[Tfug] CAT5 Cables, The Sequel

Sat Jan 19 14:58:16 MST 2008

When cutting off the outer jacket, I cut off at least 3/4 to
an inch. That way, once all the conductors are sorted,
ordered and dressed I take a pair or cutters and trim the
wires to length.

But first, I identify and untwist the pair bundles in order.
Green, Orange, Blue, Brown  (use whatever standard you
want). Sometimes the pairs don't come out of the entire
bundle neatly, so it helps to unweave them in different
directions until they dress without interfering with each
other.

Then, beyond the first 1/8 to 1/4 inch from the outer
jacket, I unwrap the pair strands in order in a fan shape.
W-GN/GN, W-O/O, W-BU/BU, W-BN/BN. (think palm tree leaves
"fan") On the ends, I straighten the waviness in the wires
because of the twisting to make the pairs.

Then, I swap the solid blue and orange strands and double
check the order. W-GN/GN, W-O/BU, W-BU/O, W-BN/BN. Then from
the fan shape of the strands, I push them together to make
them parallel with the cable axis. I use one hand to hold
the wires together (thumb and forefinger) and the other hand
to sort things and then to cut the strands to length, using
the connector as a guide

By this point, the wires are in the correct order, lined up
next to another and cut to length. After doing this, I
identify the pin one side of the connector and cram the
ordered bundle into the connector, taking care to watch how
well the wires stay lined up, and adjusting things to make
the wires line up with the channels in the connector..

Look at the side of the connector opposite of the contacts.
Are the wires crammed fully into the connector as far as
they can go? Are the wire colors still ordered properly? If
not pull the cable out of the connector and adjust things.
Usually I can pull out on the wires from the jacket to gain
some length if the wires don't go far enough into the
connector.

If I messed up, I re-cram the bundle back into the connector
and recheck. If the wire order and length are good, I check
to make sure the outer jacket extends far enough into the
connector so the jacket crimp will hold. If Ok, it's time to
crimp.

I don't recall reading what kind of crimper Don was using.
In brief there are the cheap "plier" or "hinge" type
crimpers. Then, there are the ratchet style crimpers, the
better of which usually have replaceable dies. With the
cheap crimpers, if you are forced to use them, I think it's
a good idea to squeeze the handles tightly several times
when doing a crimp.

The only times I have had trouble with non matched kit was
when using imported connectors with a genuine AMP crimper.
The AMP crimper would crush the chinese connectors.

Beyond taking care to use connectors intended for round
cable, I don't seem to have a problem using readily
available connectors with stranded or solid wire cable.

Try using different connectors or a better crimper..

How can anyone tell whether a RJ-45 style connector is
intended for solid or stranded wire by some visual clue? If
so, how?

----- Original Message ----- 
From: "Bexley Hall" <bexley401>
To: <tfug at tfug.org>
Sent: Friday, January 18, 2008 2:49 PM
Subject: [Tfug] CAT5 Cables, The Sequel

> The Crimperama Saga
> (subtitled:  How to waste an Evening without Trying)
>
> Well, buoyed by all the positive suggestions, I
> tackled the first of the CAT5 cables last night.
>
> I spent some time googling part numbers to determine
> what sorts of connector shells I had available.
> Most turned out to be for solid wire :<  *But*,
> one bag was clearly marked for *stranded* wire
> so I was (cruelly!) encouraged.
>
> Cut cable to length (plus fudge factor).  Carefully
> removed jacket making sure not to damage insulation
> on the individual conductors within.  Removed (foil)
> shield and drain wire.
>
> Separated all conductors and then, methodically,
> removed 0.1" of insulation from the tip of each.
> Reached over for soldering iron and (belatedly)
> realized I was NOT making a "solder cup" but,
> rather, an IDC cable.  <big frown>
>
> Grab side cutters and effortlessly trim off the bare
> portions of each conductor.
>
> 1st lesson learned:  do nothing when you are tired
> (besides sleep, that is!).  Or, if you do, make sure
> you remember what it is you have set out to do!!  :<
>
> Sort the conductors into the correct order.  This
> proves to be tricky -- the wires are small; stiff;
> and the colors are very hard to distinguish!
>
> 2nd lesson learned:  work in a brightly lit room
> and/or use a magnifying inspection lamp to see what
> the hell you are doing!
>
> Wires in shell, shell into crimping tool -- tah-dah!
> First connector in place.  Gee, that wasn't too bad.
>
> Proceed to other end of cable.  Foolishly assume
> second connector will go on just as easily as the
> first.  Wrong.  Getting the damn little things to
> stay "sorted" in the right order proves to be
> very difficult.  The portions of the wires that
> extend out of the cable jacket are so short that
> they tend to go where *they* want to go!  :<
>
> 3rd lesson learned:  use a pair of needle nose'
> with padded jaws to gently hold the wires side by
> side (or, grow smaller fingers!)
>
> Second connector installed.  Optimistically run
> cable around room -- behind equipment, under
> tables, through ratsnest of other cables -- to
> connect first machine in question to switch.
>
> Naively power up the machine in question and
> watch for the "link" light on the switch.  Hmmm...
> I thought it would come on as soon as the machine
> was powered up!?  No?  OK, wait until the machine
> has completely booted just in case it has to
> turn on the interface.
>
> Crap!  Nothing.  Un-route cable so I can work on
> it again.
>
> 4th lesson learned:  test cable before routing it!
>
> Figure one or both connectors are not installed
> properly.  Or, maybe a small nick in one conductor
> that I didn't see.  Better safe than sorry -- just
> replace both ends.
>
> 5th lesson learned: don't put away bag of connector
> shells until you are SURE the cable works!
>
> Reroute cable.  Turn on machine.  Crap!  Double-crap!
>
> 6th lesson learned:  see "4th lesson learned"  <:-(
>
> Before unrouting cable, drag out voltmeter and
> check cable for continuity.  Of course, contacts are
> so small, need to use paper clips to have a small
> enough "probe tip" to be able to make contact with
> them.
>
> Grrrr... *none* of the conductors in the cable
> appear to be working!  Suspecting crimps are not
> "hard enough", recrimp both ends and retest.
>
> Still nothing.  Cut off both connectors and examine
> them closely.  LIkewise, examine first set to see
> if there is something similarly "wrong" with those!
> Double check order of conductors, etc.
>
> Hmmm... maybe there is a slight difference in the
> size of connector shell expected (by this crimper)
> and the connector shells I actually *have*!  Or,
> maybe the crimper needs adjustment?  It seems
> *obvious* that the "teeth" on the connector's
> contacts must not be adequately piercing the
> insulation on the conductors.
>
> Once again, remake the cable.  (third time's the
> charm?).  Retest.  *Still* no good!  Something is
> fishy here.
>
> 7th lesson learned:  paperclips are too large of
> a diameter to reach the RECESSED contacts in the
> connector body.
>
> Replace paperclips with *pins* and recheck for
> continuity.  Gee, imagine that!  Cable is good!
>
> Reroute cable since we *know* it is good.  Power
> up machine.  *Still* no connectivity!  Crap!
> Reboot.  Still nothing.
>
> Double check cable seating in NIC and switch.  Both
> look good.
>
> Try a different port on switch.  Still nada.
>
> Get annoyed. Cram connector firmly into machine's
> NIC.  Hmmm... link LED on NIC comes on -- briefly.
> "I wonder..."
>
> After a bit more "persuasion", convince myself that
> jack on NIC is flakey.  Intermittent contact.  Note
> to self: replace NIC in that machine!
>
> 8th lesson learned:  see 1st lesson learned.  :<
>
> So, aside from a serious bout of Homer-itis, this
> was a successful -- though lengthy -- undertaking.
>
> Apparently, it *does* get easier with practice
> since the cable I made this morning only took 10
> minutes from start to finish...
>
> But, for those of you who do this "routinely" in
> your line of work, I extend heartfelt sympathies!
> It's just too damn frustrating!  :>
>
> Thanks, again, for the pointers!
>
> --don
>
>
>
> ____________________________________________________________________________________
> Be a better friend, newshound, and
> know-it-all with Yahoo! Mobile.  Try it now.
> http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>
>
> _______________________________________________
> Tucson Free Unix Group - tfug at tfug.org
> Subscription Options:
> http://www.tfug.org/mailman/listinfo/tfug_tfug.org
>
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.19.7/1232 - Release
> Date: 1/18/2008 7:32 PM
>