Failed patch panel?

HCHTech

HCHTech

Well-Known Member
Reaction score
3,848
Location
Pittsburgh, PA - USA
I was "resurrecting" a spare office space for a client of mine who will be using it temporarily while their space is being remodeled. The previous tenant took most everything when they left. There were 3 24-port patch panels hanging from the ceiling where their equipment rack used to be. Anyway, in toning & testing out the various runs, I came across an odd symptom. Every line terminated to one of those patch panels tested as having pins 3 and 6 shorted. Every one. I re-punched down a handful of the runs, but they didn't look wonky at all and it didn't change anything. I also looked into each port on the front looking for bent or crossed pins - again, everything looks ok. I put a RJ45 jack on one of the lines and it now tested correct. I've never seen a patch panel short like that before, but they're cheap enough, so I'm not going to waste any more time on it. Weird, that's all.
 
All RJ45 female terminators can and will go bad on insertion or removal of whatever's in them.

Now that being said, these symptoms I've never seen before. I've seen 1 or 2 ports go bad... but not the entire panel. If an entire panel is testing that way, honestly... I'd suspect the tester. Did you try a known good patch cable on that thing? The ports on the tool are just as fragile as everything else.

Heck I've seen a weak battery make a tester do strange things too.

Take a bad cable and connect it to a single female jack, does that fix that one run? IF so... then yeah perhaps the entire panel is bad. But I don't think so... I think it's more likely someone in the move damaged the cables moving things around above the panels. Still not something that usually kills more than a few wires though... definitely not all of them.
 
Guessing this isn't a keystone type (individual female ends snapping in).

Yeah to me, I start to look at the common denominator. If it was a bunch of male ends...the common denominator is the crimp tool. The dies can go bad, repeating the same pattern all the time. But for a patch panel, esp an old type that's all in one, where you're "punching down"...hard to envision how the assembly/manufacturing process can repeat bad connections across the board behind all the female ends.

Of course there is the "other end" of each run...but I'd guess a wall plate jack...but if just male ends....there's the possibility of a bad crimp tool.
 
I'll be back onsite today, so can test a bit more. I don't think it's the tool because using the same tool and cable to test the runs in the other two panels works absolutely fine. I found one run with a non-attached pair, but that was on the wall jack at the other end, in fact. If you think about how a patch panel is constructed, the ports aren't connected to each other in any way, so I don't see how a failure anywhere would lead to multiple ports with the same problem. I've got a second tester here in the shop, I'll take that today and compare. Maybe that panel (it was a different model than the other two) was just constructed in such a way that the cable I was using didn't mate with the pins in the jacks correctly. It was Cat5 and older, who knows. This explanation at least makes more sense.

Frustration #2 on this job, I bought a small (9U) swing-out wall rack to put everything in, and it turns out the cross-piece on the rack is just enough shorter than the cross piece on the swing-out front that the latch doesn't mate correctly. I'm taking a couple of thin washers to shim the rack about 1/16" wider. I swear, everything you buy today is crap. I've not used Navepoint in the past, but it was quickly available on Amazon and this job was a surprise with a short deadline, so I succumbed to the lure of prime shipping. Ugh.

Frustration #3 on this job, It's rare I get to set things up from scratch, so I don't often get to choose between cage nuts and the threaded pre-drilled holes for the mounting surface. I decided on using cage nuts since I had a recent experience of having to loosen and move 6 devices 'just a tiny bit' to get a new device to fit in the last remaining space on a rack using the threaded holes. I'm not sure the obnoxiousness of squeezing in those cage nuts in tight quarters was any less frustrating. Ha.
 
Oh yeah I despise cage nuts... there's a bit of a trick of the wrist that can make them tolerable. But I still hate them!
 
Ok, so I figured out the problem. This particular patch panel has little tiny indicator lights by each port. The short I'm seeing with my tester HAS to be related to how that little LED is wired in. Every run works, and if I test them with a less-fancy tester, I get full connectivity on each line.

Apparently, if you use a signal generator on the other end of one of those runs, the lights will flash - who knew? Doesn't really make sense - you would only buy one of those things if you KNEW you were going to do a half-assed job of the wiring and not label the runs. An odd use case for sure.
 
You must log in or register to reply here.

Similar threads

DanF
Help the kid with keystones
Replies
8
Views
608
YeOldeStonecat
YeOldeStonecat
Coastalbendcomputers
Xerox Network Printer-Work Stations to Not Responding When Printing
2
Replies
23
Views
4K
Coastalbendcomputers
Coastalbendcomputers
Back
Top