CS_F2012_08

Um, if not, I've got, from those of you that are here, what it is that you're going to do your modules on. So you can check with Jerry. Actually most of these should be on the website, and if you've got, you know, decent streaming speed. Otherwise you can, give you a DVD of these various things.

Today I wanted to keep going through some of these types of things that we've been talking about. Uh, Simone, you didn't bring back that uh thing from Framingham of the new field, did you? Okay that's fine, okay fine. Uh, so I may show you an actual sample of one of these things, but we talked about who writes codes and why they write codes.

And service and profit, we've talked a little bit about. Fair use, that people can use codes beyond their scope. Their scope is very specific to a particular industry and is limited, because there are other codes that cover things. But that doesn't mean you can't use the good parts of a code, the good design aspects of the code, for other things than they were intended for. And that's done all the time.

Conflicts, which governs. Well, we've talked in the cement truck case about how the federal government always trumps the state governments. And it gets complex about who, whether the federal government or the state government is going to have responsibility for things. In fact that's part of the presidential debates right now. Not necessarily for codes and standards, but you know, should health care be a state mandate or should it be a federally, you know, sponsored program and stuff. But uh, preemption is the part of the constitution that says when there is a conflict between the state law and the federal law, the federal law always takes precedent, okay.

Not all codes have force of law. What governs. Many of the codes are voluntary standards that people have developed over the years. They're developed historically. You can go and ask for interpretations. But I want to talk a little bit about collaborations and competition in codes.

It turns out, I showed you the quote from the National Bureau of Standards, if I can find it, that when they formed the National Bureau of Standards, they said in the report to Congress that there was no more valuable thing that the government could do than to essentially form the National Bureau of Standards, which would essentially write standards for industry and commerce. Remember the National Bureau of Standards is part of the Department of Commerce.

You can aid buying and selling if you have a good set of standards for people to work from. And so there are these codes and standards. There's also something called standard reference materials, which — I apologize, my department chairman came to visit me this morning and so I didn't get everything organized as well as I should.

But actual standard reference materials, there's a whole group of these things. If you look up reference material in Google and then you go to the NIST website, it will talk about reference materials and standards. But the standards they're talking about are not the codes and standards we've been talking about for the last little while. They're talking about little samples. You can get a standard which is a standard composition of a steel that you will use to analyze other steels, okay. And so there's various techniques that you can use, but this will be a steel that's been analyzed very precisely at great expense by the federal government. And you can purchase this reference material. They're not allowed to make a profit on it, and so as a result they actually do less of this than they would like to. But if they could make a profit they could actually use it to fund other standards. But Congress doesn't allow them to make any money off of it, and they'd rather spend their money on other things.

But it's a material sufficiently homogeneous and stable with respect to one or more specified properties. Are any of you, all of you familiar with hardness testing? Okay. So all of you have seen hardness testing in undergraduate class or other places. NIST has the world's most — what? Oh — most precise hardness tester. It's about half the size of this room, and it's got all kinds of instrumentation and bells and whistles, because they do a lot of research on hardness standards. They actually provide some, what they call certified standard reference materials.

There are reference materials, which is something that's homogeneous that you can use to measure properties in different ways. There are certified reference materials where someone has characterized the measurement as a valid procedure, and these are things like the STM standards. There's a reference material certificate — I showed you that with the gauge blocks that I had — a little certificate told you what the tolerances were on that. There's a NIST standard reference material which is a registered trademark, it's an SRM, okay. And this means that NIST actually manufactures these things, but they can't make a profit on it.

It used to be, for example, I didn't bring one with me, but a Charpy bar is a little notched impact bar. That's a piece of steel that you hit with a hammer, a calibrated hammer. And originally, when they started doing that around 1900 or so, those were calibrated over here at Watertown Arsenal, which is now Watertown Mall, okay. But it used to be an Army arsenal, and they did that for years because the Army was interested in steel and the impact properties of steel.

Well, about twenty years ago, maybe it's thirty years ago, forty years ago, probably about the time of the Mansfield Amendment in '72 — Mike Mansfield was a senator and he put in an amendment to a budget appropriation that the Defense Department could not fund any research on anything that didn't have defense needs, okay. And so all of a sudden the Defense Department was doing lots of services for the country because they had an interest in it, but it wasn't central to their mission, so they started sloughing it off. Turns out NIST took over the [Charpy] bar standards.

And the [Charpy] bar standards, basically it's a set of pieces of steel that have been characterized to be extremely homogeneous, heat treated just perfectly. And you put it in your machine and you break it and you see how the hammer swings, and if it swings within a couple of foot pounds of the right number, you now have qualified your machine. It costs like five thousand dollars for a set of these things to calibrate your machine. But that's actually what it costs NIST to put these things together.

And it turns out technically, if you stop the machine — if you have a steel that's so tough that the hammer swings and it just goes thud and it stops the machine — you have to recalibrate the machine. Now, when I worked for Bethlehem Steel as a research engineer, I was working on some very high toughness steels and I used to stop the machine about four times a day. They didn't like that over at the test lab. Officially they were supposed to recalibrate every time that I stopped the machine, but they didn't do that. But in any case.

So NIST provides chemical analysis samples, they provide samples of standard urine. Well why do they do that? Because people do blood tests for drugs and urine, okay. And so they will take urine and they will add some drugs to it at certain levels, and so this is what your standard is if you're trying to measure how much drug is in someone's urine or something. So they have all kinds of standards. Those standards are not the types of standards that we're talking about, okay. Those are standard reference materials. And if you go down to NIST and you're talking standards, they will generally be talking about the types of standards we're talking about, things like the ASTM standards or whatnot, and they will call the standard reference materials SRMs. Just be aware of two names for this, for different things, or the same name for two very different things, but related to similar types of things.

So it's a valuable thing for the government to do this. Why is it valuable? Well, it turns out that, for example, if you're going to build a pressure vessel, or you want to, if you're in Europe and you are — or let's say, give a more practical, you're in Japan, and you've got some of the best steel technology in the world, and you want to sell a pressure vessel in the United States, it's going to by law have to meet the ASME code and have an ASME code stamp, which means the Japanese have to build it to the ASME code, okay, if they want to sell it in this country. They could have their own rules in Japan. It turns out Japan historically has just sort of adopted the ASME code, and if you build a pressure vessel in Japan it probably by law has to meet the ASME code. The ASME code is really an international code.

That's not true in China. If they want to sell a pressure vessel in the United States, and they do sell pressure vessels in the United States, they have to build to ASME code. But if they want to build one for China, doesn't have to meet the ASME code, they can write their own laws, okay. China doesn't like to spend fifteen thousand dollars for a copy of the ASME code. They have to do it in some cases, but it is a competitive advantage for American companies who are doing, let's say, eighty percent of business in the United States. They have to build things to U.S. codes, and so this is part of their overhead. But if you're in Europe, you may have to have everything that the American codes, and all the German codes which are the DIN codes. Not every American company is going to fool around with the DIN codes unless they want to do business in Germany, okay.

So the ASME code's a great example. Another great example is the American Petroleum Institute, okay. American Petroleum Institute has lots of codes. This is the specification for line pipe. What's line pipe? It's the forty inch or forty-eight inch or thirty-six inch pipe that you bury in the ground and you transport oil. The Alaskan line pipe coming from Prudhoe Bay down to Anchorage is API 5L, okay. There's the 5L designation. There's other pipe that — other pipe is API specification 5.

Not all of the world's big oil companies are U.S. companies. What's the largest oil company? The largest two I can think of that are not U.S. — actually largest three I guess I can think of that are not the largest in the world — the largest in the world is Exxon Mobil, and then there's uh, Amoco, you know, that are U.S. companies. But there are others. The Russian oil company, which I think may be the largest in the world now. Or actually Saudi Aramco may be the largest, okay. But it's actually an American — partially American — it used to be. Yeah, but Aramco is the American Arabian Oil Company, okay, and people have kind of lost that over the years. Originally the Saudis, when they discovered oil, they weren't the ones doing the hole — it was some American companies doing worldwide exploration, and the Saudis and the Arabian American Oil Company got together, and now we just call it Saudi Aramco. But in fact it originally was a partnership. I think it's fully owned by the Saudis now, probably.

Yeah. His dad, his grandfather works in the oil business, so. Um, no, I agree with you, I think it is wholly owned by the Saudis, but originally it was a partnership with an American company. The Venezuelan oil company was all American owned until whoever it was nationalized it, okay, and took it away. British Petroleum is huge, Shell Oil is huge, the Russians are now huge, Saudis huge. They're big oil companies. Total, yeah. It's Total in English, it's Total in French. Um, and there's an Indian — uh, not Indian — uh, Italian one, okay.

In any case, there are big oil companies around the world. But it turns out there are a lot of Gulf, uh, you know, Mobil, Exxon, Amoco — anyway, a lot of big American oil companies. And the big oil business, John D. Rockefeller and things like that, they were mostly American firms. And so the American Petroleum Institute controls most of the world's oil business, even in places like China to a certain extent. And their specifications control lots of these things.

The welding specification is API 1104, and it's similar to the ASME Section 9 in many ways, but it's written specifically for pipes, okay. But a lot of the ways that you'd inspect and do x-ray and ultrasonic and everything else are very similar to what you'd see in the structural welding code or the ASME code, but there are differences.

Um, there's, within the oil business, if you're going to be welding together a pipeline anywhere in the world, there's a union in Tulsa, Oklahoma, and they will get the contract to do the field welding of that pipeline in most places, except maybe China or the Soviet Union, but sometimes in those places. Because if it's being built with — by Shell or Exxon — will be told by the union, you will not see anything built anywhere in the world unless you use our union for this new job that's coming up in wherever, okay.

That's not true in Iran anymore. It used to be that they'd be American welders from Tulsa, Oklahoma, who would spend six months in Iran putting together an oil pipeline in Iran. That's no longer true because we don't have great relations with the Iranians, because of world politics. But if you went back fifty years ago, every pipeline in the world was built, except in the former Soviet Union, was built by this Tulsa Oklahoma union.

When the Alaskan pipeline was built in the mid '70s, they had the technology to do automatic welding, but they didn't. And the reason was, the union said, if you try to bring a robot onto that job you'll never see that pipeline built. This was not written into a contract, this was something that was explained in a conference room to the CEO of the oil company, that it's not going to work, we will sabotage it, okay. This is the way business is actually done on an international basis, okay. So there are unwritten standards, like you must use those Oklahoma welders' union. It's a very powerful union. It's not as powerful as it used to be, but for example they had certain things — you had to feed them American food wherever they were in the world. Because these guys are working like sixteen-hour days in some culture, and where the water is not good you have to fly in food so that they don't get sick and can keep working efficiently. It's part of the union contract, okay.

I'm just telling you some of these things because this is the way the world works. And they found it's actually worth it. There's other things. One of the things that's good about the system, it's not written into the API code, but another part of the unwritten standards is, if a welder produces a weld, one of these circumferential welds in the field — and they will do one hundred percent x-ray inspection on most of these things — and it fails the x-ray inspection, he gets to repair it on his own time. I mean, he might be making a hundred bucks an hour when he's working on company time making the weld the first time, but if he doesn't do a good job, the repair is at his cost, okay. And so these guys really do good work, okay. They really are quite proficient.

Um, Matt's back there nodding his head, because he actually used to be a roustabout on the North Slope of Alaska, okay. You want to tell us anything else a little bit about that?

Student: Prime rib every Sunday, in all of the Alaska, anywhere you go on any rig.

So, yeah. Well, you know, another place where you get excellent food is nuclear reactors. Not new nuclear reactors, but nuclear submarines — they get the best food in the Navy, okay. Anyway, so these are not standards, but these are de facto standards if you will.

But there is — I was going to get into — there is a competition, and the country that controls the standards has the companies, have a significant competitive advantage in doing international business. And because of that, and the United States had not a monopoly, but they had a dominant position in many industries and still do, okay. Around the world, API, ASME, many, many others. The American National Standards Institute is an institute that collects standards from a bunch of other societies and will name them an American National Standard after their procedure. They don't develop the standards, but they will essentially list the standards and make them available for sale.

But it's a way to promote American business and American standards as the standard for doing commerce internationally. Well, if you're a European, you wouldn't be all that pleased with that, right? And if you're Japanese, you wouldn't be that pleased with it. The Japanese have JIS, Japanese Industrial Standard I think is what it stands for. Of course they use English letters to define it, because no one would understand the kanji. But nonetheless, the German standards are DIN, and I can't remember what the French standards are called, but the problem in Europe is, you had a lot of small countries who couldn't dominate the standards. So they got together and formed the International Standards Organization, ISO. And those guys started trying to compete with the American standards. So there's a competition, okay. Uh, with ISO standards and ANSI and things.

Um, what's the — does anyone name an ISO standard? Nine thousand, that's the one people usually get. What is ISO 9000?

No, it's not machining. It turns out it's standard procedures. And it's a paperwork standard. Under ISO 9000 you must document everything you do, okay. It's almost gotten to the point where if you want to go to the restroom you have to document the path to the restroom, okay. It's not quite that bad, but almost.

Exactly. You have to have these meetings, you have to have every procedure. I mentioned the other day, over at a Raytheon plant, the Navy was building something for the Aegis missile system, and the Navy inspector walked through, and a guy was putting a torch to a piece of heat treated aluminum. He says, where's your procedure on that? There was no written procedure. It shut down the plant for three days. Under ISO 9000 you must have a written procedure for everything in your manufacturing process.

So ISO 9000 came out sometime in the 1980s. It was part of the European Economic Community's attempt to fight this imperialistic American monopolistic view of standards, okay, that we had. Um, they came up with ISO 9000, and by 1990 Ford Motor Company said — well, actually the automotive companies have what they call tier one suppliers and tier two suppliers. Tier one supplier, if you're making body frames for Ford or General Motors, you're a tier one supplier, or you're selling them batteries, you're a tier one supplier. A tier two supplier is someone who supplies to the tier one suppliers, okay. So if you're making welding wire to weld the frames, you might be a tier two supplier.

But back in 1990 Ford said, we will not purchase from any tier one supplier that's not ISO 9000 certified. Ford's an international company, and they were pushing on their quality control in manufacturing, and that meant you could not do something unless you had documented procedures for how you did it. And this was the strongest ISO thing, and actually put ISO — I think ISO 9000 sort of put ISO on the map. In terms of, many American companies were scrambling twenty years ago to get ISO 9000 certification, because all of a sudden they couldn't deal with Ford, a major U.S. manufacturer, unless they had this ISO certification.

So, and eventually right now almost everybody is ISO, you know, except probably China. But China has to do it because they want to do business internationally. But yes, ISO 9000 has become the de facto quality control documentation standard of the world. There are books written on it, and there's ISO 9001 — and I'm not sure I can tell you what the difference is — there's ISO 9002. There's a whole series of these. But you must have documentation of what this alloy is, what the limits of composition on this alloy, everything. Anything you're using that's important to your process has to be documented. Maybe not the trip to the restroom, but anything you do has to be documented. And this is all part of manufacturing quality control.

So, getting back, this is supposed to be a manufacturing science course. Well, codes and standards are actually part of that manufacturing. I told you, you can't design anything — you have to design it, you're limited by the standards. You can't manufacture anything any way you want unless you do it by the standards. Yeah?

Student: [inaudible question about cadmium/Sweden]

Oh, okay. Well, yeah, I do know something about getting the cadmium out in places like Sweden. It's a very environmentally conscious country, as is Germany. And what you're probably telling me is there's some standard that they have for — forgetting — oh, okay. I don't know, but I'm sure it's some standard that they have for environmental things.

I mean, in part of the welding course I talk about cadmium and brazing and things. And one time the Swedes outlawed cadmium in light switches. Every light switch — if I go flip the switch, I just created some cadmium vapor in this room, because that switch is a silver cadmium electrical contact. And if you don't put the cadmium oxide in there it can only carry seven amps. That's a fifteen amp switch. The reason it can tolerate fifteen amps rather than seven amps is because the cadmium oxide suppresses the arc when you shut the thing off. And uh, you get some inductive effects and we don't have to go through what happens, but you could potentially get a little arc. If the arc lasts long enough you can melt the switches, you can melt the contacts together, and so you try to turn the switch off and it just welds itself together and it stays on and it starts fires.

Well, the Swedes about thirty years ago decided cadmium was such a terrible element they would just outlaw it, and would not be allowed in Sweden. And they, within years they were burning down houses at such a rate that they decided they would relax that standard for electrical switches, okay. So you have to be careful.

A number of years ago the Norwegians decided they wanted to outlaw chlorine. And Bob Rose, my thesis advisor, came in with a periodic table where he had Xed out chlorine, okay, gave it to me, taped it on the back of my door of my office, because he thought it was so funny that they thought they could outlaw an element of the periodic table.

Well, Congress once almost outlawed oxygen, okay. They were trying to pass a law saying that anything that destroys DNA was going to be illegal. And one of the most potent things to destroy DNA is oxygen. It oxidizes the polymer, okay, just like other polymers oxidize. But high oxygen concentrations will destroy DNA. And Congress was going to write this law that said anything that destroys DNA will be outlawed. And they didn't realize until someone finally stopped them that they were going to outlaw oxygen. That would be really effective law, wouldn't it. Um, we wouldn't have to worry about cancer anymore.

That's right, everybody'd be dead, okay. Now, this is probably the problem when politicians take over science. But in any case, there are, at the same time, good collaborations among the standard writing bodies internationally, but there is also a healthy competition that's going on, and you should be aware of that.

And it all gets down to competitive advantage. The collaborations — what — we talked about the collaborations early on when I talked about limits of precision. The French have the standard kilogram, okay. The British National Laboratory and NIST and the French laboratory are the three places in the world that really are trying to push the limits of the seven standard fundamental constants — mass, length, time, coulomb, and all these other things. That is something that knows no international boundaries, okay. Having an international standard of length is valuable to everyone in the world. We are paying for that, along with the British and the French. No one else is collaborating with us on that, to a certain extent. We're doing it because we've been doing it, we've been a rich country, but to a certain extent we want to call the shots, okay. And we don't want to get this thing into a political thing where everybody starts debating, or competing to have their definition of the meter, okay. It's sort of dysfunctional. So there are extensive collaborations and extensive competition in international standards.

Uh, we've talked about the cost ad nauseam. A purchase order — what you get — you should get what you pay for, no more. And I gave you an example of this I think before, but I'm going to actually show you, if I can find it, a standard. Oops, I may have to show it. I should — I had a copy of it, but anyway. Um, I'm sure I have had a copy, but nonetheless, it should be in here. Oops, that's inspection. Oh, it should be another inspection, okay.

So this is the structural welding code. The structural welding code, okay, AWS D1.1, used throughout the United States, Canada, Mexico, probably in a number of other countries for bridges and buildings. But some other countries have their own definitions. And this is under the chapter Inspection, General Requirements. So you can look at, just this chapter could be read as a code. This is the inspection code of AWS D1.1, and it covers the requirements for inspectors, qualifications and responsibilities, and I'll talk about that in a little bit. But in here there's contractor responsibilities, and it's talking about non-destructive testing.

And it says non-specified NDT other than visual. It says, if non-destructive testing other than visual is not specified in the original contract agreement but is subsequently requested by the owner, the contractor shall perform any requested testing or shall allow any testing to be performed in conformance with section 6.14. The owner shall be responsible for all associated costs, including handling, surface preparation, NDT, repair of discontinuities other than those described in paragraph 6.9. However — let's say — at rates mutually agreeable between owner and contractor.

So I've got a buyer and seller here. I've got an owner who wants you to build this building or this bridge, and they say, okay, we're going to require magnetic particle testing on ten percent of all the fillet welds. And that's a random sampling of fillet welds. We don't want to pay the money to have a hundred percent inspection, okay, as the purchaser, but we do want to make sure you have some quality control. So we're going to require you to do ten percent random sampling other than visual.

All welding codes require visual inspection. I mean, the welder can look at his work and see if it looks like a turd on the plate, okay, or if it really is what it should be, okay. His supervisor can look at it, an outside inspector can look at it. Doesn't cost much to look at it and say, this looks good or this looks bad. And there actually are acceptance criteria for what is acceptable or what's not. So they always require visual. In fact the code actually says it, okay. Visual inspection, then it says all welds shall be visually inspected and shall be acceptable under the criteria. It doesn't say ten percent of the welds. It is a matter of the code that everything must be visually inspected.

But if you want to do something that costs more than just looking at it, you've got to specify it in the contract document, the purchase document. And that will be factored into the price. So let's say someone builds a bridge or a building and they do the ten percent inspection, for which the owner paid for it. And then this happens all the time — I mean, probably happens, someone gives me a call twice a year on this type of problem — one of the welds that wasn't inspected, they found a defect. And the defect may have already caused a problem, may not have caused a problem, but someone noticed it, okay, whatever. And the owner comes in and says, oh, we've had a defect. And the contractor says, well, yeah, we only did ten percent inspection. Oh, but you found a defect, and so we want one hundred percent inspection.

The code has seen this happen many times before, and it turns out the contractor must do one hundred percent inspection but at the owner's cost. This is an addition to the contract, you're asking for more than you paid for in the beginning. And typically this starts a big fight, and the owner says, oh, I bought one hundred percent good welds. No, you bought welds. And statistically some welds will be good, some will be bad. And if you needed one hundred percent inspection, you would have bought one hundred percent inspection, and that would have given you a much higher probability of getting one hundred percent good welds. But if you're only going to test ten percent of them, you're going to have to accept the fact that if five percent of our welds have some defects in them, then if you only test ten percent, you're going to still have four and a half percent defective welds out there, right? Because you will have only caught ten percent of the five percent that were defective, okay.

But the thing is, the safety factors, which someone else will talk about, right, will hopefully take care of that, okay, that four and a half percent. And we've learned historically what to do. If you go to the ASME boiler pressure vessel code and you look at a drawing, they have a drawing in there in section 8 of a pressure vessel with nozzles on it, and longitudinal welds and circumferential welds. And in the text it will tell you that all the longitudinal welds must be one hundred percent inspected under these conditions, the circumferential welds can be done ten percent inspection. Why? The longitudinal welds are more critical than the circumferential welds. They're stressed at twice the level, okay. And if you learn something about safety factors and the concept of safety factors, the higher you're going to operate towards the strength, the capacity of the vessel, the more inspection you need to make sure you have nearly one hundred percent good welds, right?

So these disputes come up all the time, and the only way you can get around it — it says, however, such testing should disclose an attempt to defraud, okay, or gross non-conformance to this code, repair work shall be done at the contractor's expense. And that's the out that allows the attorneys to run up big bills, okay. People are going to start arguing because, let's say they did ten percent inspection and then they find a flaw, and then the owner says, oh, I want one hundred percent inspection, and the contractor says, okay, I'll do that for you, or you can hire someone else to come in and do it if you don't trust us, but you've got to pay for it, we're not paying for it. And so, oh, you have to talk to my attorney, okay. Well, the attorneys start talking, and now the attorneys are going to say, oh, these are so bad that this is an attempt to defraud us, okay. Well, this is the out, and so this is the gray area. And that's where the attorneys and the experts make money and everyone else loses, okay. Yeah, one comment on that, yep.

Student: Um, all of this is when the owner or the designer, the architect, actually specify an acceptance, right? There you're both — well —

Because AWS doesn't count by default. You have to ask for it. If you only ask for a quarter inch size weld, it could be — it's just got to be a piece of metal for a quarter of an inch that sort of sticks, doesn't fall off when you turn it over on its side.

That's right. But then you could get into sort of what's fair use. I asked you to make welds and you turned out something that was just nothing, okay, it wasn't a weld at all, and you didn't provide me with a weld. That's where attorneys are arguing, is what is a weld. And that's why I wrote a paper once on what is a weld. I did, actually, okay. But that's not the reason I wrote it. I wrote it because the American Welding Society couldn't give a decent definition of what a weld is, so I decided I'd take a shot at it. In any case, that's in one of the other lectures.

In any case, you should get what you pay for. But what do we mean by should? One of the advantages of having a standard, just as Dr. Belmar just stated, if in your purchase order you say that the welds will be made to AWS D1.1, you just called out a one inch thick document, okay. And all the details in here are now by reference part of your contract, okay. And so you saved yourself a lot of verbiage in your contract by just specifying AWS D1.1. But you really should say 2010 edition, because 2018 edition is not exactly the same. And if you happen to hit one of these things where it's changed, well, what is it that you should have gotten, okay, so far as that goes.

So the standards are constantly moving. One of the problems with things of force of law up here — unfortunately a lot of times when the state government or OSHA will specify that all ladders shall be built to such and such standard, and they'll put dash 1966, and they never go back to update it. And so all of a sudden in 2010 we're looking at the quality of a ladder designed to 1966 specs. Is that really what we should be getting, okay? It's best if they would just call out — and what the codes themselves do, the current edition, okay, or the edition in effect in 2007, okay. And that's why — which one was in effect in 2007? If it's ASME boiler pressure vessel code, I think the new one takes effect on July 1st, so you've got to say what date in 2007, okay.

Believe me, there's lots of money wasted, but it puts my kids through college, in trying to interpret codes and standards. And the codes and standards call out other things, other standards call out other things. The purchase order calls out a code or a standard, okay. But then which edition, and so there's all kinds of questions to these things, okay.

Let's talk a little bit about inspectors with a capital I and a little i. Chapter six. And I'll just — I can do this for the ASME code, I can do it API code, different codes have different things. The one I happen to have with me right now is the AWS code. All the requirements for the inspectors, qualifications, and responsibilities. This is what I call the Inspector with a capital I. It actually has a capital I, okay.

Now, it's not always clear. If we come over here, visual inspections, all welds shall be visually inspected and shall be acceptable. Who does that visual inspection? Often it's the inspector with the little i. It could be the welder, it could be his foreman, it could be the foreman's supervisor. But in many codes, particularly like the ASME code, there are different types of inspectors, okay.

There's inspection and contract stipulations. It can say in the contract, if you're building a ship, you're going to say the American Bureau of Shipping will be the classification society of your ship. And that means you must go and hire ABS and pay them millions of dollars to have someone on site, or maybe a team of people on site in your shipyard. Those are the capital I Inspectors, and they're going to be looking over and supervising your inspection department. If you're building a military ship, the U.S. Navy sends in what they call SupShip, Supervisors of the Shipyard, okay. And guys who came — Navy officers who came back to Course 2, and them at MIT, Course 2M — they may have just come from a SupShip where they were taking care of the inspection functions. Or they could be — when they graduate they'll be assigned to a SupShip. In one case they'll be the gopher who has to go out in the ways and look at the weld when somebody notices something. In the other case they'll be the guy who sits in the chair in the meeting about what do we do about it, okay. There's different levels of people in these things when you get to big organizations.

But in fact they're the little i inspectors, and then there are the ones that really are the code inspectors, which are kind of what I call the capital I Inspector. There's a contractor's inspection, and that is basically a test during assembly, welding, after welding. And it's basically what's done by the employees of the contractor, okay. It's the welder looking at the weld — he can tell if he made a good weld, if it looks good or not.

There's the verification inspection. The owner, or someone he hires, or someone he asks the contractor to hire — an independent party like the American Bureau of Shipping, like a local test lab. There are test labs in Boston you can go and hire them, and they will go on site. If you're building a bridge for the state of Massachusetts, they will go out and they will x-ray the welds in the middle of the night when no one's around, because otherwise you're just irradiating all the workers. And they will go do an inspection. So the owner often — the owner may have his own people. If you're Shell Oil Company, you may send your own team in, okay. You've got people who can go do that. If you're the U.S. Navy, you've got your own people. They're called naval officers, okay.

So they talk about inspectors categories. Each code will go through and define what they mean by inspector, who is the inspector, who's the independent inspector. And this is my terminology. There are little i inspectors, who are people who are supposed to be looking after their own work, whether it's the company's work or the actual guy doing the work. And then there are the outside people who come in to do the inspection. I mentioned the National Board of Pressure Vessel Inspectors — even after the thing's been built, every few years, two years or three years, a pressure vessel has to be inspected under state law, many states' law, because they want to make sure that it didn't corrode, okay.

I've got a thing up here in Salem Harbor right now where a big utility — they've just announced they're going to close it, because it was built in the 1960s, it's fifty years old now. But about seven or eight years ago they had a steam explosion. Steam came out and killed I think four guys, okay. Why? It turns out the inspector wasn't really doing his job, because it's hard to get into some of these places and inspect them. So for like fifteen years running the inspection he was supposed to do every two years, he always said, oh, we don't need to do it this time, okay, and they didn't need to do it the next time, he didn't do it the next time. In the meantime the thing's corroding, okay. And it gets thin enough that one day when some people were about two stories beneath it, it develops a steam leak. Well, this is a huge — this thing's like ten stories tall, steam generator — and when it develops a steam leak, even though the hole is only this big, it lets out enough steam that those guys probably had about two seconds to get out of the way, okay. And they would have had to get, you know, thirty yards out of the way. And they were just basically caught and cooked.

So that's a problem. The next thing, which — I don't know if I should take the time, we got five minutes — is, designed to code, good enough? Is it? Always? Just sometimes. It is.

Remember how did the code come about historically. The code was written because we learned about failures. I mean, I showed you the picture of the Brockton, Massachusetts shoe factory. And here's Henry Petroski's book — this professor, Duke University, who wrote this about 1980 I think. He's written a bunch of books since. But, copyright 1982. So he wrote this back in '82, To Engineer is Human, and he's a civil engineer. He talks about the role of failure and successful design. And I'll use an example next week out of this, the Hyatt Regency walkway collapse. None of you were born at that time I don't think, but it killed a couple hundred people in Kansas City.

Uh, big Hyatt Regency, you know, atriums and stuff that go up, you know, twenty stories. Usually they're built for people to commit suicide and stuff by jumping off the ledges. Well, in this case they had walkways going across, and the walkways collapsed while they're having a party, and everyone's down dancing under the walkways, and they just come crashing down, and a couple hundred people died. He talks about — in the very beginning he talks about bridges, famous bridge failures, because he's a civil engineer. And he talks about the Hyatt Regency collapse in here, which was like 1981 or 1980 or something.

In any case, codes were developed historically, and they tell us the design rules to keep the prior failures from happening again. That's a very valuable function. However, as we try to improve designs, we often go beyond the limits of that design, just like the galloping Gertie, that bridge that they made slender, and all of a sudden they got it so lightweight and so slender that it didn't have enough stiffness for the wind resonance. And so they go back and they rewrite the code.

Is designed to the code good enough? The simple answer is, not always. Or as Matt said, sometimes — I mean that's basically the same answer. And I don't think I have time to give you the example right now of this corrugated stainless steel tubing. Maybe we'll talk about that on next time. I'm here is Wednesday I guess, right? But uh, did you have anything you wanted to say? Okay.