Conventional connectors, where the plug and socket have metal parts which touch, suffer from the following problems:

  • Limited mating cycles or expensive plating.
  • Impedance mismatch (relevant for high speed signals only).
  • Lack of isolation.

Here's an obvious solution that would make for a useful Ethernet type connector. Instead of making the metal electrical parts mate, why not make the magnetics mate instead?

The plug would contain the primary windings, and C-shaped cores, while the socket would contain the secondary winding and more C-shaped cores. When the plug and socket mate, the C-shaped cores would touch. The advantage would be that this design doesn't wear out, and completely isolates the PCB.

Do any such connectors already exist? If not, is there some reason they are not used? Might they end up being more expensive? Might they be less reliable?

  • \$\begingroup\$ There are such transformer couplings inside HP multimeter to isolate digital section from analog. But they are soldered permanent couplings, no mechanical separation possible. The shorted wire loops pass through steel wall and have ferrite beads on each side with secondary windings. \$\endgroup\$
    – user924
    Jul 29, 2012 at 14:35
  • \$\begingroup\$ @RocketSurgeon - Yes. I'm aware that transformers are often used for isolation, but I'm wondering if they've ever been used to form the mating part of a connector. \$\endgroup\$ Jul 29, 2012 at 15:28
  • 1
    \$\begingroup\$ For signals, it seems that Fiber Optics achieves the same purpose. This doesn't work for power, unfortunately. \$\endgroup\$
    – user3624
    Jul 30, 2012 at 1:04

3 Answers 3


Apple has a patent on a technology called MagSafe that uses spring loaded pins and a magnetic shell. This handles the limited mating cycles problem to an extent. Until the springs in the pins wear out. And since it has little to no friction holding it in place (it uses the magnetic force instead) it can disconnect easily when someone trips on the cord which is seen as a safety measure. However this does not address the problems of impedance and isolation.

Inductive Charging works like an air-core transformer (though there is usually plastic in between the coils as well).This is what Paul mac was referring to when he referenced cordless toothbrushes and Barry was referring to when he mentioned electric car chargers. There is also a group known as the Wireless Power Consortium that is pushing a a standard called "qi".

I have a "qi" type back for my cell phone that allows it to be charged by placing it on a special plate. I also have a Sonicare toothbrush that uses inductive charging. Both systems work great. They give essentially unlimited mating cycles as well as provide great isolation. My toothbrush regularly gets covered in water and toothpaste without any detrimental effects.

Both these systems are used for carrying DC charging current. Both systems look as if they could be used to carry an AC signal or a digital signal as well. Though I'm not sure how high a frequency/speed you could attain using such a system. And, at the moment, I do not know of any standards/connectors that are doing so.

  • \$\begingroup\$ MagSafe isn't what's being described. It's closer to inductive-charging-in-a-connector \$\endgroup\$
    – user253751
    Nov 2, 2021 at 13:58
  • \$\begingroup\$ @user253751 He listed three problems in connectors. I start with MagSafe since it is popular and well known. I note that it does not address all of his problems. I then move to inductive charging and discuss it as well. Did you read my answer at all? Because it addresses everything you just said. \$\endgroup\$ Nov 11, 2021 at 22:42

The air gap in a mating magnetic connector would impair the performance. Magnetic force is wasted bridging the gap where the cores meet. I think a magnetic charging interface is used for some cordless toothbrushes. But in this instance efficiency is not important.

  • 3
    \$\begingroup\$ Do you know for a fact that the air gap makes this kind of connector economically infeasible, or are you speculating? \$\endgroup\$ Jul 30, 2012 at 13:25

Both of the most common connectors in use, USB and Ethernet, need to carry DC as that is part of their respective specifications. A pure magnetic connection cannot carry DC. It would be possible using an AC signal to transfer power across such a connection but this would require a major change in these protocols and additional circuitry would be needed on both ends to do the DC to AC to DC conversions. Definitely not a practical solution. Magnetic coupling is being used in some electric car chargers to avoid the need for high current metallic connectors but this is a special application and cost is high.

  • 7
    \$\begingroup\$ Standard "basic" Ethernet does not carry DC, and is already transformer coupled at both ends. PoE (Power Over Ethernet) does use DC, of course. \$\endgroup\$
    – user3624
    Jul 30, 2012 at 1:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.