User:Brian P. Josey/Notebook/2010/03/16
|Project name||Main project page|
Previous entry Next entry
I am going through my to-do list from yesterday, luckily, it gives me a lot of things to do today. This section is my follow up work on yesterday's models and data, plus the models that I am creating today.
Yesterday, when I created a plot of the magnetic field strength and its derivative from the side of the cone tip to the flat part of the yoke, I got a point that stood out. It was the last point along the line I created, and on the surface of the flat portion of the yoke. Looking at the model, it is clear that the field strength is higher in the iron than the surrounding air, and at some points it is markedly so. So I think this weird point of data is from the change in magnetic properties between the iron and the air, and nothing more. From the way I set up my graphs and derivative, MATLAB just exaggerated the change.
Permanent Magnets in FEMM
FEMM allows for permanent magnets to be modelled, but it is a little different from the normal set up. Essentially, a permanent magnet in FEMM is a ferromagnetic material surrounded by a small sheet of current that supplies the field, like copper surrounding iron. To do this, you need the coercivity of the magnet, or the strength of a magnetic field needed to reduce the magnetization down to zero after it has been saturated. This value is then put into the "H_c" box in the creating magnetic materials step. You can also determine this from the permeability of the magnetic material.
To do this you plug the rating of the magnet, E, into this formula:
This gives you the coercivity Hc of the magnet. However, it should be noted that coercivity is measured in A/m while the rating is measured in MGOe, which stands for megagauss-oersteds. While I don't have an exact conversion factor between the two, the formula accounts for this, and converts between the two. For our neodymium magnets, the rating is N50, or 50 MGOe, which gives a coercivity of 1.125*106 A/m.
Permanent Magnetic Yoke
Here's a beast: I dub him the monster It is the perfect shape, but the cross sectional area is wrong. For the bend, the area is 1 inch square, for the cylinders, and the cones, it is ~0.785 inches square. So there would be a little overlap. But I like this one for the shape, it gives a convenient 180o bend, and has a nice grove to indicate the half way point to cut at.