Question about Ethernet wiring
Solved
Needix59
Posted messages
469
Status
Member
-
Needix59 Posted messages 469 Status Member -
Needix59 Posted messages 469 Status Member -
Hello,
I really have a hard time with the concepts regarding network cabling :
- First of all, I don’t even know what its real name is, because we don’t say “RJ45 cable” because RJ45 is just the name of the connector, we also don’t say “Ethernet cable” because Ethernet is a network protocol and not a cable, and we don’t say “4 twisted pair cable” because otherwise we don’t take the RJ45 connectors into account... It’s a hassle *_*
- Next, I know that there are straight cables for equipment of the same type, and crossed cables for two different pieces of equipment, I also know that the crossed cable is disappearing, etc... But my question is why did we need to cross cables on older models, and why is that no longer the case?
- I continue in the series of misunderstandings with a bit of color this time:
But what is the difference between 568A and 568B?
- Another point
I read on the internet that today we use 4 of the 8 wires of this cable that I cannot name, the other 4 being reserved for future uses. Except that I also read that the other 4 wires serve as “ground,” so who is right/who is wrong? :(
Also, with this schematic, a question comes to my mind: Why not use 4 wires for transmission and 4 wires for reception, we would have better speeds, right?
In this same image, what do PR 1, 2, 2, 3, 4 correspond to?
Thanks in advance to anyone who answers any of these questions! I am very grateful to you
Hoping you could give me some guidance to understand all this.
Thank you!
I really have a hard time with the concepts regarding network cabling :
- First of all, I don’t even know what its real name is, because we don’t say “RJ45 cable” because RJ45 is just the name of the connector, we also don’t say “Ethernet cable” because Ethernet is a network protocol and not a cable, and we don’t say “4 twisted pair cable” because otherwise we don’t take the RJ45 connectors into account... It’s a hassle *_*
- Next, I know that there are straight cables for equipment of the same type, and crossed cables for two different pieces of equipment, I also know that the crossed cable is disappearing, etc... But my question is why did we need to cross cables on older models, and why is that no longer the case?
- I continue in the series of misunderstandings with a bit of color this time:
But what is the difference between 568A and 568B?
- Another point
I read on the internet that today we use 4 of the 8 wires of this cable that I cannot name, the other 4 being reserved for future uses. Except that I also read that the other 4 wires serve as “ground,” so who is right/who is wrong? :(
Also, with this schematic, a question comes to my mind: Why not use 4 wires for transmission and 4 wires for reception, we would have better speeds, right?
In this same image, what do PR 1, 2, 2, 3, 4 correspond to?
Thanks in advance to anyone who answers any of these questions! I am very grateful to you
Hoping you could give me some guidance to understand all this.
Thank you!
2 answers
Hi,
you are right, "ethernet cable" is a misnomer, but it's still the most fitting term.
It refers to a cable intended for ethernet networks (and others, but 99 times out of 100 today it's for ethernet) made up of a cable with 4 twisted pairs and which connects 2 RJ45 sockets (8P8C in reality).
If it makes you happy, gigabit ethernet exists, as there are many different types, and 10G uses all 4 pairs simultaneously and in full duplex (transmission and reception on the same pair).
Indeed, 2 pairs may be enough for 10baseT and 100baseT, but there was also 100baseT4, which used 4 pairs for cat3 cables.
As for the EIA/TIA568A or B wiring standard, that's indeed unnecessary; one would be sufficient, let’s say it helped create crossover cables at one time: A on one end and B on the other is crossed for 10/100baseT.
There has never been a ground in ethernet, just twisted pairs; twisted pairs, using differential signaling help reduce the effect of electromagnetic interference on signals.
If crossover cables were used before, they are no longer today because the ports adapt to the cable and perform the crossover themselves (auto mdix):
In this same image, what do PR 1, 2, 2, 3, 4 correspond to?
PR as pair:
pair 1 == 4/5 of the 8P8C
pair 2 == 3/6 of the 8P8C
pair 3 == 1/2 of the 8P8C
pair 4 == 7/8 of the 8P8C
8P8C equivalent to RJ45 means 8 positions 8 contacts,
there are 6P6C on RJ12 and 6P4C or 6P2C on RJ11, also 4P4C in RJ10 and 4P2C in RJ9
--
and ... There you go!
you are right, "ethernet cable" is a misnomer, but it's still the most fitting term.
It refers to a cable intended for ethernet networks (and others, but 99 times out of 100 today it's for ethernet) made up of a cable with 4 twisted pairs and which connects 2 RJ45 sockets (8P8C in reality).
If it makes you happy, gigabit ethernet exists, as there are many different types, and 10G uses all 4 pairs simultaneously and in full duplex (transmission and reception on the same pair).
Indeed, 2 pairs may be enough for 10baseT and 100baseT, but there was also 100baseT4, which used 4 pairs for cat3 cables.
As for the EIA/TIA568A or B wiring standard, that's indeed unnecessary; one would be sufficient, let’s say it helped create crossover cables at one time: A on one end and B on the other is crossed for 10/100baseT.
There has never been a ground in ethernet, just twisted pairs; twisted pairs, using differential signaling help reduce the effect of electromagnetic interference on signals.
If crossover cables were used before, they are no longer today because the ports adapt to the cable and perform the crossover themselves (auto mdix):
In this same image, what do PR 1, 2, 2, 3, 4 correspond to?
PR as pair:
pair 1 == 4/5 of the 8P8C
pair 2 == 3/6 of the 8P8C
pair 3 == 1/2 of the 8P8C
pair 4 == 7/8 of the 8P8C
8P8C equivalent to RJ45 means 8 positions 8 contacts,
there are 6P6C on RJ12 and 6P4C or 6P2C on RJ11, also 4P4C in RJ10 and 4P2C in RJ9
--
and ... There you go!
What is the difference between 568A and 568B: If I understand correctly, using A or B for our cable doesn't change anything, and we can mix A and B within the same network?
Why not use 4 wires for transmission and 4 wires for reception: Okay, so if we've gone from 2 pairs to 4 pairs, it's to increase our speed? Just like when we go from 4 to 8?
"full duplex (transmission and reception on the same pair)": I researched this on the web about it (half-duplex, simplex, and full-duplex): And I didn't quite get it, I was told that full duplex was two electrical signals crossing on the same wire, is that correct?
And what does my PC use? And in what cases do we use which mode of communication?
Within the same network, we put the same A or B on the fixed cables everywhere.
If you put A on one end and B on the other, it creates a crossover cable for 10/100baseT (only 2 crossed pairs).
This is never done on fixed cables, only possibly in patch cords.
The transmission occurs on one pair, see my link on differential transmission, not on one wire, so 4 wire transmission, you can forget it, at best it would be 2 pairs for transmission and 2 pairs for reception, which was done at one time on 100baseT4, but that no longer exists today.
The speed 10/100/1000 and full/half is always negotiated between 2 network cards or between a card and a switch.
On a switch port or between 2 PCs, full mode is always preferred.
By default, it's auto, but we can force a different value in case of problems or when negotiation goes wrong (not the same mode or speed at each end).
On a hub, full duplex is not possible by design, so half duplex is mandatory and the speed is 10, very rarely 100.