• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Autonegotiation at 1 Gbps with Cat6 cable.

I

iamgenius

Senior member
I made some cat6 cables and tested them with a cable tester and they worked well. However, when using them in my home network, some pc's will never auto-negotiate at 1 Gbps, but 100 Mbps only. You go replace the cable with a factory made one, and voila it will get to 1Gbps. Why is that? Can some cheap cable testers fool you? Or is it the quality or brand of the cat6 cable itself? Or, it is because I don't know how to make network cables?(But then why the tester didn't tell me anything?)

One would say if the cable is bad, it won't just work. I know that for 1Gbps, you need the 8 wires, not only four. And with the cable tester, all 8 were working fine.

What do you think? Can you please provide some insight? I can't just go and buy ready cables. I'm wiring my whole house and using different custom lengths.

Thanks.
 
Cheap cable testers often just check to see if all of the wires are present. They don't test the speed capabilities of the cable.
 
JackMDS said:
http://www.amazon.com/Fluke-Networks...ke+cat6+tester

Price Realy does Not Matter.

http://www.newark.com/fluke-networks/dtx-1800/lan-tester-lcd-cat6-autotest-9sec/dp/07J2396


😎
Click to expand...

Come on Jacky, giving me an overkill for my tiny little network. I know regular ones go between ~ 5 to 60 bucks. Although the first one might be worth it if I do cabling for a living. I love flukes since university days.
(By the way Jack, your suggestions to my ip camera thread are beautifully working, so special thanks)



That aside, I have a question, and I don't know how it should matter though. It is the termination of the wires. If the order of the wires is the same at both ends, shouldn't it just work?? Or must you follow the T568A termination order????

http://en.wikipedia.org/wiki/TIA/EIA-568


That's my problem maybe? But I don't see how it can affect things if the order is the same at both ends of the wire. If it does, how? I'd like an explanation.

Thanks again to all...
 
So long as they are ordered the same, it should work fine. I personally use T568B for my LAN. Are you sure that the cable tester is truely testing all of the wires? Is it simply a test light showing a connection? Or does it show specifically Gigabit test light (IE checking all 4 twisted pairs).

I'd bet anything some of the connectors are improperly terminated.

Is this patch cables? Keystone jacks (punch down or non-punch down style)? I've had plenty of times it looked visually like a keystone jack or RJ-45 was connected properly, but it only negotiated 100Mbps. Removed the connector, put a new one on and it worked. Of course I've also managed to improperly terminate an RJ-45 several times in a row when having a slow day.

Part of the reason I always check each and every connection with my laptop/tablet after every single new LAN drop or cable I make (using only known good bits for the rest of the path).

I've never had a bad cable before, but bad termination, oh yes, lots.
 
azazel1024 said:
So long as they are ordered the same, it should work fine. I personally use T568B for my LAN. Are you sure that the cable tester is truely testing all of the wires? Is it simply a test light showing a connection? Or does it show specifically Gigabit test light (IE checking all 4 twisted pairs).

I'd bet anything some of the connectors are improperly terminated.

Is this patch cables? Keystone jacks (punch down or non-punch down style)? I've had plenty of times it looked visually like a keystone jack or RJ-45 was connected properly, but it only negotiated 100Mbps. Removed the connector, put a new one on and it worked. Of course I've also managed to improperly terminate an RJ-45 several times in a row when having a slow day.

Part of the reason I always check each and every connection with my laptop/tablet after every single new LAN drop or cable I make (using only known good bits for the rest of the path).

I've never had a bad cable before, but bad termination, oh yes, lots.
Click to expand...

http://www.freewtc.com/images/products/network_cable_tester_lan_tester_6_38176.jpg

Look at the above photo. Numbers 1 through 8 in each piece will light simultaneously....

That what happens...is it right? Unless the G means something critical
 
It matters. Choose either 568A or B and go with it on both ends. Also, no more than 1 half inch untwisted on each end.


Fluke makes a great Ethernet Connectivity Technical Reference Poster. You can download it by clicking the link to it here:


http://www.flukenetworks.com/connectivity-poster-confirmation- emea


Assuming you have solid conductor cable, I recommend these:

http://www.monoprice.com/Product?ab...noprice.com/Search/Index?keyword=Cat6+modular


There is a nice sleeve that aligns the pair perfectly for crimping.

It's also worth while to invest in a quality cable stripper:

http://www.monoprice.com/Product?ab...w.monoprice.com/Category?c_id=105&cp_id=12204


When used in conjunction with the sleeves it allows for the maximum amount of twist in the pair right up to the termination point. It also allows a precise cleave on the end of the conductors to seat perfectly even against the inside of the modular plug.
 
Lets go back to Fardringle educational comment.

In a generic way there can be two problems with CAT cables.

1. Nome compatible wires connection. I.e,. a specific wire/wires are installed on one side on a certain pin/post and on the other side on wrong pin.

To avoid such problem it is helpful to adhere to the 568A or its alternative standard.

Miss match in pins can be debug with any cheap Tester since they measure DC continuation and will detect the wrong installation.

2. Cable is connected correctly, but there might be loose or bad connection on the pins or any other termination. Such a connection might let the DC through but will impede the network connection that is of High frequency and Not DC.

As a result a good LAN tester is based on High frequency AC analysis and thus much more expensive.

Since it seems that money is an issue it might be this one might do the job.

http://www.amazon.com/Paladin-Tools-...sim_sbs_misc_6


😎
 
Thanks to all again. But wait, regarding the termination.....does using the 568A/B really matters or no??? azazel1024 said it doesn't and then Ertaz said it matters. Electrically, I don't see how it should matter if both ends use the same order. It is not like I will only make one end and the other will be already made or something.

If I make a cable, I'll make both ends.
 
Using 586A/B DOES matter. It ensures that transmit and receive pairs are traveling down the same twisted pair. If you just wire them up so they are the same at each end disregarding the standard, then you will have a very poor performing cable. If this is how it's been wired, I'm surprised it can negotiate even at 100. Having the wires in twists helps in keeping the signal within that pair instead of migrating to other wires (crosstalk). If they are mismatched, you are promoting crosstalk and a cable that would create numerous errors when transmitting and receiving data.
 
dtgoodwin said:
Using 586A/B DOES matter. It ensures that transmit and receive pairs are traveling down the same twisted pair. If you just wire them up so they are the same at each end disregarding the standard, then you will have a very poor performing cable. If this is how it's been wired, I'm surprised it can negotiate even at 100. Having the wires in twists helps in keeping the signal within that pair instead of migrating to other wires (crosstalk). If they are mismatched, you are promoting crosstalk and a cable that would create numerous errors when transmitting and receiving data.
Click to expand...



Beat me to it.
 
iamgenius said:
Thanks to all again. But wait, regarding the termination.....does using the 568A/B really matters or no???
Click to expand...

What is important is that the wires from both side will be connected in pairs (pairing is important to reduce noise) to the pins that they intended to.

The color standard is an aid to make it easy to do so.

That means that functionally if the wires that you concocted are paired correctly and go the pins that they need to, it does not matter what color scheme you used.



😎
 
JackMDS said:
What is important is that the wires from both side will be connected in pairs (pairing is important to reduce noise) to the pins that they intended to.

The color standard is an aid to make it easy to do so.

That means that functionally if the wires that you concocted are paired correctly and go the pins that they need to, it does not matter what color scheme you used.



😎
Click to expand...

Just because you can, doesn't mean you should. You don't see people using the green for wire for hot on a 120V outlet just because they can...
 
You guys are right, I just used T568A and it is now negotiating at 1Gbps. You learn a thing everyday.

Thanks a million.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top