CS_F2012_05

Uh, this one came off a, as I remember this came off an Israeli Corvette. A Corvette's like a small frigate, which is like a small destroyer, it's like a big PT boat, okay, if you will. But anyway, and this was holding the variable pitch propellers on, okay. Nowadays they have a shaft coming down through the propeller shaft that allows you to change the pitch of the propellers depending on the speed you want to go, because there's no one pitch that's perfect for all velocities, okay.

So it turns out the bolts that they use to hold the propellers, if you lose one blade, I mean, the whole thing's out of balance and you lose the ship, okay. I mean, it's going to just tear everything apart, and that's not a good thing, particularly if you're being shot at. But even if you're not being shot at, this particular one looks like it has a fatigue crack under the head, but they were having problems with them becoming loose. But these are some of the most sophisticated bolt designs of anything I've ever seen. First of all, it's made out of a Monel as I remember, which is a high nickel alloy, good corrosion resistance in salt water.

And it's, because it was getting loose, it's why it has the fatigue crack. But you can see they taper the shank so that this thing is designed so the stress is uniform all the way along here. I mean, the threads create a stress concentration, but they've designed it so it's a little bit bigger. Uh, they control the radius down in the threads. There actually is a set screw up in here that they use to — there's actually a hole that goes all the way down, and this is precisely machined for the depth of that hole, so you can measure it before you tighten it. And then you tighten it, and you don't tighten this on torque, you tighten this on stretch. You actually measure the stretch of the bolt, okay. Very precise, because you're using Young's modulus as the measurement of stretch. And so you get very accurate measurements of your clamping force.

And I can't even remember, it's been so long, why they were actually loosening in that particular situation. But in any case, I ended up with a few of them, and we found them a few months ago, so you can pass that around. [Tom passes a bolt sample around the class.]

Okay, anybody have any questions? If not, I'm going to try to explain why we're studying manufacturing science, or why I call this manufacturing science. It can be argued — at least I've argued in the past — that there are only three things that generate wealth in society. Agriculture, mining, and manufacturing create wealth. Everything else merely redistributes it. So what does a medical doctor do? He, you can say he helps create health, but he's redistributing the wealth, okay. What does a banker do? He's just redistributing wealth. What does a lawyer do? He's redistributing wealth, okay. Uh, you could argue that an educator is also just redistributing wealth.

One person at Intel, when he saw me say this, he says, "What about knowledge? It's also something that can create wealth." Well, okay, I'm a materialist person, and I started saying this a little over twenty years ago. If you want to talk about material things, knowledge doesn't create material things. Agriculture grows things, mining takes things out of the ground, and manufacturing takes things and makes them higher value added, okay.

I got in a big argument in the mid 90s on a National Research Council committee. The chair of the committee was a guy from Pratt & Whitney, and he said, oh no, because he questioned what I meant by the meaning of wealth. To him, wealth was cash, okay. Now in this little thing I was kind of talking wealth being objects, okay. If you want to think of knowledge as wealth, the more you know, the more power, influence you might be able to exert. You could throw knowledge in there. But if you think about that, what wise — oh great, the whole thing goes away. Uh, why is manufacturing important? Because manufacturing is one of the few things that creates wealth, and it also creates some of the higher value added jobs in society, at least in the United States, okay, so far as that goes.

So I'm actually going back to what I probably should have done the first day. Dr. Belmar, when we were talking about this course in August and I told him what I wanted to do, he says, well that's not what the syllabus says. He says, the syllabus says I'm supposed to be talking about manufacturing — cuz I was talking about codes and standards — and he says, no, this is just what Tom Eagar is going to talk about. I said, okay, fair enough. So you can argue that I'm just telling stories, but in fact what we've done, we finished up yesterday on what I called — what this whole thing is supposed to be is manufacturing science, okay.

And we're interested in manufacturing not just from a material scientist point of view, but from a mechanical engineer or civil engineer's or a business person's point of view, to a certain extent. Manufacturing science — we've gone through observables and the fundamental quantities and limits to materials properties. We finished up yesterday and we're starting codes and standards. But I was thinking about it yesterday, to describe manufacturing is sort of like the six blind men and the elephant. How many people know the parable of the six blind men and the elephant?

Student: [inaudible response]

You don't know that? You know it, saying you're nodding yes. What is it like? You — right. So this actually is a fable that goes back in the Indian subcontinent, and it's attributed in different cultures to Hinduism, Buddhism, Jainism, Muslim. You know, this is a fable that goes back thousands of years, and there's actually a poem by John Geoffrey Saxe [John Godfrey Saxe], written in the 1870s or something, about the six blind men of the elephant. And if you want to look it up, you can, and I have a copy of it here, but we're not going to read it.

But the six blind men go out to look at an elephant, and one of them touches his side and he says, oh, an elephant is like a wall. Another one touches the tail and says, oh no, an elephant is like a rope. Another one touches the tusk and says, oh no, it's like a spear. Another one does the tusk and says, oh no, it's like a stake. Another one gets the leg and says, oh, it's a tree. And another one gets the ear and says, oh, it's like a fan, okay. So these are the six things on the elephant, and in fact they're all right, but they're all wrong. It's just your perspective.

So manufacturing science is sort of the same thing. I kind of made a list yesterday somewhere, and said if you went over to the Sloan School, they would tell you — one group of faculty would tell you that manufacturing is operations research. How do you manage the flow of goods through a factory, okay. Or another group might tell you it's logistics. How do you move things from here to there? Or they call supply chain management. If you go to mechanical engineering, they'll teach a course on different types of materials processing, and they'll tell you what casting is and what machining is and what extrusion is.

If you go to some other people in electrical engineering, they'll talk about statistical quality control, okay. Another group will talk about quality control and quality assurance and ISO 9000, which we might talk about. If you go up to Harvard Business School, they'll talk about operations management. You know, it's the boss who controls everything, all these other people are just — you want to get rid of all those hourly workers is the general philosophy in Detroit, so far as that goes. So everyone has a different view of manufacturing.

But in fact, one view of manufacturing is the one that the government has, or had when they started the National Bureau of Standards. If you walk into the NIST facility — this is a $700 million government laboratory in Gaithersburg, Maryland — if you walk in there, right above the reception desk in great big letters, it says: "It is therefore the unanimous opinion of your committee that no more essential aid could be given to manufacturing commerce" — which is what the Department of Commerce is supposed to be interested — "the makers of scientific apparatus, the scientific work of government, of schools, colleges and universities, than the by the establishment of the institution proposing this bill" — which is 1900. They were proposing to form the National Bureau of Standards.

And so this course was influenced a fair amount because I was on the committee to review NIST last March. I mean, I taught the course before that, but NIST had reorganized, which is a common thing — when managers don't know what to do, they reorganize. There's a great quote from Petronius Arbiter in 200 BC about, whenever it seems like we're about to make progress, management starts to reorganize us and has the appearance of making progress, but in fact it just creates demoralization and inefficiencies and things like that. Which is typical: a new manager comes in, he thinks he's going to improve things, but he goes and he just reorganizes things, and it's just rearranging the deck chairs on the Titanic.

But in fact what the government can do in manufacturing is, they can essentially create the measurement science that allows people to have a standard by which they can have trade and commerce, okay. So we're supposed to be talking right now — if you look at another outline I gave you on the first day, talking about your presentations — we should be looking at the history of codes and standards. And what's the oldest code you can think of?

Student: Hammurabi.

Hammurabi, very good. And here's one of the most complete Hammurabi codes around, okay. [Tom shows an image or reproduction of the Code of Hammurabi.] The Hammurabi code was about 1750 BC, and there were 282 laws, rules in that case. The Hammurabi code was a set of rules by which society was supposed to interact, and the most famous one is an eye for an eye and a tooth for a tooth, okay. There were other things about, if your house — if you buy a house and it falls down on top of it, kills the person, should be killed, okay. The builder should be killed. So it was sort of an ancient, sort of a vicious code, but nonetheless it was a set of rules. It's not the earliest set of rules that we know of, but it is one of the earliest set of written rules, and it's one that actually is in the Lou [Louvre] in Paris. This was found in 1900 somewhere, and it's written down, and you've got the 282 rules which cover lots of different things, not necessarily just manufacturing, but some of it is manufacturing, certainly trade and commerce.

What's another old set of rules? It's not called a code, it's actually called an oath, taken by — what, go ahead, say it.

Student: Magna Carta.

Magna Carta is old, and that's a set of rules, yes, okay. I was thinking one older than that is the Hippocratic Oath. It's the set of rules that physicians use, and that goes back to about the fifth century BC, okay. You could talk about the Magna Carta, you can talk about lots of things through the ages of different rules and regulations through which society is interacting, okay. The one I gave you the first day of class or sometime last week was the basis of lightning conventional lightning protection systems, which goes through and talks about the history of what happened after Ben Franklin.

By the way — well, I don't have to show you a picture of the Hippocratic Oath. Um, after Ben Franklin did his thing with his kite and his string — and he's lucky he didn't electrocute himself — but nonetheless, in 1752, if I blow that up, I mean, you've got a copy of this. The earliest literature available that proposes protection from lightning starts in 1752 with Ben Franklin, okay. So people started proposing to put lightning rods on tops of houses. And in fact, this thing goes through and talks about it — developed this particular history on the next page, goes through the Iowa fire marshal records, okay. Lightning was burning down a bunch of cow barns, and they were having roast beef every night they had a thunderstorm, okay, or whatever, okay.

So Underwriters Labs came up with a master label, Ontario Lightning Rod Act, eventually they came up with other codes. And the current code for lightning rods and stuff is the National Fire Protection Association 780, Standard for Installation of Lightning Protection Systems, 2011 edition, okay. And the guy who wrote this article down here happens to just happens to be the chairman of this committee, okay, that writes this thing. He's a guy in the government, works for the Army, is interested in lightning protection because they've blown up a lot of ammo dumps when they had thunderstorms.

There's a section in here on marine protection. You know, if you're out there in a sailboat, it's really a problem to go find someplace to hide. I mean, you happen to be the highest thing in the ocean, typically. And so should be — there you go, there's a picture of a, how to protect a sailboat, okay, in the lightning protection system. There's a chapter 7 — that's chapter 10 — chapter 7 in here is a code for protection for structures containing flammable vapors, flammable gases, or liquids that give off flammable vapors. So this is things like a propane tank you might have, you know, stored outside, or an oil tank if you have oil heat. And I'm sure there are examples of these things blowing up.

It turns out, I mentioned the first day, and I passed around examples of traditional black iron pipe — which we're going to talk about a little bit later today — but also passed around some pieces of CSST tubing. And we may talk about this in a little more detail, but I mentioned that this being about a tenth of an inch thick, if it gets hit by lightning, all you do is kind of melt a little surface, but it doesn't perforate it. This, if it gets hit by lightning, creates a hole. And while cleaning out some trash the other day, I found one this that I can show you. [Tom produces a piece of damaged CSST tubing.] This actually is something that's — the problem is settled, but this one has a hole in it. It's easier to see it on the back side, but there's one of the holes that was formed by lightning, and burned down a house in Maryland, okay. Cuz it's just too thin, okay. They didn't bother to read the NFPA 780 twenty years ago when they were developing the standards for that product.

If they had read 780, they would have learned — where did I put 780 — that chapter 7 says that you have — [Tom locates the page] — says metallic structures that electrically continuous, tightly sealed, okay, and have a thickness of 3/16 inch thick or greater shall be considered to be inherently self-protecting. Which means something thick like this, you know, lightning is not going to penetrate something that's 3/16 of an inch thick. Well, the negative of that is something that's not 3/16 of an inch thick needs to be protected, okay. So if you read this one way, it says something thick is safe. Well, by the converse of that, something that's thin is not as safe and should have a lightning protection system.

One of the arguments of the manufacturers of this stuff: it's more expensive to buy than the black iron pipe per foot, but you can install — you can build a whole house and install all of it in one day with this stuff because it's flexible and there's fewer joints. It takes three days to pipe a typical house with this stuff, so it's big labor savings. They say, well, the labor savings are only there if you don't have to also buy a $5,000 lightning protection system, okay. That sort of kills the savings. So you have to look at the whole system of these things.

And one of the defenses of these people, who have put a billion feet of this out there at homes and every time there's a lightning storm it's just, you know, more homes are going up with each lightning storm — they said, well, we met all the standards. Well, just because you met a standard, that's the minimum requirement. And if you say you met all the standards, what do you mean by all the standards? There's more than one standard. And standards call out other standards, which is one of the things that I want to go through. And I've lost my — not that it matters — the thing that talks about who writes codes and what's the history of codes.

I want to get into some of the engineering codes now, as opposed to the Hammurabi's code and the Hippocratic Oath and Magna Carta and stuff. A lot of the codes that we have go back to the beginnings in the industrial revolution, okay. The first research contract that was ever given by the federal government was given in the 1830s. It was given to the Franklin Institute in Pennsylvania, in Philadelphia, Pennsylvania, to figure out why steamships on the Ohio River were every now and then blowing up, exploding and killing a bunch of people. Congress wasn't happy, and so they gave some money to the Franklin Institute to study why these things were blowing up.

Well, it turns out the boilers were blowing up. These were basically cast iron boilers that might be riveted together, and they would be a fire box where you throw some wood in, and you'd have a chamber where you boil the water to make the steam to help turn paddle wheels and things like that, okay. And every now and then they would blow up. Well, we didn't have fracture mechanics, we didn't really have a real understanding of the quality of cast iron or steels because metallography hadn't been invented. We didn't really understand anything about stress concentrations and what a hole does in a plate in terms of creating a crack and stuff.

Before that, it was everything was just sort of historically built up. They had built one of these things before, whether it's a bridge or a building, and they made it this big, and someone might next time try to make it a little smaller to save a little money. And as long as it didn't fall down or break or blow up, the next time they make it a little smaller and a little smaller and a little smaller, and eventually the thing would fail, and people say, oh, made it too small, okay, too thin. And so after a while, people started saying, well, you can't go thinner than X, okay. And that's sort of the beginning of codes, okay.

One of the most important codes in the world today is the Boiler and Pressure Vessel Code, which I think in 2005 celebrated its 100th anniversary. It's the American Society of Mechanical Engineers Boiler and Pressure Vessel Code. It is used almost throughout the world, okay. But it got its real growth actually a few years after it started, right down here in Brockton, Massachusetts. And there was a shoe factory in Brockton that looked like this one day, [Tom shows before photo] and the next day it looked like this. [Tom shows after photo] Okay. It's the best picture I got. I used to have two copies of this book, they've both been borrowed on the no return plan.

In any case, the boiler blew up, and it just leveled this block in Brockton. And so this really got some legislators interested, and the American Society of Mechanical Engineers really pushed to strengthen the Boiler and Pressure Vessel Code. Today the Boiler and Pressure Vessel Code is about ten volumes or so that are this thick. This is one volume, this an old volume that I have, 1989 Section 8. Section 8 is for pressure vessels, section three is for boilers — oh no, I'm sorry, 3 is for nuclear reactors, section two is materials, section one I think is boilers. Anyway, boilers are fired pressure vessels, and an unfired pressure vessel doesn't have heat added to it.

And they have their own little stamp that goes on the vessel. In the old days, they actually stamped the steel vessel, but then they learned that created stress concentrations, probably had some failures. So now the stamp is actually something that's affixed to the vessel, and it has a little particular shape that many people can recognize all the way across the room from the little shape. Some vessels can't even be — you can't even affix it because they're inside a furnace. You know, it's heated up in a furnace, in which case the stamp consists of a piece of paper, certificate that says that an approved pressure vessel manufacturer made this vessel, this is the date, these are the design properties, and it gives you a one-page thing telling you what this thing is supposed to be used for.

Now so who writes the codes? In that case it's the American Society of Mechanical Engineers, which is a professional society. Why do they write it? They started out writing it for service: to give a history of what blew up and what didn't, and what are the design rules for building vessels. These are — that section, this is division one of Section 8, there's a division two and a division three of similar size. This one has all the design rules. If you're going to design a pressure vessel and it comes under a certain scope of the vessel, you must in most states of the United States and many other parts of the world build it to the ASME code. And here's the rules for building a pressure vessel, okay.

So that service has changed over the years to a new motivation: of profit, okay. In the last twenty years — when I bought this set back in 1989, I was probably paying a couple hundred for this set, for this book — and it had — it comes out every three years as a new addition, and there's addenda every six months, and you buy the addenda, they go with it for the first three years. So when you pay $200, you're going to get updates for three years, like buying a new car under warranty, okay. After that, you've got to buy a new code.

Well, somewhere along the line in the 1990s, a lot of the professional societies realized no one in industry can even build a pressure vessel without having a copy of the current code. And that means this is a necessary business expense, and rather than charging a $300 we should charge them $1,000 and we can make a profit off this, because — well, anyway. And so in the last twenty years, I would say most codes and standards developed by industrial organizations have gone up a factor of five or so in price. And things like the American Welding Society Structural Welding Code for steel used to come out every two or three years; in the late 90s they decided to make it come out every year. Why? You can charge — you get profit every year, you don't have to wait two years to sell a whole new set, right.

This code, which has been around for years and years, basically regulates most bridges and buildings. [Tom holds up the Structural Welding Code volume] Oh yeah, here's my example of — sometimes I bring in an old code of the structural welding code. [Tom holds up two editions side by side] And the of the structural welding code, this would have been about the size in 1988, this is 2010, okay. Yeah, it's changing. What's changing? Well, they've added addenda. Half of this now is not the code itself. It's actually not that much thicker, but all the explanation is more than half the code now, okay. So it's got annexes and appendices because these committees can add to it.

And there's — it's all these codes, or many of these codes, are a wealth of knowledge, okay. They carry the corporate history of the industry and many of the failures that have occurred. There's a book — I forgot to bring it with me — called To Engineer is Human. There's a guy, a professor down at Duke University in civil engineering, Henry Petroski. He was elected to the National Academy of Engineering at the same time I was, we're the same class of '97. And he got in because he wrote this book, To Engineer is Human. And the theory of the book is all progress in engineering is a result of mistakes, okay.

We build something, it blows up, it breaks, and we go in and we say why, and we figure out why. You know, there's a famous picture of Galloping Gertie, that bridge they teach in freshman, when they talk about freshman physics, when they talk about waves. And you see Galloping Gertie — they made that too thin, it was too slender, and so when the winds came, it wasn't stiff enough, and it would start to buckle and the waves got into resonance. And now, how many casualties in Galloping Gertie, anybody remember?

Student: One dog.

One dog. There was a guy on the bridge in his car, and he and his dog got out of the car. He was smart enough to head across the bridge and got across in time. The dog was just confused and ran around kind of in circles, and when the bridge went, now the dog went out. But nonetheless, they revised all the bridge codes for how you design bridges. And so Petroski kind of became famous and is sort of a spokesman in engineering for the whole idea of, we learn from our mistakes, okay.

And so the codes are extremely valuable because they don't necessarily tell us about our mistakes, although there is a code that tells us about our mistake. Some of you might have been around here long enough — remember on Super Bowl Sunday about four years ago, there was a big explosion down in Middletown, Connecticut. Anybody remember that? They were blowing gas out of a big new co-generation facility — not co-generation facility — a new gas-fired utility. And when they build all the piping, people leave old packs of cigarettes in the piping, or bolts and things. And they have to, if they — basically started to run this thing, and those things got into the gas turbines, they'd wipe out a $5 million gas turbine. If you start throwing bolts, you know, into the turbine right while it's operating.

So they have to blow out the lines, and they blow them out with a tremendous blast of gas. Well, if you've got, you know, 16-inch diameter pipes, where do you get a blast of gas sufficient to come out at near sonic velocities, 500 mph, to carry all that trash out of the pipes in this big facility you just built? Well, they had a 30-inch pipeline at 900 PSI that came up from Texas that had all the gas. And so the tradition was to blow the gas through and just let it blow into the air. And they actually did it successfully — they've been doing it for years successfully.

The facility was something like a $50 million or $100 million facility, but it was built with alcoves. There were three alcoves as I remember. I've only been there once or twice, but anyway. And the first time, they had a little 6-inch pipe that was going to exhaust the gas straight up in the air on this side, where there a bunch of trees, okay. They successfully blew that for about half an hour or so, until they blew all the bolts out into the trees.

The next one, the gas pipe exit was down low. This was about a 10-story building, okay, in this alcove which is sort of protected. And the 6-inch pipe was right here, and this alcove basically collected enough gas that when it found an ignition source — and no one's quite sure what the ignition source is, but it went boom, and they blew up this building. And the turbine buildings were over here, wiped out part of these, did $50 million of damage. The governor got all upset. The company that was designing it and building it wasn't all that happy. Six people died.

Why? They were doing this on a Sunday morning, but because they were behind schedule, they had workmen all scattered through here when they're doing this type of blowing. You shouldn't have anybody on site except the necessary people, and they should be in an area far enough removed that you're not going to hurt anybody. Some of the guys here had left and gone home because they were getting sick from so much gas they were breathing before the explosion. So it wasn't as if people shouldn't have known that there was a lot of gas around.

Well, because of this particular incident, General Electric, who build some of the turbines and had essentially gone along with this type of stuff, they came in: oh, no one should ever be doing this. Even though they've been doing this for thirty years, okay, at other facilities. But if you now read NFPA 56, I think it is, which is the code — there's an NFPA code that looks sort of like this, instead of 780 it's, I think it's 56 — and the introduction will actually say, it's the 2012 edition if I remember, it says in 2010 there was an incident in Middletown, Connecticut. It actually tells you what caused the big rewrite of the cleanout procedures for these pipes, okay.

There are thousands and thousands of codes out there, and they grew up historically. And some of them, whether it's lightning rods and Ben Franklin or whether it's the cleanout of pipes code, almost every one of them has sort of an interesting history, okay. And it usually goes back to some interesting failure.

But the codes, just like the one that I handed out before, okay — this one, handed it out yesterday — A106. I said we'd talk about, because I discussed a problem that they had at an oil refinery, where they bought a bunch of pipe at a time when it was hard to get pipe, and they built things. They bought it to A106. It was going into a refinery. If it's going into a refinery, the people building the refinery are going to purchase the refinery from the contractors that are going to build it. And this might be a $5 billion refinery. And so it's purchased by, in this case, a bunch of investors. It wasn't a big existing oil company building this refinery — could have been an Exxon. In many times a new refinery might be built by a consortium of oil companies. Shell and Exxon might get together in a partnership. This was actually being built by a bunch of financial investors who didn't know much about things.

But a lot of the piping would come under ASME B31.1, and this is the 2004 edition of B31.1, which is a companion to the Boiler and Pressure Vessel Code. The Boiler and Pressure Vessel Code is only for the vessels, the big reactors and stuff. The piping that's attached to them has its own code, okay. So they had — it was being built and purchased to both B31.1 and A106. Two different sets of codes. This is how you manufacture the pipe, that's how you install the pipe. Two different organizations, hopefully they don't conflict, okay.

So A106 — it turns out it will give you a scope, and we talked about that. This specification covers seamless carbon pipe, and I pointed out yesterday, an eighth of an inch to 48 inches diameter. That's a huge range for this scope. And it will call out other standards over here. This is, um, specialized carbon and alloy steel pipe. This is ultrasonic examination, eddy current examination, etching, flux leakage, which is magnetic particle examination, ANSI standards B36.1, okay, which is basically dimensions of the pipe as I remember, military standards for steel products, federal standards, other standards. They didn't mention B31.1 on installation, because this is actually — these are all the other things that tell you about, if you're going to buy a piece of pipe, here's something that a buyer can use and tell the seller, I want you to meet this specification.

And you've got a copy of this specification. It'll give dimensions, it gives compositions, tensile strength. And over here it will tell you about the hydrostatic test. This is the standard test for quality control. This is supposed to be good quality piping. It's not what I'm going to put in my basement, okay. I don't need to do a hydrostatic test in my basement. The chances of getting more than 100 PSI in my boiler are pretty slim, okay. This thing would probably take 4,000 PSI hydrostatic pressure, okay, so a factor of forty safety. I don't need to do special quality control for my basement. But when I'm talking a 20-inch pipe in a refinery, and the consequences are not blowing up one house but blowing up a half-billion-dollar refinery, well, maybe you do want to spend a little more effort in doing quality control.

Well, the quality control for hydrostatic testing is, each length of pipe — every single one sold — shall withstand without leakage through the pipe wall a hydrostatic test, except as — and it gives you the exclusions, and how to stamp it so that someone looking at the pipe can tell whether it's hydrostatically tested or not. But it says, when allowed by B31.1.2 [311.32], each pipe shall be tested by a non-destructive electric test. It's not always convenient to do this hydrostatic test. Every steel pipe mill in the world has a hydrostatic test facility, because this is the way it's usually sold. But in fact, if for some reason you want to upgrade some pipe that wasn't tested at the facility, you could do other tests. You can do non-destructive electric test.

And what does that mean? You keep reading on and you find out that means you can do eddy current testing, okay, or ultrasonic, eddy current, flux leakage — which is magnetic particle — three different electromagnetic test procedures to try to find cracks or defects. If you read the next page — I don't think it's on this page, oh here it is down here — all of them say you can't have a defect more than 12.5 percent of the wall thickness. So if it's a — let's say it's a 10-inch pipe and it's got a half-inch wall — here's your half-inch wall — 12.5 percent of that is 1/16 of an inch, if it's a half-inch wall.

And so as long as you don't have a flaw more than 1/16 of an inch, you're fine if you do the non-destructive test. If you do the hydrostatic test, this pipe might take 2,500 PSI to break it, but you may only test it at 1,250. What does that mean? That means you probably could pass the pipe if it had a flaw about halfway through. So what's the standard that you're buying it to? Well, the most common standard is to do the hydrostatic test. That's what 99 percent of all the pipe in the world is sold to refineries based on, the hydrostatic test.

You could have a flaw in that thing that's halfway through the thickness, and it can pass the hydrostatic test, it can go into service. And because you probably have, under the installation code, a factor of safety of about three or four, you'd have to be 75 percent of the way through in order to have the thing fail in service — unless you had a tremendous amount of corrosion in service. So it's okay. Not only that, most of these pipes, if you study fracture mechanics, will do a leak before break. So you get a small leak, you don't get a big explosion usually, okay, hopefully. And that is true. That's why people have learned a hydrostatic test on every single piece of pipe is good enough. It's fast enough and it's good enough.

If you decide later you want to go and check things — and that's exactly what happened in this refinery. They had — they found one failure, they were doing a test after they had cut it up and put it in service, and they found they had a little leak on their test. They did a study and found, oh, this leak went all the way through, it shouldn't have ever gotten out of the steel mill. Well, how did it get out of the steel mill? Well, sometimes people make mistakes. You know, sometimes a piece of pipe that's supposed to go into the reject pile gets put onto the ship pile. Things happen, okay. One bad pipe out of 600 doesn't make the other 599 bad.

But that's essentially the position that the refinery took, and they cut out all the rest of the pipe. And then they needed to justify doing this, so they hired a consultant out of, couple thousand miles away, who supposedly was a piping expert. And he didn't do the hydrostatic test on the pipes; he did the non-destructive electric test. And so now he was holding the pipes to a higher standard than they had been sold to, right. And that's what the whole $60 million dispute is about: did some of them fail this 1/16 of an inch, but they would have passed the hydro if they had that flaw. In fact, I spent a week up there looking at pipe, and we couldn't find a single one that failed even the 1/16th, out of 600 lengths of pipe, okay.

So now you have to go back and say, well, show us your test that failed the 16th, and even if you do show us that, what you bought was not what you're now trying to purchase, claim that you purchased. Because the codes have — you can do this, but if you can't do that for some reason, you can do this. But the two are not necessarily equivalent, okay. And you have to use a little judgment here about what did you buy, what are you entitled to get. If they had asked for a hydrostatic test at 90 percent of yield, you would have found even the 12.5 percent flaws, if they existed, okay. So that's what that dispute is about, is still ongoing, so far as that goes.

But it's a question of how do you read the code, and what does the code mean. Just because it's in writing doesn't mean everybody's going to agree on the interpretation. So in fact codes have interpretations. And this is something else the professional societies can sell, okay. This is the official book, a book of interpretations of the structural welding code. If you don't understand something in the structural welding code or the ASME Boiler and Pressure Vessel Code, you can write a letter to the code committee, and in about six or nine months you'll get an answer, okay. Because the committee has to meet. They're not going to meet next week just because you asked the question.

This is people from around the country and around the world who get together in some out-of-the-way place, like Hilton Head Island, South Carolina, okay, some resort, usually, to have their meeting and play golf. You meet in the morning, you meet in the evening — what it is, you meet in the morning, you play golf in the afternoon, you come back in the evening to have drinks and you meet, okay. I'm not kidding, this is the way these things work. Because it's sort of a voluntary service for most people. This is what they're doing as part of their profession. They get to put it on their resume, they're part of the code committee that writes the codes.

And they'll give you interpretations, and here's interpretations going back for decades. There's not a huge number of interpretations, but for example, here's one. [Tom reads from the book] The question was, this is 1984 or yeah '83, "Is it the intent of the code to prohibit level three NDT individuals from performing NDT testing if they are not qualified as level two?" okay, testers. And the answer from the code committee is yes, okay. That's an interpretation, okay. You don't even know what all that jargon means. You have to be part of the whole business to understand that.

You don't have to understand the jargon, I understand the jargon, but I've been part of this business for thirty some years, right. And I don't want to — actually I may talk, I think I do under the section on non-destructive testing, I talk about level two, level one, level two, and level three inspectors, which is very confusing things anyway. But others, they will give you lengthier things, and they'll refer you back to — this is 1990, someone asked the question, "Does the D1.1 place a minimum thickness of half-inch on materials welded within the scope?" and they say, okay read this section. Okay, here's one, they actually give you a drawing. They will give you some interpretation, and typically a lot of those interpretations will be clarified in the next edition of the code, okay, because obviously it created some question. Most people are not just cranks writing questions; they've studied the code, and they've got some question about, it wasn't clear, okay.

So who writes codes? Industry writes codes, and they started out doing it as a service. They now do it both as a service and to make a profit, okay. Codes almost always get thicker in time, because things are added to them. I have an example for you of the opposite. [Tom holds up two editions side by side] This is the 1997 structural welding code for aluminum, this is the 2003. Which one's thicker? Well, they're almost the same thickness. But in fact, the '97 was thicker than the 2003. Why? Well, in '97, chapter 2 was on how to design a welded structure out of aluminum. In 2003, that chapter got decimated in size, because the Aluminum Association came out with the Aluminum Design Manual. So this is actually what happened, and this is thicker than this, okay.

So now this code references this for design. It used to be, design was included in here, okay. So the societies are trying to work together. When one of them comes out with an improved thing, they work together. Why do they work together? Some of the people are the same people on the committees, okay, so far as that goes.

Now some codes don't increase in price, okay. If you want a 10-volume set — actually it's only nine, but the aluminum chapter is actually two volume, so it really is 10 volume — so I got this whole code. This is the Metallic Materials Properties Development and Standardization, called MMPDS. It's got the Federal Aviation Administration up here, it's MMPDS-05. These are the guidelines from April 2010. It used to be called MIL-STD-5 or MIL-Handbook-5, and it started out — this is the introduction, and in the introduction it will tell you that "since many aerospace companies manufacture both commercial and military products, the standardization of metallic materials designed which are acceptable to the government procuring or certification agencies is beneficial to those manufacturers as well as government agencies."

So what happened forty years ago is the Federal Aviation Administration, NASA, and the US Department of Defense got together and created MIL-Handbook-5, which I used to have a copy of. But when this came out, I purchased the hard copy for $115, all ten volumes, cuz that's the cost to the government of printing it. The government can't make money off standards. Private industry can, okay. So you can download this for free if you want a digital copy, okay.

And what it's got in it is — if you look at the aluminum chap — this is just for aluminum. First volume steel, next volume is — next two volumes, chapter 3, are aluminum. It's all kind of data on the strength properties, the modulus, tensile strength, the yield strength as a function of temperature, creep strength, fatigue strength of all kinds of aluminum alloys that are used in building aircraft, military or commercial. This is the standard by which you should design. And in fact, if the government says we want to buy a new aircraft, okay, they will have in the contract that it will be designed using the properties in MMPDS-5, used to be the old MIL-Handbook-5, okay.

Student: Is the government generating it?

Sometimes the government labs generate it. Sometimes — most of the time it's industry data. Boeing generated it, McDonnell Douglas generated, Alcoa generated it, okay. You can find some of the same data in handbooks written by Alcoa, written by the Aluminum Association. People who want to promote the sale and use of aluminum will generate the data, and they will make much of it if they don't see that they could hold it proprietary. Some of it was done by Boeing under government contract, in which case it is by definition part of the public knowledge. And so the DoD says, okay Boeing, we're going to put together this handbook, give us all the data that you've collected, that we paid for when we developed, you know, the X-15 or something like that. They — Bell developed the X-15, but anyway, you know, some — in some previous life, the government bought all that data.

In other cases, Alcoa is donating the data. Alcoa's got a new alloy — aluminum lithium alloys twenty-five years ago were new and different alloys, and they wanted to sell them. Can't sell them unless you can get them into the MIL handbook, right? So Alcoa generates the data and gives it to the committee, and the committee will evaluate it, make sure they accept it, and then it will go into the handbook, and now it's legal to use. Okay, legal to use, which is something else we want to talk about.

There is fair use of these things, okay. Things in codes and handbooks by almost by definition are available to anybody in the world who wants to use them. You don't have to own the handbook, you can go to a library and look at the data. You can — this Boiler and Pressure Vessel Code, for example — not this volume, but section nine is on welding procedures, okay. How to weld up steel or aluminum pressure vessels, and how do you qualify a material or a person, cuz you can do a procedure qualification record, which is qualifying the material and how you would weld that material. You can also do a welder qualification, which is qualifying a person to show that they have sufficient skill to be able to lay down a good weld bead. It's all there in section 9.

The scope of section of the Boiler and Pressure Vessel Code is going to be laid out in excruciating detail within Section 8 or Section 3 or whatever. For example, here is the scope of Section 8. [Tom locates the page] It'll all be laid out. Here's the scope for this division — this is actually division 8 — you know, you're reading it, it's the first paragraph in the whole thing, just like it was the first paragraph in the ASM spec. What does this standard cover? What does this code cover? And it says it covers containers for containment of pressure, either internal or external, which means it could be an external pressure vessel. And it goes through and it says the following classes of vessels are not considered within the scope of this division, and it goes on for a page and a full page of what's excluded.

Well, does it have to be excluded, or could I use it even for some other vessel that's not officially in the code scope? It's a good manufacturing practice. And in fact, I use ASME Section 9 all the time. Someone comes to me and says, we got this critical structure, we want you to write us a welding procedure. And I say, you know, I actually call out the code rather than writing a manual this thick for the people. I just call out the code and say, you shall weld it to ASME Section 9. Or if it's something a little less critical, I may call out the structural welding code for steel or aluminum. This also has ways to qualify welders.

ASME Boiler and Pressure Vessel Code is the most stringent of any code in the world that I know of for welding, except some nuclear codes. The bridge and building code, the Structural Welding Code of the American Welding Society, is the second most widely used. Bridges and buildings, if they collapse, don't usually explode, okay, they just collapse. They don't blow up whole city blocks. This has very stringent requirements; this has somewhat less stringent but very good manufacturing practice for bridges and buildings. The worst welding codes I know of are for roof girders, okay. We can talk about that.

But we're sort of into today, and I'll give you a case study probably tomorrow on cement trucks, and a pressure vessel that went on a cement truck, that neither the federal government said they worried about and the ASME code had it as an exclusion. So the company decided to build 100,000 of these and didn't use the code, didn't use good practice, and they killed a guy. Because they didn't do what historically was known as good design practice, because they said, oh, we don't come under any of these codes. We really have to get in here, about time.