Permanent Magnet Mistakes, Part Four

Stan Trout, Spontaneous Materials

We continue with the fourth blog in this series, describing the many types of mistakes made with permanent magnets. Again, my intent is to help engineers in the future avoid the mistakes made in the past, and not to embarrass anyone. Everyone’s experience is different, so please feel free to add your insights in the comments section below.

The fourth group of mistakes on my list are:

1. Changing the grade without considering the rest of the magnetic circuit
2. Specifying the magnet incorrectly or incompletely

Changing the grade refers to either changing the material, say from ceramic to bonded NdFeB, or even just changing the grade of the magnet within the same material type, say from N42 to N55. Designers may be motivated to make this type of change because they need a little bit more magnetic flux to achieve a performance target, and this seems like an easy way to achieve that goal, or it may be the only permissible change on an already approved design. The problem is that the magnetic flux produced by the magnet still needs to travel through the rest of the magnetic circuit in order for the device to work as it was designed. Usually this circuit involves iron or steel, which may be in the form of a motor housing or some other simple shape called the return path. The key point is that the iron or steel needs to be able to carry all the magnetic flux generated by the magnet.

When the grade is changed, typically more flux needs to be carried by the magnetic circuit. Since the magnetic circuit may not have been designed for this increased flux level, there is a distinct possibility that it cannot carry the additional flux. This is because the iron or steel saturates, meaning there is a limit to the amount of flux the material can carry, determined by the combination of material properties and their physical dimensions. A saturated return path is carrying the maximum amount of flux possible and some percentage of the flux is now present outside the return path. The magnetic field inside the device may improve, but just slightly, and the performance of the device will be less than expected. Typically, designers blame the magnet for this failure, but the real cause is that the iron or steel was not redesigned to handle the additional flux. The cure for this problem is to redesign the entire magnetic circuit to avoid saturation of the return path; don’t just replace the magnet.

Having told the cautionary tale above, I must admit that I have seen one case in forty years where changing the grade actually worked. Early in my career, I worked with a design where an alnico magnet was replaced by an SmCo₅ magnet to make the device work as expected. What was the difference? The original alnico design was poor. My theory was that the designer selected a magnet size that fit the available space, not what was actually needed. The magnet was very short in the magnetic direction for a low coercivity material like alnico. This is wrong because of the effect of self-demagnetization on alnico magnets. So replacing the alnico with a high coercivity material like SmCo₅ effectively matched the right material to the design, correcting the initial error.

The magnet specifications that appear on a drawing of a magnet are often all over the map in terms of their usefulness and helpfulness. Some drawings recite everything the designer has ever learned about magnets, at least that is my theory. This information overload is confusing because one is left wondering if all these parameters actually need to be measured for every shipment. Such extensive testing would clearly add to the cost, but it might give little or no helpful information to either the buyer or the seller about the ultimate performance of the product. This is pure overkill.

Some drawings give almost no magnetic information at all, and may just give a vague performance parameter, like a pull test. This lack of information is equally confusing. The pull test for example is tricky to do in a manner that both buyer and seller can agree on the results. Limited information is clearly not good, either.

The best drawings give what I call just the salient information, meaning that they only report the parameters that are important for the way the magnet is actually going to be used. They might also offer some idea of how the magnet will be tested for incoming inspection. This approach requires more work on the part of the design team to identify and measure the relevant parameters in each case. This extra work explains why everyone does not follow this approach. But taking this route offers two powerful advantages. Because the expectations of the buyer are clearer, we are less likely to have quality problems in production and it is usually easier to qualify additional vendors for the project, making for a more competitive and likely less expensive product.

A few years ago, I wrote a Coil Winding paper with Gary Wooten on this topic.[i] The article goes into more detail and provides a handy check list of things to consider when writing a magnet specification.

Nine down, eight to go.