High Tech From The Dark Ages
courtesy of progressive technologies, inc.2007_3_51
In eleventh-century Europe, every soldier of importance was armed with his own personal sword. Most swords were made thick and heavy to keep them from breaking in battle, since a broken sword meant immediate doom. Yet the heavier the sword, the more energy it took to swing it, slowing the user’s reactions to an opponent’s thrusts. The ideal sword was one that combined lightness and strength.
If you were a king or a knight, or had a personal fortune, you would have traveled (or sent an emissary) to Spain for your sword. The blacksmiths of Toledo had developed blades that were thin and lightweight and had beautiful proportions. They also could take and hold a keen edge. But most important, they had a characteristic unmatched by those from other sword makers: Toledo swords were so tough that they could be bent almost double, over and over, without breaking. That made them virtually indestructible in battle.
No other sword makers knew how to do this. The process was a closely guarded secret. Toledo swords became famous in medieval times and have remained so ever since. (In 1663 the poet Samuel Butler wrote in Hudibras of “The trenchant blade, Toledo trusty.”)
When swords gave way to firearms, the blacksmiths of Toledo turned to other products, but they never revealed their secret process. No description of it has ever been found. Metallurgists fruitlessly examined the blades, looking at the composition of the steel alloy, the heat treatment, and the finishing, all to no avail.
Many centuries after the Toledo sword became obsolete, automobiles appeared. I was a very young boy in the late 1920s and early 1930s, and I remember well some of the problems associated with putting America on wheels. Broken axles and broken springs were common occurrences. The roads of the time were not entirely to blame, for valve springs also failed. Clearly, metal parts had to be strengthened, and as it happened, the answer to this problem was virtually the same process that had been used on Toledo swords. But first it had to be rediscovered, and as fate would have it, the rediscovery was accidental.
Automobile springs required large amounts of steel. A thick iron oxide layer, known as scale, forms on steel as it cools down from the high temperatures required for forming and heat treatment. In automobile springs, the steel had to be descaled before it could be processed or painted. The traditional method was blast cleaning with an abrasive, usually sand.
courtesy of the author.2007_3_52
Sand was inexpensive, but it disintegrated quickly and could not be reused. Moreover, the blasting operation was dirty and dusty and could cause silicosis. What the industry needed was an abrasive with a slow breakdown rate. In the 1940s this problem was solved when methods were developed to cast fine steel shot, 1/16 inch in diameter or less, in large quantities. This shot proved to be a superior abrasive for descaling steel—safe, clean, and reusable—and it quickly replaced sand.
Soon after steel shot came into use for descaling springs, General Motors engineers noted that cars equipped with these new springs had far fewer problems with breakage. This led to a closer examination of the descaling operation. Researchers found that the tiny steel balls thrown at high speed against the springs were actually cold-working, or peening, the surface, leaving a thin layer (usually about .02 inch thick) of permanent compressive stressing. This process became known as shot peening. (Some experimental peening had been tried as early as the 1930s with chilled-iron shot, but it was not successful because the shot broke too easily.)
To understand how shot peening works, we must first look at how metals fail under load. Because of their high density, most metals can take very high compressive (push together) loads without failure. But they fail much more easily under tension (pull apart) loads. Bending a metal part produces both kinds of stress: compression on the side where the load is being applied and tension on the opposite side.
If you flex a part many times, it will ultimately fail, and the failure will occur where the tension loads are highest. The first tiny crack will originate at some minute discontinuity, such as a scratch or dent, on the part’s surface. Another name for such a flaw is a stress riser. A failure crack will propagate from a stress riser because it reduces the cross-sectional area of the part, thus raising the tension load at that point.
Shot peening prevents this kind of metal failure because it pre-stresses a part’s surface with permanent compressive stresses. These counteract the tension stresses that result from loading the part. In addition, the shot-peening process nullifies stress risers because it applies a layer of compressive stresses to the total part’s surface. This prevents a failure crack from starting.
courtesy of the author.2007_3_54
Steel is the most common peening material, but if the part’s surface is very smooth to begin with, peening with glass beads may be sufficient to prevent failures. Compressor and turbine blades for jet engines are peened with glass beads. Glass, ceramic, or stainless-steel beads are also useful when it’s important to avoid contamination of the surface with traces of iron. Another innovation is carbon-steel wire, cut into pieces that are then processed to make them nearly spherical, which is increasingly being used instead of conventionally manufactured shot.
Peening must be done with round objects; if the shot is not spherical, or nearly so, an impact may actually create surface flaws instead of suppressing them. For this reason, a critical step in shot peening is separating out broken pieces of shot before they can be reused. Shot peening must be done when the part is cold, not hot, and the surface must be fully covered with ball strikes for it to be effective. With rare exceptions, no heat treatment can be done after peening or else the effects of the pre-stressing will be lost. Shot peening does its work without increasing the weight of the part, and if the part is very hard, the fact that it was peened may not be visually obvious.
Shot peening does not improve the fatigue life of static structures, such as buildings or bridges. It works best on parts that are cyclically loaded (i.e., those in which many repeating load reversals occur), such as springs or turbine blades. In some cases, though, shot peening can help increase the fatigue life of parts that are not subject to cyclic stresses. Some high-strength materials are susceptible to catastrophic failure caused by corrosion cracking. This type of failure is the result of corrosive gases or liquids entering minute crevices between individual grain boundaries of a part that is loaded in tension. Shot peening is widely used to prevent this problem since it effectively closes the minute crevices.
Shot peening caught on quickly during the 1950s, spreading to aircraft and machine parts as well as automobiles, but knowledge of it was still mostly confined to specialists. Before I started work at the Pangborn Company in 1964, I had never even heard of it. Pangborn, of Hagerstown, Maryland, manufactured machines to blast clean castings and to descale steel using steel shot or grit (crushed shot) thrown at high speeds. It also built machines to shot-peen automobile parts.
I had previously worked at Fairchild Aircraft, also in Hagerstown, as a structures testing engineer, but I was completely unfamiliar with this new process in aircraft manufacture. I soon learned about it and became involved in lab-testing its use on automobile engine and drive-train parts as well as springs. Over the next few months, I eagerly tried to collect whatever printed information I could find about shot peening, but there wasn’t much available, since many companies used proprietary methods to gain a competitive edge and did not publish them. As a result, shot peening was seldom considered as an answer to a failure problem, because most engineers had never heard of it.
This situation persisted for years, and in an effort to remedy it, the world’s first international symposium on shot peening was held in Paris in 1981. I was invited to present a paper. That conference brought shot peening to the attention of many more people, and over the last quarter-century research and applications have expanded greatly.
A typical automobile has shot peening applied to all its springs, gears, axles, drive shafts, and torsion bars, and maybe its connecting rods and crankshaft as well. A jet engine’s compressor and turbine blades, landing gear, wing skins, and many other parts are shot-peened, as are all new railroad wheels produced in this country. Blades used in the steam turbines that drive electric generators are also shot-peened.
courtesy of metal improvement company.2007_3_55
The cost of printing the bills in your wallet was significantly lowered when the U.S. Bureau of Engraving and Printing started to shot-peen its printing plates. In the late 1970s I was called in by the National Bureau of Standards (now the National Institute of Standards and Technology) to investigate a problem with premature breakage of the steel plates that the BEP used on high-speed presses to print currency, bonds, stamps, and so forth. (“Printing plate” is an archaic term. The plate is actually a cylinder attached to the press drive. The cylinder’s internal diameter was the area with the problem, not the face, which applies the ink to the paper.)
I suggested that the NBS bring some new plates to our lab and I would work up a method for shot-peening them. Agents came up to Hagerstown the same day with six plates. Each was in a locked box, and all were constantly guarded by a Secret Service man. After a test setup, the process took only a few minutes per plate. Several months later I received a call saying that none of the test plates had failed, even when run way past previous lifetimes.
At this point the NBS recommended that the BEP buy a shot-peening machine. The director of the BEP called us to ask for a quote. I reminded him that several years before, Pangborn had done some free test work for the bureau, developing a method to destroy the engraved printing face of failed plates. In that case, we recommended one of our standard machines, whereupon the BEP took our quote, rewrote it, and submitted it throughout the industry for open bids. One of our competitors got the order for a few dollars less than our bid.
I said to the BEP man: “Pangborn has quickly solved another problem for you at no cost. Who will get the machine order this time?” He explained that as director he could sign requisitions for up to $20,000 with no further red tape. So at his suggestion, we broke down our machine quote into segments of less than $20,000 and got the order. Several years later he told me I had saved the government millions of dollars.
Yet as important as its anti-failure effects are, shot peening can do much more. In one of its most powerful applications, shot peening can actually mold the shape of a part—for example, forming aircraft wing skins to the required aero-dynamic contours. This process is called peen forming.
When you peen a sheet of metal on one side, compressive stresses will cause the center of the sheet to curve toward the incoming shot in convex fashion. That’s the opposite of what one would imagine if the shot were simply pushing the sheet metal forward, like a rock thrown at a garage door, and this counterintuitive result is the basis of peen forming. When properly controlled, the process can be extremely subtle and flexible.
Peen forming was first tried in Germany near the end of World War II. It was kept highly secret. At least one man who knew about the technique came to America after the war and formed a company to commercialize it. All negotiations and forming work were done in secrecy at his company’s plant, since the process had not been patented.
In the early 1950s the Lockheed Constellation, a piston-powered transport, was the first plane to use peen forming in portions of its wings. The technology was slow to catch on, however, since all the secrecy kept it little known and poorly understood. Still, after joining Pangborn, I became familiar with the general idea, and out of curiosity, on my own initiative, I began experimenting with peen forming during slow lab periods.
I quickly found that peening one side of a plate evenly will cause it to form an approximately hemispherical shape. Forming more complicated shapes was harder, but I found a way to do it. The method involved mechanically pre-stressing the part by clamping portions of it in a stressed position (for example, slightly bent around a block, but within the elastic limit). Ball strikes in stressed areas would affect the surface differently from those in unstressed areas, creating the desired curvature. This was my “eureka” moment, and it marked my return to involvement with the aircraft industry.
In the late 1960s I supervised original test work for Lockheed to show that peen forming could curve the entire 95-foot wing skins of Lockheed’s L-1011 wide-body transport, which was then being designed. When the testing was successful, Pangborn put me in charge of a team of engineers that designed and built machines to peen-form the L-1011’s top and bottom wing skins (the manufacturing would be done at Avco Aerostructures, in Nashville, Tennessee). Machines of this type soon became a major part of Pangborn’s business. I patented the peen-forming method and the machine design for the Avco machines.
The editor of the trade magazine Machine Design heard about Lockheed’s plans and asked me to write about the still-little-known process. My article “Peen Forming,” published on November 12, 1970, was the first description in the technical literature of how the process actually works. A decade later, returning from Rome to New York after a sales trip to Europe and Israel, I flew in an L-1011 for the first time. As I looked out the window, I had a fine view of the wing-top skin, which had been peen-formed on my machine design at Avco, and the massive landing gear, shot-peened on my machine design at Menasco Manufacturing, in Fort Worth, Texas.
Besides making better wings, with none of the bumps or bending lines that come with the old metal-bashing methods, peen forming is also simpler and more flexible. No form dies and no tooling are required; making alterations in the contour is a simple matter of changing the spray pattern and intensity; and, with technology I developed for Gates LearJet and later patented, the same machine can peen-form several different wing shapes, switching from one pre-programmed contour to another at the press of a button.
Peen forming allows an aircraft engineer to design one-piece wing skins for maximum load-carrying capability without worrying about manufacturing details. There is no penalty in weight or high-cost tooling for forming the contours, and the residual compressive stresses left in the part (on both the top and bottom surfaces) by peen forming are beneficial as well, though they are not the main point of the process. Nowadays peen forming has applications in many other industries and can even be used to correct the shape of a part that has become deformed.
Research continues in conventional shot peening as well. Today’s options include peening with a laser, lance peening (in which a slender nozzle is inserted into deep-bored holes to shot-peen their inner surfaces), peening to produce aesthetically pleasing surfaces or to reduce or control friction in specific ways, and multiple rounds of peening with successively smaller shot, to make a surface even harder.
And what of the old Toledo blades? In the 1970s a group of engineers familiar with the history of weapons began to wonder if some sort of peening was the secret. They knew that steel shot did not exist back then, but they felt the idea was worth investigating. So they re-examined a Toledo sword using modern techniques. X-ray diffraction detected an induced compressive stress layer at the surface. Acid etching removed extremely thin layers, beneath which electron microscope surveys showed subsurface compression patterns that would have been made with a ball-peen hammer. The Toledo blades had not been peened with shot, but they had been peened, and that was the secret to their strength and durability.
The serendipitous rediscovery of this millennium-old peening process has given engineers a way to design parts that will accept higher loads, or to extend their fatigue life safely. The explanation of how the process works has led to a better understanding of how metals fail. Over the decades this knowledge has been the basis for many other improvements and cost savings in metal casting, forging, and fabrication technology. That’s why I think that the basics of shot peening and peen forming should be required in all college mechanical, aeronautical, and structural engineering curricula. H
