WM_Su2015_16

Okay so this is another story of stress corrosion cracking. This is from an Army Black Hawk helicopter. This is painted gray, so what type of [Black] Hawk helicopter is this? A Navy bottom. And the Coast Guard has Jay Hawks. The presidential talker [Hawk], presidential, and if you go down to the Sikorsky with me, they actually have a separate engineering [group] from presidential Hawks. They don't make that many, [it's] the last [Black] Hawk you'd expect a special attention.

So anyway, in the early life of the program, [in the] first couple years, they had a Blackhawk go down in Arkansas, [on] night vision goggle training exercises, and six soldiers got killed, the pilots got killed, and another soldier was in a coma after the accident. And so there's a lawsuit. And it turns out this is a washer and this is the nut that holds it on. I'll pass these around in a second. But the number of holes in the washer is different than the number of holes in the nut. And what you want — they don't want to have a keyway, which is going to be a stress concentrator. And this is holding the tail rotor together, so you have a tapered shaft for the tail rotor, and you basically just squeeze the shaft onto the thing with friction and hold this thing, keep it from rotating. And there'll always be a nut being screwed [on]. So instead of a keyway, you put a screw between the washer and the nut, and that keeps the thing from [coming loose]. So don't trust any kind of wire, you know, wire or anything, or polymer inside to keep the things [from coming] loose. We can actually put a positive — yes, it's a pretty sophisticated design. This is a four-hundred-dollar washer okay.

Aluminum, and different colors for different services, but it's got a little oxide Martin hard-coating. Here's the shaft that will slide in here, which does have a little bit of wear, okay, where it cages with the rest of the thing. So anyway, what brought down — the question was why we had the failed washer. I don't have the failed washer, this is just the exemplars. And it was stress corrosion cracking. You look at the microstructure and the fracture, clearly the stress corrosion cracking. Well, [there] should have been enough stress on this thing for it to crack. This is, remember, 7075 aluminum alloy in the T6 condition. Whoever — I told you there's a problem [with the] T6 condition, has the maximum strength properties, the peak here. And it turns out Sikorsky divides their helicopter fleet stuff into different facts, different models of helicopters. And so all the white Hawks [Black Hawks], they've got an engineering group over here. The next helicopter, an MH-53, another group over here. Today, you know, they [have a] whole bunch of different models of helicopters, but each one has its own engineering group.

One of the other engineering groups had found out they were getting stress corrosion cracking of these washers, and they realized we should have made it out of T60 treatment, we should have made it out of T735 — sorry, T651, okay, the chance much thinner on them. They should have made it out of T7, supposedly out of T651. And the 51 after the six is how you stress relieve okay.

So there's actually two points to the story. So in crashes it's also about the people who died or are in comas and things like that, and Sikorsky and both the government contracts — you might note the government contractors take a little progress [precedence] here. We're going to talk about — Supreme Court, '90's almost twenty-five years ago, ruled that you can't sue a manufacturer of military hardware because military hardware is designed to be pushing the limits of technology, and you're going to have failures. Sort of like Eric Schmidt [at] Google says, if you don't have failures, you're not doing your job, you're not pushing hard enough into new technology. Plus you've got NAVAIR, or in this case the Army probably had the lead on the Black Hawk, but you've got a whole group of government scientists who are reviewing the design. Of course you have the project, [the] DOD Black Hawk and Sikorsky operative, everything else, but they have design reviews. And they come in, the Sikorsky or Boeing or whoever, would come in, they would tell what they're doing on the design, and spend several days, and have a bunch of engineers or scientists — DOD doing a bunch of engineers or scientists — and the government [is] responsible for the design. So you can't sue the manufacturer for a defective design. That's the government contractors' capacity. So Sikorsky was the government contractor, they just built what the government approved, and the government said we want the best technology, the most forward-looking advanced technology, okay. And so Congress said okay, they're going to protect the defense contractors from design flaws, okay. But they don't protect them from manufacturing flaws okay. They don't manufacture it to the spec, they're dead, okay.

Well Sikorsky basically came in, their attorneys came in with the government contractor's defense, and these military widows and stuff were going to get nothing because [they] go with contractors. The best — well, turns out two weeks before my deposition, we all agreed stress corrosion cracking, we all agreed it was the wrong T treatment. And in fact Sikorsky had discovered, because the Navy [is] the corrosion leader in the services, and [they] found these things in their plastic bag, never used, sitting on the shelf, okay, with cracks in them. They'd never been put in service, and they had little cracks.

Now this is one — a Navy one that we got — that we saw cut to measure the residual stresses. And we, by measuring the flats, the distance between the flats before the saw cut and after, we could calculate that this thing had about 10 ksi stress because its springs — its springs closed. Turns out this is even more complex, [there are] tensile stresses on the one side, [it's] compressive on the other. But so anyway, we got an idea what the stresses were. And it turns out the T16 treatment — I think this is 7075, I think 7016 or whatever — is anyone actually capable aluminum alloy, and it had residual stresses, and the stress corrosion cracking susceptibility was 7 ksi, and we had 10 ksi. That's why [it's] cracking on the shelf. You guys are out in this humid, moist, salt-laden atmosphere, and even though it has a bag, you know, little moisture humidity gets inside that bag, and a crack [grows]. So you've got — now I draw my three circles, I think you guys understand three circles now okay. You had the stress from residual stresses, you had a susceptible microstructure — it was the T16 treatment, supposedly [had] stress relief — and it turns out that they — but it had been stressful, you mentioned it wasn't certainly, but I didn't know why.

So I got the drawings, which are sending the government certified drawings — we [got the] drawings — and it turns out T653, T651 P treatment 51 means you're going to take the bar — this is made out of, let's slice it up here — take the bar and you're going to do a mechanical stretch, 1% strain, which is the plastic region, and that will relieve the stresses in the bar. And the five just means it's mechanically stress relieved okay, one percent. There's a 52 treatment where you stretch it three percent, okay, this is a one percent. And it turns out what they've done is they both — they bought a bar of material, they sliced it up into little pancakes, and they did all their anodizing and drilling holes and machining, and when they were all done they had bought the material as T651, so they figured they'd good stress relief. But after they did all this machining, they treated it but they didn't stress relieve. How do you pull on this thing in tension with the T651 P treatment? You can't grab this and pull it in tension. What they should have had on the drawing was T652 — the 2 is a compression heat treatment. I can take a disc like this and I could squeeze it by 1% or three percent, and I can stress relieve in compression. That's not what the drawing said.

And Sikorsky's chief of rotors testified the day after me, says oh we bought it in the stress relief condition. That's sort of like the guy who didn't preheat the weekend before, alright, [and welded] okay. Oh yeah, we [were] doing stress relief once. It's sort of like the New England Aquarium has stainless steel on the outside structure, supposed to look like fish scales, big sheets of stainless steel, and they bought it in the passivating condition, which grows an oxide skin, keeps it from corroding in salt air forever almost permanently. And then they go along, they do mechanical abrasion to give it some texture. Well, they bought it in the passivating condition, and they scratched off the passivation, okay. So guess what, it's not passivated anymore. So I mean, there's three examples of things — they've done the work, well we bought it in that condition and then we transformed it, so [it's] not in that condition. But we bought it in the right condition for stress corrosion resistance.

Well it turns out Sikorsky realized they had this problem before the accident in this other group, and the other group kind of communicated saying hey you've got to change out all your T6 washers in the whole fleet, 1300 Blackhawks out there. And they put them on order, [it's] going to take fifteen months to get four hundred dollars apiece — they did wrong order. The accident occurred nine months later. One of the things in process to be changed out, and this includes the presidential helicopter folks, so the presidential helicopter had lost its tail rotor authority, [it would] crash, there would have been a bigger inquiry okay. But anyway, so they didn't — they didn't tell the presidential helicopter [crew] didn't [know] back then. And it's just, you know, another business story of, you know, they actually knew what they had and stuff. The chief of rotors testified, nonetheless two weeks later Sikorsky paid ten million dollars.

Okay, so where does about something — guess what the force that would give me that compression? No, you have to — that's 6061, so it's about 40 ksi yield, [you] measure the cross-sectional area, [and] if I have to squeeze, over 30 tons of force or so. Just like — okay, well the Army could play it, [we] could be like, what about a game — they likes it to go a little bit.

And in fact this leads into — I showed you this thing before, which shows the types of heat-treated aluminum alloys. And they can be as-fabricated, the H [for] full of work, or the heat-treated alloys, solution treated, or heat treated T1 through T10. And then you can have other numbers after this. I'm not going to go through that, there are a whole book of aluminum alloys and their temper designations, okay. So if you'll read the book.

Okay so you can stress relieve by stretching or compressing, and here's actually TP52, may be as long. Well what they wanted to have was these, [they] might [have been] T752 okay, but the drawings I suspect, the drawings were T751, T752 is compressing the thing okay. And I think I did mention to you guys the back, then [we] go to the Davenport Iowa plant where they make the four to six inch thick plates. That turns out the alloy that Alcoa makes — an [Alcoa] Davenport Alcoa makes for Airbus wings and stuff is 7050, [the] 7000 series Al alloy, more modern, more corrosion resistant than 7075, higher strength too. Though entire spring, but if you go to thicknesses, you can buy this in T7651, T651, and you're going to find that the T7 is the over-aging heat treatment, so you lose a little bit of strength.

You're not — what's going on here. Oh, other — 7050 to 7075, different alloys. So the old alloy, they did over-temper and it was susceptible to stress corrosion cracking. Nowadays they developed a more stress corrosion cracking resistant alloy, and it has a little bit higher strength too because they've optimized somehow, I'll [not] know us. So in this one, they actually get better strength, better corrosion resistance, better elongation, okay. The yield strength goes from 61 up to 66, whether you're talking up to — what is each, eight, what — up to three inches thick. If you keep going on this thing, this is T7451, T7351, this is 7075-T7351, and this would be three or four inch thick plate they can make. This and up to five or six inch thick plate for bigger weeds [welds] for bigger aircraft, okay.

And this way you're going to machine away ninety percent of the weight out of this plate okay. I mean aircraft are weight critical. So you start with the thick plate and you're going to machine away ninety percent of the weight. The Air Force calls it [the] buy-to-fly [ratio]. Everybody from the industry calls — the Air Force is the big purchaser of aircraft components. You talk about the pounds of metal purchased as a plate or a big forging that you're going to machine down [into] a rotor disk or got to machine into a wing, and you could be paying five or ten bucks a pound for this aluminum plate, and you're going to — you go to machine away ninety percent of it.

Okay, you're going to machine away ninety percent of it, so now bless your material costs on your flyweight. If you pay ten dollars a pound and you machine away ninety percent of it, you're paying a hundred dollars a pound for the material you fly okay. If you're talking nickel-based superalloys, you could be paying a hundred dollars a pound, and you multiply that by ten, and your engine cost material for [is] a thousand dollars a pound of actual engine weight. And the Air Force used to have 32-to-1 buy-to-fly ratios. Most of those are down below 10 now, because in the 1980s they had a huge program called near-net-shape manufacturing. So rather than just making a big round cylinder for you and making a disc out of it, machine the way ninety percent of the way [we] accepted this other way. They now can forge something so close to the original [final] shape, they spent billions on that, [near-]net shape manufacturing, to save many billions.

Yes, okay, I hope that there's some — they collected itself, someone else, a chance — they go back to the [vacuum] melt available, because this is — in many cases triple of the air, the nickel-based alloys are triple vacuum melted, vacuum arc remelted, or [double] vacuum melted, sometimes triple vacuum melted, and vacuum induction melted, okay. So you're getting rid of all the nitrogen, hydrogen, oxygen, down to very low values. How do we make planning [a landing gear] at 250 ksi yield strength and not get hydrogen cracking? Well because we vacuum melt it three times, [there's] not a lot of hydrogen left in there. It's below — its below a half a part per million hydrogen. But if you try to do it in the air, which is where we melt most of the steel in the world, just the moisture in the air —

And when you're welding, did you know, you're never supposed to weld in a shipyard, part of the Navy spec, if they're above eighty percent humidity okay. Were you ever in [Newport] News? How many days did you shut down operations of welding because you're over eighty percent humidity? So that proves that they never get above eighty percent humidity in Newport News right? Some special burnham throw in second shift. I will say that — oh, shoot bay doors over here in the wintertime, of course all closed, it's freezing cold outside and spring [in] between, and ice and second fall, and [it's a] nice comfortable 70 degrees, it's like crazy and feels great, and they have all the doors closed and it's like why would they open [them]. Again, well, you know, that lot, big recent — all the big asses [vessels] off the levels are supposedly people containers — out all over again, little — but they don't want the atmospheric interference happening with what they're doing, so people —

Anyway, well you can actually see the fog in the summer over the ocean, okay, with oceans cooler than the air, and so sun comes down, you start seeing a fog. You don't want to blow that one hundred percent humidity inside the hangar okay. So there's a reason for it, and in fact they've set up preheat procedures and things to allow for it. But if you haven't seen that, they suspect — for 30 years but I remember the spec said do not weld more than eighty percent [humidity]. And of course the [Newport News board] chief, you know, if you're down [in] the tropics, it's never above 80 something. That starts to go worse than Pascagoula apparently. It's just the armpit down there.

So do you know how much moisture content is in the air at one hundred percent humidity? Let's say 90 degrees Fahrenheit. You learn in high school that air is seventy-eight percent nitrogen, one percent argon, twenty-one percent oxygen, right? That was tried — and about 90 degrees Fahrenheit, you can look it up in the humidity tables and the steam tables, and you'll find that about five percent of your air is moisture. What do you think rain comes from okay? You know, you hear about the rainfall in Texas and how they're having floods from the rain in Texas. One time I thought, [where the] hell, where's all this fresh water [come from], it's up there. And you start doing the calculation — for even if it's only 1% moisture in the air, which would be 107 [percent] humidity, that's what cloud is. Or it's when you get close, 107 humidity and they have clouds out there, I've seen them, I don't even have to go up there and measure okay. But figure out what the temperature is, figure out what the humidity is, and assume that you've got two miles of thickness of air with whatever the density is, this one percent moisture, and you find there's enough moisture, because you see this [weather] system go across the United States and it wets the whole country for things, 3,000 miles. That's because it's quite — [the] water can wet the whole country for tender clothing with all the moisture that's up there. There's a tremendous amount of fresh water up there in absurd okay. And when we get big heavy rainstorms of an inch, it's just a small fraction of what's up there. Do the calculation guys, don't be afraid to do the calculations, okay. Sometimes you're surprised by the answer. I'm surprised to come.

In any case, these things have five to six inches thick, they have different properties in different directions for the rolling direction okay. This is actually fracture toughness in different directions, and you have different fracture toughness — the thicker, the less mechanical work, the finer grain size you've got — the lower the properties okay. With it, you're not going to start building aircraft out of this thing unless Boeing — Boeing's not going to buy it and put it into a wing unless Alcoa has done — Alcoa and Boeing together usually have done millions of dollars [worth] of property evaluation. And it's not just strength. If you were back a hundred years ago, it was just yield strength, tensile strength, and elongation. But certainly since the 1950s, fracture toughness, which gets into fracture mechanics, that gets into fatigue, gets into critical brittle fracture and all these other things. I mean, you have to know all the properties.

When I said it cost fifty million dollars in 1960s money to develop HY80 [and] HY100, yeah they were doing full trials. But you can have a whole 200-ton heat made by US Steel for, listen, probably half a million dollars back then. Today, if you want to develop a new steel, you probably [are] in several hundreds of millions [of dollars in] different qualification tests.

And this is another problem with our pressure vessels. You guys still use ASME pressure vessel steels. Most of those, almost all of them were developed in the 1940s, and it would cost probably half a million or billion dollars to qualify new steels for pressure vessels. The Department of Energy's tried to qualify 9% nickel steels — or I'm [sorry, said] nickel-1-Moly — to use higher temperatures for some of the nuclear reactors and things. They've been doing this for 40 years, and people are still somewhat hesitant, [haven't] just got a big enough database out there until you actually build, start building prototypes okay, and get experience. And that's why the Navy, before they started with all HY100 hull, they built a couple of full-size 30-foot diameter sections for a couple of boomers back in the 90s, and put them in [service], even though then we needed HY80 [and] they wanted to get the experience with welding it. And even when they did go to a full-sized [ship], let's see, [Sea] Wolf in there in the 90s, they still had major problems even though they don't try to build it way up okay. And one of the reasons for building things like Alvin [or] the Sea Cliff, [for] things like that, was part of the prototyping research exercise. [They] build small submersibles okay, deep submergence things, but it also gave them experience with fabricating what they hoped would be the next [HY]-series alloys.

Okay so there's lots of different things. When you get to the heat-treated alloys, they put silicon in 4043 [which] is high silicon, weld filler wire, copper — that's the original Wright brothers aluminum copper alloys. The Wright brothers developed [it] back 100 years ago. We learned about pretty copper with aluminum that — copper, or that magnesium, [are a] couple of others — over [the] years [are] serious problems for corrosion fitting [pitting], because the precipitation hardening — you've got little specks of copper alloy. Copper's more noble than aluminum, so now I have little cathodes surrounding my anode. Pitting corrosion occurs [at the] anode okay. So I will actually have little galvanic cells just eat away at it.

And I mentioned [to] someone after class yesterday that — I think I mentioned in class — that does magnesium anodes look like a hot water tank at home okay. They basically extrude magnesium over a steel wire and pick the steel wire out, and this becomes your anode to grow, the sacrificial anode. Well this magnesium has to be extremely pure, less than seven parts per million. The reason is nickel and magnesium do not mix [in] equilibrium proportions. The solubility of nickel in magnesium is only about seven parts per million. If you have more nickel than that, you'll get little nearly pure nickel precipitates in here, and your magnesium anode will just consume itself by galvanic action. You'll have local galvanic cells from here to here, here, here, and [you] talk about a thing two hours later it's completely gone, and it didn't do anything to protect your vessel, your steel vessel. It just protected itself in some areas where [there was] nickel, and it corroded itself where you didn't have it, which is everywhere else, so [it's a] cathode. The Sun looks like Swiss cheese. So some of these things get to be really fairly tight on the composition control for corrosion.

So we've got magnesium — I think I told you this, the largest application of magnesium in the world is alloyed with aluminum okay. As you got 5% magnesium in some of these 5000 series, I always [be]lieve in the 7000, you got several. The 7 manganese, chromium, zinc — but in there, spraying fire titanium [is a] spring fire, don't have people in [it]. Wins zinc in the 7000 series — we actually form magnesium zinc precipitates, [you] start getting fancy in your precipitates, and the silicon can be in the precipitates and stuff. So a lot of technology goes into this stuff, nowhere near the technology we have in steel because of the different volumes that are used.

Now this is — chose [aluminum] silicon, aluminum copper, aluminum magnesium, aluminum [magnesium] silicon. So this is basically 4000 series, 2000 series, 6000 series, and 7000 series. This is a composition of the weld, and you have regions here. This is looking at a weld with the crack sensitivity. Remember I showed you a phase diagram. Most of all of these primary alloying elements in aluminum have a phase diagram that looks like that, whether you're talking about copper, magnesium, or silicon. I think they all have something looks like this. Well this is one hundred percent aluminum, and this is — we're going up as you add the alloy element. You go from 660 melting point — a little — virtually everything you put in there lowers the melting point of pure [aluminum] well. That's not a problem, this is about 555, 560 depending on which alloy [you] line on. But what you don't want for welding is to be in this range, where you have wide freezing ranges, beginning of solidification in this [liquidus, end of] solidification — this is liquid, this is liquidus, this is solidus, this is solid. You want a relatively narrow freezing range okay. And basically what they're plotting here is crack sensitivity, which is equal to the solidification — the very bottom of solidus, very bottom is a mixture of solid aluminum with a little bit, valley government [eutectic], plus these little precipitates. But these little precipitates cause pitting when you put aluminum in [a] corrosive environment. Might see what they would cause pitting if you put it in something like your Iowa sewage treatment plant okay. That's why we do plain aluminum on the surface, most of your cookware is nearly pure aluminum or 3000 series, which is [near] pure aluminum and doesn't have these little — / super strength, who needs a pot that has 60,000 pounds per square inch strength okay.

So basically there's certain ranges where you have huge amounts of cracks in the base, and on the hood is showing you, by the way, you avoid certain composition ranges. You don't want to be between one-half a percent and four percent magnesium in your welds. So you make either highly alloyed your weld or you don't alloy at all. So it turns out, [you] start looking at — yes. If you're not [in]volving high strength values, [it's] not a problem. But there's not much, the Navy doesn't know — you don't like driven structures, although as you get into composites and other things, it turns out that the two chief weld metals for [aluminum] are 1150 [and 4043, pure] aluminum. So if you basically weld with pure aluminum [as] the weld metal, you're going to be trying to work on this side, and very unlikely [an alloy] provinces [provides], the weld metal is not going to have great strength. You go with the 4000 series, so the aluminum silicon, 4043, and you're trying to work over this [side], so you're putting enough silicon in there to have a short freezing range, because the weld metal's [stiff] so much. So those are not the only weld metals — such as everything in here, sometime, somewhere [I'll] show you, for welding there's lots of choices of filler metals, and [it] includes [several] for different reasons.

And they're cracking tests for aluminum just like they're four steel. This is just making certain or patches just to count and help you know how much crack. You guys go around the base, make a little circular patch, [you'd] see a little boom, cracks horse [coarse] conceal okay. This is another practice, this is a centerline crack going all the way around okay. You pick the wrong patches. So here's 6061 made with 1100, and here's 2219 [made] with 1100 — 1100 filler wire — so if you don't [pick] the right combinations of filler wire.

So it's actually harder — when someone says — come to me, if they come to me, say I want to repair something to steel, I say okay, what's the hardness, what's the chemical composition. From that I could go to South and OD or American Welding Society, [in] no time I can write a welding procedure with those two pieces of information okay. Tells me I'll be able to calculate preheat okay, figure out what welding wire they want to use, and we can have a procedure. Aluminum, I need to know the chemical composition, the full chemical composition of each, how would you weld two things together — they don't have to be the same alloy okay. You start welding two [dis]similar aluminum alloys, it gets even tougher to pick [the] right [filler] go, okay. And the welding process has different amounts of dilution, so you have to worry about those things. But there are welding books on aluminum and you look these things up okay. Any questions?

So this is — this is a comparison of the distribution of yield strengths in 7075-T6 aluminum sheet. So out of these specimens, 4292 mil [test] okay, forty percent of them had a very narrow distribution of yield strengths. But then this is empty, this is all 4180 specimens from a single sheet, have a narrower distribution. But whether it's steel or any other metal, you're going to have a range of properties. And we usually design for specified minimum yield strength, SMYS okay. So the designers usually specify the end of design, [I'm] assuming they got the worst, of course, piece of material there okay. And that's sort of what happened with the [Sea Wolf], they thought they were welding HY100, they thought they had a filler wire that would be an HY100 filler wire, because the range of variability, [they] got something more like an 8130 filler wire. They needed tighter hydrogen control, they didn't have that, and they welded up eighteen percent [of the] ship with a filler wire that was a little [too low strength].

Let's talk a little bit [about] aluminum. One of the problems with heat-treatable alums of [aluminum] — I told you yesterday, 5083 looks like the 5000 series — we just use alloying elements to strengthen, and they're only 20 or 25 ksi, which is less than the strength of steel, but it's a lot lighter. So the overall strength-to-weight ratio is better for aluminum. You can make lightweight ships out of [non-heat-]treatable aluminum hulls. Well, you go to heat-treated alloys because you have to go through this precipitation, you know, the solutionizing and precipitation treatment, temper 8. Just like if you get the high-strength steels, you can't match that in the weld metal with aluminum alloys. You can with steels — we can weld HY80 or HY100 which has been heat-treated, and we actually have weld metals that can give us match the strength even though they [have] cast structure. I don't want to get into all the metallurgy of why, but we can match the strength in steel weld [metal] all the way up to a couple hundred [k]si on yield. Aluminum, we can't do that. It doesn't have the same properties — transformative, transforming crystal structure from FCC to BCC and stuff. Aluminum only has one crystal structure FCC through all the temperature range. So it turns out, if you've got a [heat-]treated aluminum alloy, you're going to have to play some games about getting more weld metal in that groove then you can fit in the groove okay. So what you do is you put doubler plates on, and you fill it well [thoroughly]. So now I've got more weld metal, and if I pull this in tension, I get hundreds [percent of] strength regards to joint, then you're welcome.

[Student:] Did you weld the joint before you put the floor plates on, or something about — ?

[Tom:] No, well, you may have, yes — what they have, [is] usually not okay. Because you're — you've got plenty of strength from the plates on top, you don't bother, you just fillet weld them together [or] tack them, place them on, and well in this [way], because if you made plate large enough, you don't want that extra step of welding both okay. More incentive, you can do it like this where you have skip welds on one plate slightly, you can overlap, lap weld off, and it's all you really need. But all these things add weight, but that's what you pay for in aluminum okay. If you want high-strength aluminum, you're going to have to add weight, or use [aluminum] in the first place, right? So if you want to get more, back, more lightweighting, you know, you're adding weight to the joint, and you got some [waste], right? I mean, more or less, the amount of weight savings that you have is so much more than the extra [weight] up at all right? [You'd] try to find the joints here today exactly, it's a trade-off. But if you're in the non-heat-treaties, which is what you guys are building your LCS and things like that out of, [your] superstructures and things, there's not this trade-off. This is in the heat-treat[ables]. While I've got heat-treatable to save even more weight, I can use half the thickness with double the strength, but I got to pay back about, I got to pay a tax on, which means my joint's a little clumsy okay, in terms of, you know, the not's nice smooth clean lines.

The other problem is you don't want to put your weld metal at the corners. You know, Reggie blue [Reggie Pelloux] used to track around here used to say something won't fit all of us has been welded. In 1992 I gave the [keynote address?] of the in — internet runs to the World [of Welding] in Montreal, and this is a keynote talk for this international welding conference every [year], and I got up and I said, I thought, Freddie Pelloux's view graph said something won't fail [unless] it's been welded. The dead silence in our 501, all the engineers, and then there's so leave a little [a]fter ripple through the room as they start thinking about it. But that's the attitude. In fact I teach, when you take my [sauce state] [course at MIT?], I give you an introduction to welding in the [sauce same] [course], the first time, I say hey, for [the] rules — chief rule of welding is eliminate all joints possible, because it will fail at the joint. But there's a reason we put the joints at the most highly stressed locations, of course it's going to fail there.

And many times I get some clown, I mean some expert, on the other side, [who] said oh, it's obviously defective that failed in the joint. No, it failed at the corner of the weld, just happened to be there okay, it's not a cause of the [defect]. Well, that — actually was in steel, in most cases it's just as good, but you located the stress concentration point at the corner. So in aluminum you do all kinds of things that you don't do — don't need to do in steel. And you bend the knee a little, because it's both of these two then, and then you put your wells at other locations. There's a quick little wells in these corners here, just, you know, doing twice as much welding, but you don't run into fatigue problems. Does that work?

Here are some other designs okay, you machine the surfaces — sometimes it's getting expensive okay. But now you can put a well [weld], get good backing. The guys that we were designing aluminum structures have a different mindset than the [guys] doing welding steel. The guys welding steel just take two plates, you stick them together at some angle and you throw a weld. The guys [doing] aluminum [know] it's not that simple, the metallurgy [is] different, and you're willing to pay a little more for joint, [the] proud to make sure you get the properties across the joint into one. In steel, [it's] actually very forgiving. And, well, believe it or not, we have lots of problems with it. That's good for me okay, but nonetheless, it is fairly forgiving.

I mean I can show you a bunch of rectangles, here are some things to buy clothes okay. So ordinarily in steel, you just put a weld right, you know, [for a] pipe, you know, well steel vertical base always. But in aluminum, you might actually weld it and then grind [in] question, and put a sleeve over it, and sleeve to be a split sleeve, to have some plain show. And then you fillet weld around that, and now you're going to get a full strength [joint] okay. That's full strength. You could do a double spigot joint. This is done a lot of steel pipe lines for water pipe and things like that, easy to assemble [for] fillet welds, don't have to have much lower skill. But it's also type of thing we do in aluminum quite well.

Here's, oh here's a very common one. When you have [stress] structures, you put in a gusset plate, and you weld here to your HP 2010, do you put long wells [welds] on the gusset plate, write it on a circumferential well [weld] which is the limited relief okay. Those linear gusset plate welds you make the gusset plate dig it up [big enough], and you know, get enough room even have plenty of weld-known. Even if it's half the strength, give you the full strength of the joint okay. What you want is the failure to be in the base material. In fact, that's often what I look at when I see a failure — did it fail in the base metal, or did it fail on the weld? Okay, if it fails in the weld metal, I have to start asking myself why did it fail in [that location]. I got a problem, right.

I've even seen, I've seen one picture of it. They were stacking pipe for a new pipeline out in Arizona, and this stacker, which is just like great big forklift that can carry about five or six lengths of 40-foot diameter, 40-foot long pipe — that, you know, weighs tons. It's just a huge forklift. Well it broke, [and] they built a special bracket, and was just the carbon steel weld. When I looked, [there's] no deformation [of the] I-beam. So they made this thing out — they didn't bend before the weld broke. I mean, all I had was a picture, kind of like here to the back of the room, of the failure of these things that are this size. And I called up the insurance adjuster, said defective welds. He said, how can you tell? I said, well, the [base] metal didn't bend before the weld broke. And I got a book by Lincoln Electric that's been around for 50 years that says for mild steel, properly made fillet weld will outpull the metal for any direction or magnitude of load. Mild steel — 9 — HY80, but most of steel weld is mild steel. And I've been on this a number of times. [The] actor went to the Supreme Court of Massachusetts 35 years ago, where I got in and testified. From the [start], this guy was jacking up his truck to do some work, and he was using the ICC bumper, which is the [thing] like being [thinking] about so when [a] car runs into it, doesn't undermine [under]chopper. Who I [set off, and so] the ICC bumper had failed. He was jacking by the ICC bumper, and it just lifted straight up, no bending of that little, you know, three-sided frame steel. The welds broke. And all we had was a lousy little forward, totally photo from years before, no other evidence. And I told the jury that was defective welds, and the other side took it to the Supreme Court [of Massachusetts] saying he couldn't say that. But I had the reference from the Lincoln manual okay. He says any properly made weld in mild steel will outpull the metal for any direction, size [or] load on a weld okay. [So that's the building], thank you. That's case [law] of Massachusetts down how much an expert can say. Well, they know them, you know, something you like.

What is — decidedly, this is actually a master chart [I want] to show you, the last thing here, a selection of [filler wires] alloys. You have one base plate composition here, one base plate composition here, and this tells you — in the middle, you go down this column, find an intersection, it will tell you what filler metal to use okay. The balance [is your job]. So that's something you have to worry about.

So tomorrow we'll finish up a little bit of alum[inum], we'll go to titanium, [I'll] give you my quick song and dance on corrosion, everything you need to know about corrosion in 10 minutes. Hello, you know, we really have been [covering] documents, we've been going on an overview way, and we haven't just been talking about welding. Every — I've stopped at times to tell you a little bit of metalworking with all you've got in here. This is a typical [course] in service in [the] North. [I] don't think you know enough — well, that's all about — hopefully you know enough to know when to call in the right expert, and what's caught. And there are plenty of experts in MC [NSWC?] okay. I met David Taylor — you've got a whole crew of left, fifteen welding skills, you've got about 20 corrosion experts. And yeah, it might cost the money, but if you've got a big problem okay.

However sometimes you need the chief engineer to tell NAVSEA, this is what they ought to be my Facebook for example. Remember the fire up here, Portsmouth, where the guys at the bar, while you go home early. So I get a call from a former student from this class, [an] assistant professor here, [says] we got a fire inside the sub, this was before it hit all the news okay. He says we need someone to come help us figure out if the steel has been damaged. As well, I got some engineers — previous to that problem I didn't know it [was] for another problem in time. But what he told me is, he couldn't get anyone at David Taylor to pay attention okay. He was in charge of [Po]rtsmouth — I think he was a commander — fun, you guys today would tell, they're not going to pay attention, well, she did with contract okay. They got a church at time of something. So they're not exactly someone you can call up and say hey, tell me with us, let's go.