Cat 6E cable certifier.... suggestions?

Status
Not open for further replies.

Fulthrotl

~Autocorrect is My Worst Enema.~
i have had it.

i just spent sunday evening certifying 100
nodes of 6E with a steam powered JDSU POS.

this thing is so slow..... to write to the flash memory,
if your file contains more than 10 cable records, it takes
so long to save, you want to slit your wrists.

on cable record is 41 seconds to save test results....
add 15 seconds for each additional cable... do the math... :rant:

a level eleventeen POS this thing is.

that brings me to my request for those of you who do CAT stuff....
what flavor of cable certifier do you like?

i'd like to be able to do 6E, and fiber.... and when the protocol for cat 7
comes out, it'd be nice if i didn't have to throw the thing away for a new one.

i'm partial to fluke stuff, but god, some of it's expensive. strike that...
all of it's expensive, but this JDSU pride and joy just took 5 hours to do
an hours worth of work.

suggestions? thanks......
 

grich

Senior Member
Location
MP89.5, Mason City Subdivision
Occupation
Broadcast Engineer
A couple of years ago corporate told us to buy a DTX1800 and use it. Considering the corporate push to do things differently (read: cheaper) I was surprised they would drop this much coin for test gear...I guess they saw this as the best tool for the job. We certify everything we install, including new patch cords.

So, +1 for Fluke. :)
 

texie

Senior Member
Location
Fort Collins, Colorado
Occupation
Electrician, Contractor, Inspector
i have had it.

i just spent sunday evening certifying 100
nodes of 6E with a steam powered JDSU POS.

this thing is so slow..... to write to the flash memory,
if your file contains more than 10 cable records, it takes
so long to save, you want to slit your wrists.

on cable record is 41 seconds to save test results....
add 15 seconds for each additional cable... do the math... :rant:

a level eleventeen POS this thing is.

that brings me to my request for those of you who do CAT stuff....
what flavor of cable certifier do you like?

i'd like to be able to do 6E, and fiber.... and when the protocol for cat 7
comes out, it'd be nice if i didn't have to throw the thing away for a new one.

i'm partial to fluke stuff, but god, some of it's expensive. strike that...
all of it's expensive, but this JDSU pride and joy just took 5 hours to do
an hours worth of work.

suggestions? thanks......
Well, I think you already know the answer. Fluke pretty much owns this market. But yes, to do all the things you would like, the Fluke will be real, real expensive. When I was in the contracting business my philosophy, as long as I felt the value was there in terms of the quality of results and labor impacts I would always consider the purchase price as only part of the total cost. Also, like you have said in the past, I have buried all or part of the cost of certain equipment in the cost of a job where conditions warrant.
 
Last edited:

Fulthrotl

~Autocorrect is My Worst Enema.~
Well, I think you already know the answer. Fluke pretty much owns this market. But yes, to do all the things you would like, the Fluke will be real, real expensive. When I was in the contracting business my philosophy, as long as I felt the value was there in terms of the quality of results and labor impacts I would always consider the purchase price as only part of the total cost. Also, like you have said in the past, I have buried all or part of the cost of certain equipment in the cost of a job where conditions warrant.

yeah, they do.

i was looking at used stuff, and found the whole thing with all modules on ebay for $11k.
everything priced out new was about $27k.

there is a fair bit of used stuff around, and when it's warranted, i'll probably go that way.
however, i found a workaround for my slow POS steam powered slug... made up a bunch
of test files, ten cables to a file, starting at 001, and going to 100. i only load up and have
to save ten files at a time, and the time isn't so horrible.

the fluke does a cat 6 in 9 seconds, so they say.... :rant: but if you do the math, it takes
an awful lot of cables to break even on the cost of buying a fluke and throwing this in the
dumpster.

i don't need a fluke at this point. i need three 100 node cabling jobs a month. if i had THAT,
i'd a just gone down and bought the silly thing, instead of posting a whiny post about it.

$150 a drop seems to be the rate around here for certified cable installs..... so a 100 cable
network would be $15k, cable would be about $2k, jacks and plates about $1,300, and throw
in a cheap rack for $300... say $4k for material..... and it seems i've got about 3 days to put
the last one in complete...

i can work 3 day work weeks for $11,000 a week.... :p

wonder if the fluke unit has a customer locator module that will beep when near a customer?
 

hbiss

EC, Westchester, New York NEC: 2014
Location
Hawthorne, New York NEC: 2014
Occupation
EC
The best policy is not to certify unless the customer twists your arm. I suppose if you already have spent the money for the test equipment, heck yeah sell the customer certification. But if you don't have anything I certainly wouldn't push the envelope unless the job is big enough to recoup the cost. Inexpensive wire mappers will tell you if you have terminated the runs correctly and that's about all a computer cares about. Unless you are a hack, your runs and terminations should be fine. IMO certification is just a customer's way to see that they got their money's worth so charge them up the wazoo for the service.

-Hal
 

Fulthrotl

~Autocorrect is My Worst Enema.~
The best policy is not to certify unless the customer twists your arm. I suppose if you already have spent the money for the test equipment, heck yeah sell the customer certification. But if you don't have anything I certainly wouldn't push the envelope unless the job is big enough to recoup the cost. Inexpensive wire mappers will tell you if you have terminated the runs correctly and that's about all a computer cares about. Unless you are a hack, your runs and terminations should be fine. IMO certification is just a customer's way to see that they got their money's worth so charge them up the wazoo for the service.

-Hal

i tend to agree, other than occasionally switching the brown and orange pairs
in bad light, i've had ONE jack bad, in 4 years... until yesterday.... it's weird...
104 node install, and the LAST FOUR cables, the last pull, has high skew. i've
never had cable bad, ever. and i was pulling four at a time, out of four boxes.
so each of those cables came from a different box.

normally, the skew i see goes from 0.0 to maybe 5.0, and cat 6 allows i think
for 35....... and these are running around 450..... wiring is correct on the jacks.

so, i have 100 nodes perfect, and four whacky. the four are spares out to the
warehouse, so they are coiled up in the ceiling, and are 250' long. i was just
putting jacks on them to certify they were ok for the record. the customer
doesn't know the final location yet.

so, my first thought is, they are coiled up, and the choke effect is coming
into play.... but the four bad ones are coiled with the second to the last pull,
which is four cables from the same four boxes, and those are all perfect.

and they are all in the same coil... i straightened them out on the floor today,
and checked them again, the four good ones are still good, and the four bad
ones still are wired correctly, with high skew.

just shoot me now, and put me out of my misery. i don't need a fluke tester,
and i don't need three of these jobs per month, i need a revolver with all six
chambers loaded for russian roulette. the way my luck has been running today,
i could sit there all night looking for that damn bullet.
 

hbiss

EC, Westchester, New York NEC: 2014
Location
Hawthorne, New York NEC: 2014
Occupation
EC
Exactly my point. Certifying can be a can of worms. When connected those cables will work just fine but you have some expensive piece of equipment telling you that those runs don't meet the manufacturers specs. So what now? Are you going to pull them out and re-run them? What are you going to believe? Like I said, you don't even want to go there unless you absolutely have to.

PS- Pull 250 feet out of a box and lay it on the floor. Terminate it with jacks on each end and test it. Betcha the skew will be high also.

-Hal
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Exactly my point. Certifying can be a can of worms. When connected those cables will work just fine but you have some expensive piece of equipment telling you that those runs don't meet the manufacturers specs. So what now? Are you going to pull them out and re-run them? What are you going to believe? Like I said, you don't even want to go there unless you absolutely have to.

PS- Pull 250 feet out of a box and lay it on the floor. Terminate it with jacks on each end and test it. Betcha the skew will be high also.

-Hal

If the skew numbers you are reading are in nanoseconds (ns), as they normally are, then something really extreme has happened. Typical Cat cable will have a time delay (transit time) of about 1.5ns/foot. That means that to have a 450ns. skew come about in the simplest way, one of the twisted pairs in each cable must be 300 feet longer than the shortest pair in the same cable jacket. That is almost impossible even in a 3000' run unless one of the pairs is wound in a spiral around all of the other pairs, which are run straight. (BTW, the Cat5e installation standard is 45ns overall, regardless of the length of the run, and the standard for Cat 6 cable itself is no more than 45ns per 330 foot (100 meter) length of cable.)

It seems more likely that there is some sort of defect which is causing the skew reading to be incorrect. But if they are all from different boxes.....
Now a significant difference in wire pair characteristics, such as crossing wires between two pairs, could affect the characteristic impedance and the propagation speed for both of the pairs involved. But even that seems well beyond credible error margins for a 250' cable.
(You are looking at a length difference of 300 feet in a 250 foot cable for heavens sake!)
Take a close look at cross-talk and both near-end and far-end echo return, both of which would be strongly affected by the same sort of wiring problem that might be causing the skew.
 
Last edited:

Fulthrotl

~Autocorrect is My Worst Enema.~
Which JDSU cable tester are you using?

Validator-NT905... about three years old.

new, it was about $1,600... they can be
picked up used on ebay for about $600
now.

it's funny, about 6 months ago, one of the
batteries went away, and i found that JDSU
doesn't support this any more, and i found
a battery for $125 used on ebay.

JDSU said that maybe they would have more
batteries in six months, but they couldn't say
for sure.

the client software for this, the CD got mislaid,
and that isn't available anywhere either.

when you buy cheap shit.... well, you get cheap shit.:dunce:
 

Fulthrotl

~Autocorrect is My Worst Enema.~
Take a close look at cross-talk and both near-end and far-end echo return,
both of which would be strongly affected by the same sort of wiring problem
that might be causing the skew.

i laced them up and terminated them today, and cut about 100' off the length,
as they were right under 300'......

and they certified perfect. :lol:

most cat drops are in the 150'~200' range, and i'm wondering if this certifier
gets iffy at the maximum length.

skew was 0.0, 15.3, 14.5, and 0.0 typically for the four miscreant cables.
my tester says 35 is the cutoff point for cat 6 protocol, and these now passed
easily.

 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
most cat drops are in the 150'~200' range, and i'm wondering if this certifier
gets iffy at the maximum length.

I think you may have found the reason. Just like my old bathroom scale can show that a suitcase weighs -245 pounds. (I weigh myself and get 250 pounds, then I pick up the 55 pound suitcase and the dial goes all the way around past zero to 5 for a difference of 245.)
If the time delay for one pair goes past the maximum the POS can read, it will be very different from the time delay on a pair that is just under the maximum.
 

MisterCMK

Member
Location
Twin Cities, MN
Do you need updated software for it? I have it on the server at work. Also, you can stop the long test times by disabling the save after every test option. I would have to power ours up and check where the option is.

When did they stop supporting it? We recently sent ours out for repair and JDSU fixed it.
 

Fulthrotl

~Autocorrect is My Worst Enema.~
Do you need updated software for it? I have it on the server at work. Also, you can stop the long test times by disabling the save after every test option. I would have to power ours up and check where the option is.

When did they stop supporting it? We recently sent ours out for repair and JDSU fixed it.

god, if you have the software, that would be great.
nothing but their software will read a .job file.

and you can't download it, or buy it anywhere that i've seen.
supposedly they had a version that worked on OSX so you could
use a mac, but hell, i couldn't even find the windows version.

i'll look for the save option... that would be a huge help.

when i couldn't find the software, and couldn't find the replacement batteries,
i just drew the conclusion that they aren't supporting it.

thanks so much for your help...
 

MisterCMK

Member
Location
Twin Cities, MN
god, if you have the software, that would be great.
nothing but their software will read a .job file.

and you can't download it, or buy it anywhere that i've seen.
supposedly they had a version that worked on OSX so you could
use a mac, but hell, i couldn't even find the windows version.

i'll look for the save option... that would be a huge help.

when i couldn't find the software, and couldn't find the replacement batteries,
i just drew the conclusion that they aren't supporting it.

thanks so much for your help...

I'll see what I can dig up tomorrow. Send me your e-mail address. I'll take a look at the Validator tomorrow for the save option.

I learned about the save option the hard way on a hotel. :D
 

Dan_P

Member
Location
Charlotte
Validator-NT905... about three years old.

new, it was about $1,600... they can be
picked up used on ebay for about $600
now.

it's funny, about 6 months ago, one of the
batteries went away, and i found that JDSU
doesn't support this any more, and i found
a battery for $125 used on ebay.

JDSU said that maybe they would have more
batteries in six months, but they couldn't say
for sure.

the client software for this, the CD got mislaid,
and that isn't available anywhere either.

when you buy cheap shit.... well, you get cheap shit.:dunce:

For what its worth, the Validator is not a true "Certifier", although for many years Test-um/JDSU did market it as such. The Validator fits into a segment testers generally called "Qualifiers" by manufacturers such as Fluke and Ideal (full disclosure I work for Ideal).

The primary differences are that a true Certifier performs a specified number of tests across a frequency range defined by the test standard and Category number you are testing. The current test standard for LAN cabling is the TIA-568-C, and the primary tests run are wire map, NEXT, return loss, PowerSum NEXT, ACR-F,etc... The frequency range depends on the Category tested, for Cat 5E it is 100 MHz, Cat 6 is 250 MHz and Cat 6A is 500 MHz. There are currently only three Certifiers on the market: the LanTEK II from Ideal, the DTX series from Fluke and the C40G from JDSU. Note that JDSU is leaving this market and are discontinuing all of their LAN testers including the Validator and C40G. That leaves Ideal and Fluke.

A Ceritifer is required whenever there is a manufacturer warranty being offered on the cabling system. Otherwise people usually use certifiers when an engineer or designer (RCDD) requires it.

The Validator falls into the class of tester called "Verifiers". These are application testers that test to see if the cabling can support a specific application such as gigabit Ethernet. The requirements of this class of tester is not defined by a standards organization such as the TIA so you need to be aware of what you are buying. Test-Um/JDSU used clever wording in their marketing material and say that the Validator "certifies to the TIA-568 physical interconnect standard" or something to that effect. What that basically means is that it does a wire map test according to the TIA's specification, a test you can do with a $75wire mapping tester.

With JDSU leaving the LAN Qualification tester market as I mentioned above, you have two choices for Qualification testing. Ideal offers the SignalTEK II gigabit Qualification tester and Fluke offers a product called the CableIQ. They operate in very different ways because this class of tester is not defined by a test standard.
I believe the CableIQ tests a cable by doing a quick SNR test of sorts and infers that the cable will support Ethernet by the result of that test.
The SignalTEK II operates on a different principle by establishing a live Ethernet link across the cable under test and performing a "Bit Error Rate Test" on that cable. The unit will attempt to link at 1 gigabit, but will fall back to 100 megabit if that is all the cabling will support. Then sends live traffic across the cable and passes or fails the link based on how many frames were lost. We use the IEEE 802.3ab standard for making the P/F determination. So while its a different standard than a true Certifier uses (TIA 568-C) it is still an internationally recognized standard for Ethernet. If the test passes you can rest assured that the cable will support Ethernet from 10-1000 Mbps as well as VoIP and even standard telco applications. The advantages of this class of tester is that they generally cost from $1000-2000 versus $6000-9000 for a true Certifier.

Both types allow you to test and document your installation, you just need to understand what each type of tester is telling you. Frankly, what really matters is what your customer wants. If they are asking for testing and reporting without specifically stating that they require a full certification to TIA 568-C, then you would be fine with a Verifier like the SignalTEK II or your Validator. Otherwise you need to use a Certifier such as the LanTEK II.

I hope this helps. None of this changes that fact that you are unsatisfied with the performance of your Validator, I just wanted to clear up the common confusion about what a Certifier really is. Test-Um/JDSU did a good job of muddying the water on that issue and I have spoken with more than a few contractors who tested with a Validator only to find out that the customer would not accept the results because they were not a true TIA 568-C test.

Let me know if you have any other questions about LAN testers, I am happy to help.

Dan Payerle
 

Simplex

Member
Location
nowhere
Are you pro or not

Are you pro or not

Look dude... either this is your gig or your hobby...

If you get paid to install and certify cat 5,6,7 circuits, get a fluke. I don't care what it costs. neither should you.

If you're a hobbyist then use whatever basic meter you have and deal with it. If you want o pretend to go pro as a hobbyist, buy the fluke again,

If you get paid to do this it shouldn't be a concern.

Sounds like you hobby and want to play in the big leagues.

Either ante up or sit down.

Cheers, mate
 
Last edited by a moderator:
Status
Not open for further replies.
Top