3D Printing Permanent Magnets

Researchers at TU Wien wanted to create magnets with exactly the right magnetic field for a particular application. Their solution? 3D printing of magnets. Previously, it has been difficult to produce permanent magnets with a specific shape of the magnetic field. The resulting magnets will be a boon to magnetic sensor construction.

Previously, after designing a magnet with a specific shape and magnetic field, a researcher would have to create tooling for injection molding. This is expensive and time-consuming and often not worth it for small quantities of magnets.

The new technique uses filament that contains magnetic microgranules. The filament is about 10% plastic and 90% magnetic material. While it is in the filament (and in the printer), the material is not magnetized. At the end of the print, a strong magnetic field magnetizes the finished product.

In addition to customized shapes, the printed material may open up other possibilities such as using different materials in a single magnet to create strong and weak regions.

At Hackaday, we’re mesmerised by magnets. Even when we don’t totally grok them.

23 thoughts on “3D Printing Permanent Magnets

    1. Yea, people are a lot like the elements in the periodic table. Most of them are mildly repelled by stories about magnets, but a few are very attracted to them, some quite extremely. I personally like magnet stories so much, I even make my own. If I was an element, I wouldn’t be diamagnetic like the 5lbs of Bismuth I just bought, or paramagnetic. I would definitely be ferromagnetic, making my own field all by myself. I play with magnets for fun and without prompting. :-)

    1. The filament containt 10% plastic so I don’t think it can be sintered. But there is 3D printer that use metallic powder sintered by laser. I don’t know why they didn’t use such a printer (maybe cost).

        1. Since the workpiece isn’t magnetized until after that is not going to be an issue, Depending on the plastic carrier, sintering might well be possible if the material could be pyrolyzed or driven out by a lower bake prior to the high temp portion of the cycle.

    1. I see an easy answer to this question, don´t you ?
      What material are CNC machines built of ? Can you imagine the mess of having magnetic dust all over the CNC ?
      Even if the particles are not magnetized (aligned) they individually have magnetic properties, and would stick all over

      1. No, when magnets are made they are made they are not magnetic, it is not until after the magnet is finished that they are magnetized. So you could either get unmagnetized sintered materials and just machine them yourself or make a mold and press and sinter them yourselves.

  1. Before I read it I assumed you would print with premagnetized particles onto a bed in a weak magnetic field prealigned to the desired outcome orientation, thereby coaxing the particles into alignment while the filament is fluid. This also raises the possibility of changing the alignment during printing to create a variable alignment pattern end product.

    1. Yea, that was what I was thinking. A robotic arm with an extruder on the tip would be very flexible but the code for moving it would be complex. Imagine if the extruder head had a permanent strong magnetic field that established the field direction of the molten plastic until it hardens. You would then control the field direction by applying the plastic from different directions. Simple in principle but tricky for coding. If I was actually doing this, I’d probably still go Cartesian with a special pivot feature at the tip with two axis for magnetic orientation. Then I could print magnets of almost any shape and almost any direction of magnetization granularly within. One could even use an electromagnet that gets hot when you run it and it heats the filament. Lots of technical challenges to overcome with this but it would be way-cool. :-)

    2. The alignment field cannot be as strong as the hot filaments bond in order to prevent it from pulling the deposited filament. But it would also have to increase in strength to overcome the previous layers alignment if it is different. The source of the alignment cannot be attached to the extruder for obvious reasons.

      1. Yes, this is why I think it is best to go with a design where the alignment field’s B vector is centered on the filament extrusion path. In order for that to work one would have to build in two more degrees of freedom for the print head (think pitch and yaw) Ideally one would like it to pivot with an axis point that is the point where the extruding actually happens. (lofty goal, I know) The electromagnet could even help with the heating or one could use a permanent magnet if forgoing field strength control is OK. The needed design would not be for the feint of heart to make. :-) The whole thing could be easier to implement with a multi-axis robotic arm with an extruder head/magnet assembly, but the software would be more demanding. I really think it IS possible to combine the print head with the alignment field generator. It would be tricky but probably worth it, especially for larger projects. I have spent a lot of quality time with magnets and have built five 3d printers so I feel rather qualified and confident with my opinion. Regardless, I’d still love to hear other’s thoughts and even corrections. :-)

        1. Imagine a ring of addressable field coils that maintains Z axis height and center with the extruder. Wide enough that it is out of the print area. Field strength and orientation are controlled in software. Much easier with an X, Y motion bed.

          1. That could work but wouldn’t the field coils have to be huge? (Enclosing the print job, head, and mechanical system) It sounds doable, maybe even mechanically simpler but if I follow you correctly, I imagine huge rings around the work site, moving along in sync with the print head. (Thoughts of Helmholtz coils and the machine in Carl Sagan’s movie Contact come to mind – but not THAT big) Regardless, I would think the machine would have to consume HUGE amounts of energy just to keep that strong field going unless one has an abandoned superconducting MRI magnet lying around. (Now that Could work with the whole 3d Printer on a gimbel within the uniform magnetic field of the MRI unit.) I’d sooner bet on my idea of an alignment field generator in the print head, but I sure would love to see what you seem to be talking about become a reality, but not in my basement. I kind of value the data on my hard drives. :-)

    1. not exactly, they start with blanks that they then imprint, their technology is more akin to a standard 2d printer in that it doesn’t change the shape of the object but changes the properties of the already existing blank, whereas this builds the shape from the ground up, i don’t know if it is precise enough to do stuff like halbach arrays(what polymagnet does).

      1. TU Wien folks also magnetize at the end, when the whole structure has been printed. Quote: “The end product is not yet magnetic, however, because the granulate is deployed in an unmagnetised state. At the very end of the process, the finished article is exposed to a strong external magnetic field, converting it into a permanent magnet.”

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.