2
\$\begingroup\$

I’m trying to wrap my head around electropermanent magnetism. I read this wonderful PhD Thesis by Ara Knaian as well as details the history of this concept as well as Kyle Gilpin, Ara Knaian, and Daniela Rus’ implementation with the Robot Pebbles project (also quite fascinating ).

As I naively understand, by using a neodymium magnet along side a Alnico magnet (which I believe is solely wrapped with a determined turn count of magnet wire) and sandwiched between two iron “keepers”, a positive voltage and current can be inducted through the magnet wire causing the Alnicos magnetic “polarity” to flip/change. This change can turn on or off the magnetic holding power.

electropermanent magnet concept

First off, is my understanding of this remotely correct?

The author includes a schematic outlining the H-bridge method which I think is necessary as method for providing different power polarities?

H-Bridge schematic

I’d like to try to build one of these electro perm ant magnets, yet getting caught up on the h-bridge concept as well as the general current “building” concept pulsing voltage into a capacitor (which I think builds up current).

All of this I can imagine is quite elementary for you all. Thank you you any help or guidance.

http://www.hizook.com/blog/2010/12/07/electropermanent-magnets-programmable-magnets-zero-static-power-consumption-enable-s

http://www.hizook.com/files/users/3/RobotPebbles-GilpinKnaianRus-ICRA10.pdf

http://www.hizook.com/files/users/3/Electropermanent_Magnets_Knaian.pdf

\$\endgroup\$
  • \$\begingroup\$ You only need an H-bride if you want to be able to program or un-program it in circuit. If you are willing to take it out and reverse the wires, you only need a unipolar solution. For some purposes you can also wire a DPDT relay as a polarity reverser, or even a manual switch. Though modern FET bridges are fairly available... \$\endgroup\$ – Chris Stratton Mar 9 at 17:32
  • \$\begingroup\$ Thank you, Chris! Do you know if the process of “charging” the capacitor using short pulses of voltage is meant to build current before “releasing” into the coil? \$\endgroup\$ – Nicholas Mar 9 at 17:40
  • \$\begingroup\$ I'm not familiar with the practical details of what is going on here specifically, but no, you don't really "build current" in a capacitor, as capacitors store energy in electric fields (as opposed to inductors which store it in magnetic ones). You could however build the voltage on a capacitor and then connect something through which current would flow, for a large bank of low ESR capacitors potentially mind-bogglingly huge current. \$\endgroup\$ – Chris Stratton Mar 9 at 17:44
  • \$\begingroup\$ There's no "building", the pulses are to perform the flip: it only takes a short period of current, a single pulse, to flip the magnet, and to save energy the system is designed to keep the pulses as short as possible. \$\endgroup\$ – pjc50 Mar 9 at 18:08
  • \$\begingroup\$ Thank you, pjc50 \$\endgroup\$ – Nicholas Mar 9 at 19:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.