Back to Basics

[ click here for the Back to Basics Archive ]

Penny Wise, Pound Foolish: The Dangers of Skimping on NDT

by Emmanuel P. Papadakis^*

 

 

Introduction
There is nothing more basic in NDT than having a test ready when it is needed. This "Back to Basics" article is a case history of preparing a test and finally getting it implemented. "Finally" is the right word, because the test was rejected by upper management until the night before the assembly line was to begin production. The next morning, all the equipment would be in place, all the hourly workers would be at their stations, all the raw materials would be at hand and the pistol would be fired to start the race, figuratively.

The new part in question was a powder metal connecting rod for a new four cylinder inline gasoline automobile engine. The new engine was to power a million new model compact cars in the following 12 months. The profit on those vehicles hinged upon the success of the powder metal connecting rods. Connecting rods connect the pistons to the crankshafts. The rods take all the stress of the fast burning gas mixture on the power stroke and of the gas/air being compressed on the up stroke. Throwing a rod can destroy an engine.

Originally, connecting rods were made of steel forged at red heat. Some rods were later made of nodular cast iron. Powder metal was envisioned as a strong and economical substitute for both. Powder metal parts start literally as powdered metal, which is compressed into a mold to form a preform and is then sintered to become a solid metal. For adequate strength, the piece must be coined, compressed further at high temperature in a tool and die set to give the final shape of the part. Only a minimal amount of machining is done after the coining of the near net shape part.

---

The specialty house claimed that they could produce everything perfectly and that NDT would be unnecessary.

---

Research and development was begun in the manufacturing processes laboratory at the automobile firm more than two years before production was to commence. At two years before production commencement, my NDT group was called in to join the concurrent engineering team working on the powder metal connecting rods. The chief metallurgist told us that several potential failure modes of the powder and the process had been discovered and that the engineers needed NDT methods to detect these failures should they occur in production. The failure modes included oxidation of the powder, incorrect powder composition, inadequate filling of the preform mold, cracks forming in the fragile preforms before sintering and improper temperatures. The chief metallurgist told us further that the failure of one rod in 10 000 could bankrupt the company. (This was in the hard times after the second oil embargo in 1983.) NDT was under the gun.

A scientist and an engineer in my group went to work on the problem using specimens deliberately made by the chief metallurgist's staff to exhibit those failure modes listed above. A low frequency continuous wave eddy current method was developed which was capable of sorting each type of problem specimen from the acceptable specimens. This method was written up and turned over to the engine division for implementation at an appropriate location before the first machining step. The technology transfer occurred more than a year before production was to begin. We were prepared. My NDT group went on to other projects.

A few days before production commenced, the coined parts began arriving from the powder metal processing specialty supplier. The chief metallurgist made a quick run down to the engine plant, picked out a few coined parts at random and ran metallographic tests on them to satisfy himself of the quality. By this time, he was officially out of the loop, but he wanted independent confirmation that the project was going to proceed as planned. What did he find? Precisely the metallurgical problems he had predicted in the failure mode analysis! He blew the whistle and got the attention of executives up to the vice presidential level. My NDT group was called in because we had the solution. Why, however, had it not been used?

This emergency was the first time we had heard of the actual production scenario which had been decided upon by the engine division. They had decided to outsource the powder metal parts to a specialty house which would take care of everything between the design which the auto company supplied and the delivery of the coined part. The specialty house claimed that they could produce everything perfectly and that NDT would be unnecessary. The engine division bought this assertion and did not call for the implementation of NDT. Their error was discovered by the diligent chief metallurgist just hours before the production of garbage was to commence. The error could have led to hundreds of brand new cars throwing connecting rods on interstates.

A series of high level meetings was held. I had the opportunity to explain our NDT method made available by concurrent engineering a year ahead of time. I enjoyed watching the auto executives force the powder metal specialty house to back down, swallow their words and install the necessary NDT. To bring about the implementation, I had to lend the engine division two eddy current instruments with coils - my group's entire complement of eddy current gear. One was used in the engine plant to sort the 60 000 parts already delivered. The other went directly to the powder metal specialty house and the one at the engine plant ended up there, too, after the initial sorting. They were forced to buy their own as soon as delivery could be arranged.

Production of the connecting rods, the new engine and the advanced car were all saved by concurrent engineering including NDT. If concurrent engineering had omitted NDT, then production would have been delayed a few weeks until an NDT test could have been developed on an ad hoc basis. Imagine, if you will, the loss from shutting down an engine production line and a car production line, each scheduled to run 60 units per hour for two 10 hour shifts a day, six days a week, for three weeks. If the planned profit were $5,000 per car, then the loss would be $108 million. That is penny wise but pound foolish if you consider the cost of two hourly workers and two eddy current instruments at a few thousand dollars each.

So what is basic in this lesson? First, you need the scientist and the engineer to invent the test which will become basic a few days or weeks in the future. Second, you need to involve NDT up front and not call upon it as a last ditch effort. Third, do the failure modes and effects analyses to find out what tests you may need to generate with your concurrent engineering. Finally, don't let any smooth talking salesman tell you that NDT is not needed.

 

 * Quality Systems Concepts, Inc., 379 Diem Woods Drive, New Holland, PA 17557; (717) 355-2142; fax (717) 355-2142; e-mail <papadakis@desupernet.net>.

 

Copyright © 2002 by the American Society for Nondestructive Testing, Inc. All rights reserved.

 

---

 
Copyright © 2008 by the American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement or recommendation of ASNT.

IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc.

ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.