Go Back   AnandTech Forums > Hardware and Technology > Networking

Forums
· Hardware and Technology
· CPUs and Overclocking
· Motherboards
· Video Cards and Graphics
· Memory and Storage
· Power Supplies
· Cases & Cooling
· SFF, Notebooks, Pre-Built/Barebones PCs
· Networking
· Peripherals
· General Hardware
· Highly Technical
· Computer Help
· Home Theater PCs
· Consumer Electronics
· Digital and Video Cameras
· Mobile Devices & Gadgets
· Audio/Video & Home Theater
· Software
· Software for Windows
· All Things Apple
· *nix Software
· Operating Systems
· Programming
· PC Gaming
· Console Gaming
· Distributed Computing
· Security
· Social
· Off Topic
· Politics and News
· Discussion Club
· Love and Relationships
· The Garage
· Health and Fitness
· Merchandise and Shopping
· For Sale/Trade
· Hot Deals
· Free Stuff
· Contests and Sweepstakes
· Black Friday 2013
· Forum Issues
· Technical Forum Issues
· Personal Forum Issues
· Suggestion Box
· Moderator Resources
· Moderator Discussions
   

Reply
 
Thread Tools
Old 01-21-2013, 10:46 AM   #26
Cabletek
Member
 
Join Date: Sep 2011
Posts: 171
Default

I have a hard time believing anyone but an electrical engineer could solder CATx wire and not create an impedance mismatch. Granted 15 Ohms is the tolerance level, but I seriously doubt anyone that posts here could do this in the situation described above with less than 10 Ohms in the mismatch. I'd be surprised if a single poster had the correct tool to even check the impedance in that situation and no it is not a VOM/DMM. You could check the impedance of single wires with a VOM/DMM, but not between two wires. You could even check loop resistance, but again that is not the correct diagnostic for what I am talking about.

That said, now we step out of the lab and into the real physical world, it probably works good enough.
Cabletek is offline   Reply With Quote
Old 01-21-2013, 12:11 PM   #27
ScottMac
Moderator
Networking
Elite member
 
Join Date: Mar 2001
Posts: 5,471
Default

Quote:
Originally Posted by Cabletek View Post
I have a hard time believing anyone but an electrical engineer could solder CATx wire and not create an impedance mismatch. Granted 15 Ohms is the tolerance level, but I seriously doubt anyone that posts here could do this in the situation described above with less than 10 Ohms in the mismatch. I'd be surprised if a single poster had the correct tool to even check the impedance in that situation and no it is not a VOM/DMM. You could check the impedance of single wires with a VOM/DMM, but not between two wires. You could even check loop resistance, but again that is not the correct diagnostic for what I am talking about.

That said, now we step out of the lab and into the real physical world, it probably works good enough.
You can't check any impedance directly with any DVM / VOM /DVOM ... they operate on a DC voltage to measure the resistance of the thing under test .... "impedance" implies a changing signal that interacts with capacitive and inductive components of the conductor, which typically changes with frequency.

Also, an impedance lump (i.e., immediate change of resistance/impedance), will screw a cable, even if it's within the "characteristic impedance" tolerance of the cable. The lump causes a reflection (standing wave) that destructively interferes with the primary signal.
__________________
Eschew Obfuscation! Espouse Elucidation!

Optimists are never pleasantly surprised.

Nov 4th, 2008: History is made: - Richie Daley becomes the first Mayor in history with his own pet President.
ScottMac is offline   Reply With Quote
Old 01-21-2013, 12:36 PM   #28
imagoon
Diamond Member
 
imagoon's Avatar
 
Join Date: Feb 2003
Location: Chicagoland, IL
Posts: 4,301
Default

Quote:
Originally Posted by Cabletek View Post
I have a hard time believing anyone but an electrical engineer could solder CATx wire and not create an impedance mismatch. Granted 15 Ohms is the tolerance level, but I seriously doubt anyone that posts here could do this in the situation described above with less than 10 Ohms in the mismatch. I'd be surprised if a single poster had the correct tool to even check the impedance in that situation and no it is not a VOM/DMM. You could check the impedance of single wires with a VOM/DMM, but not between two wires. You could even check loop resistance, but again that is not the correct diagnostic for what I am talking about.

That said, now we step out of the lab and into the real physical world, it probably works good enough.
I personally use this:

http://myaccount.flukenetworks.com/f...008_UPDATE.SWF

It most definitely will complain about an impedance mismatch. When it comes to soldering, I would expect it to pop up during the frequency sweep as a drop in the upper frequency response. I would expect a corresponding crosstalk increase as the frequency goes up.
imagoon is online now   Reply With Quote
Old 01-21-2013, 02:41 PM   #29
ScottMac
Moderator
Networking
Elite member
 
Join Date: Mar 2001
Posts: 5,471
Default

I also have the DTX 1800 (and love it).
__________________
Eschew Obfuscation! Espouse Elucidation!

Optimists are never pleasantly surprised.

Nov 4th, 2008: History is made: - Richie Daley becomes the first Mayor in history with his own pet President.
ScottMac is offline   Reply With Quote
Old 01-21-2013, 04:26 PM   #30
Eug
Lifer
 
Eug's Avatar
 
Join Date: Mar 2000
Posts: 16,562
Default

This splicing thread brought back some memories.

During my reno an electrician cut one of my CAT5e lines by mistake. I got home to find the two ends carefully twisted together and taped up, and an apologetic electrician. Luckily the walls weren't up yet so I just replaced the entire line.

Also, a pipe fitter torched one of my other lines. They didn't tell me but I saw it. I replaced that too.

BTW, for my house when I couldn't replace the line, I've occasionally just crimped on plugs on the two ends and connected them with a $2 Monoprice CAT6 coupler.

http://www.monoprice.com/products/pr...seq=1&format=2

Yeah, not a great solution, but hey, whatever. Worked fine for home Gigabit usage. I don't care if it doesn't pass master with you guys.
__________________

OS X & iOS: 27" iMac Core i7 870 | 13" MacBook Pro C2D 2.26 P8400 + SSD | 13" MacBook C2D 2.4 T8300 + SSD | iPad 2
Windows: X3400 Athlon II X3 435 | 11.6" 1810TZ Pentium SU4100 + SSD | Revo R3610 Atom 330 + SSD
Android: RAZR HD | Nexus 7
Eug is offline   Reply With Quote
Old 01-21-2013, 06:19 PM   #31
ScottMac
Moderator
Networking
Elite member
 
Join Date: Mar 2001
Posts: 5,471
Default

Quote:
Originally Posted by Eug View Post
This splicing thread brought back some memories.

During my reno an electrician cut one of my CAT5e lines by mistake. I got home to find the two ends carefully twisted together and taped up, and an apologetic electrician. Luckily the walls weren't up yet so I just replaced the entire line.

Also, a pipe fitter torched one of my other lines. They didn't tell me but I saw it. I replaced that too.

BTW, for my house when I couldn't replace the line, I've occasionally just crimped on plugs on the two ends and connected them with a $2 Monoprice CAT6 coupler.

http://www.monoprice.com/products/pr...seq=1&format=2

Yeah, not a great solution, but hey, whatever. Worked fine for home Gigabit usage. I don't care if it doesn't pass master with you guys.
That's a cleaner solution than soldering, because (assuming proper termination) the twist is maintained to the IDC, and the (Cat-rated) coupler should pass the signal as well as could be expected.

Scotch-Loks (Cat-rated variety) work well too, as long as the appropriate Cat-rated specs are observed.
__________________
Eschew Obfuscation! Espouse Elucidation!

Optimists are never pleasantly surprised.

Nov 4th, 2008: History is made: - Richie Daley becomes the first Mayor in history with his own pet President.
ScottMac is offline   Reply With Quote
Old 01-21-2013, 07:18 PM   #32
Cabletek
Member
 
Join Date: Sep 2011
Posts: 171
Default

Quote:
Originally Posted by ScottMac View Post
You can't check any impedance directly with any DVM / VOM /DVOM ... they operate on a DC voltage to measure the resistance of the thing under test .... "impedance" implies a changing signal that interacts with capacitive and inductive components of the conductor, which typically changes with frequency.

Also, an impedance lump (i.e., immediate change of resistance/impedance), will screw a cable, even if it's within the "characteristic impedance" tolerance of the cable. The lump causes a reflection (standing wave) that destructively interferes with the primary signal.
As to VOM's:
Given a spool of twisted pair wire, I can take both ends of one wire, place the red lead on one end and the black lead on the other and measure the impedance of that one wire. What use is this? Given a impedance per length formula you can estimate how much cable length is left, that's about all it is good for, or if you get a weird number you can assume damage. The more realistic way to test is to short the pair at one end or use a UTP [100 Ohm] terminator, and test the LOOP resistance at the other end on the pair red on one, black on other [in coax one is the conductor/stinger the other is the shield and you use a 50/75 ohm terminator depending on your rating] [this also makes a good toner system, find more than 100 ohms and you found the terminated line], however, as I stated neither of these are what I am talkign about on the splices, so a VOM will not test the correct impedance as I was referring to the 100 Ohm Characteristic or surge impedance which requires special tools to actually test, and is based the materials involved and the distances between wires supplied. What you need is TDR which basically sends out a wave and look for reflections of that wave, determined the time it took to receive them and then using an attenuation and velocity of propagation formula estimates the distance to the impedance mismatch which is what causes the reflections.

Your tool does characteristic impedance or Time Domain Reflectometry and for that price it better, no? And I am surprised to find someone that does carry something like that around, but then it says it does FIBER too, so your doing rather higher up corporate runs I take it, small business IT is what I usually run into and I am lucky if they know they can tone UTP with a standard VOM, they always insist on going back to get the broadcaster half of the phone timer they left somewhere. Or running up to lowes/home depot to get another one...

I always wondered if you TDR'd a properly terminated cable, will it crash form doing indefinite calculations or not, heh. Look slike yours comes with ends. But I bet they have some osrt of response system in them too.
Cabletek is offline   Reply With Quote
Old 01-21-2013, 10:29 PM   #33
ScottMac
Moderator
Networking
Elite member
 
Join Date: Mar 2001
Posts: 5,471
Default

Quote:
Originally Posted by Cabletek View Post
As to VOM's:
Given a spool of twisted pair wire, I can take both ends of one wire, place the red lead on one end and the black lead on the other and measure the impedance of that one wire. What use is this? Given a impedance per length formula you can estimate how much cable length is left, that's about all it is good for, or if you get a weird number you can assume damage. The more realistic way to test is to short the pair at one end or use a UTP [100 Ohm] terminator, and test the LOOP resistance at the other end on the pair red on one, black on other [in coax one is the conductor/stinger the other is the shield and you use a 50/75 ohm terminator depending on your rating] [this also makes a good toner system, find more than 100 ohms and you found the terminated line], however, as I stated neither of these are what I am talkign about on the splices, so a VOM will not test the correct impedance as I was referring to the 100 Ohm Characteristic or surge impedance which requires special tools to actually test, and is based the materials involved and the distances between wires supplied. What you need is TDR which basically sends out a wave and look for reflections of that wave, determined the time it took to receive them and then using an attenuation and velocity of propagation formula estimates the distance to the impedance mismatch which is what causes the reflections.

Your tool does characteristic impedance or Time Domain Reflectometry and for that price it better, no? And I am surprised to find someone that does carry something like that around, but then it says it does FIBER too, so your doing rather higher up corporate runs I take it, small business IT is what I usually run into and I am lucky if they know they can tone UTP with a standard VOM, they always insist on going back to get the broadcaster half of the phone timer they left somewhere. Or running up to lowes/home depot to get another one...

I always wondered if you TDR'd a properly terminated cable, will it crash form doing indefinite calculations or not, heh. Look slike yours comes with ends. But I bet they have some osrt of response system in them too.
Shorting one end and measuring it with a VOM (or similar meter), you are only reading loop RESISTANCE, not impedance. Impedance will change with frequency, usually increasing with frequency.

You CAN do polarity and continuity, and if you're clever, you can check shield continuity of shielded / screened cable and the relative effectiveness of the grounding point.

If you TDR a properly terminated cable / pair, it sees in infinite length, as there is no reflection to time. you can still do return loss and some other measurement.

The 1800 is pretty versatile. I use it primarily for copper and use other stuff for OTDR, power, spectral dispersion and the rest. There's a special test set for PON / GPON.
__________________
Eschew Obfuscation! Espouse Elucidation!

Optimists are never pleasantly surprised.

Nov 4th, 2008: History is made: - Richie Daley becomes the first Mayor in history with his own pet President.
ScottMac is offline   Reply With Quote
Old 01-22-2013, 06:42 PM   #34
Red Squirrel
Lifer
 
Red Squirrel's Avatar
 
Join Date: May 2003
Location: Canada
Posts: 25,384
Default

All this gives me a neat idea. I want to see what is the "breaking point" of ethernet, and going to try all sorts of crazy stuff to a cable until I notice transfer speed differences. All these fancy meters may show things, but at the end of the day, it's all about what is noticable in a real world situation, or when it's being pushed to the limit in a real world situation (high speed transfer). I think this could be fun.
__________________
~Red Squirrel~
486dx2 @66Mhz turbo, 8MB ram, 512MB HDD, sound blaster 16 + 2x cdrom, Trident 1MB video card @ 640*480, 56k high speed modem.

Last edited by Red Squirrel; 01-22-2013 at 06:46 PM.
Red Squirrel is offline   Reply With Quote
Old 01-22-2013, 09:23 PM   #35
ScottMac
Moderator
Networking
Elite member
 
Join Date: Mar 2001
Posts: 5,471
Default

Quote:
Originally Posted by Red Squirrel View Post
All this gives me a neat idea. I want to see what is the "breaking point" of ethernet, and going to try all sorts of crazy stuff to a cable until I notice transfer speed differences. All these fancy meters may show things, but at the end of the day, it's all about what is noticable in a real world situation, or when it's being pushed to the limit in a real world situation (high speed transfer). I think this could be fun.
It would be interesting. Try to find a stable TX/RX pair of hosts or test suite.

If you decide to expand it, try a vendor-homogeneous (all components from the same class and product line of one vendor), the "cheap mix" i.e., the cheapest batch of mixed components, etc ...

We did this a long time ago (~Cat5) ... just after we did the SOHO switch versus commercial-grade switch testing.
__________________
Eschew Obfuscation! Espouse Elucidation!

Optimists are never pleasantly surprised.

Nov 4th, 2008: History is made: - Richie Daley becomes the first Mayor in history with his own pet President.
ScottMac is offline   Reply With Quote
Old 01-22-2013, 09:39 PM   #36
imagoon
Diamond Member
 
imagoon's Avatar
 
Join Date: Feb 2003
Location: Chicagoland, IL
Posts: 4,301
Default

It would be interesting and depend of your needs. VoIP is rather intolerant of loss. 2% is generally noticeable in conversations. Surfing the web may not matter. Something like netflix buffers so it may not show.
imagoon is online now   Reply With Quote
Old 01-22-2013, 11:42 PM   #37
wirednuts
Diamond Member
 
Join Date: Jan 2007
Posts: 7,121
Default

Quote:
Originally Posted by Red Squirrel View Post
All this gives me a neat idea. I want to see what is the "breaking point" of ethernet, and going to try all sorts of crazy stuff to a cable until I notice transfer speed differences. All these fancy meters may show things, but at the end of the day, it's all about what is noticable in a real world situation, or when it's being pushed to the limit in a real world situation (high speed transfer). I think this could be fun.
with cat5, you really cant hurt speeds. tie it in tight knots even, you will still get full bandwidth.

i know this because we tried it in class.

above 1Gb speeds though, its a whole different story, and even when you use cat6 cable you need to have all the kinks out to get best performance. reason being that cat6 really isnt that much different then 5. you get more twists, and some crosstalk buffering if you get the augmented stuff... but for the most part, there isnt much to cat6 compared to cat5. utp cable just doesnt carry frequencies above 100mhz (1Gb speed) easily...
wirednuts is offline   Reply With Quote
Old 01-24-2013, 08:57 AM   #38
Eug
Lifer
 
Eug's Avatar
 
Join Date: Mar 2000
Posts: 16,562
Default

Quote:
Originally Posted by Red Squirrel View Post
All this gives me a neat idea. I want to see what is the "breaking point" of ethernet, and going to try all sorts of crazy stuff to a cable until I notice transfer speed differences. All these fancy meters may show things, but at the end of the day, it's all about what is noticable in a real world situation, or when it's being pushed to the limit in a real world situation (high speed transfer). I think this could be fun.
I ran Gigabit Ethernet over a home made CAT3 patch cable and it worked fine. However, it was only a couple of meters long.

I couldn't get Gigabit Ethernet to work over a cheap flat retractable patch cable though. My home network hardware down-spec'd itself to 100 Mbps Ethernet... but that worked fine. However, that cable was CAT-nothing, because basically it was a flat ribbon with all the wires running parallel to each other, about 1 or 1.5 m long when fully extended. It looked like this:



BTW, IIRC Mellanox supports (or has supported) 10GigE over CAT5e, over shorter than usual spec distances that CAT6a would support though.

I'm hopeful that when home 10GigE lands, that means it will work well enough for home use on my installed CAT5e and CAT6 (not 6a). I think the longest cable length in my house is probably about 75 feet or so. Note though my CAT5e is mostly good quality Belden, but the CAT6 isn't. If anything I wouldn't be surprised if my CAT5e is better than my "CAT6". BTW, some of it is labelled as "CAT6e". It's splined Ethernet cable that they claim exceeds the CAT6 (not 6a) spec, but I'm not sure I believe it. It all works great for Gigabit though.
__________________

OS X & iOS: 27" iMac Core i7 870 | 13" MacBook Pro C2D 2.26 P8400 + SSD | 13" MacBook C2D 2.4 T8300 + SSD | iPad 2
Windows: X3400 Athlon II X3 435 | 11.6" 1810TZ Pentium SU4100 + SSD | Revo R3610 Atom 330 + SSD
Android: RAZR HD | Nexus 7

Last edited by Eug; 01-24-2013 at 09:05 AM.
Eug is offline   Reply With Quote
Old 01-24-2013, 09:39 AM   #39
skyking
Lifer
 
skyking's Avatar
 
Join Date: Nov 2001
Posts: 15,814
Default

What in the world would you need 10 for at home? "I can move an ISO at the blink of an eye".
__________________
Heat

skyking is offline   Reply With Quote
Old 01-24-2013, 09:48 AM   #40
Eug
Lifer
 
Eug's Avatar
 
Join Date: Mar 2000
Posts: 16,562
Default

Quote:
Originally Posted by skyking View Post
What in the world would you need 10 for at home? "I can move an ISO at the blink of an eye".
Actually, that's about right. It's rather annoying waiting for say a 15 GB file transfer from my desktop to my laptop.

However, Gigabit is OK. 100 Mbps Ethernet was quite painful though. 10 years ago people were asking "What in the world wold you need 1000 for at home?"
__________________

OS X & iOS: 27" iMac Core i7 870 | 13" MacBook Pro C2D 2.26 P8400 + SSD | 13" MacBook C2D 2.4 T8300 + SSD | iPad 2
Windows: X3400 Athlon II X3 435 | 11.6" 1810TZ Pentium SU4100 + SSD | Revo R3610 Atom 330 + SSD
Android: RAZR HD | Nexus 7
Eug is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 08:37 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.