Steel 1008 and 1018 BH curves

I recently had an Internet experience that I have repeated far too often. I needed saturation data for 1008 and 1018 steel for a magnet calculation. These are two of the most common materials used in electromagnets, motors and transformers. The devices are clearly useful — their design involves 100s of thousands of people around the globe. Therefore I was sure it shouldn’t take but a few seconds to find what I needed. Will I ever learn?

The reality was two hours of sifting through chaff, downloading worthless PDF documents and sinking into sites willing to sell me the arcane data for only $30. It’s amazing what the search term STEEL 1018 BH CURVE will dredge up.

I finally succeeded, but the disproportionate effort required to get such fundamental data set me to reflecting. When will we admit that the Search Engine model applied to the Internet repository is a failure when you need “knowledge”. It’s true that the Internet is a valuable sources of “information,” like the telephone number of your nearest ZipLube or the shotputter with the unusual name in the 1952 Olympics that you need for the Sunday crossword. But substance? Never!

Perhaps I am always disappointed with Internet searches because my expectations are not diminished enough. I can remember going to the two-room library in the small town where I grew up and finding exactly what I wanted by flipping through the catalog cards. In contrast , the Internet is equivalent to filling that library with 3 feet of dead leaves and adding a catalog entry for each leaf. To make matters worse, a leaf could get pushed toward the top of the list by making a small contribution.

In any case, this article is not intended to be a complaint; rather, it is an experiment. I am actually putting useful information information on the Internet: those BH curves that were so difficult to obtain. How many people will find them? To help, I picked a can’t-miss title for the article. Here are is a link to download the data:

The tables are averages over the instances I found. They are in a format for use in our PerMag program. The raw values of B0 = ?0*H versus B are listed as comment lines (starting with an asterisk). The code uses a table of the form B versus ?r = B/B0. To ensure convergence in numerical codes, I extended the tables analytically to high values of B using the method described in a previous article:

For more information on PerMag and Magnum, please see and

Comments are closed.