CAT5e still shows 10/100...

[koonthul]

Hello all - thanks in advance for the help. I'm a beginner, so go easy. I've done a lot of research and need guidance on next steps in order to solve my problem as I'm not sure what to do next.

Problem:

I'm trying to run a simple CAT5e cable from my router to an endpoint (laptop with wired connection) and get Gigabit speeds. Can't seem to get it to register at 1 gigabit though, only says 10/100 (in windows). Trying to figure out what to do next.

Some more information:

1. The cable i'm using is an old Belkin cable and it's labeled for 5e on its side. I cut it in half and then terminated both ends MYSELF to verify all 4 twisted pairs are tight with no gaps between the cable and termination. My cable tester (although cheap) registers all 8 wires as connecting properly on both ends. Cable tester does not test for speed or anything else. That all checks out so i'm pretty sure my terminations with RJ45 plugs (3 prong) using the T568B standard are good. 1 half of it registers gigabit, so I don't THINK the cable is the issue. Here's what the side of the cable has printed on it...

E158012 Verified (UL) & (ETL) CMR Category 5E 350 MHZ TO TIA/EIA-568B . 2 24 AWG 4 UTP C(UL) 06728FT

2. The original cable itself was too long, so I cut it in half as I only needed about 50 feet or so. As a test, I took the other half (longer) of the cable that I didn't run, terminated both ends and tested it between router and end point. REGISTERS 1 GIGABIT just fine.

3. The part of the cable that I did run is only through the floor boards of my house, around some of the duct work for the HVAC system and alongside some other (dead) wiring. It is not parallel to any particular AC cabling and only crosses some AC cabling at one point.

4. I've verified on more than one laptop that the bad cable still registers at 10/100 rather than gigabit.

5. I just noticed the 6728 FT part on the cable, but I know this cable is WAY shorter than that. I think total we are talking 150 ft maybe, not even. That's BEFORE i cut it in half (almost). I know I had this cable as far back as 2008 or so...I think.

Questions on next steps...

I'd really like to NOT re-run this cable and would like to figure it out. Here are some guesses.

1. When I ran the cable, I may have crimped/bent it when running it over the HVAC system. How would i test for broken or bad connections in the middle of the cable? My internet sleuthing has turned up a test that I could do with a standard multimeter (which i have) by shorting the ends of each twisted pair and seeing its resistance (ohms) readout on the other end, but I'm wondering if there is anything I can try using my PC/ Laptop and router first that would be easier? Again, my cable tester didn't show anything broken so I'm hesitant to try this.

2. My terminations are bad, even though my tester says they pinout correctly. Thoughts on this? I could re-terminate each end, but am trying to be smart about this. Reason I haven't is because the tester tells me my pins are lined up, so again i'm leaning away from this option.

3. How would i figure out if there is interference of some kind? Is there a way to test for this other than ripping the cable out? I suppose the ohms test in #1 might show me whats up? Again, don't want to cut off my terminations again but I'm leaning this way for both #1 and this test.

Here's the small crimping kit and tester kit that I bought for this project. I really like the crimper, as it does passthrough and feels solid (to me at least). Tester is garbage cheap but does what I wanted.

Amazon.com

Thanks for any advice. My last resort (obviously) is to re-run the cable but I'd rather learn something first

 

[drinkingbird]

Hello all - thanks in advance for the help. I'm a beginner, so go easy. I've done a lot of research and need guidance on next steps in order to solve my problem as I'm not sure what to do next.

Problem:

I'm trying to run a simple CAT5e cable from my router to an endpoint (laptop with wired connection) and get Gigabit speeds. Can't seem to get it to register at 1 gigabit though, only says 10/100 (in windows). Trying to figure out what to do next.

Some more information:

1. The cable i'm using is an old Belkin cable and it's labeled for 5e on its side. I cut it in half and then terminated both ends MYSELF to verify all 4 twisted pairs are tight with no gaps between the cable and termination. My cable tester (although cheap) registers all 8 wires as connecting properly on both ends. Cable tester does not test for speed or anything else. That all checks out so i'm pretty sure my terminations with RJ45 plugs (3 prong) using the T568B standard are good. 1 half of it registers gigabit, so I don't THINK the cable is the issue. Here's what the side of the cable has printed on it...



2. The original cable itself was too long, so I cut it in half as I only needed about 50 feet or so. As a test, I took the other half (longer) of the cable that I didn't run, terminated both ends and tested it between router and end point. REGISTERS 1 GIGABIT just fine.

3. The part of the cable that I did run is only through the floor boards of my house, around some of the duct work for the HVAC system and alongside some other (dead) wiring. It is not parallel to any particular AC cabling and only crosses some AC cabling at one point.

4. I've verified on more than one laptop that the bad cable still registers at 10/100 rather than gigabit.

5. I just noticed the 6728 FT part on the cable, but I know this cable is WAY shorter than that. I think total we are talking 150 ft maybe, not even. That's BEFORE i cut it in half (almost). I know I had this cable as far back as 2008 or so...I think.

Questions on next steps...

I'd really like to NOT re-run this cable and would like to figure it out. Here are some guesses.

1. When I ran the cable, I may have crimped/bent it when running it over the HVAC system. How would i test for broken or bad connections in the middle of the cable? My internet sleuthing has turned up a test that I could do with a standard multimeter (which i have) by shorting the ends of each twisted pair and seeing its resistance (ohms) readout on the other end, but I'm wondering if there is anything I can try using my PC/ Laptop and router first that would be easier? Again, my cable tester didn't show anything broken so I'm hesitant to try this.

2. My terminations are bad, even though my tester says they pinout correctly. Thoughts on this? I could re-terminate each end, but am trying to be smart about this. Reason I haven't is because the tester tells me my pins are lined up, so again i'm leaning away from this option.

3. How would i figure out if there is interference of some kind? Is there a way to test for this other than ripping the cable out? I suppose the ohms test in #1 might show me whats up? Again, don't want to cut off my terminations again but I'm leaning this way for both #1 and this test.

Here's the small crimping kit and tester kit that I bought for this project. I really like the crimper, as it does passthrough and feels solid (to me at least). Tester is garbage cheap but does what I wanted.

Amazon.com

Thanks for any advice. My last resort (obviously) is to re-run the cable but I'd rather learn something first

It is very unlikely that interference would drop you to 10/100 (which is it, 10 or 100?). Possible, but never seen it happen, it would have to be pretty severe. Typically you'd just have lots of packet loss, errors, etc.

Are you testing with the same two endpoints on the test cable vs the one through the floor?

BTW Spools of cable do not start at 0 feet. You deduct the number at one end from the number at the other to see how long the run is. It can be any random number, usually they go 00000 through 99999 and start over but I've seen higher ones too.

Did you re-terminate both ends of the cable that you ran through the floor or leave one of the original ends on it? The tester just needs some sort of continuity so maybe one pin is making poor contact but enough to register on the tester.

But other than that, your next step would be testing pairs of wires with one end shorted together and the multimeter on the other to see if there is high resistance in any of them. If you have a female keystone jack you could make your own loopback terminator.

By far the most common reason that two gig capable endpoints come up at 100M is one or more of the 8 wires is not connected or swapped. If it is a garbage cheap tester maybe it is not telling you that a pair is reversed. Look at both ends visually and make sure they are all the way in and the order of the colors is right. Those testers will tell you the cable is fine if you've made the same mistake on both sides. While technically the cable will work and is still "straight through" the crosstalk can be really bad when you mix pins up. Pairs are meant to be + and - twisted together, not + from one pair and - from another.

The color code for B should start with orange/white striped on the left when looking at the connector from the pin side (not the clip side). If you've reversed it, the crosstalk will be really bad and could explain it not connecting over a longer distance.

EDIT - realized that wasn't specific enough. Pin 1 (orange/white) is on the left when the gold pins are facing up toward the sky and are furthest away from you (cable is facing toward you with the connector further away from you). If you use a magnifying glass or have really good eyes, one side will be marked "1" and sometimes the other side marked "8".

What do you mean by 3 prong RJ45 plugs?

Worst case, cut and re-terminate both ends again. Those no-name RJ45 connectors can be hit or miss. Maybe one is making just enough contact to register on the tester but not good contact in the PC.

Picture is worth 1000 words, make sure both ends look like this

 

[koonthul]

It is very unlikely that interference would drop you to 10/100 (which is it, 10 or 100?)

Great question! I can pull down around 80 so I'm pretty sure I'm capped at 100.

Are you testing with the same two endpoints on the test cable vs the one through the floor?

Not sure what you mean here. Each cable has two terminations that are RJ45 plugs that are all the same, and i've used the same test setup for both (router and pc) as end points. The only difference (that I can see) is that the working test cable is not run through the floor and around the duct work. The other one that is not working is "installed" in the floor.

Did you re-terminate both ends of the cable that you ran through the floor or leave one of the original ends on it? The tester just needs some sort of continuity so maybe one pin is making poor contact but enough to register on the tester.

I cut off all the old terminations and re-did them myself. I get your point about the connection being weak, so I might just suck it up and re-terminate again.

What do you mean by 3 prong RJ45 plugs?

If you look at the uncrimped RJ45 plugs that I got with my set, they have 3 points of contact when crimped with the wire (3 points dig into the wires when crimped), not 2. I read somewhere that 2 points was okay for solid but you need 3 for stranded. More is better here? or not?

@drinkingbird - thanks for the image, I have several cheat sheets that I've been using for the wiring standard and my crimpers even have a cheat sheet printed on them. I will double check again though for bad pairs though.

Worst case, cut and re-terminate both ends again. Those no-name RJ45 connectors can be hit or miss. Maybe one is making just enough contact to register on the tester but not good contact in the PC.

I hadn't thought about the actual gold contacts on the plug making good contact with the inside of the NIC on the machines (router and PC). I will check that next. As for the crimping part, can you OVERCRIMP? As in, will that damage the contacts and not make a connection? Really wondering if I did that here or not.

 

[drinkingbird]

If you look at the uncrimped RJ45 plugs that I got with my set, they have 3 points of contact when crimped with the wire (3 points dig into the wires when crimped), not 2. I read somewhere that 2 points was okay for solid but you need 3 for stranded. More is better here? or not?

Ah sorry, I call those teeth but prong makes sense too. All good RJ45 ends have 3 teeth. If you ever have to terminate stranded... its an art form, prepare to have some patience.

@drinkingbird - thanks for the image, I have several cheat sheets that I've been using for the wiring standard and my crimpers even have a cheat sheet printed on them. I will double check again though for bad pairs though.

Yeah I would definitely double check both ends. Make sure the wires are all in the correct order and pushed all the way in.

I hadn't thought about the actual gold contacts on the plug making good contact with the inside of the NIC on the machines (router and PC). I will check that next. As for the crimping part, can you OVERCRIMP? As in, will that damage the contacts and not make a connection? Really wondering if I did that here or not.

With a decent crimper and ends, over crimping should not be possible, but that set looks fairly generic, so never know. But your other test cable that you crimped works fine. It is possible the test set has pins that stick down further and are making contact with a pin on the RJ45 that is too deep. Or it is just very sensitive and a very weak contact is making it light up, but that isn't enough to pass full signal.

I stick with only recognized brands for RJ45 ends. I've been using Tripp Lite ends for years and never had a bad crimp, I've used Ideal in a pinch too and never an issue there either. Used to grab handfuls from work too, don't recall what brand, probably Commscope or Panduit. Those are the best but they're pricey, the Tripps have been the best value I've found.

If everything looks good, I'd cut your ends off, loop/short each pair and test with a multimeter and see if one pair is higher resistance than the rest. If so, you may have cracked one or more wires if you bent it too hard (especially if it is CCA however if it is old it probably is not, but copper can still crack/break). In that case you have no choice but to pull a new one through. If not, re-terminate and see if it is better this time, if so you had a bad connector or crimp.

While most "continuity only" test sets are pretty similar some are a lot better in that they will not light up if they don't get enough voltage or have high resistance.

The fact that you are getting 80 megs over 100M tells me it probably isn't terrible interference, but rather one or more bad wires out of the 8. Theoretically you should be able to get up to about 95 megs but you probably would need IPERF to get that high, 80 over an internet speed test from a 100M connection is reasonable.

Since most cannot afford a Fluke test set (which tests quality) or even the cheaper brands with quality/throughput tests (which are still expensive), there are a few tricks. A couple PCs with IPERF can tell you if you're getting 95% throughput or not. In linux, check ifconfig on both machines after and look for errors, discards, etc. In windows use netstat -e and look for errors/discards. Or better yet open powershell and do "Get-NetAdapterStatistics | fl *" without the quotes. That has much more detailed counters. There are a couple others for TCP/UDP/ICMP stats but that can be misleading as some errors/discards are normal there so I'd stick with the layer 2 ones above.

Actually you could run those now, before and after one of your 80M speed tests, and see if you're getting any errors or not.

If you put a smart switch in the path that has port counters on it, you can check for errors on that after the iperf test too but just looking at the ones on the machines themselves is better. Some smart switches have a "cable test" feature, but it is fairly rudimentary. They claim to be able to tell you the distance to the bad spot but I've never had a bad cable to test it with. I have seen a Fluke do it and it was within a couple feet of accurate (on a 100+ foot run), quite impressive (should be considering how much they cost).

Sorry that's just sort of a brain dump and not really in any particular order, just some stuff to try if you're looking to learn and troubleshoot rather than just replace the cable. Unless you're passing it through a major source of EMI I think that is the least likely cause, but never impossible. This definitely sounds like a physical problem though.

 

[bbunge]

Instead of using crimp connectors on your solid CAT5e cable, you might consider something like this:
https://www.amazon.com/dp/B005E2Y9RY/?tag=snbforums-20
Then use a good patch cable to connect to your devices.
You have a punch down tool in your "kit" from Amazon.

 

[koonthul]

Welp... this is where it gets tough I guess.

So for my first test I went and clipped off both my terminations and stripped down all 8 wires on both ends. Tested for continuity on all open ends to make sure there were no crossed wires. That was all fine. Then ran the resistance tests by twisting 4 pairs together on one end and then with my multimeter tested the other ends and got consistent results on all 4 pairs. Ohms readings were around 004 each time so nothing fishy or surprising. This makes me feel good (I think) and that the cable checks out.

Next I re-terminated both ends again, crimped "less" forcefully and rechecked pairs/pin outs with the crappy cable tester. All checked out again (although something was off at first, see fishy thing 1 below.)

Replugged in my termination to the PC laptop and the other into the router, then restarted windows. Established a wired connection and ... BOOO ... 100/100 shows up. But that is fishy thing number 2 (below).

Fishy thing number 1 = When I first tested with my crappy cable tester, I noticed that when the lights flashed down through each wire as it sent a signal through, it MISSED number 8. I thought this was odd, so I wiggled the wire in the port of the tester and noticed quite a bit of play. This seemed to "fix" the problem. After cycle 2, it registered wire 8 and never missed it again. Crappy tester is a crappy tester I guess. Checking the other end registered everything okay. Switched the tester around to the other end and didn't see anything odd anymore. Wires match up. I'm thinking these RJ45 plugs are now my main culprit but...they worked fine on the other cable so ... ?

Fishy thing number 2 = I could have sworn that my first test (before re-terminating end points) registered 10/100 in windows as a link speed. NOW it reads 100/100 ??!?!?! WTF? It's like the cable is MESSING with me. What does this even mean? Or did I just misread the first time?

Definitely can't go the Fluke route. I looked those up and they are awesome but I don't do this enough to warrant that kind of cost.

So now I'm really stuck. I am not sure why this cable would be still registering poorly. I'm gonna sleep on it and try one more thing with the termination end that goes into the router to see if this is just some sort of poor connection issue with cheap RJ45 ends. Any other ideas? Getting closer to just ripping out the cable and starting fresh though.

Since most cannot afford a Fluke test set (which tests quality) or even the cheaper brands with quality/throughput tests (which are still expensive), there are a few tricks. A couple PCs with IPERF can tell you if you're getting 95% throughput or not. In linux, check ifconfig on both machines after and look for errors, discards, etc. In windows use netstat -e and look for errors/discards. Or better yet open powershell and do "Get-NetAdapterStatistics | fl *" without the quotes. That has much more detailed counters. There are a couple others for TCP/UDP/ICMP stats but that can be misleading as some errors/discards are normal there so I'd stick with the layer 2 ones above.

By "a couple of pc's" you mean 2, but both connected to the same router to transfer data back and forth?

For IPERF, are you talking about this?

iPerf - The TCP, UDP and SCTP network bandwidth measurement tool

iPerf3 binaries - measuring TCP, UDP and SCTP bandwidth performance

iperf.fr

Guess I can bang away on Powershell for a while too, no harm there.

 

[koonthul]

@bbunge - Yes i could do that but i have wall jacks and cable management that I am trying to make look "pretty" so I'd rather not go that route. Good to know that would be how to terminate one end female instead of male. I could look for a wall jack that is a punch down keystone I guess, but I did have some trouble finding those at a decent price. I'll ponder it.

 

[drinkingbird]

Welp... this is where it gets tough I guess.

So for my first test I went and clipped off both my terminations and stripped down all 8 wires on both ends. Tested for continuity on all open ends to make sure there were no crossed wires. That was all fine. Then ran the resistance tests by twisting 4 pairs together on one end and then with my multimeter tested the other ends and got consistent results on all 4 pairs. Ohms readings were around 004 each time so nothing fishy or surprising. This makes me feel good (I think) and that the cable checks out.

Next I re-terminated both ends again, crimped "less" forcefully and rechecked pairs/pin outs with the crappy cable tester. All checked out again (although something was off at first, see fishy thing 1 below.)

Replugged in my termination to the PC laptop and the other into the router, then restarted windows. Established a wired connection and ... BOOO ... 100/100 shows up. But that is fishy thing number 2 (below).

Fishy thing number 1 = When I first tested with my crappy cable tester, I noticed that when the lights flashed down through each wire as it sent a signal through, it MISSED number 8. I thought this was odd, so I wiggled the wire in the port of the tester and noticed quite a bit of play. This seemed to "fix" the problem. After cycle 2, it registered wire 8 and never missed it again. Crappy tester is a crappy tester I guess. Checking the other end registered everything okay. Switched the tester around to the other end and didn't see anything odd anymore. Wires match up. I'm thinking these RJ45 plugs are now my main culprit but...they worked fine on the other cable so ... ?

Fishy thing number 2 = I could have sworn that my first test (before re-terminating end points) registered 10/100 in windows as a link speed. NOW it reads 100/100 ??!?!?! WTF? It's like the cable is MESSING with me. What does this even mean? Or did I just misread the first time?

Definitely can't go the Fluke route. I looked those up and they are awesome but I don't do this enough to warrant that kind of cost.

So now I'm really stuck. I am not sure why this cable would be still registering poorly. I'm gonna sleep on it and try one more thing with the termination end that goes into the router to see if this is just some sort of poor connection issue with cheap RJ45 ends. Any other ideas? Getting closer to just ripping out the cable and starting fresh though.




By "a couple of pc's" you mean 2, but both connected to the same router to transfer data back and forth?

For IPERF, are you talking about this?

iPerf - The TCP, UDP and SCTP network bandwidth measurement tool

iPerf3 binaries - measuring TCP, UDP and SCTP bandwidth performance

iperf.fr

Guess I can bang away on Powershell for a while too, no harm there.

There is no "less forceful" - it is either a complete crimp or incomplete crimp..... A crimper should not release until it has completed crimping.

If pin 8 is loose enough that wiggling it changes the test it is not crimped properly.

 

[koonthul]

Well the plot sickens, and then ends. Here's what I finally ended up doing.

Got sick of dealing with the cable I ran initially (half of an old cat5e cable), so I ripped it out and re-ran another cable that was the second half (the one I initially tested to be working at gigabit) of the first. Hooked it up and....100/100. Would not negotiate Gigabit. Q!@#$$%

At this point I was very frustrated, as the cable DID work at one point. Now I'm thinking that it's the fact that the cable is so old and was so COILED up tight that when I ran it through the floor boards and around stuff (and straightening it out), no matter how careful I was (and I was VERY careful with it) it still ended up ruining a few of the connections. I also still had the suspicion that my terminations were bad, and to try and elminate a few variables, I went and used @bbunge trick of terminating each end with female punch down keystones. This ended up making things VERY obvious that everything was worse, as I was getting intermittent connectivity on pins 1 and 8 and 3 when using the crappy cable tester. So I think, somehow, at this point, no matter what I could do, this ancient cable was not going to play nicely, regardless of what terminations I was using.

So I ripped that out and replaced with a brand new Cat6 cable that was already terminated. End of story. Threw away the other cable and said good riddance. I still need to test out my terminations on a GOOD cable so I at least don't feel like a complete idiot and that my hardware is good. Considering I did get it to work at one point, I'm thinking that they are fine. Either way, that other cable is history.

Thanks for the help, I'm going to experiment again with a newer cable and my current RJ45 plugs. Hopefully with better results.

 

[bbunge]

Well the plot sickens, and then ends. Here's what I finally ended up doing.

Got sick of dealing with the cable I ran initially (half of an old cat5e cable), so I ripped it out and re-ran another cable that was the second half (the one I initially tested to be working at gigabit) of the first. Hooked it up and....100/100. Would not negotiate Gigabit. Q!@#$$%

At this point I was very frustrated, as the cable DID work at one point. Now I'm thinking that it's the fact that the cable is so old and was so COILED up tight that when I ran it through the floor boards and around stuff (and straightening it out), no matter how careful I was (and I was VERY careful with it) it still ended up ruining a few of the connections. I also still had the suspicion that my terminations were bad, and to try and elminate a few variables, I went and used @bbunge trick of terminating each end with punch down keytsones. This ended up making things VERY obvious that everything was worse, as I was getting intermittent connectivity on pins 1 and 8 and 3 when using the crappy cable tester. So I think, somehow, at this point, no matter what I could do, this ancient cable was not going to play nicely, regardless of what terminations I was using.

So I ripped that out and replaced with a brand new Cat6 cable that was already terminated. End of story. Threw away the other cable and said good riddance. I still need to test out my terminations on a GOOD cable so I at least don't feel like a complete idiot and that my hardware is good. Considering I did get it to work at one point, I'm thinking that they are fine. Either way, that other cable is history.

Thanks for the help, I'm going to experiment again with a newer cable and my current RJ45 plugs. Hopefully with better results. End thread.

Yup, kinks in UTP cable can be troublesome. Today I carefully pulled a CAT5 cable I had run to the garage and reused it to wire a ROKU in the basement near the treadmill. Was careful to coil it in large loops and to straighten it when pulling through the floor joists. Terminated one end in a formally used CAT5e wall jack. Used my "junk" cable tester for continuity. Works great!

 

[drinkingbird]

Yup, kinks in UTP cable can be troublesome. Today I carefully pulled a CAT5 cable I had run to the garage and reused it to wire a ROKU in the basement near the treadmill. Was careful to coil it in large loops and to straighten it when pulling through the floor joists. Terminated one end in a formally used CAT5e wall jack. Used my "junk" cable tester for continuity. Works great!

You have to kink copper cable pretty damn tight to crack or break it. I've run tens of thousands of feet, as long as it doesn't snag and you start yanking the hell out of it and creasing it, pretty tough to do.

It can be a bit easier with CCA wire but still don't have to treat it like fiber optic. Obviously as cable ages or if it gets damp and corroded that can have an impact too.

 

[drinkingbird]

Well the plot sickens, and then ends. Here's what I finally ended up doing.

Got sick of dealing with the cable I ran initially (half of an old cat5e cable), so I ripped it out and re-ran another cable that was the second half (the one I initially tested to be working at gigabit) of the first. Hooked it up and....100/100. Would not negotiate Gigabit. Q!@#$$%

At this point I was very frustrated, as the cable DID work at one point. Now I'm thinking that it's the fact that the cable is so old and was so COILED up tight that when I ran it through the floor boards and around stuff (and straightening it out), no matter how careful I was (and I was VERY careful with it) it still ended up ruining a few of the connections. I also still had the suspicion that my terminations were bad, and to try and elminate a few variables, I went and used @bbunge trick of terminating each end with punch down keytsones. This ended up making things VERY obvious that everything was worse, as I was getting intermittent connectivity on pins 1 and 8 and 3 when using the crappy cable tester. So I think, somehow, at this point, no matter what I could do, this ancient cable was not going to play nicely, regardless of what terminations I was using.

So I ripped that out and replaced with a brand new Cat6 cable that was already terminated. End of story. Threw away the other cable and said good riddance. I still need to test out my terminations on a GOOD cable so I at least don't feel like a complete idiot and that my hardware is good. Considering I did get it to work at one point, I'm thinking that they are fine. Either way, that other cable is history.

Thanks for the help, I'm going to experiment again with a newer cable and my current RJ45 plugs. Hopefully with better results. End thread.

I'm assuming the existing cable you cut in half was stranded. You cannot punch down stranded cable onto keystone (or anything for that matter), it cuts through it, as you experienced.

Crimping RJ45 on stranded is very difficult to do too, but it is doable.

Punching solid cable onto keystone jacks or 110 block is almost foolproof, as long as someone shows you how to do it right once.

 

[dosborne]

Putting an rj45 connector on the end of a cable isn't always as easy as a couple of pictures make it out to be. In learning, I went through dozens of connectors until practice make it perfect. it takes a while to get things to fit just right the wires have to be trimmed straight so that they can be inserted all the way to the base of the connector, the outer sheath has to be cut just right to be held in place but not obstruct the jack, etc etc.

Bix mount is certainly harder to screw up as punching down the wires is easier, they get trimmed properly, and you can easily pull it off and do it again....

 

[koonthul]

You have to kink copper cable pretty damn tight to crack or break it. I've run tens of thousands of feet, as long as it doesn't snag and you start yanking the hell out of it and creasing it, pretty tough to do.

It can be a bit easier with CCA wire but still don't have to treat it like fiber optic. Obviously as cable ages or if it gets damp and corroded that can have an impact too.

This is what I thought too, after many years of dealing with cable (but not actually having to terminate it). I'm still wondering if that was indeed the issue, but after adding the keystones and seeing what happened after running it, I'm pretty sure the cable was the problem and "running" it was what ruined it. I know the first part of the cable that I ran (first attempt) definitely got kinked, but I don't remember yanking on it that hard. The second part I know I didn't kink, so I think that whole entire wire was just brittle.

I'm assuming the existing cable you cut in half was stranded. You cannot punch down stranded cable onto keystone (or anything for that matter), it cuts through it, as you experienced.

Crimping RJ45 on stranded is very difficult to do too, but it is doable.

Punching solid cable onto keystone jacks or 110 block is almost foolproof, as long as someone shows you how to do it right once.

No actually, the cable I initially used WAS solid copper. Another reason why i'm so baffled it became so problematic after running it.

Putting an rj45 connector on the end of a cable isn't always as easy as a couple of pictures make it out to be. In learning, I went through dozens of connectors until practice make it perfect. it takes a while to get things to fit just right the wires have to be trimmed straight so that they can be inserted all the way to the base of the connector, the outer sheath has to be cut just right to be held in place but not obstruct the jack, etc etc.

Bix mount is certainly harder to screw up as punching down the wires is easier, they get trimmed properly, and you can easily pull it off and do it again....

I'm tempted to take a video of my process because I've done over 50 connections now and have been playing with it non-stop. I'm guessing I'm still doing several things wrong without knowing it. Here are my guesses.

1. When I go to strip the jacketing off, I have a stripper that is very cheap and could be cutting into the actual wiring. Need to be careful here.
2. To eliminate some variables/problems, I'm only using PASS THROUGH RJ45 connectors and crimper, so I know i have the wiring pairs just right. I really don't think this is my issue.
3. When I go to crimp, i have the jacketing VERY tight up against the actual connections being pushed into the pins. So when the crimper comes down and pushes the back half of the RJ45 connector closed (the part where it cinches the jacketing down on top of the twisted pairs) I'm wondering if it's too close to the strands and since they are still twisted pairs, they get squashed, rather than lay flat.

Any insights here would be good. I'll try and take some pictures and a video to show you my process and what I might be doing wrong.

 

[drinkingbird]

This is what I thought too, after many years of dealing with cable (but not actually having to terminate it). I'm still wondering if that was indeed the issue, but after adding the keystones and seeing what happened after running it, I'm pretty sure the cable was the problem and "running" it was what ruined it. I know the first part of the cable that I ran (first attempt) definitely got kinked, but I don't remember yanking on it that hard. The second part I know I didn't kink, so I think that whole entire wire was just brittle.

I guess it could have been bad cable. But I've got a 20 year old spool of cat 5e (home depot brand, nothing special) that sits in my scorching hot attic. Just ran some for a friend the other day and tested out good (standard test and throughput). Copper cable lasts a really long time, the only thing that can really hurt it (other than major physical stress) is water and it would have to be subject to a lot of water, not just humidity.

No actually, the cable I initially used WAS solid copper. Another reason why i'm so baffled it became so problematic after running it.

You implied that it was a cable that was already terminated previously so assumed it was a patch cord (which are stranded). So now the question is, what was that cable used for previously, did someone just yank the hell out of it when it was taken out of commission?

If it was solid wire and it was not punching down reliably, now I'm starting to wonder about that cable tester or the patch you were using with the tester. Even if the wire was shot, you shouldn't get varying results with the tester, would be always bad or always good, unless you happen to be flexing the very part where it is internally broken (at least 2 spots on the 2 different sections of cable you've tried). Or maybe just your punch methods. I always give at least 2 strikes, sometimes 3. Who knows, maybe that cord was used by some kids to play tug-o war at some point

1. When I go to strip the jacketing off, I have a stripper that is very cheap and could be cutting into the actual wiring. Need to be careful here.

Never cut through the jacket. You only score it, then pull it off, breaking at the score. Cutting through will almost always cut the jacket of one of the internal wires. That being said, that usually isn't enough to actually hurt anything unless you cut it really bad and it breaks when you terminate it, or two cuts short together. Personally I like to use a razor to score it, not a tool, so I can feel exactly how much pressure (very little) I'm applying.

2. To eliminate some variables/problems, I'm only using PASS THROUGH RJ45 connectors and crimper, so I know i have the wiring pairs just right. I really don't think this is my issue.

Honestly, the few times I've seen those connectors, they were total junk. Almost as bad as the clamshell ones. What kind of keystones were you using - decent or no name?

3. When I go to crimp, i have the jacketing VERY tight up against the actual connections being pushed into the pins. So when the crimper comes down and pushes the back half of the RJ45 connector closed (the part where it cinches the jacketing down on top of the twisted pairs) I'm wondering if it's too close to the strands and since they are still twisted pairs, they get squashed, rather than lay flat.

No concerns there, they are meant to get squashed, unless it is the totally wrong connector type for the cable (unlikely with 5e) it won't break or damage them. The retainer is meant to hold the wires themselves, not just the jacket. You want as little exposed wire as possible inside the connector, since exposed means untwisted and more prone to interference. I always strip off about 1" of jacket, get everything lined up right, then trim the whole end flat leaving about 1/2". Pinch the jacket as you push it into the connector so the wires don't retreat on you. Once you see all wires are all bottomed out (and double check they are the correct order) put it in the crimper, stop pinching the jacket, and push in on the jacket while you crimp down.

Find a scrap of cable not from that same piece and try terminating and testing it, should give you some idea if it was the cable or your methods (or your tester). Then cut the ends off and punch both ends down, and test again. I learned (too long ago to admit) by trial and error, nobody at the place I was working knew anything about it. Hell the guy before me ran flat 8 pin pre-terminated telephone cable to all the PCs as ethernet for gods sake.

 

[dosborne]

I'm m old school. Ive never used a wire cutter / stripper to remove the sheathing, I always use a utility knife to score it, then remove it by hand.

I've also never used a pass thru connector. I like the idea, maybe not valid, that the wire is not exposed to moisture for longevity purposes.

The sheathing only needs to go in as far as the plastic tabs that hold it in place (with a bit extra).

My crimper is also old school. I big clunky heavy duty ratchet style.

Lots of great videos on YouTube, but I'm sure at this point you've seen them all. In general it sounds like you are doing it right. The only part you don't want to go cheap on really is the connector. The tester, cutter, crimper can all be pretty basic.

Another trick, the test the quality of the work, is if you cut a connector off, use it as quality control. Cut it 6 inches from the end and get an rj45 coupler. You can then expose the wires (remove the sheath) and plug 2 into the coupler and test with a multimeter or twist the wires together and test with a tester. It gives you and idea as to how good you are

 

[drinkingbird]

I'm m old school. Ive never used a wire cutter / stripper to remove the sheathing, I always use a utility knife to score it, then remove it by hand.

That's not old school, that's proper school

I did see a fancy one that one of our data center guys had that you stick the end in and it scores and pulls off the sheath for you. Not something I do enough to invest in something like that, nor did it really save much time (and it still left the damn string). I mean seriously, has anyone ever actually used the rip cord?? It just gets in the way.

I've also never used a pass thru connector. I like the idea, maybe not valid, that the wire is not exposed to moisture for longevity purposes.

I'm also wondering how you get it cut close enough after to not interfere with inserting the plug? I wonder if that could possibly explain some of @koonthul 's strangeness? Even most end cutters leave a bit. You'd have to cut it with a razor to get it near flush. Maybe the excess is preventing full insertion (that's what she said).

My crimper is also old school. I big clunky heavy duty ratchet style.

Stop calling me old.