Handmade vs. Commercially Produced Ethernet Cables 837
An anonymous reader writes "We have a T1 line coming into our satellite office and we rely fairly heavily on it to transfer large amounts of data over a VPN to the head office across the country. Recently, we decided to upgrade to a 20 Mbit line. Being the lone IT guy here, it fell on me to run cable from the ISP's box to our server room so I went out and bought a spool of Cat6. I mentioned the purchase and the plan to run the cable myself to my boss in head office and in an emailed response he stated that it's next to impossible to create quality cable (ie: cable that will pass a Time Domain Reflectometer test) by hand without expensive dies, special Ethernet jacks and special cable. He even went so far as to say that handmade cable couldn't compare to even the cheapest Belkin cables. I've never once ran into a problem with handmade patch cables. Do you create your own cable or do you bite the bullet and buy it from some place?"
How much is your time worth (Score:5, Insightful)
Just do what your boss wants (Score:2, Insightful)
It's clearly not your company's core business to make their own patch cables. It may be fun for you to wittle down your own toothpics from lincoln logs but if it's not in your job description it ain't going to fly. Seriously, just buy the damn stuff and do what your boss has asked.
Not sure if I'm stating the obvious here but (Score:2, Insightful)
If you get the rated speed and it's reliable, need we delve further?
If the Belkin cable fails, you can blame Belkin (Score:5, Insightful)
Doubt he's correct, but believe him anyway! (Score:5, Insightful)
Wait a minute. Your boss is telling you to buy cables instead of toiling to make your own, and you're _complaining_? I don't think a self-terminated link of CAT6 will have the slightest trouble maintaining 20 megabits, but that's not the point.
Word of advice, take his word for him and nod. If he's willing to spend money to make your job easier, then keep that job!
Not worth the responsibility (Score:5, Insightful)
Re:your boss sucks at making ethernet cables (Score:5, Insightful)
Obviously your boss isn't good at making cables. While if you lack the skill to do something like make cables with care you're going to have problems, there's no reason that you can't make your own cables and have them perform just as well as the ones made by a machine in a factory.
They can put it under the "Tech" section if they like, but this is really another disappointing Ask Slashdot. It's disappointing because too many of these have this format: "my boss at work wants me to do X, but I'd really rather do Y; what are the merits of X versus Y?" All of them need to be summarily rejected, with a polite e-mail sent to the submitter which says "within the bounds of the law, you need to do what your boss asks you to do whether or not you necessarily agree with it. If you cannot convince your boss to do otherwise, and this is a problem for you, perhaps you should consider working elsewhere."
The other disappointing category of Ask Slashdot-type submissions are those questions that are factual in nature and have only one correct (and rather well-known, easily researched) answer. Asking a large group with varying levels of expertise makes a lot of sense when there are multiple possible solutions to a problem and there is room to be creative. It makes no sense when it's more of a yes/no question -- remember the recent Ask Slashdot that asked whether spam is increased by trying to opt-out of spam e-mails? That's an excellent case in point, and not atypical either. That should have been an "Ask Google", not an "Ask Slashdot".
I think it's a shame that the quality of these particular submissions are on the decline. There's nothing inherently wrong with the "Ask Slashdot" format and there are a lot of very knowledgable people who browse this site. I'd love to see how creative they can be. It's just never going to be as good as it easily could be when it's handled this way.
Just pretend that the question is... (Score:3, Insightful)
It's disappointing because too many of these have this format: "my boss at work wants me to do X, but I'd really rather do Y; what are the merits of X versus Y?" All of them need to be summarily rejected, with a polite e-mail sent to the submitter which says "within the bounds of the law, you need to do what your boss asks you to do whether or not you necessarily agree with it. If you cannot convince your boss to do otherwise, and this is a problem for you, perhaps you should consider working elsewhere."
Just pretend that the question is "how should I convince my boss that Y is better than X?". It's like asking legal questions on Ask Slashdot: the real question is "what should I know before my appointment with a lawyer?".
Re:your boss sucks at making ethernet cables (Score:2, Insightful)
Re:your boss sucks at making ethernet cables (Score:5, Insightful)
Bull! Entirely aside from what the submitter should do to protect their job, it is topical on slashdot to question whether DIY ethernet cables are any good, just as people on a home repair DIY site might discuss whether doing drywall yourself is worthwhile.
When the only answer slashdotters can imagine is "just pay somebody else to do it," that is the day there is no point reading here.
Re:meh, easy... (Score:2, Insightful)
Monster cables are nothing but a name. I can make higher quality cables out of old telephone wire and snickers bars than anything Monster puts out. Seriously, that whole gold plated thing is a bunch of bullshit. Who gives a damn if your connectors are gold plated, the wire running through the rest of the cable is still plain old wire. The only thing monstrous about them is the price.
DJCalarco
Re:How much is your time worth (Score:5, Insightful)
It depends on the situation and the reason for the cable.
Sometimes there are restrictions for routing the cable that makes a prefabricated cable unusable.
And you may sometimes run into problems with a handmade cable, but often it does work just fine. If you get problems - just remake one contact at a time. If you have a decently modern intelligent switch you can also monitor the port for data errors, and if you don't have any errors it's good enough.
As for cabling quality - all the outlets in buildings are usually contacted by the cable jocks from the installation company and they do a simple test and then moves on to the next. I doubt that the quality from a hand made cable and those outlets are much different.
Field Crimping Cat6? (Score:3, Insightful)
Re:Whatever saves time (Score:3, Insightful)
The problem with the "grab whatever if it's temporary" is that temporary solutions oftentimes become more permanent than anything. I have had many experiences where fixing a problem in the server room exposes some "temporary" fix from years ago that I never had time to make permanent (and since it worked, nobody thought twice about the problem it had fixed).
Or when developing web applications, somebody implements that "quick function" that does X, intended only for internal stuff. Another feature comes along, and pretty soon we're using that temporary function as the core of a new system... and sometimes it even gets embedded into the core of the system. But remember, it was only temporary.
Re:Just do what your boss wants (Score:3, Insightful)
And why not? The only reason I ever bothered to learn how to make patch cables was because I had bosses who didn't want to spend the extra money to buy commercially produced stuff. If your boss is giving you the OK for the money, then place an order online and have a whole bunch delivered straight to your door. Nothing could be easier.
I can produce cables all day long and something like 99% of them will be easily good enough for my needs. Still, I wouldn't doubt that a company like Belkin, running an ethernet cable factory or whatever, has better equipment to create and test the cable than I do, and they create better quality cables with greater efficiency than I would. Fewer problem cables, and they catch a greater percentage of the bad ones in their QA.
Re:your boss sucks at making ethernet cables (Score:5, Insightful)
They can put it under the "Tech" section if they like, but this is really another disappointing Ask Slashdot. It's disappointing because too many of these have this format: "my boss at work wants me to do X, but I'd really rather do Y; what are the merits of X versus Y?" All of them need to be summarily rejected, with a polite e-mail sent to the submitter which says "within the bounds of the law, you need to do what your boss asks you to do whether or not you necessarily agree with it. If you cannot convince your boss to do otherwise, and this is a problem for you, perhaps you should consider working elsewhere."
I don't know about you, but I was hired for my technical expertise. It is part of my job description to let management know when they are making bad technical decisions. If they still insist after that, then sure, I gotta shrug my shoulders and do it; but until then, it's my job to find out the facts and make sure they stay informed...which is exactly what the questioner is doing.
Save the "shut up and do as you're told" bit for McDonald's burger-flippers. We're professionals here.
What kind of distance are we talking about? (Score:2, Insightful)
Drill test (Score:5, Insightful)
Best advice I've ever heard on cabling:
If you have to drill holes to run it, make your own. If you don't buy it premade.
Second best advice:
Test it all. Even if it comes in a shrink wrap package.
Bad Attitude (Score:5, Insightful)
This principle of going with the provider you can sue over the one you can rely on is becoming far too prevalent.
I'm not saying that there's anything wrong with Belkin, and I think in this situation the pre-made cables are the better option.
However, in a more general sense, I'd prefer that my systems didn't go down rather than being able to point the finger when they do. If you are the front end provider of a service your customers are not going to be placated by the fact that, even though all their data is gone, you are currently seeking glorious retribution from the guy that solders the LEDs onto your motherboards (or whatever).
On top of this, when things go tits up at three o'clock in the morning - you can be sure the Belkin shop won't be open.
Re:How much is your time worth (Score:2, Insightful)
keep drinking the koolaid, I think you can do better at repeating the Belkin marketing FUD than that.
Re:Just do what your boss wants (Score:5, Insightful)
It's clearly not your company's core business to make their own patch cables. It may be fun for you to wittle down your own toothpics from lincoln logs but if it's not in your job description it ain't going to fly. Seriously, just buy the damn stuff and do what your boss has asked.
Exactly.
We buy all our patch cables in bulk. There's no reason for me to assemble a new cable every time we want to patch in another machine. It may not take long to throw together a 6' cable, but why waste any time at all? Break open a package and plug it in. Done.
We do keep a couple spools on hand, and some wiring tools. If we absolutely need a 15' cable and we don't have any on hand we'll throw one together...
Or if we really need a new wall jack somewhere... It's good to have the capability to do some of your own wiring if necessary... But for anything substantial we'll contract someone else to do that, too.
But, really, that isn't what I'm paid to do.
Re:How much is your time worth (Score:5, Insightful)
time out! How does making one cable run constitute reason for determining ROI on making cables based on cables per hour etc?
There are some of us who can whip out a standard 15ft Cat5 cable in about twice the time it takes to unwrap a commercially purchased cable. If you need custom sized cables, it's far easier and cheaper except for very small number of situations. If you have one cable run to do and the parts available, it makes no sense to go buy one. It's not like you're going to run out and get a 39 meter Cat6 cable at lunch time in the Mall.
Now, if you need 50 of them? perhaps a different story, but same story goes when you need 5000 of them. Savings get bigger with bulk. In this case, it was for a single cable. My suspicion is that his boss has never actually seen a well made cable created in front of his eyes. Remember grasshopper, all great Samurai swords were made by hand, not a machine with expensive dies.
Whatever happened to quality custom workmanship? It's almost as if people expect that it can't be done anymore? WTF?
Re:How much is your time worth (Score:3, Insightful)
It does when the provider uses a home-made cable as justification for line errors. Stranger things have happened.
This is not a time/money issue (Score:5, Insightful)
This is a CYA issue. Your boss does not want to explain to HIS boss, when a cable goes bad and the company is losing $large_number per hour until it is diagnosed and fixed, that he authorized one of his tech guys to use "homemade" cables.
Re:How much is your time worth (Score:5, Insightful)
Re:How much is your time worth (Score:5, Insightful)
Uh, what part of unshielded twisted pair don't you understand? The whole idea of the twists assuages any reasonable amount of both reception and transmission externally. Modern endpoint tranceivers are really good at signal to noise problems; you can usually make cables quite a bit longer than the IEEE specs call for with total impunity. And Cat6 isn't necessary, either, just quality Cat5/5e is fine.
The only place where fast transmission cables have problem with Ethernet is at the connectors. Crosstalk there, and ONLY THERE, can be a problem unless you do something pretty unnatural to the cable. RJ-45s simpily suck because of the parallel tines. Crimp according to directions. Takes about 30sec a side. If you're worried, shoot a TDR down the line. Fluke and others make some pretty cheapo testers, or bug a cable guy to test it for you. This is not rocket science. Don't wimp out and have someone else do it. Buy quality connectors and cable, and just do it. Get the color code, follow it, and move on.
Not as such... (Score:5, Insightful)
They are made at specific lengths for marketing reasons. All of the "transmission line" characteristics of Ethernet cable have been solved for every length within the specified maximum.
I have a whole data center (~32 rows of 22 racks) fully cabled with lengths ranging from 100 meters to 5 inches (crossover between 1U boxes). They are cut to custom lengths, source to destination. Where their port is on the router and where they were placed in the tray add and subtract inches here and there. They run to the patchpanels in bundles about 7 inches in diameter. We have no problems with crosstalk, reflections, intermod and what have you.
If this were coaxial Ethernet we could have a fun discussion... but those days are well behind us.
Re:How much is your time worth (Score:5, Insightful)
>Welcome to the real world, where standing waves on transmission lines do exist, and you can choose lengths carefully based on the frequency going down it.
That may be true for RF, but the whole point of communication is that you get more than a single frequency - a frequency spectrum. So matching your cable length is no good for Ethernet.
In practice, there are hardly issues with standing waves, because the cable is unidirectional. So two reflections have to occur, and there is additional damping in the cable. So unless your impedance is way off, you should be fine. Even a 10% mismatch is perfectly harmless.
Re:How much is your time worth (Score:4, Insightful)
Re:How much is your time worth (Score:4, Insightful)
But ethernet is a broadband signal with a huge range of frequencies. You might optimise out one class of standing wave, only to bring in another. Also, the 100m length restriction isn't really to do with loss increasing wrt frequency; it's a restriction on the time of flight of a pulse. If you're trying to prevent collisions, then the time-of-flight cannot be too long, otherwise the response speed of the switches at either end is limited. (USB has a 5m restriction for the same reason).
Re:Create your own but TEST the cables... (Score:2, Insightful)
I used to work for an IT consulting / do-it all shop in a small town. One of our big things was wiring businesses.
For switch to outlet wiring we made our own cables. It was a lot easier than guessing what lengths of cable we needed to go up the wall, across the ceiling, and wherever else. At the switch we'd put a plug at the end and at the wall we'd put an outlet or plug (if it was going into a pre-wired cubical).
All walled cables were tested using a basic 6 light tester. I'd recommend that be done even if you bought a 200ft commercial cable and put it into the wall. It is a lot easier to test it now and fix it than have to come back later and fix it.
For wall - PC patch cables and switch to box patch cables we'd use basic commercial 3,6,9 ft 5e cables. Generally purchased from Greybar in bulk. If you use proper cable routing techniques (with proper cable holders) in the server room you can keep it clean. If you use custom length cables in there you end up having to make new cables every time you need to change the connections around.
Re:If the Belkin cable fails, you can blame Belkin (Score:5, Insightful)
If something happens with the Belkin cable, you can blame Belkin.
That presumes:
Besides, do you really want to crimp your own cables?
Yeah, but I'm one of those crazy people who fix their own fences, hang their own ceiling fans, build treehouses for their kids, and generally like to do things not conducive to the strictly consumer lifestyle.
Re:How much is your time worth (Score:3, Insightful)
The wavelength of the high frequency waves your generating in the cable is going to be on the order of ... 10^-12m (I don't have a calculator in front of me, I might be off by some orders of magnitude). So, you would want to have cables that were some integer (or half-integer?) multiple of your wavelength to take care of harmonics, right? How the hell can you do that on a 20m cable?
I only have a B.S. in Physics, but I have to question your wavelength calculation. The waves travel at nearly c, so a 1GHz wave would then have a wavelength of nearly 30 cm, wouldn't it? Still, I agree in that the wavelength is small enough that cable length shouldn't matter unless it's a very short run. Also, for a CONNECTED cable, I have to wonder how much signal reflection you're going to see.
Re:your boss sucks at making ethernet cables (Score:5, Insightful)
Here's how people screw up cables:
1. They just match colors at either end and don't pick a standard pinout (T568A or B). That might result in a usable 10mbps link, but it won't work at 100 or 1000.
2. They mix 568A and 568B - usually wiring A in the wall, and using premade B patch cables. Instant crosstalk. OK on very short runs, but anything longer than 80' to 100' will become problematic with many NICs.
3. They score the insulation. Use the right tools, and adjust the tension on the stripper.
4. They only strip 1/2" of insulation and try to organize the cables and jam it in, so you don't get a clean connection on all the conductors - or might miss one completely. Strip the insulation back 2", then you have room to sort the conductors, trim them neatly then you can slide them all the way to the end of the terminator, then the clips will "bite" each conductor twice - cleanly, resulting in a good connection and a strong link.
Your boss sounds like an idiot and a hack.
If you're going from the wall to the PC I'd say yeah, buy a premade cable and save on labor. Just buy a good one. Believe it or not monoprice's cheap stuff is extremely good.
From the patch panel to the switch, it depends on the length. Will a premade cable fit the length well? Go premade. If not, then make the cables.
Ultimately though, your boss is the boss. If he insists on hanging patch cables loose through the wall, etc. just document it in email, send it to him with your recommendation, etc. so that when he comes back to you complaining that it looks hack or that the cabling was overpriced, you can remind him that you recommended otherwise.
Re:How much is your time worth (Score:4, Insightful)
If you're wiring a patch panel for the first time, however, order a hundred or so cables of various length and save yourself the hassle.
... or hire an unpaid summer intern from a local high school who wants some experience in the IT biz. Lets be honest here. Its people can't find jobs right now, and if the kids is even remotely interested in IT, he will choose pulling cable through your dusty attic over flipping burgers any day.
Re:How much is your time worth (Score:5, Insightful)
The real acid test would be to get one of these TDR units, buy 10 cables from each of two or three reputable companies and compare it to the results from 10 cables done in-house.
All this talk without an objective stress test is pretty pointless.
Re:How much is your time worth (Score:5, Insightful)
But are they exactly the right length to pull from A to B, or do you have some coiled up uglies?
A forward thinking IT professional knows to leave some slack in the cables, otherwise plate tectonics makes fools of us all.
Re:This is not a time/money issue (Score:3, Insightful)
I think you're missing my point, and from your tone, I suspect it's on purpose. I'm sure the submitter can make a good cable. But the boss doesn't know that. And should he put his own ass on the line just because the submitter CLAIMS he can make quality cables?
Look: when something goes wrong in a company, somebody gets blamed. They call it all sorts of things, like "root cause analysis" or some other jargon bullshit, but it's really sticking someone with the blame. If you're the boss, you really, really don't want the blame to land on you OR your department.
If a bought cable goes bad, the blame only goes to you if you made a bad choice of supplier, which is why you buy from ESTABLISHED companies. On the other hand, if a homemade cable goes bad, it's your subordinate's skills, and more importantly, YOUR JUDGMENT that's called into question.
You don't want that, and you know that nobody will ever question your decision to spend a little extra buying cables instead of making them. What decision would you make?
Re:This is not a time/money issue (Score:5, Insightful)
Doesn't help when the first question is, "who supplied that cable"? I don't know if you've been in a "root cause" meeting, but they DO get all Sherlock Holmes on that shit, I assure you.
Re:How much is your time worth (Score:4, Insightful)
And I believe this is the principle on which the submitter's boss operates.
Sure, you can save a lot of money custom fitting and testing every cable, but someone, somewhere is going to be late for lunch, or thinking about their hot date the night before, or busy socializing while they work, and forget to test... or overlook a bad test result, or depending on your karma, even intentionally sabotage a connection.
And while it's aesthetic to have custom-fitted and dressed cables, it's not necessary. In fact, having a meter or two of extra cable can save you some major headaches down the line when you need to alter the installation somehow. And if you're right on the bleeding edge of cable length such that an extra few meters is going to limit your performance, then you're either using the wrong cable, or the wrong protocol.
The bigger your installation, the more nines of reliability you need and connectors are the place where shit goes bad most often, so that's the easiest place to apply money to improve reliability. It's worth it to pay more for the better reliability that automated production, both lot tested and unit acceptance tested, will give you. If you're in a small company or a mom-and-pop-shop, then handmaking all your cables isn't a problem. But if you're in a large-staff or large-volume facility, then buy as much as you can, don't handcraft it.
Finally, when you compare the cost of the factory cable vs. professional-crafted, don't leave out the opportunity cost of the work that your professional is NOT otherwise doing while he's doing a job a robot could do. And realize that you're assuming that a supervisor is NEVER going to say "we need that professional to go do something else... I'll give the tedious job that robots can do to the chronically-stoned intern."
Re:How much is your time worth (Score:3, Insightful)
It is really just common sense to leave some slack in the cable so the next guy who comes along and has to move things around doesn't end up burning you in effigy.
Re:How much is your time worth (Score:5, Insightful)
Make one long cable of the total length of all of the cables you want, and terminate the two ends and test the cable. Then, you know those two ends are good. Then, for your first "finished" cable, snip off the length you want, and terminate the snipped end. Then, test the cable. If it fails, you know which end needs fixing.
Then, with the remaining slightly-less-long cable, terminate the snipped end of that, and test. Then, snip off the next length you need, etc...
I used to just pull off the length I needed from a spool, crimp the two ends, and test. But, if the cable failed the test, and I couldn't see where the problem was, I'd have to flip a coin to decide which end to re-do first. The above method avoids the coin flip. You'll know which end you have to re-do.
Re:This is not a time/money issue (Score:3, Insightful)
But the guy's boss isn't the one Asking Slashdot, speaking of missing points. I don't think what the boss *should* do is germane to this discussion.
We could talk all day about what bosses are like in Happy Land, but the submitter has a question about his real-life job, and mostly he's getting good advice: do what the boss says, and if you can't handle that, update your resume.
Re:How much is your time worth (Score:3, Insightful)
Any decent tester will be able to tell you which end you fubar'd though..