Introduction

Since the term "alloy steel" is used very frequently on this site, we thought that some of our readers might be interested in learning more about the alloys used for tool making. We've divided the material here into two sections, the first covering the industry standard alloys commonly used for tools, and the second with information gathered by testing the alloy content of actual production tools.

Standard Alloys for Tool Making

Table 1 below lists some of the AISI standard alloy steels used for making tools. The AISI number is a commonly used reference for steels and can be used to find more information about the alloy.

The chemical elements in the alloys are listed by their standard abbreviations, in particular carbon (C), manganese (Mn), chromium (Cr), nickel (Ni), molybdenum (Mo), and vanadium (V). Some additional elements are reported, including phosphorus (P), sulfur (S), and silicon (Si). Phosphorus and sulfur are usually regarded as impurities whose content must be strictly controlled.

Alloy Analysis Using X-Ray Fluorescence

X-Ray Fluorescence (XRF) is a technique widely used for measuring the metallic element content of steel alloys and other substances. The testing is non-destructive, quick, and inexpensive, making it ideal for checking the content of various alloys.

Principles of X-Ray Fluorescence

The basic principle of X-ray fluorescence (XRF) is based on the fact that most of the chemical elements (including all of the metals) will emit radiation when excited with sufficiently energetic X-rays. This secondary radiation (termed fluorescence) is emitted at precisely defined wavelengths (or energies) characteristic of each specific element, and is generally in the X-ray spectrum as well. Thus the most basic XRF analyzer would consist of a source of X-rays and a detector capable of determining the wavelength and intensity of the emitted radiation.

For more information, the Wikipedia article X-Ray Fluorescence[External Link] provides an excellent introduction.

Measured Composition for Tools

We were able to have a small number of tools tested on an X-ray fluorescence (XRF) analyzer and have reported the test results in the table below. The XRF machine used for the analysis was set up for measuring the specific metal content of scrap metal and did not report the carbon content, as this was not relevant for the application.

The items tested were actual finished tools that can be seen elsewhere on this site. As finished tools, several of the examples retained at least partial chrome or nickel finishes, which has resulted in skewed measurements for chromium and nickel content in some cases. The suspiciously high readings have been noted with an asterisk (*) in the table.

Anyone wishing to do comparable testing on their own tools would be well advised to select examples that are unfinished, whether originally or courtesy of extensive rust, or that have been ground down such that the finish is no longer present in some areas. XRF testers generally look at only a small spot, so if the finish is missing from that area, the results should indicate the base metal.

Discussion of Results

Before we begin discussing the results of the testing, several caveats should be noted. The first is that we do not have any information on the calibration of the machine, and did not bring examples of known alloys to provide a calibration test. The machine was in active use at a business and presumably was in proper operating condition for their requirements, but for more rigorous testing some knowledge of the calibration of the machine would be important.

A second major caveat is that only a single measurement of each sample was made, so we have no information on the accuracy and repeatability of the measurements. If at some point we are able to secure access to an XRF tester on a regular basis, one of the first priorities would be to characterize the accuracy and repeatability, so that basic statistical tests could be used to infer differences in samples.

Now that we have these rather major limitations out of the way, let's see what can be said about the individual tests.

1. Williams 1027

   The results of this test show a chrome-nickel-moly alloy similar to AISI 8640, though with somewhat higher manganese (Mn) content. The wrench itself has been identified as probable late wartime production, based on its "Alloy" marking and plain finish, and interestingly the markings show traces of a prior "Chrome-Alloy" incision on the forging die. (See the discussion of example 4 below for further information.)

   Thus this wrench serves to show that even after tool markings were changed to indicate a generic alloy, production continued to use high-quality alloys when they were available.

2. Williams 1029

   As with the previous example, the results of this test show a chrome-nickel-moly alloy similar to AISI 8640, with somewhat higher manganese (Mn) content. The wrench has been identified as probable late wartime production, based on its "Alloy" marking and plain finish.

3. Williams 1732

   The wrench used for this test was marked "Chrome-Alloy" and had a plated finish, which had been mechanically removed in the tested area. The results show a chrome-nickel-moly alloy but with a suspiciously high nickel content, suggesting that the remnants of the finish may have skewed the results.

4. Williams 3731

   The results of this test are easily the most significant (even startling!) of all the examples. The wrench is one of the Williams 3xxx series, which are hypothesized to have been made only during WWII, based on the plain or cadmium finishes and the absence of catalog listings. With the forged-in "Chrome-Alloy" markings on the wrench, we expected to find a chrome-vanadium or chrome-moly steel, but the results show no chromium at all! Instead of chromium we find a significant amount of cobalt (Co), a somewhat exotic and expensive metal usually reserved for high-speed steel.

   With a little imagination we can sketch the situation that resulted in this unusual and expensive alloy showing up in a rather ordinary wrench. Due to wartime shortages, the Williams factory must have run out of the standard alloy stock for wrench making, and with a production quota to meet, the foreman ordered the use of the special cobalt-molybdenum steel. And with the wrench markings incised into the forging die, there was no time to modify the markings, resulting in tools that not only wasted special alloy steel, but also failed to meet the nominal "Chrome-Alloy" specification marked on the wrench!

   With its unexpected alloy content, this wrench is the proverbial "smoking gun" to explain why tool markings were changed from specific alloys to a generic "Alloy Steel" equivalent during the wartime years. With material shortages making it impossible to guarantee production with specific alloys, companies opted to switch to generic markings on tools.

5. Williams 1034

   This wrench is an example of Williams' early "Chrome-Molybdenum" line and the results are easy to interpret. With no finish to skew the results, the measurements show chromium (Cr) and molybdenum (Mo) readings similar to AISI 4140 steel, although with a slightly lower proportion of manganese (Mn) than expected.

6. Williams 1723

   The wrench used for this test was marked "Chrome-Molybdenum" and had a partially worn plated finish. The results show a chrome-nickel-moly alloy but with a very high nickel content, suggesting that the remnants of the finish may have skewed the results.

7. Armstrong 7729-A

   This tool was marked "Hi-Tensile", a very generic indication of steel properties without any indication of specific alloy content. We were curious to see what was in it, and the results did not disappoint. This example is basically a nickel-chrome-moly steel, but with a rather high proportion of nickel, and with titanium (Ti) as well. It appears to be a specialized steel, probably with very desirable properties, and definitely not a cheap substitute for the regular steel.

8. Armstrong 2426 Special

   This example was intended to check the composition for a tool marked for chrome vanadium steel, one of the most popular alloys from the 1920s onward. The results show reasonable readings for manganese (Mn) and Chromium (Cr) that might indicate AISI 6150 steel, but surprisingly no vanadium (V) was detected. This points to the need for calibration information, as a typical chrome-vanadium steel has only a small amount (e.g. 0.1%) of vanadium, and this might have been below the detection limit for the machine.

9. Billings M-1029

   This wrench was marked "Vitalloy" and we had hoped to discover the typical composition used by Billings for such production. The extremely high nickel reading though points to the futility of trying to make measurements on plated tools, and about all we can say is that it seems to not have any molybdenum. We're inclined to completely disregard this test, and will try again at some point with an unplated example.

10. Bonney 3120

    This wrench was marked "Zenel" and we were hoping to discover the "secret sauce" in Bonney's special steel. But as with the previous example, the presence of a partial finish has skewed the nickel reading badly. It does show a reasonable value for the molybdenum (Mb) content though, confirming that Zenel is some kind of chrome-moly steel. Further testing needs to be done with unplated examples.

References and Resources

Information on particular alloy steels was obtained from Machinery's Handbook, Revised 21st Edition, published in 1979 (and many other editions) by Industrial Press Inc. (New York). This tome of 2,400+ pages is a standard reference for machinists, mechanical engineers, and anyone needing information on machine shop practice.

The interested reader will find numerous online articles on XRF available via a Google search. A good starting point is the Wikipedia article X-Ray Fluorescence [External Link], which provides an excellent background on the physics of fluorescence, as well as a discussion of the applications and links to manufacturers of XRF analyzers.

Another good reference on XRF is available at Geochemical Instrumentation and Analysis [External Link].