CS_F2012_13

So I'll lecture today and then Dr. Belmar will do tomorrow. But I wanted to talk about, first, just mentioning these shaving of safety factors, and then I want to talk about the opposite, where people tend to bulk up safety factors, and why they do that.

So far as shaving of safety factors, I've mentioned roof joists a couple of times, and how in the mid 80s computer programs got to the point where they could start calculating the stresses, and they could decide rather than having a uniform beam they could put layers, doubler plates and things on these things, and they could still meet the code of 1.67 safety factor, five-thirds times the expected stresses. And the problem there is the computer programs couldn't deal with asymmetries in erection. They also couldn't deal with the fact that some of the joist welds might not be any good. And so you have a big heavy snowstorm, which tends to exceed the safety factor — anyway, it doesn't exceed the safety factor but it exceeds the expected snow load. So all of a sudden you kind of get beyond the safety factor and a number of things come together and the roof falls in, and hopefully no one's around to get hurt other than just a lot of property damage.

Although I did have one at one time in one of these things. It was in a battery plant, and it was a very heavy snowstorm. In fact most of the plant was closed. And the problem was when the thing caved in, there's lots of plastic and stuff in a battery plant, and I think it went through a natural gas line or something. It was a big plant. And the snow trucks, the fire trucks couldn't get in because of the snow. It was like six inches — six feet of snow on the ground in Pennsylvania, which was not normal. And they just, fire trucks are out there on the highway and about a half a mile away was this great big building about the size of a football field. They sort of watched it burn down because they couldn't get in there to do anything. It's surrounded by snow, but you know all it did was melt the snow, and you know, the ashes got a little bit wet. So it was a pretty big property loss, but no one got hurt.

The X-33 space plane, I mentioned a couple of times, they had a factor of two safety, but in fact they had a manufacturing problem they knew about it. And so they just did all these calculations and started doing fracture mechanics and said oh we think we're okay, we can do this, and we have a 1.05 safety factor. Well, they didn't. Something else went wrong in the thermal stresses when they cooled, froze this liquid nitrogen tank, and the thing popped open, and they ended up canceling a 1.3 billion dollar project because of this failure. No one got hurt, but— And actually the neat thing, I got to see, they built this thing in the hangar where the F-117, the stealth fighter, original stealth fighter, had been built out at Lockheed Martin Skunk Works. So I got to go visit the hangar. And if you want a two-story carbon fiber composite that cost 50 million dollars, I know where you can probably buy the second one cheap from NASA. Anyway, they didn't have a whole lot of use for it afterwards.

But the bulking up of safety factors is another matter. To give you an example, and an example I've given a number of times, is the Liberty ships. And the classic picture of the Liberty ships is what's going on here. Hmm, my image is rotated. Okay. I got this upside down, yeah, I got it upside down. So there's the classic picture of the Schenectady. It was brand new, it was sitting in port, and they used to build these things in less than two weeks from laying the keel to floating them out. And in World War II, this is actually a tanker rather than a true Liberty ship, but the T1 tankers and Liberty ships had basically the same, except some of them carried oil. Anyway, had a crack all the way through. And there's another picture that I prefer, of the Esso [SS Schenectady-class — Tom searching]... I can find it, where the same thing happened, except with the Esso, it was sitting out in the middle of the ocean when it happened. Okay, it's a lot safer to have it happen in the dry dock. Anyway, at the end of all of this, let me rotate the images back.

In 1946 the Navy did an investigation and found that the reason was that these things didn't have enough toughness. And in fact, back in the 1880s, people learned to do tensile tests. Okay, and they would, they learned that a piece of steel, you could pull the piece of steel and you could measure how much strength it had, in terms of a tensile test. And that's the force of fracture. But what they didn't really realize until the Liberty ships, you should also be interested in the energy of fracture. If something doesn't have enough toughness, which is a measure of the energy of fracture, then a notch can cause something to fracture quite rapidly and catastrophically. And we'll talk about that for several things. And the example I always give of the effect of a notch: you can pull on a piece of paper with pounds of force, but if you put a notch in it, it takes ounces of force to rip it.

Well, so far as looking at the effects of notches, people for the last 105 years have looked at Charpy [Sharpie] tests. And we don't have any Charpy machines around here anymore. There actually was one around here once when I was a student. It takes a lot of money to run Charpy tests, and it's not something you do at a university too much anymore. But anyway, it's just a calibrated hammer on a pendulum. And here's the hammer, you put a specimen down here. The specimen is a little piece of steel, one centimeter on the side, 10 centimeters long, with a two millimeter notch. And you hit it with the hammer and you see how much it swings up, and you measure the rebound on the pendulum. And if there's no energy absorbed in the fracture, this thing will swing all the way up to the same height as the original starting position, because there's no energy absorbed. If you absorbed energy, it won't swing up as far. And if this thing is really tough, you'll stop the hammer. Okay, but if you do that, you should go through and recalibrate your whole machine to see if your anvils that hold the sample have been deformed in any way. And so it costs you about five hundred, a thousand dollars to get another set of calibration samples to test.

This is one that stopped the machine. This one looks like it stopped the machine because it didn't hit it quite square for some reason. Here's one that's very brittle, didn't absorb much energy. If you look in Don[?]... there's a tough one that stopped the machine. Here are some that are fairly brittle. This one is 300 joules of energy, this one's 22 joules of energy, so far as that goes.

So the Charpy test had been around for 50 years, but, and people used it from time to time, but it wasn't a design criterion. So after the Liberty ships, in this report in 1946, it turns out, people started worrying about it. And they found that what happened, this report, they went out and measured the plates where the cracks had started, and they found all the plates— and they had an incredible number of ships that had major cracks. They didn't all split in two like the Esso and Schenectady, but 4,700 ships built during World War II, 970, okay 22 percent, had major casualties involved in fractures. 24 vessels sustained complete fracture of the strength deck, which is the top deck. Okay, so if you have something like that, obviously it's not good, because fatigue could take it the rest of the way. So anyway, they found that all the ones that had major failures, the fracture toughness — or the Charpy energy, not fracture toughness — was less than 10 foot-pounds.

And so what do you do? They put a safety factor on it. They said from now on they wouldn't build Navy ships and most other structures unless they had 15 foot-pounds. So this was, let's say we're back up to around 1950 now, give them a few years to implement the 1946 recommendation. By 1960 the Coast Guard decided well, that's not enough safety factor, and so they essentially said we'll make it 20 foot-pounds. By the time I started working in the early 70s, it was not only 20 foot-pounds, it was, you had to measure the weld— that's the Charpy specimen. But if you look at a weld you would have to take out Charpy specimens at the weld metal, the fusion line, and three millimeters from the fusion line. Okay, so now you had to take four specimens where you used to have to take one. And now it was 20 foot-pounds, which was double what the Navy found for the Liberty ships was the thing that caused critical fractures. That was because people were concerned, and brittle fractures were still occurring.

However, I can remember, down here in Quincy Shipyard — which is no longer, but there was a shipyard down here in Quincy that they built, I think in World War I, but certainly it was busy in World War II building Liberty ships and things. In the mid 70s when I worked for a steel company, they were building liquid natural gas tankers, okay, and they put in this mammoth crane to be able to lift these big aluminum spheres that were up to six to eight inches thick of aluminum, that were made in South Carolina and brought up by barge. And it had this huge crane that would drop five of these into the hull of the ship. And the skirt, which was basically just a cylinder that would hold the sphere, had to be a low temperature steel, minus 60 Fahrenheit fracture toughness. And there wasn't a really good steel, so I was working on developing a new steel for that skirt.

And they were using a grade of steel A537 that was, they would normalize and temper, which means you stick it in the furnace and you let it air cool to get a finer grain size, and that gives you better toughness and strength. And they were finding to get their 20 foot-pounds, they would have to, I think half the plates had to be double tempered. They would roll the plate, they'd measure the toughness, it was less than 20 foot-pounds after they had normalized and tempered. So they would put it back in and they'd renormalize it and re-temper it to get an even finer grain size. And after a double temper, most of the plates would pass. And the average was 20.5 foot-pounds was the average passing toughness, okay, after a double temper. A few of them they did a triple temper. It wasn't economical to go and temper something a fourth time. But every now and then they would get a plate that would have like 30 foot-pounds as the basic toughness of that steel.

And what would they do? The shipyard — these were all marked, and the shipyard would go out there and they would cut that up into the samples, okay. This is how they could pass. Because the weld and the heat affected zone would typically degrade the toughness. And so you'd fail at the fusion line or the one millimeter position unless you started with something that was really good. If you were starting with 20-and-a-half and you lost very much, you'd flunk when you actually made the weld. So you weld the big plate on the ship, and then you had what they called runoff tabs, which were a little — when I say little, they were probably six inches by 12 inches plates — that at the end of the weld on the actual ship they would just continue the weld. They would cut this off and they'd go and machine this for fracture toughness. So this might have 20.5 in the base plate, this might have 30, and this one would pass and this one really never got tested, right? Isn't that a neat thing? Okay, but that's the way the test was done, and that's the way they made it meet the spec for the Coast Guard, which is maybe why the Coast Guard had found people were playing games like this. So they went from 15 to 20, okay, to increase the safety factor, because people will find ways around things, okay.

None of you are parents yet, but you will find children will learn ways, whatever the discipline the parents use, children will find a way around it. Yes?

Student: [inaudible — apparently asking about higher and lower toughness variability]

Higher and lower— well, this was in the days before low sulfur steel was common, okay. Nowadays we make low sulfur steel all the time. In that time, in the mid 70s, when we made steel it was what we called regular sulfur steel, which is 0.025 sulfur. Nowadays you can get double-O-5 sulfur, five times less, at no extra cost. If you wanted to pay a 10 percent premium for your steel in 1975, they could lower the sulfur, but the U.S. steel companies wanted a 10 percent premium because they had to do some extra processing to lower the sulfur. It was sort of a shock when the Japanese started selling garden variety A36 plate on the West Coast that was low sulfur for no increase in price, okay, in the mid 70s. Because the United States we're getting a 10 percent premium and telling you it was ultra high quality steel, and they were selling the garden variety steel. But that's because they had built brand new steel mills then, they just incorporated the low sulfur processing into the whole thing. They didn't make anything but low sulfur, okay. Nowadays people are like the Japanese were then, and everybody makes low sulfur, okay.

So it was primarily the sulfur. You get a heat with low sulfur, you get another heat that has just the right amount of residuals so you get the finest grain size or whatever. But there's just natural variability in these things, and toughness going from 15 foot-pounds to 30 foot-pounds in toughness, you can still have that range of variability, okay. But the thing is, what's your mean, okay. And when your mean was only half a foot-pound above your minimum, that's not good. If we were at 30 foot-pounds mean, we never would have had a problem. And they never had a problem on these things, because if you actually look at the fracture mechanics, 20 foot-pounds was a lot more than you really needed.

They had been creeping up. So what happened, right as I was leaving, they had just finished building the Alaskan Pipeline. And there was still a big fury in the press about what if we have a fracture in the pipeline. And I remember as a young assistant professor giving some lectures and explaining to the students that the oil company has more interest in not having a big fracture in the pipeline than any of the environmentalists do. And the reason is, if you got two million barrels a day coming down from Prudhoe Bay to Anchorage, and back then it might have been thirty dollars a barrel, okay, that's 60 million dollars a day. Today that would be of what, two times a hundred, so it's going to be 200 million dollars a day coming through that pipeline. If you have a fracture— and unfortunately the people who go scaring people about this— the fracture in a pressurized pipeline like that, the fracture might be 100 feet long, okay, when it starts when it pops open. Okay well maybe less, it may only be 30 feet long. But you can't just go down to the hardware store and pick up another piece of pipe. And if it's in the middle of nowhere in Alaska, it's not as if you can get the piece of pipe that you got in storage in Anchorage to the site overnight. You don't FedEx a 40 foot long length of pipe, and then you got to get the welders from Tulsa up there and everything else, and you got to have the heavy equipment.

If it was the military and national security you'd be airlifting things on C-5As. And in fact the oil companies get to almost that. I mean I've seen things where they have Learjets flying parts around the world, okay, because of some failure. Because the dollars per day that you're losing easily runs in the tens of millions of dollars, sometimes hundreds of millions of dollars per day. And so you could lose billions, okay. And so the oil companies had done everything they could to try to prevent a brittle fracture, but they were starting to talk about building a gas pipeline, because we still have as much gas up on the North Slope today as all the oil we brought back from the North Slope over the last 40 years. The problem is you can't put gas in the oil pipeline, there's different fracture criteria. And it turns out a gas pipeline and an oil pipeline, if you have a rupture— it's basically hydraulically loaded, it's a fluid, and the stored energy is not very great. And so you might get a 30 foot long crack, you might get 100 foot long crack as the crack runs.

In a gas pipeline it turns out the decompression velocity of the gas is slower than the speed of the brittle fracture, and so the brittle fracture is always pressurized. It's running faster than the decompression velocity in the gas, so it's like there was no crack or no release of gas. The tip of the crack is always seeing full pressure. And they have had, in smaller pipelines like six inch pipeline, they've had cracks run for 30 miles. Okay now you don't even store enough pipe for a 30 mile fracture, and how long would it take to replace 30 miles of pipeline?

So what they wanted is— they were doing a number of things. You could put crack arrestors in, where you basically put a piece of steel maybe 18 inches long, a ring of steel, that was twice as thick and a super toughness steel. And they had those steels — A710 I think was the one we used to talk about — it was a copper-bearing, highly alloyed copper-bearing steel, super tough, give you a couple hundred foot-pounds. And if a running crack ran into that, even though it was a brittle fracture it ran into this super tough steel, it would slow down and hopefully stop it, okay. And so they were doing tests for that. The other thing they were looking at is, why don't we just make instead of having this 300 joule ring of toughness that would stop the crack, why don't we just get to a toughness in the base plate, because we can easily, in these pipeline steels getting low sulfur from the Japanese and controlled rolling technology at the time, we could get some pretty good toughness, this is in the 40 and 50 foot-pound range.

By the way, joules is 1.33 times the foot-pounds. So 20 joules is 15 foot-pounds. Okay, so 60 foot-pounds is, uh, 80 joules right, okay. So they were looking for 80 foot-pounds in the plate to stop a running crack. Except if you're at 80 foot-pounds, you never get a brittle fracture anyway, so it never made sense to me why they needed that. But they were looking at trying to develop plate that had 80 foot-pounds, not the 10 that the Navy originally found was necessary, okay. And so this is sort of creeping up over time. Over about 40 years we went from 10 foot-pounds to 15 to 20 — at multiple locations people learned to cheat on that, and they were now getting up where they were really concerned. It wasn't a question of cheating, this was a question of trying to make sure you didn't have a failure in this pipeline. And they were talking about 80 foot-pounds. I don't know that— well actually today I think we probably do have plates that are 80 foot-pounds in the steel, okay. But back 35 years ago, that was actually still a challenge. In any case, there's the creep of increasing the safety factors.

There's some other safety factors that kind of creep up on us. One is— where'd I put my notes... no... helicopters, okay. The first helicopter, Igor Sikorsky, in the 1930s — I think it's 1932 or 1936 — flew a helicopter. And you know, it went up to like 30 feet in the air for 60 seconds or something and came back down, and didn't come down hard, came back down, yeah, at a reasonable speed. But at the time, when you're developing new aircraft, you often are just barely able to move the person. In fact, if you look at the man-powered flight they do over in the AeroAstro department, you know, they got a guy who's a lightweight bicyclist, and they put him in as the man-powered flight, and that was sort of it.

The original helicopters in the Korean War: two men in the cockpit and then a stretcher on the tail boom to bring someone back, an injured soldier back, and that was it. You could kind of lift three people, okay, and those things— well, nowadays we have, you know, helicopters that can lift an M1 tank. You know, if maybe not an M1, but they can lift pretty heavy vehicles. I mean the Army's got some heavy lift helicopters that can lift some pretty heavy vehicles. But in any case, helicopters, how many people have ever flown a helicopter? One? Two more, gain with me. It's sort of interesting when you get in the helicopter. If you start touching the doors and everything else, it's sort of flimsy, right? And so, it's really flimsy. I mean, it's composite material, it's really light. And you know, I wouldn't go push, try to put my fist through it, because it's probably strong enough, you might hurt your hand, but it's probably very strong. But it's not very heavy and it's not very stiff.

And so first time I ever got in a helicopter I thought, oh I don't know if I like this. But I've worked on a lot of helicopter engine and rotor things, and when I've done the stress analysis I find those parts actually have about a factor of 10 safety, okay. They're really much bulkier than they need to be. So the helicopter is sort of an anomaly in the sense that the shell that the people go in is sort of lightweight and probably doesn't have a safety factor much more than one and a half, but the moving components that keep you up in the air actually have tremendous safety factors. Because lots of things can go wrong in a helicopter, it's not very forgiving, okay.

It actually is— in a regular plane you can glide. You know, if you lose your engine, you can, if you get the right slope, you can still have enough airspeed that the wings still develop some lift and you can hopefully glide and find a landing spot. In the helicopter, they actually do have something similar called auto rotation. If you have enough forward momentum, you can basically tilt the helicopter a little bit and let the rotors free wheel — it's called auto rotation — and the wind coming up through the rotors will cause the thing to slow down and fall to the ground. You actually can see this on some little kids' toys that are little rotors, that you go spinning up there and it kind of comes floating down to the ground because it's basically auto rotating to the ground. The problem is if you're in a hover, you have no forward momentum in order to allow you to push the air through the rotor disc, and you just kind of fall like a rock, okay. And with this structure that has no real crashworthiness to speak of, it's not a good day.

But there are some— well, what's the, I guess it is probably the most expensive helicopter you can think of, I think we've mentioned it before — the V-22 Osprey. Oops, can't even say it. [Tom struggles with a slide that has gone off-screen.] This is interesting, I've never had this happen. In any case, the V-22 Osprey looks like this. It's a tilt-rotor, dual rotor airplane. It's both a helicopter and an airplane. And I think I may have— you know okay, so this thing's back, it went off into never-land.

So there's the Osprey. The Marines, or the Defense Department, got interested in a different type of helicopter after the Iranians took over the American Embassy in 1979 or whatever. And they tried to rescue some of them, and the helicopters weren't successful. Helicopters tend to announce their arrival — they're fairly noisy, okay, and that's inherent with the blade whipping around. The tip of the blade on the advancing— if you're moving forward with the helicopter, the tip of the rotor disc that's advancing will be approaching the speed of sound, it gets very noisy. The one that's receding is not. If you could tilt that and turn it into a rotor for an airplane, the tip speed is never getting close to the speed of sound, and they're much quieter, or they can be much quieter.

So it turns out the XV-15 was a prototype that NASA was developing with Bell at the time. And it was a tilt-rotor, two-blade, two-disc helicopter, that basically you could rotate these things down and you could fly in airplane mode. And the advantage is, instead of being limited 200 mile an hour airspeed — or actually 180 — you could go 300 miles an hour, and you can do it quietly. And what was the famous military action in the past two years where they used an Osprey? The assassination of bin Laden, okay. They had other— secret — they call them, they said they had Blackhawks, but they were Blackhawks like you've never seen, these were Blackhawks that were stealth Blackhawks, okay, that are classified that we even have them, okay. But nonetheless, the main one was, the Blackhawks went in as the stealth fighters, if you will, they were helicopters. But the one that brought all the SEAL team in was basically an Osprey.

And the Osprey can still auto rotate even if it loses one engine, because it has a drive shaft in between, so that if one engine is operating, let's say the far engine goes out and this one's operating, the torque from this engine is driven by the drive shaft over to the turbine, or to the transmission, and you can basically rotate both rotors so that you can maintain flight, okay. Well it turns out in Ship 4, I think it was Ship 4, I can't remember right now anymore, it was coming in to put— Patuxent River Naval Air Station — after having gone through tremendous environmental tests for six months down in Eglin Air Force Base in Florida. And they basically had seen its full useful life in those six months. They have a huge chamber, the Air Force can put whole planes in, and it can cause it to snow, to rain, hurricane force winds, it's an environmental test chamber large enough to put a full-size aircraft in.

And this Ship 4 had just finished six months of full-scale testing under every weather condition, and it basically was sort of worn out, okay, but it had gotten in six months most of its useful life. Well it turns out they were coming in and there were a number of generals and others to welcome this back to Patuxent River. And it was taking a victory lap sort of around the field, and it had come in airplane mode, and it was getting ready to go into landing, and it was moving the cells up to helicopter mode to come in and land. And one of the engines caught fire — actually the engine didn't catch fire, there was a hydraulic leak and that caught fire and got into the engine, blew up. And they crashed and eight men died, in front of all the spectators, okay, who are engaged to welcome them home.

Well what happened, the reason they had a problem — you should have been able to lose one of your engines, but the problem was, they had to make this thing super light. This was the first aircraft ever built out of all composite, essentially, because, just like Igor Sikorsky's first helicopter, there was not a whole lot of payload left over if you made it out of conventional aluminum, okay. You wouldn't have any payload left over. In fact some people would say this thing never would have flown if it weren't for composites. You get on this thing and you kind of touch — same thing, you touch the side wall of things, and you say, ooh, this is sort of flimsy. But it's all carbon fiber composite. And the drive shaft between here and here was also carbon fiber composite. And in the engine fire, the drive shaft burned up within seconds, and so it did try to transfer torque but the drive shaft just tore itself in two. And so afterwards they had to redesign it, and they switched the drive shafts to titanium, increasing the safety factor in this case against fire. It's not they hadn't thought about fire, they just didn't particularly think of the engine becoming a fire-breathing dragon, which is essentially what happened, okay.

There's another reason for bulking up safety factors. I remember once, probably around 1990 or so, I'd taken a trip to Japan, and I got back in my office, and I had a note to call Engelhard — which makes platinum, right — and they needed me to come down because they were making titanium housings to put catalytic converters on. And this was going to go on the new 747-400, okay. The 747-400 was a lighter weight 747, redesigned to take, you know, thousands of pounds off so they could put thousands of pounds more fuel, and they could go further around the globe and go, you know, direct flights to further distances. The first time I ever went to Japan, we always had to stop and refuel in Anchorage, Alaska, okay, because you couldn't make it all the way to Tokyo from the West Coast — or even the East Coast. Certainly now you can get direct flights on the new 787, supposedly from New York to Tokyo, okay, which we never could do before. As planes get lighter and have more fuel and more fuel-efficient engines, you now can go essentially almost halfway around the globe, which is as far as you need to go, okay. Don't need to go any further, unless you're the U.S. Air Force, they want to go all the way around. Because they're going to drop bombs when they get halfway and then they want to come back, that's another story.

And I said, why do you need— I actually asked that when I was on a committee, I said why do you need to go all the way around the globe? They said because we assume we're not going to have any air bases anywhere except the United States. You know, we're going to be kicked out of the rest of the world, and so if we take off the United States, we're gonna have to land in the United States.

So in any case, what was I talking about... oh, so this was going— they had switched from— well, actually, where's the catalytic converter on an aircraft, on a Boeing jet? It's in the air you breathe. If you get up at 40,000 feet, a significant fraction of the air up there is ozone. And if you breathe ozone for very long, you get a tremendous headache and it's also not very healthy. So they actually have catalytic converters for the air you breathe. And of course on big aircraft, something like this commercial, they have redundant systems, so they have two systems and two catalytic converters. And these were pretty good size, they're like eight or ten inches across and about this big. But just like a catalytic converter on a car, it's just passing air through to get rid of the ozone.

And they were welding these things, and Boeing had a spec, they had to meet the spec. Engelhard was making it in a — it had a titanium tube coming in, and then you have the catalytic converter space, which had to be larger diameter to hold the converter. So you had to make a weld here and a weld here, a weld here and a weld here. And it's on titanium, it's like 40,000 [40-thousandths] millimeter [inch] thick titanium. It wasn't very heavy. It has no real pressure, I mean it's one atmosphere of pressure is the maximum it would ever see, and probably something less than that. And this was, I don't remember, four inches, and this was like eight or ten inches. And Boeing had an x-ray spec that you could have no flaw larger than ten thousandths of an inch, okay.

Well, anything greater than ten thousandths of an inch on an x-ray of these welds was going to reject the weld. And they said, and you can only repair it once. So you got two shots. And it turns out they were making these welds — this is a gas tungsten arc weld, wasn't a big deal, it's easy to make it — but that's a very small flaw. And they were taking x-rays and they were failing half the welds. Now if you have to make four welds, and you can only get one re-weld, a repair weld, what's the probability of getting one of these things out? Pretty low, okay, because two of them are going to fail, and then one of those two is going to fail, right? And you can't repair it after that. And this turns out, they'd been working for weeks, and it was now holding up the whole flight of the new 747-400, because they didn't have the catalytic converters. They're way behind schedule because they had this spec, the standard, that Boeing had written that no defect can be larger than ten thousandths of an inch.

Wait a second, folks. This is only carrying air. And if it leaks air, I mean first of all it's pressurized air, so the leak is mostly out, not in, okay. And even if you did leak in a little bit of air, how much ozone would be in that if you got past the catalytic converter? I mean, I said, where did you get the spec? They said it was a Boeing spec. I said well, you know, it makes no sense. The only way it made sense is, that is the smallest flaw you can find. If you magnify your x-rays by a factor of five, you can find about a fifty-thousandths flaw, okay. Or you can reliably detect a fifty-thousandths flaw at about a fifty percent level. Actually we won't get into the statistics of detecting flaws on x-rays. But whoever had written this specification for Boeing had just picked out the most stringent requirement they could think of, okay. It had nothing to do with the actual application of this part, okay.

It was just a stupid spec. I said, why don't you go back to Boeing and tell them, you know, this is just stupid. Well, we couldn't do that. You know, it would take, in the Boeing bureaucracy it would take a year and a half, and you probably wouldn't even win then, because there'll be someone there who is not willing to give up CYA, okay, for saying oh it's okay, all you're going to do is leak a little air into an airstream, okay. It wasn't going to blow up, I mean there wasn't enough stress on it, there's no way this was going to be harmful. But that's what the spec was, and everyone was afraid to go and tell Boeing that their spec was stupid. It was stupid, but no one was going to tell them that.

Another thing, a bulking up of— another reason for bulking up of safety factors. These MAP gas cylinders, okay. And the MAP gas cylinders, actually they're polypropylene nowadays, but— these are one pound, they have about one pound of gas in them. You can— it's okay to have them stored indoors. It's not okay to store a 20 pound cylinder like on your gas grill outside— a 20 pound cylinder, one that's about this big. Does anybody know the reason why you can store these indoors and you can't store the 20 pound propanes indoors? If it leaks in a room 10 by 8 or 10 by 10, a thousand cubic feet, I could throw this — doesn't have any gas in it but I could throw this across the room — burst it open, and you might burn up that corner of the room. If anyone is standing there they'd be burned fairly badly, but I wouldn't be burned if it was, if I was 10 feet away. I mean I might be singed or something, but I wouldn't be harmed. You might blow the windows out from the overpressure and stuff, but there's not enough gas in this to blow up the whole room. You start taking a 20 pound cylinder and you could have an explosion in this room that would blow up the whole, you know, the whole building, if you have a leak.

So that's why the plumbers can have these torches. Now one of the problems — well, it's not a problem — this thing has to be able to hold at the temperatures you would expect, you know, you don't, at something like 150 degrees Fahrenheit, and you don't expect these things to get hotter than that. You would expect no more than 130 psi. And so they have to test them, I think at 300 psi or something like that, which is a pretty good safety factor. But in fact, if you want to blow one of these things up under gas pressure yourself, it'll take closer to 900 psi. Why? Because shoot, it's only about thirty thousandths inch thick steel, you can't make it any thinner, you'd be able to dent it with your fingernail, you know, with your thumb. Just the inherent strength of the material is such that you have plenty of pressure strength. It's got much bigger safety factor than you'd ever need.

Sometimes these things fail. Why do they fail? Well this particular one, we tried to torque it off, and we found it took like 80 foot-pounds, which is, you know, two people pulling on a bar, a reasonably long bar, and we bent it when we did it. But in fact these things do get bent, okay. In one case it was bent by a guy, a plumber, and he was lying on his stomach doing some work. He was lying on his stomach on the grass out, and by the curb, and there was a— he was working on a water pipe, and he was going to solder the water pipe. And he had a cigar in his mouth — he's lying on his stomach, cigar in his mouth, torching his left hand, cell phone in his right hand, talking on the cell phone, trying to solder this pipe. And the elbow started slipping off, so he started using this as a hammer to knock the thing back on. And we're not sure that's actually one event, but in fact, out of the hole— there was a woman walking her granddaughter on the sidewalk just about a block away. This was in a residential neighborhood. And she sees this eight foot blast of flame shooting right up in his face as he's looking in there with his lit cigar and his cell phone that he's talking on. Now, is that the cylinder's fault? Probably not, okay. But he collected a fair amount, because his injuries were bad — not from a jury but from settlement.

Other cases on these things, people do throw them, you know. A guy will be doing his plumbing and he'll reach over and he misses, he has it lit on the table and he reaches over and he doesn't grab low enough, he grabs here and they pick— you know, or whatever, and they throw it to the ground. And I've calculated with about a 45 mile an hour— it's not a real fastball, but a 45 mile an hour throw, which is a good get-it-out-of-my-hand throw, you can hit the ground, you can bend this thing over. This one's bent but it doesn't have a crack. It has a little crack but it doesn't go all the way through. You need about a 25 or 30 degree bend and you actually can split it up here. So as far as overpressure, it's got a safety valve, but they can be abused, okay, so far as tools go.

Another tool that gets abused fairly often, this is a simple little sprinkler head adapter — another sprinkler head adapter — but a sprinkler head, okay, you know, fire sprinkler. Now in some hotel rooms they will have these sitting out sideways, or even sitting here in the ceiling, and they will have signs around them saying don't hang your clothes from this, okay. It turns out people love to, you know, take the coat hanger and hang it up from— they figure this is a handy hook to hang things from. It turns out this is actually a precision piece of equipment, okay. It cannot be stored at above 100 degrees Fahrenheit, because some of them use Wood's metal as the melting sensor at 160 degrees. And if it stays 100 degrees for longer than a certain amount of time, it'll creep, and these things can fail at less than their rated temperature. And so the code, the standard, National Fire Protection Association says, I shall not use these or store these above 100 degrees. They actually have to ship these in refrigerated trucks in many cases, just because it can get hot in a truck in Texas, okay, where this one was made.

In any case, these things come in, they're shipped in sort of an egg crate type of thing, so they're not just thrown around like typical plumbing components like these, okay, where people just, you know, they throw them around like this. However some plumbers don't realize that, and they throw these around and damage the mechanism. It actually is a fairly precise mechanism, and you can jam it so that it doesn't go off when you want it to go off. So in some ways— well, we'll get into that not too much more.

Another safety factor thing where people have bulked it up, and I mentioned this I think the very first day of class, I brought this corrugated stainless steel tubing, and I've shipped, sent around one that has a hole in it from lightning strikes. People never even thought about the problem of lightning strikes on these things. They gave you the Ben Franklin lightning protection system things. They used to use black iron pipe. If you get a lightning arc to a black iron pipe, you might melt up to 10 or 20 or 30 thousandths deep, a little divot, but you won't penetrate, because it's a tenth of an inch thick. The lightning protection code says that if anything's 3/16 of an inch thick, it's considered self-protecting. Well, officially this thing's not even that thick, but no one has ever heard— well, people claim they've heard, but no one has ever seen one of these that was perforated by lightning. We get two or three hundred of these a year that are perforated by lightning, okay.

So what, and people never looked at the code. When they developed these things in the 1990s, they didn't worry about things until about 1998. They started hearing rumors of failures. By 2001 they knew it was lightning for sure, and they started developing an improved product. The improved product is this. It basically is the same corrugated stainless steel, ten thousandths of an inch thick. It's got a carbon-filled plastic as opposed to the yellow plastic. They had to get permission to do that, to make it black, because the National, International fuel gas color for gas piping is yellow. Right, see, fuel gas tends to be— not all, I mean, propane sometimes red, sometimes blue, in these things, but in general yellow. If you go see yellow piping in a room, don't start sticking nails through it and stuff.

But what they did, first they filled it with carbon so it was slightly conductive, and they went from a tenth of a coulomb to about five coulombs — six, well, five coulombs. Coulombs the measure of the energy in the lightning strike. A typical lightning strike's three to five coulombs. This original stuff was good for one-fiftieth of that. The new stuff with the black stuff was good for basically the average strike. This stuff, which has perforated aluminum sheet, basically is supposedly good for 80 coulombs. There's not a lot of lightning strikes that are more than 80 coulombs, so this is probably a safe product. But it took them about 10 years to come up with something that won't burn down your house, okay. In the meantime, there's a billion feet of this out there, and they got two or three hundred fires a year from this.

So bulking up for the safety standards in this case, because they didn't have a good enough safety standard, and they kind of went through several different iterations and finally ended up with something that probably is safe. By the way, for a long time they were selling this product eight percent less than this type of product, which was sort of a dumb thing to do, okay. If you actually told people that you had something that's 50 times safer, and you were selling for eight percent less — you can buy something that you're dead in a lightning storm, okay, or for eight percent more you can buy something that you might be safe, okay, which would you buy, okay. It wasn't until September a year ago they took this junk off the market in the United States, in North America. You can still buy it anywhere else you want, okay. But anyway, okay, Simone will be here tomorrow.