How a bad RJ45 termination can ruin a cable
bluejeanscable.com
bluejeanscable.com
An example of this is thinking that the order of wires in your crimps doesn't matter so long as it is consistent at both ends. This is emphatically not the case! Each twisted pair in a cat5 cable is intended to carry a signal and it's inverse. If you add a signal to it's inverse you should get nothing, but if noise has skewed both signal and inverse, the result of addition will be the isolated interference you need to subtract from your signal. Cat5 connections rely on this noise cancellation to work, as the wires used have little or no shielding. If you put wires into the wrong order so that signals are not paired with their inverses, this noise cancellation system is totally fubar'd and will likely make things worse than if there was no noise cancellation at all. It might work over a few meters, but longer cables are almost guaranteed to fail.
Pick a wiring standard (568-A or 568-B) and stick to it. If you do this, it's pretty hard to mess up anything else so badly that your cables won't work. Bluejeanscables is a cable manufacturer, so they're probably exaggerating how hard it is to make a good cable. In particular, their "buy american" schtick is not very applicable to cabling. Monoprice is functionally equivalent, even if it's from China.
Note: If you're wiring your home, be sure to use plenum grade cable. Other cables may be flammable and fire codes tend to disapprove of having flammable cords running through your home.
Also Note: I am not an electrician. I just found out the hard way by making bad cables for my home. If you buy one $50 spool of cable, $5 of terminations, and a $20 crimp tool you'll never have to pay for network cables again. I'm not sure if it would be worth it now, but it certainly was 10 years ago!
And after reading your comment also realized it is probablhy not up to code as well...
The tricky bit I've found (and I also used a Fluke tester on all cables I manually terminated) is that when terminating cat-6 cables, the inner two pairs had to be separated and terminated perfectly symmetrically. With cat-5, or 5e, you seem to be able to get away with some sloppiness, but with cat-6 you need to think like a seamstress. Those wires need to be exactly in place, and exactly the right length.
I used to find cable termination quite a zen-like experience. I'd do a hundred cable terminations, then fifty cable tests, and very rarely get to smile at a perfect result.
Plenum cabling is typically only required when the cabling is in a space open to the air (e.g., a t-bar grid ceiling with HVAC cold air returns in it). If sealed behind sheetrock or something else you can use the standard stuff, which is less expensive.
Also, almost as important as the proper ordering of the individual strands is preserving as much as the twisting as possible. Hard kinks in the wire and stripping too much at the termination ends can affect that.
Running too near other wiring can severely affect the noise as well. If you need to cross it, perpendicular is the best.
And lastly, be aware of the maximum rated length for the wiring when doing long runs. It can only reliably carry the signal so far.
That means pin 1 is the same on both sides, the devices using this cable have to be different, like a network card on one side and a switch on the other. Internally the two devices have opposite meaning of what wire is receive and transmit.
Or you can use one standard on one end, and the other standard on the other. This makes a crossover cable, where the receive and transmit pairs are swapped.
This is good when connecting two devices that both have the same belief about which pin is receive and transmit.
These days people got tired of dealing with that and most devices auto negotiate which pin is which, and the type of cable doesn't matter.
If you try to use two wires from different pairs for one signal, you lose this property because now the wires are further apart, and possibly worse, twisted with another signal.
https://www.google.de/search?q=ethernet+transformer&tbm=isch
On an old 10MBit or 100MBit/s card (10BaseT, 100BaseT) on each device there will be one pair to send data out, and one to receive data. This uses up 4 pins on your RJ45 connector, the other 2 are unused. The pair that is used on your network card to send out data are pins 1 and 2. The pair that is used on your network card to receive data is on pins 3 and 6.
The 2nd thing you mention are the two standards TIA/EIA-568 assigns colors to pins, and they have a version A and B where the colours of pair 1/2 and pair 3/6 are swapped.
https://en.wikipedia.org/wiki/TIA/EIA-568
Hence, when you have a cable where one side is wired according to "A" and the other is wired according to "B" you have something that, in the old days, was called a cross-cable, which you could use to directly connecto two computers with each other, without having a hub or switch in between. Which is very useful, especially if you'd connect e.g. two switches in two buildings with each other.
Tx 568A 568B Tx
1 o---white----\ /---white---o
2 o---green---- \ /----orange--o
X
3 o---white---- / \----white---o
6 o---orange---/ \---green---o
Rx Rx
With 1000BaseT (Gigabit-Ethernet) data is transmitted and received on all pairs simultaneously, and also this miswiring is automatically detected and taken into account.
This. The article has some very decent and scientifically valid advice regarding terminations, and they unnecessarily throw in this protectionist voodoo about Chinese cable being "almost always terrible" (with an American jobs reference to boot), without backing up their assertions with tests of cables. This makes me somewhat mistrustful of the rest of their article, for no good reason.
So, given the context they may have actual conclusions they reached about Chinese made cables.*
I have had alot of experience in the recent few years with Chinese made wire, and I am unsurprised when it comes out bad. The most common issue I have had is buying a particular gauge and getting something smaller, but marked as the gauge I'm expecting, ex: buying 22awg and getting 24awg.
I don't have much recent experience with network cables being bad, mainly because I only use cables from 3 manufacturers (Belkin, Carlisle Interconnect and Ortronics). But my takeaway from the world of patchcables is that alot are more flakey than people really notice.
* The problem with the phrase "Chinese made" and its like is that between two things made in China you might see wildly different quality. The real thing that drives how well something is made is the quality standards the goods are expected to meet. Typically the case with bad Chinese made goods is that the quality was not met, either it was not expected, it was not tested or not enough pressure was put on the manufacturer to meet the standards.
What the previous test BJC did mainly reveals is that cables are not being tested to meet the standards they are expected to meet. A number of factors are probably contributing to this, such as the industry not really caring to test cables and the price points driving down QA expectations.
To add to the whole Chinese made issue is that factory scrap and seconds is often resold as product. This is stuff that failed QA and should have been torn down to be disposed of but is instead either sold as is or refurbished on a second shift.
High speed cabling is susceptible to all sorts of nasty "gremlins". I work with copper fibre channel runs, RF and long 1394 runs regularly. I am still preaching the gospel to other technicians and engineers that ringing this cabling out with a multimeter is NOT enough to guarantee that the signaling will get through! This article is the perfect demonstration of this. Those terminations would have rung through using a multimeter, which is just sending a very tiny DC signal, but the TDR analysis showed poor terminations. I have encountered the exact same scenario in my own work.
I'd expect this to be the minimum level of diligence when installing cables on a military aircraft, or shouldn't I? ;-)
Blue Jeans Cable was mentioned on HN a few months ago in the context of Monster Cable sending them a letter threatening to sue them for patent and trademark infringement. An HN poster characterized their response letter to Monster Cable as: "That. letter. is. glorious."
Here are just a few snippets:
if you file on this sort of basis, you are in
Rule 11 frivolous-claim territory
...
You are required, as a matter of legal ethics,
to display good faith and professional candor
in your dealings with adverse parties, and you
have fallen miserably short of your ethical
responsibilities
...
Read the patents narrowly, and Monster loses;
read them broadly, and Monster loses.
...
I spent nineteen years in litigation practice
...
I am "uncompromising" in the most literal sense
of the word. If Monster Cable proceeds with
litigation against me I will pursue the same
merits-driven approach: I do not compromise with
bullies and I would rather spend fifty thousand
dollars on defense than give you a dollar of
unmerited settlement funds.
...
Not only am I unintimidated by litigation;
I sometimes rather miss it.
This can't be stressed enough, especially when running horizontal cabling through buildings. We've had issues in taking ownership of network infrastructure that was installed by "reliable" electricians.
Some advice: Make sure that you keep a decent bend radius on the connections and that you're able to sustain (near) peak transfer over the connection. Cable testers and certifiers add to the confidence that you've got a solid connection. Modern switches will also report on CRC errors and issues that come from physical layer problems.
Planning was undone by execution unfortunately. The installers claimed they had worked with fiber optics before and could do a great job. They installed the cat5 cables with such tight turns that only half of them work. The fiber optic cables had the same tight turns, and fiber is more sensitive to bends. The loss between benches just 4 meters apart is in excess of 30 dB! Totally unusable for practically all experiments.
After I got the better switch, it was able to tell me approximately how far along the cable the fault was, which was enough to narrow down the problem. I checked bend radius and re-terminated, and problem solved.
So nice to work with good equipment. It turns voodoo into engineering.
Rather Reaganesque: If they think you're crazier than they are, they'll leave you alone.
(Note: Doesn't work if you have no way to hurt them and they know it, or if they're convinced you have no way to hurt them.)
I wasn't using structured cabling, just ordinary Cat5e cable.
But I think for a security camera, as in the parent comment, which may be getting PoE, the tidiest connection would use a crimp-down male RJ45. I would not enjoy being up on a ladder fussing with a crimping tool and 8 stubby wires.
People sometimes find that LEDs and even florescent lights glow dimly even with the switch off.
The is caused by magnetic induction in lines, which is in turn caused by not running the hot and neutral near each other.
[1] "Good, Better, Best -- Or Not?"
http://www.bluejeanscable.com/articles/good-better-best-or-n...
- Make sure your wires are extending far enough into the plug, you should see the wire fully overlap the brass spades.
- Confirm you have the right wire layout by looking through the plug before you crimp the cable.
- As mentioned in the article, the tail part of the connector is meant to cover the tip of the cable's insulation to provide a bit of stress relief, make sure it's not too short.
I used to make a lot of Ethernet cables, but pretty much stopped when I went to Cat 6. Beyond that, you need to be careful to remain consistent 568B or 568A, etc., and it gets messy. Almost all 10GE I've ever used has been fiber, though.
Almost had a physical altercation with someone who wired a building straight through vs correctly. Amazingly you could still get a 10M link on longer runs, or a 100M link on patches.
(Not straight vs crossover)
It seriously looks like they're ripping off telephoneparts.com: http://telephoneparts.com/index.cgi?category=EZRJ45 (For example, compare http://www.ezrj45.com/ezrj45plugs.php to http://telephoneparts.com/index.cgi?product=EZRJ45/Plugs/CAT... ).
- if nothing else, buy a cheap crimping kit on amazon that includes a cable tester and basic crimper.
- invest in good patch cable. the cheap stuff is much harder to work with and you'll realize the dollar you saved wasn't worth it. I've noticed the boxes (e.g., 1000ft) seem of much better quality than short length patch cables that you re-purpose. $80 for a 1000ft that will last for many years has been a good investment for me.
- cut off an inch or more of outer cable jacket
- cut off that annoying fiber string
- before you untwist the pairs, spread them wide apart in the order they'll line up: when looking at the clip side of the plug, it will normally be Brown, Green, Blue, Orange. I've memorized this order because it's the order of colors going from ground to sky (brown ground, green grass, blue sky, orange sun).
- now untwist the pairs, solid color goes on the left, striped color on the right.
- green striped wire is the exception that gets separated from its pair; move it between the blue and orange wires.
- hold the well-ordered wires tightly between your thumb and index finger and use a rocking side-to-side and back-and-forth motion to bend and comport with its final positioning. This lessens the chance it will mis-align when inserted into the plug.
- Use scissors or some other tool to clip the wire ends so they are of equal length and can extend inside the plug completely. after years of practice this has ended up being the length of my thumbnail, which is convenient since that's where I'm holding the wire as I cut.
- Put wiring into plug and view that wires enter their correct slot. Be sure to enter via clip side facing you (as that's how wires were oriented above. I prefer the clip side because it's easier to see that wires are entering properly.)
- Push cable jacket snugly into plug and rock back and forth to ensure wiring enters full length. Cable jacket should be well inside the plug past the crimp wedge.
- Crimp!
- Check that contacts have all been seated properly. As the crimper gets older it will have a tendency not to seat the contacts properly. This is a sign that it's time to invest in a new $10 crimper.
- Check that opening side of plug has wedged on top of cable jacket and not the wiring. Cable jacket should not be separable from the plug. If you don't do this, over time the jacket will separate from the plug and the wiring will eventually fail.
- When both ends are crimped check with the cable tester.
https://cs.uwaterloo.ca/~brecht/servers/docs/PowerEdge-2600/...
The link has to run in 1000BaseT-Mode to make the tools show any meaningful output, so have a known-good switch connected to the "other" side. I had used them in a distant past (must have been 10 years ago or so) to find the "best" links in an otherwise messed up building installation that could be reliabily be used for higher transmission speeds.
It's of course much less information that what you'd get from a proper ethernet cable analyzer, but much, much more than what a simple "yes, I can ping the other side" can give you.
Practice is the only way to get really good at it. I did several thousand RJ-45 terminations and a similar number of punches into 110-style jacks back in the late 90's, early 2000's. After you've done enough muscle-memory kicks-in and you get consistent, good results. If you just need a few crimps purchase pre-made certified patch cables and you'll save time (money) and frustration.
The fact that differences can be seen in the analogue domain is also not a direct correspondence to how the cable will perform digitally. Ethernet is digital, and as long as the signals pass the thresholds at the receiver, there will be no difference.
The most unusual thing here is that they didn't mention at all whether they actually solved any of the problems the customer originally had, which would be the true validator of their theory. (If they did, wouldn't it be a great thing to mention?) "Network performance issues" are vague - I was expecting to see tests of throughput/packet loss between the original and reterminated cables.
Edit: downvotes. Care to explain...?
A lot of the audiophile nonsense is only nonsense in its application to the ridiculously low frequency world of audio. But when we're dealing with transmission lines, yes, cables absolutely do matter.
I agree, but as long as the receiver interprets the values correctly (i.e. the 1 is above its threshold, and the 0 is below its threshold) it doesn't matter. Among other things I've watched what signals like USB 2.0 HS really look like on an oscilloscope, so I'm well aware that they don't look anything like the nicely-drawn diagrams in textbooks.
Poor cables can absolutely cause high BER and lower performance.
But was that the case here? They identified a cable that had sub-par signal characteristics, and without looking at the actual effects it had on network performance, made the premature conclusion that it was. We don't know whether or not it was really the cause of the customer's problem (imagine them getting back this nicely terminated cable and seeing that it has the same performance... I would not be surprised.)
From the fine article:
"[Our customer is] dealing with network performance issues at a small company, and has come to suspect that bad patch cords, made by an electrician, account for some of the problems he's seeing."
Looks like the customer who sent them the cable had already done the troubleshooting required to determine that the cable was probably bad.
Also, you should probably read: http://www.bluejeanscable.com/articles/channel-certified-eth... Each part in an Ethernet network has well-specified tolerances for a reason. If one part of that network causes far too much signal degradation then all sections of the network that flow through that part can fail to function.
They didn't need to look at the "actual effects".
The very smart engineers who created the high speed Ethernet standards have already studied the "actual effects". They spent years designing, studying, analyzing, and characterizing.
The designers distilled their efforts into some simpler rules such as "near end crosstalk (NEXT) shall not exceed ...".
Those simpler rules allowed a company like Fluke to build a test instrument that checks for things like crosstalk and delay skew and report PASS and FAIL and even report how far off the cable was from meeting the requirements.
That's all there is to it. It's ludicrous to ask for an ab initio analysis of every network problem that could be possibly encountered.
It may eventually be necessary to study "actual effects" in a particular network, but only after dispensing with low hanging fruit such as poorly terminated cables.
Perhaps this mfgr card will work on a almost-but-not-quite cat5 cable run, but that new one won't, or that other one only works half the time, and slowly, or that one won't work when that other fluorescent light is on. The labor cost of this kind of troubleshooting can be pretty spectacular compared to a cheap and simple "must conform to cat-5 minimum standards".
Its rather like AC line voltage. True, some stuff, sometimes, will work at 100 volts or 150 volts, but I'd call an electrician and get it fixed before you spend a lifetime troubleshooting individual things that won't work reliably and consistently at 100 or 150 volts.
It's clear that you're "ignorant" of the black magic involved in high speed communication. I felt that your post amounted to "I don't understand this stuff ... therefore it is probably a scam".
As others have pointed out to you, at high enough speeds things become very analog instead of digital. You can literally fit entire Ethernet packets into a twisted pair cable. Packets can be short enough that they exist completely "in the wire". That's a lot different than what a "premium audiophile cable" does.
If you ever want to learn just how much "black magic" there is, read Howard Johnson's books.[1] The information is somewhat dated, nowadays things are even weirder.
[1] https://en.wikipedia.org/wiki/Howard_Johnson_%28electrical_e...
"ignorant"? That's really jumping to conclusions...
I've worked with DDR, DDR2, PCI, PCIe, and USB (2.0 only, but that's still 480MHz), in mass-produced designs. Also some proprietary busses operating in the 600-800MHz range. It doesn't have to be perfect. That is what is so great about digital signaling.
My post is more of a "I know from experience how much you can get away with, and fancy test equipment that can tell the difference does not always reflect how something performs in practice."
On the other hand, I won't comment on high-frequency true analogue stuff like microwave/RF.
Now I don't happen to agree with 'userbinator's skepticism -- indeed I think that a lot of engineers who work with digital stuff tend to forget that it is analog under the hood, with all the associated crosstalk and BER problems. But creating this atmosphere of "Ooooooh, it is something you probably don't understand" isn't very productive, if you know what I mean. I wish Johnson had titled his books, "High-Speed Digital Design: Quite Simple If You Do the Math and Physics".
I intentionally used the words "black magic" because I was going to link to Johnson's books. While Johnson's publisher (not Johnson) probably named the books, I like the titles. The words you suggest may be more appropriate but the actual titles are definitely "catchier".
It's still true that there are voltage thresholds at the receiver, but for high-speed digital comms there are now _time_ thresholds. In a transmission line, discontinuities cause partial signal reflections that can show up at the receiver to mess up your day. A more intuitive way to show cable quality is the eye diagram, which tells you if the threshold and timing requirements are being violated.
http://en.wikipedia.org/wiki/Eye_pattern
Edit: But this is getting away from the point of the article, which is that bad terminations cause crosstalk. There's nothing mystical about cables failing ISO specs for crosstalk. I agree with you though that some before/after packet loss numbers would've strengthened their case.
This does not mean that all digital cables are fine then - with very long, low quality and/or improperly terminated network cables, you tend to see higher packet loss, sometimes you see dropouts, and sometimes the equipment negotiates to a lower speed. This kind of thing has a lot to do with what's happening in the analogue domain, because interference and signal integrity problems interferes with decoding the analogue signal back into the digital data, so you have packets that fail the checksum validation because they have been corrupted mid-flight.