§1. Course structure and module selection [00:02]
You're supposed to take a welding course. If you look at what the course catalog puts in there, it's welding metallurgy. Like I've told you a couple times, if the only tool you have is a hammer, you see everything as a nail. The welding people at TAC historically, many of them came out of Lehigh or Rensselaer Polytechnic, or some midwestern schools, Ohio. John Appel was at Lehigh, and a couple of us in welding metallurgy have been working a bit beyond those subjects.
I decided in my first year as a faculty member that I wasn't going to compete with all those high-powered welding metallurgists out there. I was going to work on welding physics and chemistry and processes. And I felt my own courses, which were solid-state diffusion — for a number of years that was Navy officers — instead of welding metallurgy. Then Professor Massa, who had been teaching welding, was in ocean engineering, and he helped build submarines in Japan during World War Two. His expertise was building real ships.
So it evolved. If you go back and look — Summer 2013, Summer 2014, and further back, you'll find I taught solid-state welding in Summer 2012. It was sort of agreed that that's what you guys would take. But you have the same choice. You might be interested in materials selection, or codes and standards. In fact, I think it would be better, with ten of you, to pick five different things, take two of them, and then you can get together and talk about it.
None of this is going to turn you into a corrosion expert or a welding expert. You're going to be managers some day. You need to know enough to be dangerous, okay, and ask the people who are responsible some tough questions.
So I don't care which ones you take. But you are supposed to take two more modules. There are the YouTube videos from this list, and I would encourage you — even though you're getting excited about your math courses now because they have quizzes — to start watching these things. You can take tomorrow off, you can take next week off, but after the fourth, start watching. Don't put it all the way up to August. Don't procrastinate. All you have to do is watch it. One student told me he used to watch it while his wife was fixing dinner, like the evening news. That's not going to be good.
Seriously, I keep telling you you're supposed to take this and that, but I don't care what you think. And that's a story too. When Apple computers first came out, you had the ability to change the little buzzer or click you hear when you hit the wrong key. You could record some message. So one of my graduate students showed me how to do it, and I recorded "who cares?" as one of the alerts. Because that's what I always told my graduate students. They were all worried about things. So which course you take, you pick.
§2. Hydrogen porosity in aluminum and titanium welds [05:51]
One of the things you should know about aluminum is that aluminum, just like steel and most metals in the molten state, will dissolve lots of hydrogen. But it doesn't do the same thing as hydrogen embrittlement in steel, because it's face-centered cubic. Face-centered cubic metals can store lots of hydrogen. But in the liquid state, you've got ten times the solubility limit. When it solidifies at 1220 degrees Fahrenheit, the more moisture in the air, the more hydrogen, and it comes out as bubbles on solidification.
There's an ASTM standard for aluminum castings and aluminum welds — they have reference radiographs. They'll have eight different radiographs. These are x-rays of welds. This would be an eight, this would be a one, this might be a three in terms of the severity scale, and that's how you do it — a visual comparison. So I'd say this is between a four and a five. Usually for high-quality aircraft-quality aluminum castings or high-quality welding, it must be class two or less. You never have aluminum castings that don't have a little bit of porosity.
In fact, you don't have titanium welds that don't have a little bit of porosity. Some of these things can be ten thousandths of an inch, and that's the Boeing story I'm going to tell you about. When Boeing was coming out with the 747-400 — this is the extended range, lighten the weight, increase the fuel capacity, an extra thousand miles, because at that point they were at eight or nine thousand miles, and you'd like to have 11,500, which is the mileage on a great circle from the two largest cities, the greatest distance on a great circle for major cities of the world. Kind of like going from New York to Sydney, Australia. There are only a couple of routes. You'd like to be able to supply that distance, and of course you have a little safety factor in case you get weather.
So they were trying to lightweight it. There's ozone up there, and if you breathe ozone, you get a headache. It's also considered a carcinogen. So they have catalytic converters for the air coming into the cabin, to get rid of the ozone. They were using stainless steel ductwork, but they were switching to titanium. And they called me from Engelhard in New Jersey, the contractor — it was about to hold up the whole plane. This was getting to be the critical path for rolling out the 747-400. Engelhard was going to be the supplier.
I didn't have time — I was gone for two weeks — and they had porosity from hydrogen. The same thing. I said, look, all you have to do is clean it really well, and don't use acetone. Most commercial-grade acetone leaves a little oil residue — you need to use reagent grade. I got the exact same story up at General Electric Lynn, where they were welding seventy-thousand-dollar titanium compressor parts by electron beam welding. I got the same story twice, several years apart. Get some reagent grade acetone, clean it right before you weld, and you'll be fine.
They tried to do it and they couldn't. The Boeing spec was that they could not tolerate any porosity, a single pore greater than ten thousandths of an inch. But a four pore is not going to cause the thing to fail. You do the fracture mechanics on it. It's not going to cause a brittle fracture. This thing doesn't have any stress on it. It's just ductwork. We've got 14 psi on this.
So they called me back. I said I'd be down in New Jersey for something else, and I'd come by. I came by at six o'clock, they walked me through it, I showed them how to clean it, and I left. Their next four welds were defect-free. They were all delighted, called me up. Then they called me back: "I think we're back to our same failure rate." I said, guys, did you follow the cleaning procedures I showed you? They'd already fallen off the procedures. They went back to cleaning properly and they finally got it out.
Exact same thing up here on great big seventy-thousand-dollar compressor parts. It was an old engine — I was told I would research for General Electric at the time, and the guys said, "Tom, can you help us out?" I said sure, get some reagent grade acetone, because if you leave a little oil film on there, that's where hydrogen comes from. He said, by the way, this is an old engine, we don't have any development money, can you do it for free? I go up there and spend a half a morning. There were 17 managers and engineers from General Electric watching me do the inspection, as the technician cleaned the surfaces to prep them for putting in the electron beam chamber.
After he cleaned it, I noticed he wasn't using reagent grade acetone. I got down with a 10x magnifier, and I went back over the area he'd just cleaned, and I noticed white specks that weren't there 60 seconds before. I said, guys, this room is raining dust, white dust on here. The technician would already have put it together into the chamber. At that point, all the engineers and managers figured I was just a buffoon and quit paying attention to me. I went over to a glass slide on a bench, cleaned it as well as I could, took some of the acetone they had in an old squeeze bottle, squirted it on, let it evaporate, held it up, and you could see the rings of color from the residue.
Anyway, I told him six weeks ago to get some reagent grade acetone. He got some reagent grade acetone, the next three or four parts they welded, no problem. He called me up and told me, by using reagent grade acetone — which I told him for free six weeks before — he solved the problem, and he got a letter of commendation. I said, I didn't even get a letter, and I did it for free. So you get porosity from hydrogen in titanium. That's the Boeing inspector story — they over-specify the system. But there are some specs you have to meet.
In fact, I have a student at Caterpillar, and he's been told to come up with a new physical inspection technique to inspect filler welds. Everybody's already taken all of the electromagnetic spectrum and the mechanical wave frequency spectrum to do inspection of welds. But he's supposed to find some new physical thing. They emailed me this week — he went out to operations and talked to this manager, and she said she didn't really want a new inspection technique, because it would be a new specification that she would have to meet. That happened to me 30 years ago at Electric Boat. I had a better way to inspect welds, and they threw me out of the shipyard practically, because the last thing they wanted was a new criterion they had to meet.
So how bad is porosity in aluminum? Here's tensile strength versus amount of porosity. This is 7039, the aerospace alloy. You lose some strength pretty much linearly, up to thirty percent. The elongation goes down because each one of these pores is a fracture site. But in fact, you can tolerate five percent porosity. It's not a big deal.
Student: Is that the summation of all the bubbles, or is it the largest one as a volume fraction?
They're all going to be small, because they form on solidification. You saw a variation by a factor of three or five in size — some of that is projection. They might vary by a factor of three to five, but they're all sub-millimeter size. So in general, porosity is not that big a deal.
§3. Titanium metallurgy and Watertown Arsenal [18:04]
Titanium is of interest because it has excellent corrosion resistance and it's lightweight. In fact, it's immensely expensive to fabricate. Titanium has no phase transformation like steel — well, it does have phase transformation, body-centered cubic at high temperature, hexagonal close-packed at low — alpha and beta. Sometimes you add an alloying element such as aluminum, oxygen, nitrogen, or carbon, and you raise the alpha stability. If you add vanadium, molybdenum, niobium, tantalum, you decrease it. These are called beta stabilizers; those are alpha stabilizers. Most of the time you want to make it neutral, so you throw in a little aluminum and vanadium. The workhorse alloy for titanium is Ti-6Al-4V. Just like 304 is the workhorse for stainless steel, and 6061 for aluminum.
If any of you have been to the Watertown Mall, which used to be the US Army's Watertown Arsenal, back in the days of the Civil War and even until about 15 years ago at the BRAC closing, the Army had a research reactor in this building. When they were in the first BRAC closing, they said, you can't close us, it will cost 50 million dollars to shut down the reactor and decontaminate the area. The DoD said, so what? They closed them, and now they've turned it into a mall. That's where Ti-6Al-4V was developed, at Watertown Arsenal, just after World War Two.
Recently — I used to help them solve failure analyses over there, because they remembered me — I would stop by on the way home, some rocket motor case or something had failed, and help their failure analyst. I remember walking through the reactor areas, and it wasn't operating. I remember when they were starting to clean it up, they had one-meter grid lines everywhere, because they had to take samples square meter by square meter. I'd been walking through there.
If you look at the composition and metallurgy of titanium alloys, they fall in three groups for three different industries. The first group is commercially pure, unalloyed titanium. Some of the grades have a little bit of palladium in them — a precious metal at gold prices, but only a quarter percent max — and these have excellent corrosion resistance in certain types of acids in chemical plants. Just a little bit of palladium significantly improves things. You've got these different grades which change their oxygen content. With more oxygen you get higher strength.
Here's a plot of the unalloyed grades — yield strength in ksi, 25, 40, 55, 70. It's increasing directly with the oxygen content, just like carbon in steel. Oxygen hardens titanium. So oxygen is an alloying element, particularly for the commercially pure grades, which are usually used as sheet, or tubes for heat exchangers and things like that. We were talking about schedule 10 titanium piping — it's pretty thin stuff. It's thin because you're trying to save money. The alloyed grades might be thirty dollars a pound, the commercially pure stuff costs less per pound after you roll it into plate.
§4. Sea Cliff, Alvin, and the heavy-section titanium problem [23:37]
If you get down to the alpha and near-alpha alloys, this is where Ti-6Al-4V sits, around 120 ksi tensile. The Navy's alloy is Ti-100, just like HY-100 or HY-80 or HY-130. The Navy's alloy is Ti-6Al-2Nb-1Ta — six aluminum, two niobium, one tantalum. Rather than vanadium, they use that combination. It's about 100 ksi strength. It's called Ti-100. It's lower strength than the 6-4 that's the aerospace alloy.
These are heavy forgings and plates. The Sea Cliff was built out of Ti-100. They prototype larger things first — they start making little laboratory specimens the size of a notebook, test those for properties, and then do a prototype development. The Alvin, the original Alvin research submarine, was made with a steel hull, a high-strength steel — HY-80 or HY-130 — because it was back in the '60s. By the early '70s, they switched and built the new Alvin with Ti-6-4. They just got rid of the old Alvin and built a new one — the same titanium hull was used for about 40 years. The new hull is also titanium.
In about 1980 or so, they built the Sea Cliff, which was a little bit larger. You can actually stand up in it. Alvin was about a 7-foot sphere, and the Sea Cliff was 8-foot diameter. Draper Lab had one a few years ago — they had to build two of them, and these were actually hot-dog-shaped subs, not spheres. They were going to make them out of titanium, quarter-inch, four feet diameter. They were autonomous vehicles.
They came to me — Draper hired me — and as I remember, on DoD rates they paid me fifty dollars an hour. They were on a rapid prototyping schedule. This was the early-to-mid '90s. After the first Gulf War, rapid prototyping was a big thing, because they had certain equipment they needed in order to fight, and they arrived for the first war. One of the reasons they waited six months in Desert Shield was because they were trying to get these new systems for desert fighting operational. So they had a big program in DARPA. They were rapid-prototyping two autonomous 4-foot-diameter subs, I think 50 feet long. I didn't know anything about the mission, but it was public knowledge they were building these subs. If they're quarter-inch, you can probably figure out how deep it would go. I never had any idea what the mission was.
We ended up building them in aerospace. We went down to Pratt & Whitney, places like that, where they're building titanium parts for big-diameter jet engines. They basically used aerospace fabrication facilities to make the titanium for those prototypes.
The Sea Cliff — one of the things they wanted to do was weld heavy-section titanium for the Sea Cliff, around two inches, maybe two and a quarter inches thick. It was supposed to go to 6,000 meters or even deeper. That's a fair part of the ocean. They were building at Mare Island, and they were trying to use gas metal arc welding, which is what David Taylor in Annapolis had been working on for years — gas metal arc with a wire feed. They couldn't do it, and they finally gave up and went back to what we call hot wire TIG. That's how they're building it.
The difference is, hot wire TIG is two pounds an hour deposition rate, really slow. Gas metal arc is 10 pounds an hour, five times the productivity. But they couldn't do it.
§5. The Soviet program and the Gurevich book [29:30]
One of the problems in the titanium business is that it's not a big enough industry to sustain its own rolling and melting facilities. They borrow time, rent time on a steel mill — they go in one day in a steel mill to roll the titanium they need for a month. They can't afford a half-a-billion-dollar rolling mill. They have to go roll in a dirty little steel mill, which creates its own problems. We never had a big titanium industry. The Soviets did. And the Soviets built the Alpha subs.
The first time I ever went abroad was 1979. I first read about the Alpha subs on the front page of the International Herald Tribune flying back from Europe. That was either '79 or '80. My first research contract at MIT, almost 13 months to the day after I started as an assistant professor, was to do submerged arc welding of titanium for the Navy. These are some of those welds we made in the late '70s, submerged arc welding titanium, one pass. [Tom passes samples around.] This isn't too bad-looking — it has lousy contact angle, instead of 90 degrees, that just kind of blends in. This is another one with different fluxes, was all rubble.
It turns out the Soviets had been publishing work — a guy named Gurevich. [Tom holds up the book.] Here's Gurevich's book. I can't read Russian or Ukrainian. He was at the Paton Welding Institute in the Soviet Union, in the Ukrainian capital. In April 1980, the US Air Force translated that book, Metallurgy and Technology of Welding Titanium. It became my bible. It has a nice little letter on the inside from Boris Movchan, who was one of the top five scientists — "we're glad to send you the book of Metallurgy of Welding Titanium." I still get Christmas cards from Movchan. They're nice people.
Jimmy Carter, President Carter, had an exchange program with the Soviets. Besides them coming to MIT and RPI, the guy who ran the whole program with a professor in this department, Nick Grant — he had set it up with the National Science Foundation for the whole country. The first time, 30 American scientists went over to the Paton Institute. The next time they had an exchange, I think eight went. And the next time, two. This was 1980. President Reagan was trying to shut this down. Carter had reached out, and Reagan didn't see any value in cooperating with the Soviets. I was one of those two people. I was still an untenured professor, but Julius Szekely, who has now passed away, was in this department. He still had an NSF contract, and the State Department was sponsoring it. He wanted someone to go with him.
I can give a lecture about being followed constantly by the KGB — about 40 people followed us. But I got to talk to Gurevich, who'd been publishing all this stuff in the 1960s. In about 1972, all of a sudden, he quit publishing. That was when they decided to build the Alpha submarine. All the work would be done — I estimate, if you converted it to US dollars in 1972, the work he did from the '60s to the early '70s was probably a hundred million dollars of research on welding titanium. More than we ever spent. But he did it.
§6. Submerged arc, electroslag, and semi-submerged arc welding [34:31]
We had determined that we should be using submerged arc welding. Submerged arc welding is used for steel all the time. Big line pipe like the Alaskan pipeline was done by submerged arc. You have two plates, an inch thick, two inches thick, whatever, and you essentially feed in granular flux that looks sort of like sand. A wire feeds in, submerged underneath the flux of sand. You have an arc right there and you'll be standing next to it with no shielding, because it's all under the sand. It's a very special sand — like the coatings on these electrodes but in granular form. The process was developed in the United States about 1936. There's a letter from Churchill to FDR at the beginning of World War Two about a welding process — it was the Union Carbide Unionmelt process developed in 1936. This was one of the ways we built ships.
Tremendous advantage — you can lay down 30 pounds an hour, not two or four pounds an hour, but 30 pounds an hour. Very seductive. Only works in horizontal position in big welds. So we figured this would be a way to build a heavy-section titanium submarine. That was my first research contract, to develop the fluxes. If you go through the Gurevich book, half this book is on how to make the fluxes. I ended up using optical-quality calcium chloride crystals that I was buying for a hundred dollars a pound, and crushing them up. [Tom passes a sample.] I was crushing them up into a sand and welding underneath. This is optical-grade calcium chloride, 8 to 12 mesh. It had 0.14% oxygen, 0.015% nitrogen — very good chemistry.
In fact, if you use optical-quality calcium chloride, you can actually refine the oxygen out of the weld pool. My first student ever, John Galata, who ended up an engineer at Electric Boat — that was his bachelor's thesis. He made those welds. We very quickly learned that you couldn't afford it. It was going to cost about $300 a foot of weld, just for the flux. That didn't sound very good.
About the same time, we started learning about the Soviets' sub. They had funded conferences — they weren't classified, but they couldn't get the clearances through quick enough. We went down to David Taylor in Annapolis. The Navy was not happy. The Soviets had put a dog in space before we ever got any living creature up there. That created a big awareness and increased the research funding. The Navy was unhappy that the Soviets had leapfrogged us in submarine technology. So we were having conferences right here.
I was at one of these conferences, David Taylor, listening to other people's presentations and thinking back about papers, the background on Gurevich. All of a sudden, on about the second day of a three-day conference — Gurevich, in '79, had published a lot of work on electroslag. This is a welding process that had been perfected back in World War Two for steel at the Paton Institute by Paton's father — Boris Paton's father — who was a hero of the Soviet Union. You've seen his pictures, because he used to weld the Soviet tanks back together to get them back to the Russian front. He had created a lot of welding technology, successfully welding the armor steels in World War Two — they didn't have enough tanks to keep the Germans away.
You have water-cooled copper dams. You set two vertical plates — these things weld vertically — and instead of gas metal arc, you basically put thick plates in, anywhere from one to six inches thick. You feed a wire in, and you put some flux in the bottom. So now you're using half a pound of flux, where you might use 10 pounds of flux for submerged arc. I can afford 15 or 30 dollars of flux and make a weld 20 feet tall on the same batch of flux, in principle.
I didn't have Gurevich's book at the time, but I realized this would be a very efficient process. They had certainly spent lots of time and money on it. This was the flux-shielded process we never even thought about, because electroslag on steel sort of has a problem. Turns out it's a great process for titanium. [Tom hands out a sample.] This is a titanium electroslag weld, 2-inch-thick plate. The weld between the two-inch-thick plates, no distortion. It can only be made in the vertical direction. The first titanium electroslag welds made in the free world were made in the room right next door. It was terrible — no fusion — but we made it. I came back to the conference and showed it. Boy, you can really lay down titanium metal in these flexes very quickly.
At that point, the Navy said, you can't work on it anymore. And they sent it out to Oregon Graduate Center with — what's-his-name, used to be at the University of Tennessee. So electroslag — you've got two plates, this is the side walls, you've got a molten flux, a liquid, you've got a wire coming in, and you have a metal pool that floats underneath, okay. So it's not an arc weld — you're resistively heating through the resistive flux. The flux has a certain electrical resistance. The wire feeds in, it melts and develops the weld, melting through. You're just barely making fusion to the side walls. At the end, you could take a hammer, hit it — you might knock the plate apart, because you're just barely fusing the sidewalls.
We learned that was one of the ways. I shouldn't tell you too much about how I know, but I've had to use a security clearance previously. They did get some foreign technology. I went down to David Taylor afterward, and I had — I won't say how you can get something that they had — someone basically robbed a scrapyard, a shipyard. This is 30 years ago, 35 years ago, so I don't mind telling you that much. But they had the technology. They had welds, and I came down to look at them, and a lot of them were electroslag welds.
Then they had some other welds made by a process that Gurevich had also published on, back in the early '70s. We called it deep TIG — in Russian it was "shielded inert gas heating," and when translated into English it was always called semi-submerged arc weld. It wasn't really submerged arc like the one I showed you, which is one of the things that confused the Soviet papers — they never give you all the details. Semi-submerged arc — Gurevich had found that if you put just a little painted layer, half a millimeter thick, on the top surface of a piece of titanium, and you use a gas tungsten arc torch, with the flux, you can get a deep weld. Whereas if you have no flux there, you get a very shallow weld. [Tom passes a sample.] This was made in our lab with flux shielding on this one. This is 30-year-old technology.
Okay, this weld on this side was made with TIG that had a flux on the surface. [Tom indicates the sample.] That's the fusion line. This weld was made with TIG with no flux, and you got a big heat-affected zone but only a shallow penetration. Exact same welding parameters — same current, volts, travel speed. What happens? There was a lot of research done on this for stainless steels in the early '70s. You have a change in your convection patterns when the weld pool cools. Regular TIG has a weld pool that's very shallow, the circulation pattern looks like that — hot metal comes up, the hottest metal underneath the arc gets carried off to the side, and that widens the weld pool. You don't get a lot of penetration.
When you put the flux there, you get — there's research from 18th-century surface tension driven flow. You change the surface tension of the metal, and all of a sudden the convection takes the hot stuff right under the arc and pushes it down to the bottom, and you get a much deeper weld. This became a big area of research. A guy at the National Physical Lab, Ken Mills, published a 12-page paper saying he thought this was why these things — not in titanium but in stainless steels — were changing the shape of the weld pools. People started doing — they're doing research on this in the '70s, less about titanium. I was interested, but I hadn't applied it to titanium until after the Alpha sub stuff, when we were trying to figure out how they built that.
There was getting to be lots of research in this. Professor Szekely and I, in 1983, wrote a paper where we did some computer modeling of the convection flows in weld pools, and we won an award for our paper. We were the first people to model that. The next 15 years, everybody in the world who had a computer and didn't know how to do an experiment but could run a finite element program was doing bigger and better finite elements of convection in weld pools, and didn't improve anything we hadn't proved. Ken Mills did the first hypothesis. We did the calculations to show it. But in fact, this is what happens — when you put a very thin paint layer, only twenty-thousandths-of-an-inch thick, it's kind of like welding through paint, and you get a very deep penetration.
Then Ohio State University, which became one of the big welding schools — they had a welding engineering department — Lamar University in Louisiana, Texas, most people haven't heard of it, has a welding program, a four-year welding engineering program. But Ohio State started doing a lot of stuff on this. I decided not to. It became too popular — I don't like to work on popular things. There's a lot of smart competition out there. So I would walk away from things like that. You work on something else when no one else is working — it's a lot easier to make a big impression when there's no competition.
§7. Putting it together: how the Soviets built the Alpha [48:22]
If you were trying to weld a piece of 2-inch-thick titanium in the flat position, in a shipyard, what you might do is a joint prep like this — two plates, put some little flecks of flux in here, use a tungsten electrode, six millimeters, a quarter-inch diameter, and a thousand amperes. You get a great big deep pool like that. Then you flip it over and weld another one like this, and you come along and you do a few weld passes. That's what I might do — it's the horizontal flat position, sort of like submerged arc, except it's not submerged beneath the flux, it's just blowing through the flux and changing the convection pattern.
If I were a Soviet trying to design a sub with vertical welds by submerged arc and flat-position welds, I still have to have a few other welds. I have to build into the plate with gas tungsten arc, and the way you do that in some cases — this is my favorite picture. Here are guys inside of an argon-filled vacuum chamber, wearing oxygen masks, to do welding of titanium components in argon, because you cannot tolerate air contamination.
At the Naval Air Rework Facility in 1968 in Norfolk, Virginia, they would have a glove bag — you put your hands in like a blood bank to repair the titanium disks. They used gas tungsten arc to repair, very slow. TIG welding of the Sea Cliff was how they did the whole thing, two inches thick, because they didn't have a better way. They wanted to do it by gas metal arc, which is like submerged arc with the wire going in.
But I had a chance in 1980 — as one of the two people who went over with Szekely. I told them I wanted to go to Kiev, and I wanted to talk to Gurevich. They were so eager to continue the exchange that Carter had set up — Reagan was now shutting it down — they granted us anything we wanted. I spent two hours with Gurevich, with a translator. He wasn't part of the system, he was just a scientist, and he'd been told to quit publishing in 1972 because "we're going to use the technology you've been working on, we're going to build something, and we never heard of him again outside the Soviet Union." He's a wonderful man. He told me anything I wanted to know.
I had done enough work on it for a few years, I knew when he was telling me the truth. I asked him, well, how do you do gas metal arc? Translator in between, and at first he didn't — I had to ask the question about three times. No use. They had done lots of work — I'm guessing this is what David Taylor Annapolis had focused all their research money on, gas metal arc, because it was the most productive thing for steel. "If the only tool you have is a hammer, you see everything as a nail." It turns out we never saw any evidence that they actually used gas metal arc building anything.
I had a couple students in the early '80s. I told you about Dan, who's in your program — eventually retired as Captain at Pearl Harbor. Dan did his master's thesis on, I think, high-speed movies, looking at the metal transfer, how little drops of metal melt off — the unstable processes that wouldn't work. Ten years later, that was retired between us. Hydrogen cracking, due to creep-fatigue interaction cracking — not hydrogen cracking, but the fact that the metal, if you put it under a constant compressive stress and then cycle it, you build broad cracks in the middle. So the metal itself, that's what you find out.
The Naval Research Laboratory had been working for years, and at this three-day conference down in Annapolis, they were going to talk about how they solved the fatigue interaction. They didn't know. We found out a couple years later, the Soviets didn't know either. They may not have even known about this, because we can do a better scientific job on the fundamentals in many cases. NRL is a top-flight laboratory. GPS was developed at NRL.
Student: At a shipyard, they take such pains to make sure they're getting a good clean weld — they have a layer they're playing with, they flip it, they back-gouge it. They go through such pains.
Same thing here. They do anything they can to get rid of oxygen. They have to grind it. When you're doing Monel in a submarine shipyard, they basically have to grind off the top layer because you get nickel oxide. They have to grind the top surface of every weld, every pass. If you're doing piping that doesn't need beads, and if you're doing good TIG welding, they don't have to grind on piping. If you're doing heavy-section nickel-based alloys, submerged arc and things like that, grind off the top surface, then you don't have ten thousandths of that oxide layer on the surface — otherwise it's melted in and creates inclusions, dirties the weld.
We've done lots of details — you'll learn some of it. You'll learn when you take the solid-state and the fusion welding modules — those really get into the process business: why you use a liquid positive, why you use negative on this process, how fast can you go with laser, as opposed to the Navy spent millions, not 100 million, trying to figure out how to use lasers for high productivity on the ship. You may have some little laser welding cell the size of this room, welding components in some places. But you're not going to use them about the ways.
§8. Lasers, radiation, and armor penetration [56:13]
Student: [inaudible question about lasers]
Yeah, of course. But with the laser, you have to cordon off the whole area, you have to automate it, because the laser goes so fast with this high power density that no human can control it. You can't have a torch and aim the laser by hand, plus you have the reflected light. If you go to General Motors with a laser, it'll be orange. Boeing, they might be using a laser, it'll be a locked-out area. And there are interlocks — no one can go in the area while the laser's on. You've seen the laboratories around here — flashing lights, laser on, do not enter the room. It's sort of like taking x-rays at the shipyard — you've got to have 15 people standing watch to make sure some rat doesn't run in there. Maybe not a rat, maybe an employee, but nonetheless.
Student: I have a question — bear with me. Have you seen the movie I, Robot? There's a guy who's working in a robotics facility, and he goes in there to radiate these metal frames. I just always wondered — is that, do they actually harden things up with radiation? Is radiation ever used to harden something?
Well, radiation — a nuclear reactor hardens the steel vessel. You have to scrap it after 30 years because it's been embrittled. So yes, radiation will change the crystal — the location of vacancies and interstitials and the atoms in the crystal. But as far as hardening it, I don't know of something specifically hardening with radiation. We have used lasers — in fact, Ben Wilcox, back in the '60s, when he was working at Pratt, used high-powered pulse lasers. If you hit something hard enough you can actually blast it apart. You send stress waves of 300 ksi through the material, and you can stress-relieve. We did some of that in the '90s, trying to use laser stress relief. You blast it, do little pulses, and all you do is leave a little divot on the surface, which you can grind off, but you hit it so fast that you send a shockwave through the metal, and that shockwave is a mechanical stress.
Usually this order of heating with photons, the power density is 10^8 watts per square centimeter. A laser, electron beam, never goes above 10^6 to 10^7. You go up to 10^8, 10^9, you're hitting the material with heat so fast that it can't melt — it just sublimes off the surface. For every action there's an equal and opposite reaction, and you start getting to 10^8, 10^9, and you can knock missiles off course with the reaction of hitting them with a pulse laser beam. Lasers in space — they're not going to melt the warhead. They might blast a hole through it, which could mess up some of the control circuitry, but one of the things they do is knock it off course with the shockwave.
We didn't talk about this with armor. The Army can hit — suppose a tank round, it could be a 36-inch-long, three-quarter-inch-diameter rod with depleted uranium. They use it because the high density, but what penetrates is the kinetic energy. Kinetic energy penetrators, just like your rail guns. It actually sharpens itself as it's penetrating, and you can go through as much steel as the length of the shell. I've seen down at Aberdeen Proving Ground, 36 inches of steel penetrated, one shot.
I remember as a student, working over at Watertown Arsenal for the Army, they told me the story that in about 1984, '85, they perfected this so that there was no armor that couldn't be penetrated. They lined up three old tanks on the battlefield, and they had an artillery general — I don't know what stars he had — and they shot right through all six layers, through the side of all six tanks lined up. Apparently he tossed his cookies right there, because he was just afraid that some shoulder-mounted RPG was going to be able to wipe out his tanks. Well, within two years, the full thing flipped, so there was no armor — there was no munition that couldn't be defeated by the proper armor.
That's why they developed active armor. You basically have a surface of explosive on the outer part of your armor. Maybe your armor is two or three inches thick — it's got a quarter-inch layer of explosive. When the sabot round hits it — it's going four or five thousand feet per second, supersonic — it hits, and the explosive goes off, and sends a shockwave through that 36-inch penetrator. That shockwave, in addition to the force, shatters it. It's already got stresses of a couple hundred thousand psi in it, and you send another shockwave of a couple hundred thousand psi, it shatters it. They have high-speed movies, and you can see it breaks up into little pieces.
Student: [inaudible — about speed of sound?]
Yes, well, the speed of sound is 5,000 meters per second. You do the calculation. I used to think the Army was sort of backwards in technology compared to the Air Force and the Navy, but when I got on a couple of Army science boards and actually saw what they were doing — they don't advertise a lot. When you get into the weapons area and armor and penetrators, it's better than Star Wars. The Air Force has ideas — their guys are watching too much Sci-Fi, and I think they can make time travel work. The Army actually comes up with practical things. One of the things they did — how do you defeat active armor? You have a sabot round, the cap comes off, but you have two penetrators. The first penetrator goes in, yes, it's shattered, the other one falls through the hole. No more active armor.
Right now, one of the best armors is actually electromagnetic armor — just hanging wires off the side. The thing comes in, you set up magnetic fields. I can't remember exactly how it works. But I'll tell you the armor on the MRAPs — I can tell you what it's made out of. They have about six inches of, six pieces at three-quarter inch, glass. I can't tell you why it works — many of them can't tell you why it works — but they test it. That's an interesting story.
The Secretary of the Army — we were losing all kinds of soldiers after we took over Iraq, to the weapons of mass destruction. We went there to get weapons of mass destruction, but they couldn't find them — they found IEDs. So we were losing soldiers to IEDs. The Humvees were not resistant enough. Secretary of the Army goes down to Aberdeen and says, you've got to solve this problem. Within a year and a half, we're not losing any more soldiers. One of the things they did is they had a four o'clock conference call every Friday afternoon with the commanders.
So the scientists — most of these laboratories, if they're given a strong mission, they can produce. But then just after they've solved that problem, they just continue to exist, and then they're just a trashcan where we throw money in and burn. They don't have a mission. I used to see that at Watertown Arsenal. I think the Secretary of the Army came in and said, "you guys solve this problem." They have resources — the Army has most of the resources in the budget. They knew what their problem was, and they had assault. They sent these things off through all kinds of different methods. They didn't use just one technique, but they got the feedback and put some things very quickly into the field — trial, get feedback on whether they work or not, what was wrong with them, from the captains and lieutenants who are out there searching for the devices.
Student: [inaudible — about RPG vehicles getting punctured]
MRAP vehicles or equivalent Humvees — they were getting punctured as well. EFP — explosively formed penetrator — copper, performing very low — okay, go ahead.
Student: I was wondering if anyone had solved that problem.
Yes, they did. In fact, that's where the glass came in. In this rush program, the Secretary of the Army told Aberdeen, "solve this problem." They tried titanium, steel, aluminum, glass — five or six things. I was on a review panel and they were telling us the story about how they did all this. There were 40 or 50 people around, half of them Army, half of them — we did the panel for the National Academies, I was on that. This woman explained how she first did a screening test: let's take some of these RPGs and fire them at these layers of armor and see what works, and what we're going to put on the bottom of our MRAPs. What worked best was glass. You don't usually think a parental material is going to be your best thing. The reason I can tell you this is — any Army soldier out there running around in an MRAP can see it's fading glass. But why it works is classified. It does work — it defeats things even better. Why don't we go ahead and take a little break, I'll go get one of those.