On Saturday, November 5, 2022 at 2:38:34 AM UTC-7, Ricky wrote:> I think the CAT5e cables won't work for 1000baseT, even with all 8 wires populated, no?Oh, it works fine; 100baseT used two-level signalling, one pair, and 1000baseT went to five-level signalling, four pairs at a time, to get an order of magnitude speedup. But, because part of that speedup was using up the noise margin (from two-level to five) and the rest was paralleling multiple (independently clocked) paths, it doesn't need any more bandwidth than 100baseT; so, all the Cat5 and Cat5e wire was immediately compatible. The termination accuracy and crosstalk minimization gets a tad more important. 1000baseT checks the pairs for connectivity, so an old crossover cable... might just work, or a two-pair cable even. It's compatible, so it can fall back to two-level and work with the 10baseT protocols, and with a media adapter, that means it can talk to old thinwire (10base2) or thickwire (10base5) subnets.
TIA/EIA 568 A vs. B
Started by ●November 4, 2022
Reply by ●November 5, 20222022-11-05
Reply by ●November 6, 20222022-11-06
On 2022-11-05, whit3rd <whit3rd@gmail.com> wrote:> On Friday, November 4, 2022 at 6:22:03 PM UTC-7, Ricky wrote: >> I was looking up the wiring diagrams for Ethernet cables and found TIA/EIA 568A and TIA/EIA 568B >> wiring diagrams. They seem to be wired the same, but for the colors used. They swap green and orange. >> At first I thought I was looking at a crossover cable, but no, they clearly say they should be >> connected the same at each end. One uses orange and green pairs, while the other uses >> green and orange pairs for half the signals. > > > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use.the place I normally shop only has "A". https://cdlnz.com/search?q=t568b&adv=false&cid=0&isc=false&sid=false&pagenumber=2&pf=&pt= no patch cables https://cdlnz.com/search?q=t568a&adv=false&cid=0&isc=false&sid=false&pagenumber=21&pf=&pt= 19 pages of patch cables. (I'm assuming the hardwere in the "B" results is also scatterd in the A results somewhere) -- Jasen.
Reply by ●November 6, 20222022-11-06
On 2022-11-05, Ricky <gnuarm.deletethisbit@gmail.com> wrote:> I was looking up the wiring diagrams for Ethernet cables and found TIA/EIA 568A and TIA/EIA 568B wiring diagrams. They seem to be wired the same, but for the colors used. They swap green and orange. At first I thought I was looking at a crossover cable, but no, they clearly say they should be connected the same at each end. One uses orange and green pairs, while the other uses green and orange pairs for half the signals. > > This reminds me of the Star Trek episode where the Black/White guy was fighting the White/Black guy. > > Anyone know what's going on with TIA/EIA 568 A vs. B?Wikipedia says "B" is for compatability with "AT&T 258A" and "A" for compatability with modular phone jacks https://en.wikipedia.org/wiki/ANSI/TIA-568#T568A_and_T568B_termination -- Jasen.
Reply by ●November 6, 20222022-11-06
On 11/5/2022 8:51 PM, Jasen Betts wrote:>> All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. > > the place I normally shop only has "A". > > https://cdlnz.com/search?q=t568b&adv=false&cid=0&isc=false&sid=false&pagenumber=2&pf=&pt= > no patch cables > > https://cdlnz.com/search?q=t568a&adv=false&cid=0&isc=false&sid=false&pagenumber=21&pf=&pt= > 19 pages of patch cables. > > (I'm assuming the hardwere in the "B" results is also scatterd in the A results somewhere)Patch cords don't care about color codes. The only time it is of interest is when you have to terminate to an 8P8C jack. One could conceivably wire a patch cord with all white wires if your manufacturing process could guarantee each conductor ended up where it should (and was twisted with the appropriate mate) Installers fasten wires to jacks -- most of which have a color code printed on them. You can defy this, as well -- but at some peril as The Next Guy will likely wonder why you've got wh/bl where the grn wire should be, etc. He's not going to want to chase down your wire map to find the other end of *that* cable to see if you've been consistent in your color scheme "violations". [The biggest advantage of following a color code in a patch cord is that it gives you a handy way of identifying pin 1 of the 8P8C without having to consult printed reference material (it's easy to remember the color order but always a challenge to recall which side of the connector to start at)]
Reply by ●November 6, 20222022-11-06
On Sunday, 6 November 2022 at 01:46:07 UTC, whit3rd wrote:> On Saturday, November 5, 2022 at 2:38:34 AM UTC-7, Ricky wrote: > > > I think the CAT5e cables won't work for 1000baseT, even with all 8 wires populated, no? > Oh, it works fine; 100baseT used two-level signalling, one pair, and 1000baseT > went to five-level signalling, four pairs at a time, to get an order of magnitude speedup. > But, because part of that speedup was using up the noise margin (from two-level to five) > and the rest was paralleling multiple (independently clocked) paths, it doesn't need any > more bandwidth than 100baseT; so, all the Cat5 and Cat5e wire was immediately > compatible. The termination accuracy and crosstalk minimization gets a tad more important.CAT5e is also fine for 2.5Gbit/s ethernet.> 1000baseT checks the pairs for connectivity, so an old crossover cable... might > just work, or a two-pair cable even. It's compatible, so it can fall back to two-level > and work with the 10baseT protocols, and with a media adapter, that means it > can talk to old thinwire (10base2) or thickwire (10base5) subnets.
Reply by ●November 12, 20222022-11-12
Ricky <gnuarm.deletethisbit@gmail.com> wrote:> On Saturday, November 5, 2022 at 6:42:11 AM UTC-4, Don Y wrote: >> On 11/4/2022 11:24 PM, whit3rd wrote: >> > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. >> +42 >> >> You also want to be sure to be buying connector terminations that carry the >> matching color code, lest folks get even MORE confused. > > Confused about what??? The color of the wires in the cable are totally irrelevant to anything done using cables with RJ-45 connectors on each end. What is important is that the cable be straight through if that's what you need, or crossover, if that's what you need. I will never see the colors in the cables. > > >> > POE, I'm not sure of. >> When power is delivered over the data pairs (Orange/Green), polarity is >> important. So, you want to know whether you've adopted 568A or B. >> Unless you are always probing an 8P8C by pin number (and never actual >> conductors). > > Again, the only difference is the color of the wires in the cables. > You can use either type of cable and they are identically the same > unless you cut one open and look at the colors.This is mostly true. The amount of twist is different between the colors. There's two amounts of twist used, strip the sheath off a cable and look at the pairs. It's obvious once you see it. Wiring up patch cables with odd color sequences may not meet specs. I use 568B for panels as that's what AT&T/lucent/Avaya/whoeverit is now ones silkscreen right on the front. Commercially made patch cables will almost always be 568B.>> The "spare pairs" (Blue/Brown) feed diode bridges so don't really care about >> polarity (unless you're a stickler for standards). >> >> But, 8P8C's are shit connectors. Even shrouded, the damn latches seem >> to find a way to snap off ("plastic fatigue"?)Stop buying junk cables. Panduit 688 series plugs have been the best for decades. They don't snag and they don't break apart.
Reply by ●November 13, 20222022-11-13
On Saturday, November 12, 2022 at 11:09:42 PM UTC-4, Cydrome Leader wrote:> Ricky <gnuarm.del...@gmail.com> wrote: > > On Saturday, November 5, 2022 at 6:42:11 AM UTC-4, Don Y wrote: > >> On 11/4/2022 11:24 PM, whit3rd wrote: > >> > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. > >> +42 > >> > >> You also want to be sure to be buying connector terminations that carry the > >> matching color code, lest folks get even MORE confused. > > > > Confused about what??? The color of the wires in the cable are totally irrelevant to anything done using cables with RJ-45 connectors on each end. What is important is that the cable be straight through if that's what you need, or crossover, if that's what you need. I will never see the colors in the cables. > > > > > >> > POE, I'm not sure of. > >> When power is delivered over the data pairs (Orange/Green), polarity is > >> important. So, you want to know whether you've adopted 568A or B. > >> Unless you are always probing an 8P8C by pin number (and never actual > >> conductors). > > > > Again, the only difference is the color of the wires in the cables. > > You can use either type of cable and they are identically the same > > unless you cut one open and look at the colors. > This is mostly true. The amount of twist is different between the colors. > There's two amounts of twist used, strip the sheath off a cable and look > at the pairs. It's obvious once you see it. Wiring up patch cables with > odd color sequences may not meet specs.The reference I found said there were four twists used, and also indicated the amount of twist is not in the standard. So no point in assuming any correlation between twist and color.> I use 568B for panels as that's what AT&T/lucent/Avaya/whoeverit is now > ones silkscreen right on the front. > > Commercially made patch cables will almost always be 568B.They don't seem to indicate that, but I'll never know, even if I buy them. I suppose I could use a lens and look at the connectors carefully. I think they are mostly clear. -- Rick C. -+- Get 1,000 miles of free Supercharging -+- Tesla referral code - https://ts.la/richard11209
Reply by ●November 13, 20222022-11-13
søndag den 13. november 2022 kl. 05.23.57 UTC+1 skrev Ricky:> On Saturday, November 12, 2022 at 11:09:42 PM UTC-4, Cydrome Leader wrote: > > Ricky <gnuarm.del...@gmail.com> wrote: > > > On Saturday, November 5, 2022 at 6:42:11 AM UTC-4, Don Y wrote: > > >> On 11/4/2022 11:24 PM, whit3rd wrote: > > >> > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. > > >> +42 > > >> > > >> You also want to be sure to be buying connector terminations that carry the > > >> matching color code, lest folks get even MORE confused. > > > > > > Confused about what??? The color of the wires in the cable are totally irrelevant to anything done using cables with RJ-45 connectors on each end. What is important is that the cable be straight through if that's what you need, or crossover, if that's what you need. I will never see the colors in the cables. > > > > > > > > >> > POE, I'm not sure of. > > >> When power is delivered over the data pairs (Orange/Green), polarity is > > >> important. So, you want to know whether you've adopted 568A or B. > > >> Unless you are always probing an 8P8C by pin number (and never actual > > >> conductors). > > > > > > Again, the only difference is the color of the wires in the cables. > > > You can use either type of cable and they are identically the same > > > unless you cut one open and look at the colors. > > This is mostly true. The amount of twist is different between the colors. > > There's two amounts of twist used, strip the sheath off a cable and look > > at the pairs. It's obvious once you see it. Wiring up patch cables with > > odd color sequences may not meet specs. > The reference I found said there were four twists used, and also indicated the amount of twist is not in the standard. So no point in assuming any correlation between twist and color.each pair have a different number of twists per length to reduce crosstalk
Reply by ●November 13, 20222022-11-13
On Sunday, 13 November 2022 at 14:49:18 UTC, lang...@fonz.dk wrote:> søndag den 13. november 2022 kl. 05.23.57 UTC+1 skrev Ricky: > > On Saturday, November 12, 2022 at 11:09:42 PM UTC-4, Cydrome Leader wrote: > > > Ricky <gnuarm.del...@gmail.com> wrote: > > > > On Saturday, November 5, 2022 at 6:42:11 AM UTC-4, Don Y wrote: > > > >> On 11/4/2022 11:24 PM, whit3rd wrote: > > > >> > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. > > > >> +42 > > > >> > > > >> You also want to be sure to be buying connector terminations that carry the > > > >> matching color code, lest folks get even MORE confused. > > > > > > > > Confused about what??? The color of the wires in the cable are totally irrelevant to anything done using cables with RJ-45 connectors on each end. What is important is that the cable be straight through if that's what you need, or crossover, if that's what you need. I will never see the colors in the cables. > > > > > > > > > > > >> > POE, I'm not sure of. > > > >> When power is delivered over the data pairs (Orange/Green), polarity is > > > >> important. So, you want to know whether you've adopted 568A or B. > > > >> Unless you are always probing an 8P8C by pin number (and never actual > > > >> conductors). > > > > > > > > Again, the only difference is the color of the wires in the cables. > > > > You can use either type of cable and they are identically the same > > > > unless you cut one open and look at the colors. > > > This is mostly true. The amount of twist is different between the colors. > > > There's two amounts of twist used, strip the sheath off a cable and look > > > at the pairs. It's obvious once you see it. Wiring up patch cables with > > > odd color sequences may not meet specs. > > The reference I found said there were four twists used, and also indicated the amount of twist is not in the standard. So no point in assuming any correlation between twist and color. > each pair have a different number of twists per length to reduce crosstalkAccording to para 5.8.2 of TIA 568.C.2 "Because of their identical pair groupings, patch cords terminated in either T568A or T568B may be used interchangeably, provided that both ends are terminated with the same pin/pair scheme." As for the number of twists per unit length, the standard simply states: "5.6.2 Pair assembly The pair twist lengths shall be chosen to ensure compliance with the transmission requirements of this Standard." In other words, it is entirely up to the cable manufacturer to choose how many twists per unit length are used so long as the completed cable is within specification. So different patch cables could have totally different internal construction and still meet the standard. John
Reply by ●November 13, 20222022-11-13
On Sunday, November 13, 2022 at 10:49:18 AM UTC-4, lang...@fonz.dk wrote:> søndag den 13. november 2022 kl. 05.23.57 UTC+1 skrev Ricky: > > On Saturday, November 12, 2022 at 11:09:42 PM UTC-4, Cydrome Leader wrote: > > > Ricky <gnuarm.del...@gmail.com> wrote: > > > > On Saturday, November 5, 2022 at 6:42:11 AM UTC-4, Don Y wrote: > > > >> On 11/4/2022 11:24 PM, whit3rd wrote: > > > >> > All 'normal' patch cables are 568B, so that's the only spec I'd ever want to use. > > > >> +42 > > > >> > > > >> You also want to be sure to be buying connector terminations that carry the > > > >> matching color code, lest folks get even MORE confused. > > > > > > > > Confused about what??? The color of the wires in the cable are totally irrelevant to anything done using cables with RJ-45 connectors on each end. What is important is that the cable be straight through if that's what you need, or crossover, if that's what you need. I will never see the colors in the cables. > > > > > > > > > > > >> > POE, I'm not sure of. > > > >> When power is delivered over the data pairs (Orange/Green), polarity is > > > >> important. So, you want to know whether you've adopted 568A or B. > > > >> Unless you are always probing an 8P8C by pin number (and never actual > > > >> conductors). > > > > > > > > Again, the only difference is the color of the wires in the cables. > > > > You can use either type of cable and they are identically the same > > > > unless you cut one open and look at the colors. > > > This is mostly true. The amount of twist is different between the colors. > > > There's two amounts of twist used, strip the sheath off a cable and look > > > at the pairs. It's obvious once you see it. Wiring up patch cables with > > > odd color sequences may not meet specs. > > The reference I found said there were four twists used, and also indicated the amount of twist is not in the standard. So no point in assuming any correlation between twist and color. > each pair have a different number of twists per length to reduce crosstalkYes, different from one another and each an unknown quantity. So no point in preserving any wiring scheme thinking the specific twist associated with a color is significant. -- Rick C. -++ Get 1,000 miles of free Supercharging -++ Tesla referral code - https://ts.la/richard11209
