Cable miswire diagnostics, poor performance

[Alan Bergman]

First I want to introduce myself. I’m not experienced at all in networking. I can build and fix practically anything, but home networks is pretty darn new to me. YouTube only gets you so far and sometime clarification is needed, so please bare with me on my lack of knowledge and expertise of the topic.

Ok, in the process of building our “smart net zero homestead” I wanted to set up a legit home network.

I ran conduit out from my home to my shop for power and a separate conduit run for data.

Pulled through the data conduit is a shielded Cat 6 rated for direct burial, pretty heavy duty stuff. This Cat 6 line runs from my router to a PoE switch in my shop. Currently it runs a Unifi Access Point and TV.

In the process of adding a second access point (for outdoors) and two PoE cameras I discovered I was only getting 10 mb/s in the shop, standing next to the access point.

In the house, I’d get 200+mb/s. So, being the colorblind individual I am, I crimped in two new terminals figuring I messed it up. No change.

The cable is shielded and drain wire is grounded. Volt meter doesn’t indicate any noise. The only other item in the conduit is a signal wire for the CT sensors for the solar side of my energy monitor. The total run of the Cat 6 cable is about 150 feet.

I went and got a cheap Southwire 550 tester and it says miswire.

I crimped new terminals on again...and again, triple checking to make sure it was correct. Even had the wife verify. Same issue, no change.

I tried plugging the end directly into my laptop and still slow, so that should eliminate the switch or WAP. I tried changing ports on the router, no change.

Tester says miswire but I’m clearly seeing that they are accurate. Ideal brand crimp tool and ideal brand shielded pass through connectors.

So the only thing I can think of is that the cable got damaged while pulling it through the conduit. It was lub’d up and Pulled nicely, straight through.

I don’t have a distance tester but can anyone think of any other possibility? Or way I can diagnose further before pulling a new cable?

I thought that maybe my crimps were bad but I crimped a small patch cable together and it tested good.

 

[ColinTaylor]

You say the cable tester indicates a miswire. Does it not also tell you which wire numbers are wrong?

You have created a straight-through cable and not a crossover haven't you?

 

[Alan Bergman]

You say the cable tester indicates a miswire. Does it not also tell you which wire numbers are wrong?

You have created a straight-through cable and not a crossover haven't you?

Yes it’s a straight through cable. 1-2 & 4-5 are flashing. I’ve redone these terminations multiple times, no change. My only thought is that the cable got damaged unless it could be something else that I’m not thinking of?

I’m torn on ordering a distance tester on Amazon for $135 or just pulling new cable. I’d hate to spend that time and money on either and it not be the issue.

 

[ColinTaylor]

Strange. The plug wiring looks OK to me. I'll leave it for others to come up with more ideas.

 

[coxhaus]

So how did you pull the cable through the conduit? If you pulled it through a 90 degree turn you could of damaged it. What did you use for conduit?

If you are getting a 10 meg connection it only needs 2 wires and the distance can be over 100 meters. If you are getting a 100 meg connection then you only need up to 4 wires and you are limited to 100 meters.

I would say you need a better tester if you can not figure out what you are doing wrong. My guess is your crimps are the problem unless you really abused the wire pulling it. On long pulls they make cable lube.

 

[ColinTaylor]

The strange thing from your picture is that it isn't showing a short on any on the wires and it isn't showing a reversed connection. It also suggests that there isn't a break in the wire because the LED would be off rather than flashing. The only fault scenario I can think of that would match those lights would be if wire pair 1-2 was swapped with pair 4-5 at the other end of the cable. Are you brown-green colour blind? (I am, but not to that extreme .)

I see from the manual that your tester can do a debug test if you press the TEST button for 2 seconds. Maybe that might be more informative.

 

[CaptainSTX]

Does your switch have the option to run a cable test?

My TP-Link smart switch has that option and if yours does then it might give you some information.

Also for grins just take a known good factory made cable and test it on your tester and see what the results are. Also while my cheap tester works perfectly on 568B cables if the cable is pinned 568A it shows an error.

 

[Alan Bergman]

So how did you pull the cable through the conduit? If you pulled it through a 90 degree turn you could of damaged it. What did you use for conduit?

If you are getting a 10 meg connection it only needs 2 wires and the distance can be over 100 meters. If you are getting a 100 meg connection then you only need up to 4 wires and you are limited to 100 meters.

I would say you need a better tester if you can not figure out what you are doing wrong. My guess is your crimps are the problem unless you really abused the wire pulling it. On long pulls they make cable lube.

cable was pulled through 1-1/2” PVC conduit, there was probably about a total of 270 degrees in bend. I pull by soaking a rag in lube tied ahead of the cable on the pull cord. That way the rag lubes and cleans as I pull. The cable pulled with little resistance, I don’t remember there being any snags or difficultly, I was by myself so maybe there was some damage as it came off the spool.

It’s been rainy here, but supposed to be clear tomorrow. I’m going to go ahead and pull new cable and see what I find.

I truly appreciate all the help. I really wanted to seek some pro advice before I pulled new cable Incase there was something I was overlooking.

I’ve done several new terminals on each end carefully making sure they were right. At this point, that cable had to get damaged somehow, it just had to. We’ll see what I find ‍

 

[Alan Bergman]

The strange thing from your picture is that it isn't showing a short on any on the wires and it isn't showing a reversed connection. It also suggests that there isn't a break in the wire because the LED would be off rather than flashing. The only fault scenario I can think of that would match those lights would be if wire pair 1-2 was swapped with pair 4-5 at the other end of the cable. Are you brown-green colour blind? (I am, but not to that extreme .)

I see from the manual that your tester can do a debug test if you press the TEST button for 2 seconds. Maybe that might be more informative.

Im red green color blind. The green and brown look identical in anything other then bright daylight to me. I had the wife and another buddy verify for my too. I guess this is why I couldn’t do EOD?

Anyway, yes, there is a debug feature. I don’t understand how to decipher what it’s telling me, lol. I wish I ordered the more expensive tester that had a digital read out and gives distance. There’s one on Amazon for $135 but don’t think it’d be worth it for just this, not when I can just pull more cable relatively easily. What other uses are those testers good for? I might justify it if it can do more than just network/coax cable ‍

Y’all are great, truly appreciate the help. I’ll report my findings.

 

[coxhaus]

Another thing I thought of did you buy the right ends for the cable you are using? Twisted and solid wire take different ends. I am talking about in wall wire and patch cable wire.

 

[degrub]

Is the other end a male or female termination ?
The issue might be on the other end ?

 

[Alan Bergman]

I believe I found out my problem.

I was reading the specs of my cable. It said .57mm 23awg solid conductor. When I read that, I remembered that the terminals stated .48mm max. The cable is a heavy duty, direct burial rated cable. I think what’s been happening is that the pins were not fully penetrating the insulation. Maybe enough to pass current over short distances but it was probably too much resistance over long distances. Which would explain why short cables tested just fine.

So. I am going to pick up some keystone jacks and a punch pin tool in the morning and see what results I get.

I should mention that I discovered this AFTER I pulled a new cable today along with purchasing a new, more expensive pair of crimps

Gentleman. Thank you for all the help. I’ve been losing my mind over this! I’ll confirm back when I try the keystone jacks.

 

[ColinTaylor]

If the 0.48 mm you are referring to is that shown in your second picture then you have misread what it says. That says the size range of the conductor and the insulation is 0.040 to 0.048 inches (not mm). So that is equivalent to 1.02 to 1.22 mm. Your cable spec says the "Conductor Insulation Dia." is 1.02 mm. So those numbers match perfectly.

The keystone jacks are a very good idea though.

 

[Greg72]

See this. https://signaleer.wordpress.com/2010/04/28/cabling-for-the-color-blind/

 

[Greg72]

http://colorblindpal.com/

 

[ddaenen1]

So, from the picture, it seems like you are using the T-568B color coding standard. It may sound ridiculous but you are using the same color scheme on both ends, correct? I have encountered this before where the ISP had used T-568A on one end and i applied T-568B on the socket end and thus had a miss-wire.