connections. All pins on all RJ45's are straight through. Just don't terminate your cables as 568A on one end and 568B on the other.

Today, most network cables are terminated to 568B. At work we use 568B for network cables and 568A for phones. We use CAT5 for our phone cables and do not use USOC standards because they only allow for 3 pair.

Hope this helps,

DrHudd

[Dennis Cherry]

I just bought a cable tester on EBay for $1.00. It comes from China. It took 2 week for delivery and got it this past Saturday.

Nice basic tester, it scans the connections and uses LED indicators on both ends to show you the connections.

http://cgi.ebay.com/USB-LAN-Network-Phone-Cable-Tester-RJ11-RJ12-RJ45-Cat5_W0QQitemZ160410302759QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item255932f927

Cost about $3.00 shipping.

[brownjm74]

Thanks guys for the info. I just ordered the tester from E-bay. I was searching a little more into the forum and found info on making your own cables however I was looking under cat5e not cat5. Thanks again!

[ErnieHorning]

In the end, it really doesn’t matter whether you follow any standard as long as the wires are in the same order on both ends. The data rate of LOR, or any light controller for that matter, is so slow compared to what the cable was designed for.



That being said, its recommended that you use the T568A or T568B standard just incase you should ever use one of these cables for an actual Ethernet connection; you’d thwart a lot of frustration.

[-klb-]

You know, I had a coworker that figured that as long as it was consistent from end to end, it did not matter how they were pinned out for ethernet either.. He just could not figure out why, when the cheap led test tool said the cable was fine, the cables were not reliable for ethernet.. The reason was that he had pairs on 1-2, 3-4, 5-6, and 7-8. Ethernet (and LOR) expect that a pair is on 4-5, and 3-6. So the way he wired things resulted in the pairs being split up from their intended uses, or effectively straight wire on those 4 pins. LOR really only cares that 4-5 is a pair, but if you don't respect that, you might as well just use phone cable. You won't get the extra noise immunity that you are paying for with cat 5, unless 4-5 is wired as a pair.

Using either 468B, or 468A will respect that requirement.

[KenL_MCSE]

-klb- wrote:

You know, I had a coworker that figured that as long as it was consistent from end to end, it did not matter how they were pinned out for ethernet either.. He just could not figure out why, when the cheap led test tool said the cable was fine, the cables were not reliable for ethernet..

Tell your co-worker the reason they tested fine but were not fine is noise.

the reason the why 3 and 6 are split is to stop cross talk between the pairs. The cheap testers like the one listed for $1 on ebay send a 9v power down the line to turn on a led bulb. A real cable tester checks for cross talk (noise) between the cables.

I never realized LOR only uses two wires to send signals from box to box. Interesting!

Just for general knowledge to all a LAN only uses 4 of the wires. A lot of us network guys have been known to split the cables to have two nics use 1 cat5 cable (not recommended in production!!!)

[bwaldrep]

KenL_MCSE wrote:

Just for general knowledge to all a LAN only uses 4 of the wires. A lot of us network guys have been known to split the cables to have two nics use 1 cat5 cable (not recommended in production!!!)

An additional note, a 100-T LAN uses 4 wires (2 pairs), a 1000-T (Gig) uses all 8 wires (4 pairs)

[rgardner518]

brownjm74 wrote:

Thanks guys for the info. I just ordered the tester from E-bay. I was searching a little more into the forum and found info on making your own cables however I was looking under cat5e not cat5. Thanks again!

Whats the difference between cat5 and cat5e? Don't you want cat5e? That's what LOR sells you when you buy it from them.

[KenL_MCSE]

5e is just like cat6 made to handle gigabyte transfers.... Overkill for what we are doing



http://www.networkcablingdirectory.com/articles/structured-network-cabling-id_1151.htm

[brownjm74]

I'm using 5e just because I was able to get 1000' of green for $60.00!

Jeremy B.

[rgardner518]

brownjm74 wrote:

I'm using 5e just because I was able to get 1000' of green for $60.00!

Jeremy B.

Were! Spread the news! I don't know whats better to have. Shielded or unshielded?

But I just found 1000' unshielded for $43.25 with free shipping on Ebay!

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=400080148848&_trksid=p2759.l1259

[rgardner518]

KenL_MCSE wrote:

5e is just like cat6 made to handle gigabyte transfers.... Overkill for what we are doing
http://www.networkcablingdirectory.com/articles/structured-network-cabling-id_1151.htm

That's for the tip! It Seriously educated me in ways I "really" never new. Going fourth, I now since having started with Cat5e I'm going to keep going that way.

One more thing. "Is there a way to tell the difference between the to 5's"

[ErnieHorning]

ny_yankee_25 wrote:

"Is there a way to tell the difference between the to 5's"

CAT5e is typically labeled with the word ‘Enhanced’, hence the letter ‘e’. Otherwise the manufactures part number along with their name or logo might help identify whether it’s CAT5e.



The cabling isn’t going to matter, go with whatever is cheapest; CAT5e, CAT5 or even CAT3. They’re all way over the 1.2 mega bit rate required by the LOR communications.



A previous post implied that the twisting in the wires would help with noise. If the cable is use in an unmodified way, there would be a slight increase in immunity due to the small amount of extra capacitance from forcing the wires closer together because of the twisting. The twists are there only to prevent crosstalk from the other pairs in the cable, which aren’t being used by LOR anyway. Induced magnetic noise from outside the cable is nullified by the differential nature of RS485.

[brownjm74]

ny_yankee_25 wrote:

brownjm74 wrote:

I'm using 5e just because I was able to get 1000' of green for $60.00!

Jeremy B.

Were! Spread the news! I don't know whats better to have. Shielded or unshielded?

But I just found 1000' unshielded for $43.25 with free shipping on Ebay!

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=400080148848&_trksid=p2759.l1259

A local commucations company went out of business and I know the owner and she made me a deal. Sorry just a one time deal.

[-klb-]

Let me know when you can find a solid source that indicates that twisted pair cabling does not substantially improve EMI signal rejection on differential signalling. Yes, it is more important for reducing cross talk, as your interference sources are tightly bundled, but it still will reduce other sources of EMI by the same mechanism. The things I have read to self check seem to indicate that while twisted pair doesn't do anything for single ended communication, it actually does more for foreign noise immunity for differential signalling than differential mode itself.