Cat 5e twists

Cables are built with different twists in the pairs, to minimize crosstalk. All pairs are good enough to run ethernet.

Reply to
James Knott
Loading thread data ...

Other than with Gigabit, how is delay skew a concern? At 10 & 100 Mb, only one pair is used in each direction.

Reply to
James Knott

Is there a difference between the pairs of Cat 5e?

I've been doing cabling (not exclusively) for a few years but today while drawing a temp makeshift patch cord from my study to my bedroom i noticed that some pairs are more tightly twisted than others. Is it just cos this happens to be a dirt cheap cable?

B.

Reply to
Brendon Caligari

Yes.

Nope, thats exactly the way it should be.

Different twists mean that that radiation pattern for that pair is slightly different than the next pair, reducing cross-talk along the whole length of the cable between the different pairs.

Reply to
Doug McIntyre

Thanks :D

Hmmmm......should the 'rate of twist' change over distance to eventually even out? The twists of this particular cable were quite pronounced and guess that if it was the same all along one pair drawn out would be substantially longer than on another pair.

B.

Reply to
Brendon Caligari

This concern is taken care of by specifying a parameter that's called "delay skew". Up to the maximum distance the signal can travel (100 meters), the length of the "stretched out" pairs (or simply copper length) cannot be different more than in takes light to travel 45 ns in copper (approx. 0.67 speed in vacuum). So, multiplying it all up we get 4.5E-08 sec x 3E08 m/sec = 13.5 meters or

44 feet 3 inches if you like. That's roughly how much copper length difference a pair of devices talking over that cable can tolerate. This, in turn, means that the designers of those devices should allow enough memory buffer to hold the parts of the data stream for at least 45 ns.
Reply to
Dmitri(Cabling-Design.com

James, you are absolutely correct. There is no arguing about that.

10BASE-T and 100BASE-TX devices don't care about delay skew, FEXT, PSNEXT etc. Gigabit Ethernet, however, is not some obscure application that only exists in a lab. A Dell PC you'd buy today will come with 10/100/1000 NIC by default. 10G Ethernet is only couple years away from general acceptance. So, we better start worrying about delay skew today as the cables we lay today will definitely see at the very least Gigabit Ethernet (and most likely 10G) before the end of their life cycle.

With best regards,

Reply to
Dmitri(Cabling-Design.com

Interpair slew (different pair lengths) is a big problem for multi-pair video (YC or RGB).

I don't think any of the big cable mfrs are anywhere near 45 ns out on 100m, but could reach that proportion over shorter

1-5m lengths.

-- Robert

Reply to
Robert Redelmeier

Not quite true - 10Base-T and 100Base-T do care about delay skew. The problem actually surfaced about 10 years ago now. There was a world-wide shortage of teflon used in Plenum cables so some manufacturers started making cables with different insulation on some pairs. These were known as 2+2 and 3+1 cables. The Green / Orange pairs were teflon, and sometimes the Blue pair as well. I think the other pair(s) were polyolefin or similar.

As different insulations were used, the insulation thickness was not the same on all the pairs resulting in all kinds of cabke geometry and interconductor capacitance problems. This was about the time 110VG AnyLan was coming out and getting established, so the amount of delay between one pair and another became a real issue.

Perhaps some of the biggest problems were when using what many people considered a standard as far as cable quality was concerned, AT&T. AT&T changed to a 3+1 cable construction and didn't change the model number of the cable. Their 1062 cable had been pretty solid, although pricey, for several years and now it was made as a 3+1 without the model changing. When people tried using the cable for 100VG, it bombed

- big time. AT&T, later Lucent, admitted they changed the material and started providing date codes on when the cable changed. I don't think they ever recovered from that fiasco and became one of the "avoid unless speced" suppliers within the cabling industry.

There is still some of the 3+1 and 2+2 cables out there in walls being used for 100Base-T and working fine, but if they ever attempt to try to use all four pair, then they will have to replace the run.

Reply to
Justin Time

I think 10baseT & 100baseTX are immune to delay skew because only one pair is use for each of Rx & Tx. 100VG and 100baseT4 could suffer since they use more than one pair for Rx & Tx

Do you have any data that 3+1 or 2+2 won't run 1000baseT (Gigabit)? Don't the controllers have sufficient buffers to compensate for slew?

-- Robert

Reply to
Robert Redelmeier

Not quite true - 100BASE-T4 and 100VG-AnyLan do, but 10BASE-T and

100BASE-TX still don't care about delay skew.

Polypropylene that is (in most cases). The cable manufacturer's are well aware of the difference in dielectric constant between Teflon (2.1) and solid polypropylene (2.25). If you take it directly, the skew due to insulation material difference can be calculated at 10 ns. However, the "slower" polypropylene can be foamed to achieve the constant reduction to any level desired down to 1.55 range to match that of Teflon's. Besides, the electromagnetic wave's path is not exclusively going through the insulation (partly through the 'skin" of the copper conductor), so the difference is going to be even smaller.

The insulation diameter has noting to do with delay skew but impedance of the cable and, consecutively, return loss.

Delay skew requirements for category cables were not yet established/standardized back in 1995. Having based their equipment on the missing specification, HP practically gambled with the life of the application. By the time delay skew was standardized in 1999, 100VG-AnyLan was dead.

The correct part number was 2061 for AT&T SYSTIMAX plenum cable. Has been

2061C to indicate "true" 4+4 plenum insulation and overall performance improvements since the smokes of the FEP shortage have disappeared in 1996 if memory serves me right.

Considering (lack of) acceptance of 100VG-AnyLan I'd call it "puffed" as oppose to "bombed". Most adverse effects of the 2+2 and 3+1 plenum cable designs were associated with fire code compliance, and not performance. Again, cable manufacturers were not stupid justifying the design change: from the point of view of the cabling standard there was NO performance difference.

There has never been any conspiracy as the way you put it implies. If one was interested, one could have asked and gotten the straight answer. True, the info did not make it into sales brochures as the shortage was (appropriately) considered temporary.

One can only talk about personal perception of a product this way. I'm not sure why you single SYSTIMAX out of all the rest of the cabling world, but you are talking about nine lo-o-o-ng years here. They have recovered (if ever suffered), had few great years, then fell prey to stock market plunge and BAD management, then been pushed around and eventually got bought by Commscope, but mid-1990s FEP shortage has very little to do with it. BTW, the way SYSTIMAX has generally been specified is rather "use unless cannot afford" before the shortage and after.

The best approach to deal with installed cable plant of unknown origin/quality is to re-certify it on CAT5E standard. That usually uncovers any potential problem, including possible cable design issues.

Didn't mean to come down hard on you, Justin (forgot your real name, sorry). Just feel cranky this morning ;-)

Reply to
Dmitri(Cabling-Design.com

Thanks to all. I just posted a curiosity more than anything and got an avalanche of relevant meaningful knowledge back.

B.

Reply to
Brendon Caligari

Dimitri,

Thank you for the corrections and illuminations. It has been several years since I was directly involved in the pulling and ordering of cable. Just one or two other points though. The problem with delay skew on 10Base and 100Base-T usually cropped up when you had people splitting a run and putting a second Ethernet on the "unused" pair. Looking back over the history, this was more frequent than a lot of us would like to admit, what with the spot price of teflon cable being up to $300 a box.

While it is true the makeup of the poly can be changed to correct for the differences in delay skew, it was something that wasn't measured between '94 and '96, so a lot of people didn't know it existed as a problem. All I remember about it was that because of the job we were working, we specifically called out all four pair had to be insulated with teflon because of the problem.

If people want to do a google search on the topic of delay skew as it began to unfold in the newsgroup, they will see there was a lot of discussion on the topic. But as far as AT&T not divulging the fact they changed their formulation of the 2062 (I stand corrected) going to a 3+1 and then a 2+2, it all depended on who you were talking to. I was never able to get an answer from AT&T directly, I finally got the answer by going to one of the component manufacturers about their jacks and panels.

Rodgers Platt

Reply to
Justin Time

Hi Rodgers!

One day someone will put together a memoir about the Internet boom era, and we both can contribute a paragraph or to into it ;-) Sort of looking at the same problem from the different sides of the fence (plus some very interesting history prospective, looking back). At the time I was working for the company you just happen to mention as involved into the issue, hence some personal emotions showing in my previous post. Besides, I have to admit I would not know exactly how difficult it would have been to get info on the insulation material from the outside as I had access to this data from inside. I guess, I assumed it is quite open, but I don't know exactly how much sales folks told the customers. Even these days I still get to read it in some cabling project specs that plenum cables has to be 4+4 and not 2+2 or 3+1, which today reads as a kind of technical curiosity. Someone needs to update their spec templates!

Reply to
Dmitri(Cabling-Design.com

Why would running two circuits over the same cable, cause a problem with skew? You're still running on only one pair in each direction, for each circuit.

Reply to
James Knott

Why would running two circuits over the same cable, cause a problem with skew? You're still running on only one pair in each direction, for each circuit.

James,

If I remember my Ethernet right, and it has been too many years since I delved into all the specifications, there is a finite amount of time between when a unit transmits and when it expects to hear something. If the signal is delayed in one direction, then the timers begin to kick in and another node begins to think the pathway is clear and transmits. The signal that had been delayed is then stepped on and the entire system goes into a wait state because a collision has been detected. The clearing of the system after a collision causes the entire network to slow down as it is a function of both hardware and software waiting for the first node to do its collision detection timeout.

Does this help?

Rodgers Platt

Reply to
Justin Time

Skew refers to differences in parallel signals. For example, in gigabit ethernet, you're using all 4 pairs in both directions. This means that if one pair took significantly longer than another to propogate the signal, there would be a significant relative delay between the two, which might cause problems. When you've only got one signal, there's no relative delay to worry about. Collisions are another matter, which does not apply to switches. Back in the days of hubs or coax based ethernet, collisions were a concern, which is why the spec limits the maximum return distance of collision domain to 512 bit times. This provides a process by which collisions may be recognized and handled by the NIC. Collisions have nothing to do with skew.

Bottom line, skew may be an issue with gigabit ethernet, where all four pairs are used.. It is not an issue with 10 or 100 Mb, where only one pair is used in each direction.

Reply to
James Knott

James,

I think we are beating different ends of the same dead horse. The issues regarding delay skew and 10/100 Base-T date from when hubs were more common than switches. As a historical note, delay skew did affect a hub as a hub did not retime the signals.

Rodgers Platt

Reply to
Justin Time

Once again. How do you have skew, when you don't have parallel signal paths? Whether using a hub or switch, 10 & 100 Mb ethernet only use one pair in each direction. This eliminates the possibility of skew. On the other hand, gigabit ethernet uses all 4 pairs in each direction and as a result, the possibility of skew exists.

Reply to
James Knott

Get off the poor animal! ;-)

In the context of the current cabling standards "delay skew" is referred to as a UNIDIRECTIONAL parameter, so I think that you are talking about DELAY, not DELAY SKEW. Of course, different delays on transmitting and receiving pairs, if not controlled, can accumulate, especially across the possible four segments of the Ethernet network based on hubs. I can imagine that, with a cable bad enough, it could possibly get you to the point where devices dis-synchronize, but I don't have an Ethernet textbook handy to check on the actual timing values.

Reply to
Dmitri(Cabling-Design.com

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.