Maker Pro
Maker Pro

Cat 6 Ethernet Cable data restriction?

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
I made up two ethernet cat 6 cables for internet access. On my internet service provider's plan the d/L speed should be at 300mbs. Both my ethernet cables are showing a through put of 94 d/l from the router to the other end of the cable. The network provider has checked the router. It is able to supply 300d/l. Using commercially produced ethernet cables running from the router to his tester the network supplier's engineer has just told me the full 300Mbs speed is being transmitted.

When I crimped the rj-45s I checked the cabling was correct using an RJ45 cable continuity testor. Today I checked it again. Each strand is correctly connected at each end as each strand has continuity to the corresponding pin on the RJ-45..

I did not follow the standard cabling colour scheme, but ensured that each colour goes to the same position on both RJ45's and the tester confirms I did it correctly.

What do you suggest is the reason for the restricted speed on the two cables I made?
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
What length have you made?
180’ (feet) seems to be max.
Also, is your cable twisted pairs?.

Martin
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
CAT cable is twisted pairs and they MUST be made up according to the correct colour wires to the correct pins in the RJ45 connector else you lose the 'twist' which is there to eliminate crosstalk and interference and maximise speed.

Your own wiring 'system' is not good enough.
 

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
Thank you all for your helpful comments.

The cable length is less than the working max quoted above.

I did not realise that there could be such a large data speed loss due to the matching wires not being close for two of the colours.

Why I moved the white orange around:

There were 8 wires but one of the other solid colours, say brown that should have had its pair in white brown instead had it's twisted pair miscoloured to white orange. It was a production error in the cable.

I kept the two white oranges far apart in order to prevent the wrong white orange being accidentally matched with the solid orange.

I will redo the RJ-45s making sure the twisted pairs are kept together in the socket. I will use a marker pen to reduce the risk of accidentally reversing the orange white wires. I hope that redoing the RJ-45's wiring in the correct order will clear up the data loss.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
There were 8 wires but one of the other solid colours, say brown that should have had its pair in white brown instead had it's twisted pair miscoloured to white orange. It was a production error in the cable.
Done a lot of cat cable terminations over the years and must say I've never come across that.
Thought about getting your eyes tested..??:oops:
 

Kiwi

Jan 28, 2013
471
Joined
Jan 28, 2013
Messages
471
My thoughts, in no particular order.
1 - Possible colour blindness. Sorry, I had to put this in.
2 - Incorrect crimping. May pass continuity test, but not able to give full throughput.
3 - Faulty/counterfeit cable. Incorrect colours may be a clue.

What length are your cables?
 

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
I have removed the rj-45's on the cable between the router and the computer, and replaced the RJ45's with new ones. I have confirmed the 8 wires have continuity and that there is no mismatch on the cable connections. I have just done a speed test at the computer and I am getting 94 ul /57 dl.

The ISP engineer came down during the week and confirmed that the router is 300 dl 150ul. He also checked my two cables and found that the speed on both the cables I had made up are transmitting at 94UL / 57 DL roughly. I had a spare short commercially made spare cable that he also checked. It was giving the output 300/150.

The LAN RJ45 socket on the computer can go to 1Gps.

What can I do next?

PS
I am not colour blind.
The cable length is probably about 60 feet at the most.
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
Clearly the issue is the cable. Is it SCREENED cable? Where did you source it? Sure it's 'legit' cable (lots of fake stuff exists). Some so-called copper wire cables are actually aluminium with a copper coating! Can you compare the commercial cable, physically, with yours? Try making up a shorter length and see of there is an improvement.

Cut back some cable sleeve and compare the 'twist' in the pairs - the better cable has more twists/inch (whatever).

Finally, but not specifically, what is your quests for 100% download/upload for? Is it a strict necessity for the tasks you have in mind - if so that's not a good approach anyway as you really need to account for other issues that could restrict speed. I very much doubt anyone has a 'need' for the max down/upload speeds routers can typically claim to be capable of - in most cases the ISP won't deliver the full spec speed anyway - they always state 'up to....' rather than guaranteeing their maximum 100% of the time.
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
Clearly something is not right.
Is it rigid or stranded?. Stranded, as already pointed out, could be fake.
Here is a good page with a good explanation of colour codes. Your earlier post about orange and orange/white being separated might make some sense.

Martin
 

bertus

Moderator
Nov 8, 2019
3,303
Joined
Nov 8, 2019
Messages
3,303
Hello,

What kind of Ehternet cable are you using?
This page shows you the difference between all available types:

Bertus
 

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
Clearly the issue is the cable. Is it SCREENED cable? Where did you source it? Sure it's 'legit' cable (lots of fake stuff exists). Some so-called copper wire cables are actually aluminium with a copper coating! Can you compare the commercial cable, physically, with yours? Try making up a shorter length and see of there is an improvement.

Cut back some cable sleeve and compare the 'twist' in the pairs - the better cable has more twists/inch (whatever).

Finally, but not specifically, what is your quests for 100% download/upload for? Is it a strict necessity for the tasks you have in mind - if so that's not a good approach anyway as you really need to account for other issues that could restrict speed. I very much doubt anyone has a 'need' for the max down/upload speeds routers can typically claim to be capable of - in most cases the ISP won't deliver the full spec speed anyway - they always state 'up to....' rather than guaranteeing their maximum 100% of the time.
The cable is not screened.
I have checked a wire and it is silver inside but copper looking outside. Is there a requirement for cat 6 to be pure copper?

My ISP supplies fiber to the router. It is at the speed high I mentioned above and it has been checked as such.
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
The cable is not screened.
I have checked a wire and it is silver inside but copper looking outside. Is there a requirement for cat 6 to be pure copper?

My ISP supplies fiber to the router. It is at the speed high I mentioned above and it has been checked as such.
You certainly don't have 'quality' cable then and are seeing the consequences thereof.

Regardless of the speed supplied, what is it you're doing that requires the full rate? I know of no tasks whatsoever that require 300Mb/s - even downloading (streaming) HD from Netflix only requires 5Mb/s. Do you plan watching 60 such channels simultaneously?

If you own a car that has 140mph on the speedo, do you necessarily need to always drive at 140mph? It's called 'chasing rainbows' and I can GUARANTEE that your ISP won't maintain 300Mb/s 100% of the time so what do you plan to do - monitor it 24/7/365 and call the engineer every time it drops to 299Mb/s?

Get the proper cable and connectors - no cost-cutting (or purchase one made up to ensure it's the right quality) and stop agonising over the theoretical maximum speed possible. You're wasting your efforts.
 

bertus

Moderator
Nov 8, 2019
3,303
Joined
Nov 8, 2019
Messages
3,303
Hello,

How are the driver settings in the computer?
Will an other type of cable give more speed?

Bertus
 

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
Hello,

How are the driver settings in the computer?
Will an other type of cable give more speed?

Bertus
The LAN RJ45 socket on the computer can go to 1Gps. As for driver settings I am not aware of restrictions (Ubuntu 22.04)

I do not know if another type of cable will give more speed. It did on the commercially made cables. It should give the router output speed which has been confirmed at 300/150.
I signed up for something more than just Broadband, it is true fibre to the router. I am just trying to get what I signed up for. I am of course aware that often those quoted speeds are maximums, in this case it seems different. The company dug up the streets and laid their own fibre. The service can provide a much higher speed so 300 is low for them.
 

bertus

Moderator
Nov 8, 2019
3,303
Joined
Nov 8, 2019
Messages
3,303
Hello,

The following apge will tell you a lot about ethernet configuration on ubuntu:

Bertus
 

R791945

Jun 19, 2015
92
Joined
Jun 19, 2015
Messages
92
Hello,

The following apge will tell you a lot about ethernet configuration on ubuntu:

Bertus
Thank you for the link to the ethernet configuration article on Ubuntu.
When the ISP engineer visited me, he came with test equipment and without using my computer he found the following:
My self made Cat 6 cables were not giving anything like the same speed as the router was supplying (not withstanding the confirmed continuity and connections of my cables).
A short commercial cable did show the full speed on his test equipment.

I confirm that the wired network setting in Ubuntu is connected at a link speed - 1000 Mb/s. I also understand that my Cat 6 cables should have given a much higher speed than I have attained as the cable length is not theoretically sufficient to have caused an attenuation of the speed or loss of signal.
 
Top