§1. Nickel, HY-steels, and the case against HY-130 [00:03]
If they don't have the capacity, the automotive designers are not going to design with a lot of aluminum for fear of a shortage. Then the prices are going to skyrocket, and the people who make aluminum don't want to spend the money to put in the capacity unless they know there's a market. So they try to work that out. The Department of Energy tries to get in and help things out, but they're not much good at that either.
If you look at some of these other things — they once wanted to use an alloy which is very similar, actually it's the predator [predecessor] alloy, to what is now HSLA-80. There's HY-80 that was developed in the 1940s and 50s by U.S. Steel, used in the Nautilus submarine, and used until recently — recently being twenty years — on other subs. They built a couple of HY-100, which is actually the same composition, just a different heat treatment, to give you HY-100. The Navy in the 1960s developed HY-130, and we've never really used it for a full-size ship. We can talk about why.
I had to go meet with the chief engineer of the Navy once and tell him why he shouldn't. They were going to promise Congress around 1990 that the third or fourth ship in the Sea Wolf class — this is before peace had broken out with the former Soviets — they were going to start making HY-130 ships. The people at David Taylor had been trying to tell the people at NAVSEA that they couldn't do it. People now say, well, we spent fifty million dollars in 1960 dollars developing HY-130, why can't we weld it? If you weld only a certain thickness under certain conditions you can weld it, but a real submarine has to have different thicknesses, and not necessarily under those conditions. So from a practical point of view, HY-130, you can use for specialized applications, but you couldn't build a whole ship out of it.
So the Navy started looking in the 1980s and 90s at developing new steels — the HSLA steels. You have HSLA-60, HSLA-80, HSLA-100. The sub guys will not use anything but the HY steels for the hulls because they're conservative, and the surface ship guys will use whatever will give them more surface ships, which is the HSLA steels because they're more weldable. Back when I was working for Bethlehem Steel in the mid-70s, they wanted to use the precursor to the HSLA steels — which had one percent nickel in it — to build the Alaskan Pipeline. Nickel adds toughness, so you won't get brittle fractures. It would have used a substantial fraction of the nickel in the world to build that one pipeline. You can build a few ships, but when you start building whole pipelines it actually takes more steel than to build a few ships.
Nickel is expensive. And you can come down here to neodymium. Anyone know why I put neodymium in there? First of all, I found the data. But neodymium is how you make very strong magnets — neodymium-iron-boron magnets, that's another story. So we have strength, toughness, availability, and relative cost. Steel has a unique combination of strength, toughness, low cost, ready availability, fabricability, and repairability. It's sort of a sweet-spot material, which is why people make one and a half billion tons in the world. We like to make steel, and we like to use it, because it's cheap.
§2. Why this course spends so much time on steel [04:29]
So now let's start getting into welding metallurgy. We will talk about aluminum and titanium and stainless steels, but you're probably going to say over the next few hours of lecture, why is he spending so much time on steels? This is why.
Here's another handout. How many people have ever seen a phase diagram? About half of you. You don't have to learn about phase diagrams because you're not going to be quizzed on them, but they can be useful things. I'll pass out the phase diagram, but before I do that, here is one of the other ones I just passed out: Materials Research to Meet 21st Century Defense Needs. Some committee I served on back in the early 2000s. The Defense Department decided, why don't we figure out what materials we're going to use in the future. It turns out people are always debating this. You read in the Wall Street Journal about some new material — about nanotechnology and all this other stuff.
In general nanotechnology is good if you're going to make some sort of fancy optical material or electronic material — a functional material. But nanotechnology is useless for structural materials, which is what this course is supposed to be focusing on — how do you build something big. You don't make something big out of something that's super small, right? But you'll read all kinds of people: oh well, we can come up with something that's ten times stronger than any material we have right now. Yeah, and we did that sixty years ago. We made iron whiskers that had strengths of 2 million PSI. HY-80 is only 80,000 PSI. The problem is the only way you can do that is if you have a single screw dislocation right up the center and no other dislocations, because dislocations are what makes steel malleable. If you get rid of the dislocations, the inherent strength of the steel is about 2 million PSI.
You read about Bucky balls and Bucky tubes and carbon fibers — the physicists have calculated the strength of these things will be fantastic, millions of PSI. Well, physicists and material scientists did those calculations in the 1930s. Some guy who wasn't even born in the 1930s comes along and he says, oh, I just calculated the inherent strength of a carbon-carbon bond and it could be millions of PSI. Yeah — we knew that in the 1920s and 1930s with the beginning of quantum mechanics. And you don't even need quantum mechanics. You just need to understand surface energy and the formation of surfaces, which we were starting to understand in the 1920s. You can predict that the inherent strength of a carbon-carbon bond is about 3 million PSI — great, as long as it doesn't have a defect in it. A little microscopic defect like a missing atom, in which case that 3 million PSI becomes 100,000 PSI, and it rips. The physicists have done that because they haven't tested any of these things, they just calculated. So they put it all in the Wall Street Journal, they get Congress and everybody else to give them all kinds of money, and it's money down a rat hole. So I'm not real big on nanotechnology. There are some uses for it, but not in structural materials.
I was on this committee, and I wrote up my little submission. My submission was sort of what I was teaching to my students about material selection. This is entirely written by me — Appendix C, "Integration of Material Systems and Structures Development." If you want to read a whole module of my introduction to structural materials, material selection and design, here it is. The four-page synopsis, so you don't have to read anything else. It's about cost, relative cost, which is an important thing when you're building big things.
§3. The value of a pound saved [09:19]
Anybody know the value of a pound saved in an automobile? I was driving in this morning — we're having problems with the highway trust fund going to run out of money because we've increased fuel efficiency, so they're not paying as much gasoline tax, so we can't rebuild the roads. The trust fund's going to run out of money in the middle of the summer, so they're going to take it from the post office. Oh, well, this is clever. The post office had a five-billion-dollar deficit last year, so — huh, maybe they're going to take it from the pension fund of the post office, I don't know. These accountants are great, aren't they.
Anyway, what is the value of a pound saved in the weight of an automobile over the life of an automobile, which is about 100,000 miles? You'll find it's about two dollars. For every pound you can get out of that car you save two dollars worth of gas.
What is the value of a pound saved on a Boeing aircraft over a 100,000-hour life of a typical commercial jetliner? Two hundred dollars. On jetliners they actually have vice presidents at US Airways who decide how many magazines you can carry on the airline, because magazines add weight, and at two hundred dollars a pound that's a significant amount of money. And a spacecraft, you'll find in this article, is twenty thousand dollars a pound. To put a pound of payload up in orbit is about twenty thousand dollars a pound. By the way, ships — to transport a pound of goods by ship across the ocean costs how much? Twenty cents. So you go from twenty cents a pound for railroads and ships, to two dollars in transportation of cars, to two hundred in commercial aircraft. Military aircraft might be a thousand dollars a pound. And then it starts depending on where you're taking the weight out. If you're taking it out of the calipers for the disc brakes, that's worth more, because that's non-sprung weight, and it vibrates, so it moves faster. You're not going to get this lecture in full — if you wanted to take this course again and take different modules, I'm giving you a thumbnail version of some of those things, and that's what this material selection is.
But the real point is, if I'm going to make something and sell it for less than two dollars a pound, my metal choice comes down to steel. Another rule of thumb is that the metal cost for the fabricated structure is about ten percent of the overall structural cost. One time NAVSEA, back in the days when the aircraft carriers were starting to gain weight — the Nimitz-class carriers gain weight at 250 tons a year. That doesn't sound like a whole lot, but if you plot the weight of a Nimitz carrier over the sixty or seventy years of the hull, they gained 250 tons a year. And it's not down in the keel, it's mostly up top. That's not good. That means you have to put more lead in the bottom. Trying to save all this weight in the ship and then you put more lead in the bottom anyway.
So they decided they wanted to go to HSLA-60 for a lot of the lower-strength steel components. Newport News estimated they could save 2,000 tons on the weight of the next-class carrier if they had an HSLA-60 available. David Taylor with the steel companies developed an HSLA-60, which wasn't too hard to do — there were commercial versions, it was good dual-use technology. They said, well, we need to do a study to find out, if we go to this change, what's it actually going to cost as fabricated. My rule of thumb is, if the material costs you a dollar a pound, it will cost you ten dollars a pound as fabricated. Let's face it, you start out with these rectangular plates and you cut away thirty percent of them and send them back to the steel mill for scrap. That just increased your price by fifty percent per pound of what you buy. The Air Force calls it a buy-to-fly ratio — how many pounds do you buy versus how many actually fly in the airplane.
After spending a million dollars of NAVSEA money and a million dollars of Newport News money, they found that if the HSLA-60 was going to cost them a dollar a pound, two thousand dollars a ton from the steel mill, it would cost them ten dollars a pound as fabricated. All they had to do was read my little thing, it's right there in that paper. Now, some things, like the pipeline industry, the material cost is thirty percent. But just as a rule of thumb — and I've got the breakdown in there — you spend as much on non-destructive testing as you do on welding, you spend as much on design as you do on welding, you spend as much on forming and shaping and machining as you do on welding. The breakdown is in there. And you've got to pay the manager, too.
So the point is, if I want to sell a car for two dollars a pound, or if I want to save weight on a car and it's only worth two dollars a pound over the long run, I can't use a material that costs more than about twenty cents a pound. And that's steel. You can't afford aluminum. Back when I was developing all this stuff ten years ago we had gasoline at two dollars a gallon — now at four dollars a gallon, maybe you can afford steel. But guess what, the price of steel has gone up. It's now about forty cents a pound for the sheet steel they use in automobiles, so the numbers all still sort of work out. It does make sense if you're buying gasoline in Norway, which is probably at eight dollars a gallon — not because they don't have a lot of gas, but because the Norwegians want to conserve, it's a national policy. The only reason the world makes steel cars is because they want to sell into the North American market, and that's us. Aluminum cars make lots of sense everywhere else because the price of gasoline is higher. But everybody wants to make something that's going to be competitive with the Ford Taurus. Even Ford looked at an all-aluminum Ford Taurus years ago — instead of a $25,000 car it becomes a $35,000 car.
Student: Truck?
Yeah, they're coming out with it now. They're coming out with an all-aluminum truck, but we haven't seen the price yet, and I'll bet you it's a $40,000 truck even though the steel one you can get for $30,000. When I was saying all this twenty-five years ago, I was saying we wouldn't see all-aluminum vehicles for another twenty years. I wasn't talking about all-aluminum Audis at the time. I told you yesterday or the day before, anybody can build a $100,000 vehicle out of aluminum. How do you build a $25,000 vehicle out of aluminum?
§4. Aluminum superstructures and the Belknap fire [17:41]
When the Navy, about twelve or fifteen years ago, decided they were going to the littoral battlefield, they had to have fifty-knot ships, and they had to be aluminum. Well, they didn't have a lot of experience with aluminum in the Navy, and the experience they had wasn't very good. It was superstructures catching fire, and becoming a big flare. I told you about the Belknap disaster. Anybody else know about the Belknap disaster? It's in the book on disasters. If you're finished with that I'll take it back sometime. I do want it back, otherwise I have to buy another one. If you want one for the library you can buy yourself one — there's the Air Force disasters too. There's a series of these books.
The Belknap was a destroyer that had an all-aluminum superstructure — that was what they were building in the 70s and the early 80s. On some maneuvers it collided with the original JFK, the original Kennedy, not the new Kennedy. The original JFK was launched about 1970. I went to Virginia Beach High School, and one of my classmates' father was the captain. He lived in Virginia Beach, but he was the assigned captain while the ship was being completed. They had a collision a few years later, right underneath one of the hangar deck elevators, and some jet fuel fell on top of the Belknap, the aluminum superstructure caught fire, and it wiped out the whole ship.
The joke at the time was, a gallon of jet fuel could wipe out any ship in the fleet. Of course they knew about that because of the Sheffield. I've had several people in this class do presentations on the Falcon [Falkland] Islands war and how the British cruiser Sheffield got hit by an Exocet missile, and people say the aluminum superstructure caught fire. There are other things now you can find on the web that say that wasn't the cause of the fire on the Sheffield, it wasn't the Exocet missile — whether it was or not, it's still a debate. I've asked some of my friends at Carderock and they'll tell me there's some debate about it. But let me tell you, the Navy believed it was the aluminum catching fire, because by the mid-80s they were trying to replace the superstructures in surface ships from aluminum to waffle membrane steel, high-strength low-alloy steel. They were making a waffle composite of steel to get rid of the aluminum. Well, why were they spending all that money if they didn't think the aluminum would catch fire?
I should have gotten it from Harold Larson, but he won't be in until tomorrow. One of the students here had recovered a piece from one of these ships where you have the corroded holes through the wall. He was in the head — the toilet on the ship — and he noticed a breeze coming through the wall. On the other side was steam from the engine room, and the breeze was coming through because it had completely corroded away. It was at a steel-aluminum transition joint where there's tremendous corrosion. He just picked this up out of the scrap yard and took it home. So Harold Larson probably still has it. I have a little piece that we polished and you can see the corrosion — it just corroded through the steel like mad.
So where was I — talking about the cost of steel. You end up finding that you can't afford these other things. Now, submarines you can afford a little more than you can for surface ships. That's why they still use HY-80. But there are welding problems with all these things, and we'll talk about some of those welding problems in a little bit.
§5. The Diet Sprite demonstration — hydrogen out of steel [22:06]
Before we take our break, let me start my demonstration, because we're going to talk about hydrogen eventually. I showed some of you the video before we got started, but I have to do a demonstration that you'll really remember. This is only the second time I've ever done this in class. Nothing dangerous about it. Diet Sprite — two cans of Diet Sprite.
[Tom opens a can of Diet Sprite.] If I pop the top, there's CO2 in the Diet Sprite. Essentially what you're seeing is gas coming out of a liquid, whereas when you weld steel with hydrogen in the atmosphere you will put hydrogen into the steel, and the hydrogen will come out. I showed you the video from the 1950s from University of Delft, and you can see how rapidly the CO2 is coming out, right? You've poured sodas before. So here it is — it dies down very quickly. You still see bubbles in here. I can shake it up a little bit and generate some more bubbles, but they die down pretty quickly.
Before the end of the next hour you'll no longer see bubbles. But when I shake it up at the end of the next hour I can still generate bubbles. There's still CO2 in there, but it comes out very rapidly in the first few seconds. By tomorrow this will be completely flat, no more CO2. It will slowly evaporate away. That is almost a perfect analogy for how fast hydrogen comes out of steel. This is out of a liquid; out of steel it's coming out of a solid. If you saw the video — I can play it at the break for some people if you want — it said you had to put that thing under the microscope slide within the first ten minutes or an hour, or you wouldn't see the bubbles coming out. It turns out that if you're welding a high-strength steel to build a bridge, the American Welding Society says you must wait 48 hours, if you use the 100 KSI steel, before you do your non-destructive testing. Because otherwise, if you do your non-destructive testing in the first hour or two, the cracks may not have formed yet. They might form ten hours later.
For nuclear submarines you have to wait a week from the time you complete the weld to the time you do your non-destructive testing, because with higher restraint on the submarine you might get hydrogen cracks days later. Usually the hydrogen cracks are only a problem in the first week or a couple of days. I'll tell you a story one of these days of where it lasted for nine months. But it was frozen in the High Sierra mountains in the winter — so you can trap it if you freeze it. We'll talk about that too.
It shows up with all processes. The Sea Wolf submarine, the problem was with MIG welding, gas metal arc welding, where you had no flux. I'll show you a plot of how much hydrogen you get with different welding processes. So we're going to spend a fair amount of time on steel. We will get to aluminum, but that'll be probably next week. Let's take a break until about 8:39.