The DNA Project 

Before William came along, there was a big problem in determining the properties of cast iron and how strong it was after manufacture. Cast iron is a lot different than steel.  It can’t be welded or forged.  It hasn’t been purified like steel, so it contains a lot of the impurities of the ore it was taken from and from the fuel that was used in its production.  Therefore, one piece of cast iron can have very different properties than another piece.

To make matters worse, when a piece of cast iron is broken off from a larger piece, the smaller piece can be very different in makeup of strength and other respects than the larger section from which it was taken. 

There is no way to tell the difference by looking, and testing up to that time was either ineffective, impractical, or simply didn’t work.  William set aside all of these tests and substituted one, his Keep’s Test, where any casting made in one foundry using his system could be compared to the castings made in any other foundry using the same system, something not possible before.

WILLIAM JOHN KEEP

Metallurgist, Inventor

William John Keep was born June 3, 1842, at Oberlin, Ohio, USA, and died on September 30, 1918, the son of Theodore John and Mary Thompson Keep.  He attended Oberlin College and received his degree in Mechanical Engineering at Union College, Schenectady.  More of his background appears at the end of this tribute to a Keep who is very famous in his field.  His fame is demonstrated by the fact that many of his books are still in print, though he died in 1918.

It was said that foundrymen could understand William’s inventions if they put their minds to it.  This tribute to William by necessity, then, has to be simplified for those of us not in the metallurgy line of work, and thus the consequent brief generalizations.   In his book, Cast Iron:  A Record of Original Research, it took him 200 pages to explain his process for dealing with the problems of making cast iron.

 Dead Load Testing Machine 

The Keep’s Test, very simply put, involved shrinkage of the test material by temperature change, and stress testing the samples to the point of failure (breakage), all recorded automatically on paper in the form of graphs.  The result was analyzed and placed in one of about 20 categories, thus making it possible for comparison with other results at the same or different foundries.

William’s invention was thoroughly tested by others and to the dismay of the skeptics was found to be exactly as presented.  A search of Mechanical Analysis on the internet, which William named in place of his Keep’s Test, will yield many modern devices that appear to do similar but of course infinitely more modern and varied testing.

In Margo Keep’s Descendants of John Keep of Longmeadow and on numerous websites, we learn that before entering Union College, William was determined to learn the metallurgy trade and worked for some years at the Globe Iron Works in Cleveland.  After graduation, he became superintendant of Hubbell & Brother stove works of Buffalo, New York, and then superintendant of Fuller, Warren & Company of Troy.

Probably the greatest achievement in his working career was becoming general superintendant of the world’s largest stove manufacturing business, the Michigan Stove Company of Detroit, Michigan, at one point producing stoves that numbered in the millions per year. The famous Garland Stove was one of its products.

One can imagine the number of patents he held and the number of associations and organizations to which he belonged.  He was also a genealogist and had a big part in the compilation of the original Frank Best  genealogy, Descendants of John Keep of Longmeadow.   We are told that he was a Presbyterian and a Republican.

MORE Science, Technology, and Industry on the Next Page