Picture Picture

Hardness Testing:

Absolutely Not Absolute

Common misconceptions of fundamental hardness testing principles can lead to erroneous measurements.

by Dirk Dusharme


She's tough as nails. This bed is as hard as a rock. His head's like a brick.

Correctly answering the question "What do these three statements have in common?" means you also understand what hardness testing is all about: comparing the hardness of one object to another.

Unlike other measurements that directly relate to a material's physical properties (e.g., weight is directly related to mass), hardness is a comparative measurement. It's typically defined as the resistance of a test specimen to permanent indentation. Using standard testing methods, operators compare the results of a test specimen to a known specimen -- i.e., a hardness test block -- whose hardness they'd like to match.

How hard can this be? Actually, not hard at all if done properly. Unfortunately, hardness tests and testers often are misused, leading to erroneous results. The problem may lie in the user's assumption that the tester is measuring an absolute physical characteristic such as weight or length. While it's always true that a piece of brass that is 1.0023" long is the same length as a piece of steel 1.0023" long, it's not always true that a piece of brass which gives a particular Rockwel1 hardness value has the same mechanical properties as a piece of steel with the same Rockwell hardness.

An even bigger mistake is assuming that converting from one hardness scale to another is the same as converting from inches to millimeters. Incorrectly specified parts, bad procedures, conversion charts and unfamiliarity with the nature of hardness testing leads to incorrect measurements and, ultimately, to product failure.

To examine some of the mistakes made during hardness testing, this article will begin with the basics of hardness measurement tools and then explore how a misunderstanding of these basics leads to misuse and erroneous measurements.

Hardness 101 -- the tools

For the purposes of this article, we'll look at the more popular hardness testing tools. The first group consists of indentation testers: Brinell, Vickers, Micro-Vickers and Knoop. These tools work by pressing an indenter into the specimen and then measuring the indent's characteristics, either manually, with a microscope or automatically, with sophisticated vision measurement systems. These dimensional measurements are plugged into formulas to derive a hardness number. The second group is depth-of-penetration tools: Rockwell and Rockwell Superficial, which measure the depth that an indenter penetrates the specimen.

Brinell -- In the United States, the Brinell test typically consists of a 10 mm diameter tungsten carbide or steel ball pressed into a specimen with a force of from 500 kg to 3,000 kg (1,100 lbs. to 6,600 lbs.). The diameter of the indent is measured with a microscope in two perpendicular axes.

Because of the heavy force and large indenter, this test is very suited to non-homogeneous metals, where one test is more representative of the sample being tested.

Vickers and Micro-Vickers -- A pyramid-shaped diamond indenter with a 136 degree angle between opposite faces is pressed into the specimen with a force of from 1 kg to 50 kg for the Vickers, or from 1 gm to 1 kg for the Micro-Vickers. Both diagonals of the indent are measured with a microscope. The advantage of this test is that on homogeneous metals, the hardness value is considered not load-dependent.

Knoop -- This test can be performed using a Micro-Vickers tester but substituting a Knoop indenter, which has a rhombic base with one set of opposing faces at 130 degrees and another set at 172.5 degrees. Unlike the Vickers tests, the Knoop test requires that only one diagonal be measured. An advantage is that it's possible to measure closer to the edge of a sample or to adjacent test areas. The disadvantage is that materials with grain structure will produce different measurements depending on the orientation of the diagonal being measured to the grain.

Rockwell and Superficial Rock-well -- Unlike the tools listed above, Rockwell testers measure the depth that the indenter goes into the specimen. Because this tester measures penetration depth directly from an indicator rather than making subjective measurements from a microscope, it's an easy instrument to teach an operator to use. Rockwell tests work best on homogeneous metals with smooth surfaces.

A Standard Rockwell hardness tester typically employs a 120½ conical-shaped diamond-tipped indenter with a 0.2 mm radius tip. The indenter is first pressed into the specimen using a minor load of 10 kg. The instrument is zeroed. A major load of 60, 100 or 150 kg is applied for a certain amount of time (known as dwell time) and then removed, leaving the minor load in place. Time is allowed for the elastic recovery of the specimen and the spring of the tester's casting. The hardness number is proportional to the depth of the indenter from after the minor load is first applied to after the major load is removed. Depending on the material to be measured, 1/16" to 1/2" ball indenters may be used instead of the diamond-tipped indenter.

The Superficial Rockwell test uses a     3 kg minor load and 15, 30 or 45 kg major loads. This test is used for thin materials or where surface hardness needs to be tested.

Now that we have an idea of how the tools work, let's look at some of the problems.

Tool damage

Misconception -- All indenters are alike. In a Standard Rockwell or Brinell test, the user drives a ball indenter into  the specimen under test and then measures the indent. The indent depends on the indenter's roundness. If the indenter loses its roundness, the hardness measurement will be wrong.

Unfortunately, because of the difference in cost, some test departments choose cheaper steel balls for their indenters rather than the harder and more costly tungsten carbide balls. It's easy to see why: A tungsten carbide ball costs from $20 to $30, whereas a steel ball costs just pennies.

The cost savings is worthwhile only if the tested materials are softer than Rockwell C47 hardness or Brinell 444BHN hardness. This is the equivalent of steel used on automobile springs. If you test material harder than this using a steel ball indenter, you'll more than likely flatten the indenter's surface. This has two consequences. First, the material will appear to be harder than it actually is because a flattened indenter won't penetrate as far. Worse, subsequent measurements will be wrong even on softer material where a steel indenter normally would be fine.

Switching to tungsten carbide indenters will eliminate the worry about a tested material's hardness. The initial cost is high but not nearly as high as the cost of catastrophic equipment failure because a part tested harder than it actually was.

If you do use steel balls, be sure to inspect the balls for roundness on a regular basis. This should be done as part of the regular maintenance for your tester. This maintenance should also include daily checking with hardness test blocks to ensure that your tester reads accurately.

Improper test methodology

Misconception -- Material preparation and test setup have little effect on hardness readings. The characteristics of various metals or the thickness of the specimen will determine the type of hardness test an engineer specifies. A knowledgeable operator should verify that a specified test makes sense for a given sample.

Observe the following basics when running a test:

  The test surface should be representative of the material. It does no good to test the center of a saw blade when you're looking for the hardness of its cutting edge.

  The test surface should be milled, ground or polished so the indentation is clearly defined. Particularly at lower forces, a rough surface, a carburized or decarburized surface, or a surface affected by grinding or filing will give inaccurate results.

  The test surface should be parallel and perpendicular to the indenter.

  Use the highest load possible without breaking one of the following rules. Higher loads create larger indents, which better represent the material under test.

  Tests shouldn't be taken closer than three indenter diameters from the previous test edge or from the sample's edge. Too close to the edge or to another indent, and you're not measuring the hardness of the material, you're measuring the hardness of air.

  The tested material's minimum thickness should be 10 times the depth of penetration. If the material is thinner than this, you aren't measuring the test specimen; you're measuring the anvil it's sitting on.

Engineers are always right …

Misconception -- I know more than the engineer. When an engineer specifies a hardness in Rockwell or Brinell numbers, there's probably a reason for it. A skilled engineer understands the properties of the part to be tested and the characteristics of the various hardness testing tools.

A common hardness inspection error occurs when the part drawing specifies a particular hardness test and the inspector chooses to substitute a different test. To a certain extent, the manufacturers of hardness testers are to blame. They provide their customers with conversion charts that allow them to correlate between Rockwell and Brinell, Micro-Vickers and Knoop, or what have you. However, unlike converting from English to metric units, there is no direct mathematical correlation between hardness units. In general, conversions must be used as a last resort and with a great deal of caution.

As an example, a company recently received some spent uranium for use in manufacturing an armor-piercing projectile. This material sticks and conforms to the armor in front of the piercing part of the projectile, helping to focus its energy. The uranium's hardness is critical. The specification called for a Vickers hardness test, which requires a polished sample of the material and an experienced Vickers operator. Needless to say, it also requires a Vickers tester. Whether for lack of a tester, an experienced operator or because of an unwillingness to take the time to prepare a sample, the company ran a Rockwell test, consulted a conversion chart and looked up the Vickers value that corresponded to the Rockwell number obtained.

The problem was that a Rockwell test depends on the dwell time of the major load, which is normally specified. Depending on the dwell time, the Rockwell value could vary by as much as seven numbers, meaning the company didn't really know whether the material was usable. Although the problem was caught, the error could have meant the difference between a projectile that pierces armor and one that doesn't.

… Except when they're wrong

Misconception -- The engineer knows more than I. Even engineers can make mistakes. Take the case of a small instrument spring specification that called out a Brinell hardness number. This obviously can't be done. If you put a 500 kg force through a 10 mm ball onto a small instrument spring, you'll get a crushed spring. The engineer in this case probably took the Brinell number specified for the original block of material used to manufacture the spring and carried that specification through to the finished product.

The solution here is not to call the engineer a fool and then blithely perform a Rockwell test, convert it to a Brinell number and expect accurate results. A Rockwell test is alloy-sensitive, and the number could be meaningless.

The solution to both an incorrectly specified part or a specification that can't be tested is to go back to the expert, i.e., the engineer, and ask him or her to respecify the part using a test that you can perform.

A portable checker is not a tester

Misconception -- A portable hardness checker is simply a handheld hardness tester. Due to their relatively low cost and portability, portable hardness checkers have become quite popular. Unfortunately, many people use portables under the assumption that they are miniature versions of a bench tester. This may be the most important warning in this article: A handheld portable hardness checker is not the same as a bench hardness tester.

Too often, technicians use a portable checker to test a part against a part specification, thinking that if the checker reads out in a Rockwell hardness value, it's a true Rockwell hardness value. Here's the warning: Because a portable tester reads out in Rockwell, Brinell        or Vickers units, don't assume that it has performed a Rockwell, Brinell or Vickers test. It hasn't.

Each portable uses its own technology -- not Rockwell, Vickers or Brinell technology -- to measure hardness in internal units that are unique to that tester. The tester then performs a calculation to convert its units into units that users recognize. They suffer from the same type of conversion errors discussed above for bench testers except that the conversion is performed by the portable tester, and the user isn't aware that a conversion is taking place. A portable reads in Brinell, Vickers or Rockwell numbers more for marketing reasons than for any other. Most of us wouldn't buy a portable hardness checker that read out in Smith or Jones, but we would buy a portable that read out in Rockwell or Brinell.

Portable hardness checkers are very useful for their intended purpose -- as comparison checkers for parts that, for one reason or another, can't be bench-tested on a standard hardness tester. The key word here is comparison. Portables are correctly used to compare an unknown specimen to a known specimen of the same type -- a known-good turbine blade to an unknown turbine blade, or a known-good landing strut to an unknown landing strut -- preferably under the same conditions.

To properly use a portable, take a reading from a known-good specimen. That reading becomes the benchmark for tests on unknown specimens. If the portable reads less on the unknown specimen than it did for the known-good part, you know that the part under test is less hard. To know what hardness it was, you'd have to measure it on a bench tester.

It's not so hard

We depend on hardness every day. Carpenters expect their hammers to pound nails without chipping or falling apart. Metalworkers expect their machine tools to be harder than the material they're working. And a pilot expects the landing gear and other critical parts of a 747 to withstand the force of the giant aircraft touching the ground at 200 mph.

How critical is it that the right hardness test be performed the right way? Tom Farrell, hardness expert and test equipment product manager at Mitutoyo, likes to pose the following question when training Mitutoyo personnel: If you were on a 747 and knew that the hardness specification for a critical component of the landing gear was specified in Brinell units, would you prefer the hardness test to have been done with a Mitutoyo Rockwell tester, a Mitutoyo Vickers tester, a portable tester reading in Brinell units or a properly maintained Brinell tester made by any manufacturer on the planet?


1. Rockwell refers to the hardness testing technique developed by Stanley Rockwell in 1921. It is a registered trademark of Wilson Instruments, a division of Instron Corp.


About the author

Dirk Dusharme is Quality Digest's technology editor. Thanks to these hardness testing experts for sharing 60 years' worth of expertise: Tom Farrell, test equipment product manager, Mitutoyo; Richard Ellis, president, David L. Ellis Inc., manufacturers of hardness test blocks; and Robert Ellis, David L. Ellis Inc. Thanks also to Bill Wilde, director of marketing, Mitutoyo, for supplying information on Mitutoyo products.


[Home Page] [Current] [ISO 9000 Database ] [Daily News] [Phil’s Journal]
[Quality Management] ['98 Past Issues] [Resources] [Advertising]
[Subscribe] [Guestbook]

Copyright 1998 QCI International. All rights reserved. Quality Digest can be reached by phone at (530) 893-4095.

Please contact our Webmaster with questions or comments.

ISO9000 ISO 9000 TQM management quality QC QA teams QS9000 QS-9000 quality digest juran deming baldrige ISO9000 ISO 9000 TQM management quality QC QA teams QS9000 QS-9000 quality digest juran deming baldrige ISO9000 ISO 9000 TQM management quality QC QA teams QS9000 QS-9000 quality digest juran deming baldrige ISO9000 ISO 9000 TQM management quality QC QA teams QS9000 QS-9000 quality digest juran deming baldrige ISO9000 ISO 9000 TQM management quality QC QA teams QS9000 QS-9000 quality digest juran deming baldrige


e-mail Quality Digest