What is a signal transformer used for?

Why are Ethernet / RJ45 sockets magnetically coupled?

As the title suggests, why do Ethernet sockets have to be magnetically coupled? I have a basic understanding of electronics, but most of the time I can't find the right keywords to Google this properly.


The correct answer is because the Ethernet specification requires it .

Although you didn't ask, others may be wondering why this connection method was chosen for this type of ethernet. Note that this only applies to the point-to-point Ethernet variants such as 10base-T and 100base-T, not to the original Ethernet or ThinLan Ethernet.

The problem is that Ethernet can support longer uptime, allowing devices on different ends to receive power from remote branches of the power distribution network within a building, or even from different buildings. This means that there is a significant amount of space between the Ethernet nodes Mass offset exist can. This is a problem with ground based communication schemes like RS-232.

There are several ways to deal with ground misalignment in communication lines, the two most common being opto-isolation and transformer coupling. Transformer coupling was the right choice for Ethernet given the tradeoffs between methods and what Ethernet was trying to achieve. Even the earliest version of Ethernet that uses a transformer coupling works at 10 Mbit / s.This means that the overall channel must support at least 10 MHz digital signals, although in practice it actually needs twice as much with the coding scheme used. Even a 10 MHz square wave has levels that only last 50 ns. This goes very quickly for optocouplers. There are light transmitters out there that go much faster, but they're not cheap or easy on either end like the Ethernet pulse transformers.

A disadvantage of transformer coupling is that direct current is lost. It's actually not that difficult. They ensure that all information is transported through the transformers quickly enough through modulation. If you look at the ethernet signaling you will see how this has been taken into account.

Transformers also have nice advantages, such as very good common-mode rejection. A transformer only "sees" the voltage across its windings, not the common voltage that both ends of the winding are being driven to at the same time. You get a differential front end with no intentional switching, just basic physics.

Once the decision to couple the transformer was made, it was easy to specify a high isolation voltage without creating a large load. Making a transformer that isolates the primary and secondary voltages by a few 100V will do pretty much anything if you don't try to do it. Getting it to 1000V isn't much more difficult or expensive. With this in mind, Ethernet can be used to communicate between two nodes that are actively driven at significantly different voltages, rather than just to handle a few volts of ground offset. For example, it is perfectly fine and within the standard to operate one node on one power line phase while the other is on neutral.

  1. Insulation. So if the cable is shorted to a high voltage, your board will not explode.
  2. It is needed because the other end may have a different reason. This is a special case of isolation, but it is also required during normal operation.

Isolation is a very good idea for communication systems that connect many different hardware components together over a wide area. You don't want residual currents / voltages in the power cords or devices to spread to your communication cables.

Basically there are two options for isolation, opto and transformer. Transformer isolation has several major advantages. First, the signal power is passed through the transformer so you don't need to power the "isolated" side of the barrier. Second, transformers can generate and receive differential signals very well and at the same time offer high common-mode rejection. This makes them a good combination with twisted pair cables. Third, it is easy to design high frequency transformers (also known as high speed transformers) as optocouplers.

Transformer coupling has a few drawbacks: transformers don't work on DC power, and small transformers that work well at high frequencies don't work as well at low frequencies.

Another important seamless feature that is often forgotten is impedance matching:

The signal transformer adjusts the impedance of the PHY side (type 100 Ohm diff) with the impedance of the mains side (type 150 Ohm diff).

SOME EXPLANATIONS after Kevin's comment:

from here :

Some names for different types of cables:

  • UTP = unshielded, twisted (symmetrical) 4-pair cable, 100 ohms
  • STP = total foil / braid Shielded 2-pair cable with single shield, 150 Ohm
  • FTP = total foil shielded 4-pair cable, 100 Ohm
  • ScTP = total foil / braid shielded cable, 100 or 120 ohms

The standard also mentions 100-ohm UPT and 150-ohm STP as medium - see IEEE 802.3, subsection 24.1.2, point d).

Hence it is clear to say that the impedance of the signal transformer on the PHY side (type 100 Ohm diff) matches the impedance on the line side (can be different) .

We use cookies and other tracking technologies to improve your browsing experience on our website, to show you personalized content and targeted ads, to analyze our website traffic, and to understand where our visitors are coming from.

By continuing, you consent to our use of cookies and other tracking technologies and affirm you're at least 16 years old or have consent from a parent or guardian.

You can read details in our Cookie policy and Privacy policy.