enter image description hereI want to build a strong elctromagnet. The design is very intuitive and simple, shown in the above figure. The metal core is in a cylindrical shape.

The current design is 100 turns and 30 Ampere, with a radius of 7cm. With the air core (without any metal core inside), the magnetic field is only 20-40 mT. I want to realize a magnetic field of 0.6T. I think the only choice I have is too use a metal core, which has a high permeability and saturation. The problem is where to purchase such material?

The metal core should be large, of radius around 7cm and length around 5cm. I purchased a 1018 steel rod from McMaster-Carr and it turned out this metal core is awful. The 1018 core has only doubled the magnetic field, far behind my expectation. I have checked the B-H curve of 1018, and if the curve if correct, with a H field of 20-40mT, the magnetization of the metal should be close to 2T. I'm confused!

Can anyone kindly tell me why the 1018 steel core doesnot work? And where should I purchase the metal core? Which material should I choose, pure iron, special alloy or silicon steel? Thank you very much!

  • \$\begingroup\$ A microwave oven power transformer MAY do what you want. Grind off the welds that hold the core together. Remove the magnetic shunt. The windings are usually on bobbins which are easily removed. || I made an electromagnet this way for my schoolteacher son. Lots of fun. \$\endgroup\$
    – Russell McMahon
    Dec 18 '20 at 8:03
  • \$\begingroup\$ Regardless of the alloy, tabulated magnetic properties depend on annealing; heat the core to red heat and cool slowly, then retest. \$\endgroup\$
    – Whit3rd
    Dec 18 '20 at 8:30
  • \$\begingroup\$ Sketch your current coil and iron please. \$\endgroup\$
    – Andy aka
    Dec 18 '20 at 9:41

The 1018 core is working just fine, doubling the field is exactly what's expected.

The problem is the remaining air-gap in the magnetic circuit. Using the steel core inside the solenoid has more or less halved the length of air in the circuit.

To raise the B field further with the same H field (H measured in A/m, not T) you have to replace more of the air with iron.

In the case of a transformer, the iron core forms a complete loop, with no airgap. In the case of a lifting magnet, the iron forms an E, and the scrap metal that it's lifting sticks to the open face and closes the magnetic circuit.

Once you have a core and a return path with no airgaps, your 1018 will return the figures you see in the B-H curve.

  • \$\begingroup\$ Hi, Neil, I didn't consider the air gap in my design. As you can see in the attached figure, the design is rather intuive and simple. I major in mechanical engineering and I don't have a background in electromagnet. Could you give me some advises? I want to realize the lifting magnet. When current is applied, I want a strong magnetic field. When the dc current is removed, the magnetic field should be as small as possible. Thank you! \$\endgroup\$
    – jim1124
    Dec 19 '20 at 4:30
  • \$\begingroup\$ @jim1124 Sphero has a picuture of how a lifting magnet is normally arranged, which nicely illustrates my last but one paragraph. \$\endgroup\$
    – Neil_UK
    Dec 19 '20 at 6:39

You want to do something more like this:

enter image description here

The magnetic path is closed completely when you place it against a flat ferromagnetic surface.

In the case of the above design, the winding is in the potted section surrounding the center pole. The magnetic field passes through the center pole, radially out through the end, and back through the outer shell.

You can also make a U shape, or a E shape like a transformer if the I part was removed from E-I laminations.


I want to realize a magnetic field of 0.6T

This is a huge number that normally requires special equipment. You're not going to achieve it with normal electromagnet construction.

  • \$\begingroup\$ I understand that the magnetic field would decrease drastically with the increased distance. It's great if I can achieve a field 0.6T on the surface of the electromagnet. \$\endgroup\$
    – jim1124
    Dec 19 '20 at 4:35

Your Answer

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

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