Lt Col John Stapp aboard the rocket sled at Edwards Air Force Base in California The first mannedrun of the rocket sled took place in December 1947 and a total of 74 manned tests were conducted through 1951
Lt. Col. John Stapp aboard the rocket sled at Edwards Air Force Base in California. The first manned-run of the rocket sled took place in December 1947, and a total of 74 manned tests were conducted through 1951

What Could Go Wrong? Calibration Answers That Question

Murphy’s Law Spilled milk and complex systems +9,600 gauges, instruments per plant Paying for answers

The science of calibration dates to 1949, when the U.S. Air Force was figuring out how many G's a human being could endure. That is, they wanted to know how to calculate gravitational force, G-force, which feels like weight to the test subject but actually results from acceleration. By definition, it is the effect of force on an object per unit of mass.

The USAF experiment was known as MX981: Human Deceleration Tests. At the time, most experts believed that 18 G's was the absolute maximum that anyone could survive. Captain John Paul Stapp, who headed the MX981 project, thought otherwise.

At Muroc Army Air Field in California, the MX981 team strapped test subjects (including Stapp) to a rocket sled nicknamed “Gee Whiz,” propelled them at 200 miles per hour along a half mile track, and then brought the subjects to a stop in less than a second. Then, they monitored the results.

Often, the results were not good. Captain Edward A. Murphy, Jr., an engineer on the experiment, arrived at the idea to plant strain gauges in the sled harness to quantify the force of gravity inflicted on Gee Whiz’s riders.

After the first test run that made use of these gauges, Murphy noticed that they had failed to record anything. After examining the gauges, he discovered that one of his two assistants had installed them backward. According to the story, Murphy said, “If there’s any way they can do it wrong, they will.”

A few weeks later, when the press questioned Stapp about safety concerns at MX981, he reportedly answered, “We do all of our work in consideration of Murphy’s Law.” Pressed to explain, Stapp gave some iteration of the idea, “if something can happen, it will happen.” Today, thanks to Murphy and Stapp, and MX981, we have the indispensible axiom, “Anything that can go wrong will go wrong.”

The Lasting Effect of Cautiousness

Murphy went on to design safety-escape systems for the North American X-15 and the SR-71 Blackbird aircraft, and Murphy’s Law continued to serve as his philosophy for safety-conscious engineering. Murphy’s Law was not intended to explain the commonplace catastrophes that happen to everyone, the dropped vase for example, or the spilled glass of milk. It was intended to define the low tolerance for mistakes that should prevail among researchers and manufacturers when complex systems are in place. 

The crisis of calibration is the extent to which modern manufacturing tempts Murphy’s Law.

Whereas Murphy had just four sensors to worry about, modern manufacturing operations may have as many as 25,000 gauges and instruments at one plant. In other words, there are at least 25,000 things that can go wrong. At Advanced Technology Services (ATS), we’ve found that the average manufacturing operation has more than 9,600 gauges and instruments, and the majority of these require calibration multiple times per year.

The basis of Murphy’s Law was an experiment with just four innocuous components improperly installed in a complex, dangerous system. Had the goliath breaks on Gee Whiz’s track failed or a harness detached, Murphy might have had far more to lament than a missed gravity reading.

Today, in many industries we pay people to find a solution to a problem. Murphy’s Law and the MX981story suggest that we should be rewarding people for preventing a problem. Operating problem-to-problem, crisis-to-crisis, and failure-to-failure is simply less efficient than addressing problems before they happen. It is the difference between waiting for cancer and then hitting it with chemotherapy, or making simple, healthy choices in order to reduce the possibility of cancer from occurring.

In industries like aerospace, we know that documented calibration audits are not only crucial for safety but also required by law. When an aviation accident occurs, it is one of the only ways investigators can explore and hopefully rule out a manufacturing defect.        

Manufacturers with a lesser safety burden still stand to gain from calibration. In 2008, when ATS was wondering how much calibration issues might cost a company, we asked AC Nielsen for assistance. We learned that, on average, poor-quality calibration costs manufacturers $1,734,000 per year. At companies with revenues over $1 billion, poor calibration can take a $4,000,000 toll annually. Our automotive industry survey conducted in 2006 found that stopped-production costs an average of $22,000 per minute.

At worst, when manufacturers fail to calibrate effectively, they endanger the people who drive their cars, operate their machinery, and rely on myriad mechanical and digital devices for safety.

The legal costs can be ruinous, too. Even when poor calibration leads to minor, fixable flaws, the recalls, the supply-chain disruptions and public scrutiny still damage a manufacture’s credibility and reputation. And, when calibration mistakes are caught before goods leave the plant or warehouse, the operation has wasted hours, days, or weeks making flawed products that cannot be sold.

The story of Murphy’s Law begins with four flawed gauges that wasted an experiment and irritated Captain Murphy. The calibration crisis, however, happens whenever someone forgets to relate Murphy’s insight to the tragedies and business crises that can result directly from a few flawed gauges. The lesson of knowing that things will go wrong is to do everything imaginable to prevent them from going wrong.  

Treg Goss is the manager of Calibration Services at Advanced Technology Services Inc., which performs managed services for production equipment maintenance, industrial parts services, and IT. Contact him at [email protected]

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish