Hello everyone!
Thanks, Antti, for creating this forum; I think this is a great idea!
My name is Marko, and I am a PhD student at the University of Split. I have a bunch of theoretical questions, but currently, I am in the process of building a prototype of an axial flux electric motor, so I will start with some practical questions about magnets.
In my SPM axial motor, I originally planned to glue the magnets to the rotor surface to prevent magnet detachment due to high centrifugal forces acting on them. So, I bought a couple of super strong two-component epoxy glues and glued NdFeB magnets with a NiCuNi coating to steel plates. Then, I tried to separate the magnet to simulate the shear force that will act in the real motor. It turned out that the glue didn’t break, but rather, the NiCuNi layer slipped and separated from the base material of the magnet.
I tried to contact the manufacturer of the magnets to learn about their experience with the adhesion strength of the NiCuNi layer to the base material. I don’t know whether this is normal or if something went wrong in the manufacturing process. They claim that the epoxy coating has the best adhesion to the base material, but they don’t have any values for the adhesion strength that is supposed to be achieved.
My question is: Do you know what value of adhesion of different coating layers to the base material is considered normal?
Additionally, I am wondering—does a magnet permanently lose part of its magnetic properties if exposed to surrounding air in the process of placing the magnet on the rotor. This question comes to my mind after looking at load line curves of magnets, where for decreasing ratio of magnet thickness to air thickness in the motor, angle of the load line increases. I am wondering does the load line reaches the knee in the second quadrant of BH curve when magent is placed in surrounding air?
In general, can you recommend some magnet manufacturers that you are sure produce magnets with the mechanical, thermal, and electromagnetic properties stated in their datasheets? I talked with some motor manufacturers in Croatia, and they claim that it is really hard to find magnet producers which produce trustworthy magnets, especially when considering operation at elevated temperatures. I plan to use SH magnets, and I would like to be sure that they will operate up to 150 degC and that B and H will reduce with claimed coefficients.
Thank you for your responses!
NdFeB magnets
Re: NdFeB magnets
I can only give you an 'it depends' answer to the demagnetization-in-air question: it is indeed possible if the magnet material has a poor-enough demagnetization resistance. After all, demagnetization more or less starts to happen when a part of the magnet material is subjected to a large-enough demagnetizing field, i.e. negative values of H in the magnet datasheet curves. The limit value can be eyeballed from the curves, corresponding to a point at which the B(H) or J(H) curves start dropping below linear behaviour.
Note: The intrinsic coercivity HcJ is often quoted as the demag limit. HcJ (if I remember right) is the field strength at which the magnetization curve crosses 0 T. Bad things are indeed happening here; however said bad things start happening sometime before that, depending on the squareness of the BJ loop.
Now, in a real magnet floating in air, the magnitude of H obviously depends on the coordinate, and the magnet shape. You can probably indeed get estimates from some analytical formulae; just never done it myself. Failing that, 3D analysis should be fast enough to set up and run, just a piece of magnet floating in air, using linear material properties to speed up things and then checking if we fall out of the linear part in post. Probably overly pessimistic, tho.
Regarding the suppliers, that's a common issue. Maybe some of the expensive Western options, like Arnold or Bomatec could guarantee some values?
Note: The intrinsic coercivity HcJ is often quoted as the demag limit. HcJ (if I remember right) is the field strength at which the magnetization curve crosses 0 T. Bad things are indeed happening here; however said bad things start happening sometime before that, depending on the squareness of the BJ loop.
Now, in a real magnet floating in air, the magnitude of H obviously depends on the coordinate, and the magnet shape. You can probably indeed get estimates from some analytical formulae; just never done it myself. Failing that, 3D analysis should be fast enough to set up and run, just a piece of magnet floating in air, using linear material properties to speed up things and then checking if we fall out of the linear part in post. Probably overly pessimistic, tho.
Regarding the suppliers, that's a common issue. Maybe some of the expensive Western options, like Arnold or Bomatec could guarantee some values?
Re: NdFeB magnets
I've seen in this paper that the adhesion strength between the Ni-P coating and the substrate is around 2.3 MPa. While this might be depending on a lot of factors from manufacturing techniques used in the process to temperature, I hope it can give a rough idea.
Regarding the other question, I believe exposing the magnet material to air and humidity would worsen the magnetic performance as corrosion will change the magnetic properties.
Wang, Y., Deng, Y., Ma, Y., & Gao, F. (2011). Improving adhesion of electroless Ni–P coating on sintered NdFeB magnet. Surface and Coatings Technology, 206(6), 1203–1210. doi:10.1016/j.surfcoat.2011.08.027
Regarding the other question, I believe exposing the magnet material to air and humidity would worsen the magnetic performance as corrosion will change the magnetic properties.
Wang, Y., Deng, Y., Ma, Y., & Gao, F. (2011). Improving adhesion of electroless Ni–P coating on sintered NdFeB magnet. Surface and Coatings Technology, 206(6), 1203–1210. doi:10.1016/j.surfcoat.2011.08.027