WM_Su2015_08

...or nearly, you think, we've been piping to insulate it, use the earth to your advantage, not on the military base, okay, because expected — not today, nowadays they hit you with a warhead, don't wipe out the whole basement. Let's give the old age, they didn't have really high explosives like that. So the other thing, I sort of, well, why do they do these things, and it wasn't, I came to apologize and know whatever you told me, I just, one day I was thinking, oh, okay. And I've had a number of failures over the years where people forgot what an internal expansion joint is, okay. So what questions about Nate's [presentation]? Yeah, so that's what you're getting, your regular questions. One of these were good presentations for you to hear about what other people have gone through.

Student: [inaudible question about lagging on both of these, particularly the stainless steel one]

There's something called corrosion under insulation, which if you take a corrosion course they fought under the positive tags, or crevice corrosion or something. But what you got, lagging, but anytime you put insulation around something, if that insulation gets wet and moisture gets in there, you have a perfect little environment, if the moisture gets deep in there, to create a nice concentration of corrosion cell. And in carbon steel it's not so bad — some of those for twenty-five years and before something bad might start happening. But in stainless steel it can really be bad.

I had a situation, the company that makes Manwich, you know, the supplements with the ground beef, okay. They were making Manwich, and it turns out you have to — if you shut the system down you have to keep your product above 200 degrees Fahrenheit or you have to scrap it, just throw out all the food. And now people need it, but they don't like to do that. So they have a jacket that had this hot water tank basically, and they, people design these things such, it was a hundred gallon hot water tank, and it was not to operate above 204 degrees, effectively supposed to operate at 200 degrees Fahrenheit, but it had to be operating between 200 and 210 because you had to keep the product above 200 or you're going to have to scrap it, okay. Are you doing maintenance? Well every day this was a plant food and they had to disinfect. So what do they use? Bleach. So do you spray it on? Right. It kills bugs. It also corrodes stainless steel like gangbusters, okay.

My mother, when she lived with us for seventeen years until she passed away, but she, one night decided she wanted, she used to always want to do the dishes, okay, so I let her do it. You're taking, oh man, see her turn the hot water on, let her go on to the next hour and a half. But nonetheless, she for the remaining I guess, but anyway, one night she didn't have many things, so she decided to put the stainless steel dinnerware in a little bit of water and some bleach to sanitize it rather than running the dishwasher. Well overnight, an oxygenated chloride environment will pit stainless steel in twenty-four hours. So we had this big pit in that part of the same, so that's why that time that device a new things dinnerware. But it's amazing how well.

So they had insulation, lagging, around this hot water tank, and they'd been washing it down with a dilute bleach foam on a regular basis until one day — and part of the problem was the controller, it was really quite clear if it was really maintaining things, balances would go to 12, but it has gotten a little bit of love, but in case it would open, it kills, and you got off much hot water under some pressure and it expands, of any body area. Same thing that happens to people, behooves you, okay. So, corrosion under lagging.

So when you talk about in the in-ring designs of the stainless steel, I mentioned something about the pre-real shearing system of CARE-34 [CRA-3/4?], for three or four smokers, who will get bloody. Stainless steel is a serious problem because it's stress corrosion cracking. Carbon steel, you don't get stress corrosion cracking, you get general corrosion. And one of the problems the Navy has is originally back in the 1950s they built the ships to last 30 years. By the 1980s they were starting to push them out to 40 years, and now what have you actually even said, it's 45, you said, going for another 25, I think. Is that right? People are looking at 50 years for a ship. Do they actually have, is the port that evolved to the places of minutes? Then someone finally after 15 years to the bulletin spend 50 minutes also, they don't ship other was another program, one floating now from that, that get in the water but cosmic. So they have those, you go to 64, okay. So they finally did, from they just couldn't afford our skating later.

In any case, the show is the Florida heavier displacement Rabinowitz, whenever it was, was getting up over 95,000, I think over 100, 110. Yeah, no, it's definitely, okay. In any case, we're going — oh, so anyway, they decided, they decided to extend the life of the ships, and as they've done that, there was sort of a transition region there in the 1990s where they hadn't really replaced all the piping with more corrosion-resistant piping. For carbon steel piping, you could let it go for 30 years and you can have enough original thickness from the general corrosion you'd get. Well after 30 years you just scrap it. But people who are taking care of these ships in the 90s and 2000s, these ships that have been built for 30 years of life, they were just replacing carbon steel all over the place. And that's why they've gone to that pain seawater piping on the new ships. So they still, carbon steel, it's probably not, but you're not going to get 50 years of harvesting, you've got to be down and replace the level piping system, which other ship. Hey, so there's some thoughts there.

Okay, let's, you want to start the deal, take me, okay. So what I'm doing right now is I want to just talk enough about hydrogen [embrittlement]. There were a couple points that, basically I'm taking things out of, and this is on your syllabus, but this is out of Jud Olds's [Jud Olson's?] old book of welding metallurgy. And actually he's got a very good chapter on hydrogen embrittlement, okay. And there's, what basically happens is, the hydrogen goes to the crack tip — I think I've discussed this in the joining picture — but you have a triaxial stress region, and that's why it's delayed cracking, it takes hydrogen time to diffuse to the tip of the crack. It concentrates at the tip of the crack and it embrittles it. And I told you, people have taken videos — it polishes a crack tip, look out of the microscope, charge it with hydrogen in a corrosion process, and then they just stretch it in the microscope and you can see the height of the bubbles come around.

This is photo session last week, we have the video really putting Lucille hide my gal, we talked about, you know, it's a very rapid process. There was a really tail-base cracking, so this is a slower, so where the hydrogen coming from, the slow process is crack introducing — like, no, this is all occurring within a couple of days. Oh, this was so loud this event. It's so rapid on an atomic scale, the diffusivity, go through the numbers sometimes if I watch eats it right here. But I put up some things that show, at room temperature the diffusivity of hydrogen, it's about 20 × 10⁻⁵ centimeter squared per second.

It turns out a guy named Albert Einstein looked at all the diffusion equations. You got the diffusion equation for viscosity, okay, and that's called the Navier-Stokes equation, okay. You've got it for mass, which is Fick's laws of diffusion, mass diffusivity. You've got it for heat diffusion, which is Fourier's law. Those are all the 19th century. So Einstein comes along and he says, well, these are all just diffusion processes. And if you look at the math — and I actually took in graduate school a full course out of the math department all dealing with one equation — and Einstein showed, sometimes called the Einstein number, that you can have a dimensionless number talking about mass diffusivity. It's d-squared over — or no, it's x-squared over Dt, diffusivity. If you're talking about Fourier's law it's x-squared over alpha t, where alpha is the thermal diffusivity, okay.

Now Navier-Stokes, this little messier with viscosity convection and things like that, but still, you can talk about the diffusion of shear stress, okay. And if you don't have any convection you'll end up with a similar formula, that all three of these things — momentum, which is Navier-Stokes, mass, or heat — can all be described by a diffusion equation. But the argument of the diffusion equation is always going to be of the form x-squared over Dt or alpha t or whatever, and that is dimensionless because D is centimeter squared per second times — centimeter squared divided by, divided into centimeter squared, dimensionless, okay.

So if I have D minus, 5 × 10⁻⁵ centimeter squared per second, and I want to know how long it takes for hydrogen to go one millimeter, which is the tenth of a centimeter, and I square that, I get point oh one, as the distance, over Dt, so by 5 times time, and time times — wait, so time is equal to 10 to the third, right. A thousand seconds. Three hours — no, twenty minutes. For hydrogen to diffuse one millimeter, this was equal to one. Yeah, if it was conventional, as long as you say yeah set it equal to one. Actually, in a lot of the forms of the equation that went, MIA for here might be a buy here, there's different depending on the job you're, what you're looking at.

There are old books written by mathematicians. One's Crank, Mathematics of Diffusion — it's just solutions to the diffusion equation in different geometries, plates, thick plates, thin plates, infinite plates, cylinders, spheres, okay. There's another one, Carslaw and Jaeger, Conduction of Heat in Solids, and it's just the same math. Crank was for mass diffusion. And these are about 50- or 60-year-old books, but mathematicians can solve this. And that's where you get into starting learning what a Bessel function is, okay. The solution of the diffusion equation in cylindrical coordinates are sort of, are for this, you've got Bessel functions, thin plates give you an infinite series, sum types of, thank you, guys ordered without now. So you, I could be teaching you all that stuff, go back to my youth, okay.

But in any case, if I, × 4, so it's an hour and a half rather than 20 minutes, who cares, okay. The point is it takes time for hydrogen to diffuse a millimeter, okay. Well, millimeters not very deep. Let's say I have a piece of steel that is two inches thick. That means to get to the surface the stuff in the middle has to diffuse one inch, which is 25 millimeters, squared, 25, this is 625, okay. 25 × 25, 625. And now this becomes 6.25 times an hour and a half, okay. Now × 4 (h thick), latex going, has the square of the distance, and now it's going to take a whole day for something inside a 4-inch-thick plate.

And so it turns out, when they were doing some of the foundations and G.E. Wilson things, they could put in half an inch of weld metal, they had to stop and make sure they keep the preheat on for like three or four hours just to allow the hydrogen out before you put more on top, okay. If you're doing armor steel back in the World War II, that's 17 inches thick, they had to go slow. Actually they had to go, so they went slow anyway because then stick electrodes and where clause and will die quickly, but if you didn't allow time for the hydrogen to diffuse out you had a problem.

I can't remember — oh, it was the Fore River Bridge where Quincy Shipping are — you still being state of Massachusetts is building a new — we don't, rich — drawbridge. And they got some things, ships at heart, from 8 inches thick of a high-strength steel, highly alloyed, very difficult to weld. And I told them, you've got to stop every half-inch for two hours and diffuse the hydrogen atoms. At what the, think people never heard of, going to very thick steel you've got to stop to give time to the hydrogen to diffuse out, otherwise you just, traffic will be trapped forever. You start going through the diffusion equation, you're not going to get it out of there. We're going to get cracks before it all happens. So you have to worry about steel thickness as well.

Part of the problem, and I think I pose this, is, people would call me and say, well, we want to weld type-of-touch [type of such-and-such], how do we do it? And I'd ask, what's the composition, what's the thickness — those are the two things I needed to know.

Another thing that we talked about, then we're going to go over again, is this microstructure, tensile strength of stress, and hydrogen. Here's a nice Venn diagram to tell you what the situation is, in terms of, you want — each one of these circles, so it's going to be an intersection where everything's on track. Now we have this one. This is actually a plot that tells you how much hydrogen you can expect from different welding processes. You go to some of the books on welding steel without hydrogen cracking, this is well, hydrogen level, in either milliliters per 100 grams weld deposit, essentially within ten percent. This is the same number as parts per million hydrogen in the steel deposit. Here's your shielded metal arc, these stick electrodes, way up here. Even with two-tenths of a percent moisture here — even with two-tenths of a percent moisture, which is the lowest level you can get, you're going to be up around the potential, well over 100 parts per million. But in fact, if you have your freaking everything, you can still get things now, long before, okay. Typical basic welds, these are baked electrodes, at 350 to 450 °C. Titanium dioxide electrodes or lime electrodes, different types of coatings, you can be much higher, and you'll get much higher, like 10 parts per million, 20 parts per million hydrogen, and you're going to get cracking. This will crack HY-80 [steel], some of this will crack HY-100 [steel]. If you really want to do a 100, you should be down to this range.

We have flux core welding that built bridges. I remember the electric boat brought in down here 25 years ago, and they built a new kind of administration building where the cafeteria is right down the road there. And they took me to lunch in a couple of welding engineers, this pretty nice building, this is, dad was welded with what score. So don't jump up and down, okay. Flux work is a hike, it's a solid wire process, sumby gas metal arc, accepted this number one, song work, the wire is hollow, and inside it has some net flux. So here are the Flex is on the outside, here in flexes on the inside, much more productive process, also can be fairly high in hydrogen. The lowest hydrogen is gas metal arc, this is what you typically use, picking on the hole of the ships now. It's not true 30 years ago, but today closely guess how r, and you can get fairly little hydrogen because it only thought you from grease and moisture or looking on the water to places like that. So we keep pushing for better controls, which some air gasping our code wire and speeding through that.

Student: [question about where the water's coming from]

Yeah, I actually, you have a little nozzle, the copper tip that has a hole in it, and you have a wire feeder, a little spool of wire, you have some rollers pushing it through this little nozzle, and the arc is struck to that wire just like this is a non-continuous wire. With gas tungsten arc, they could have a hundred-pound drum of wire. See, where it's like the tungsten Bednar in that, that's gas tungsten arc, is that in the same part. That's even lower than this part, okay, because you weld much slower. I've got something in here that the deposition rates, I'm just wearing the same spot on that, charges you can think of it as the same spot as gas-metal-arc-welding is sort of, sir overlaps in terms of central high ovation.

You know, we haven't gotten into it yet, but there's lots of formulas for what we call carbon equivalent. It's a measure of the hardenability. So you take all these alloying elements and you know carbon will give you hardness and alloy elements give you hardenability. Well for welding, people talk about carbon equivalent, and over the last 80 years there are lots of different formulas. This is not all, okay, which is the right one to use. When this was developed for high-strength low-alloy steels, it's got niobium, because that's how they get the strength is in the low alloy. Everybody wants to come up with their own thing in terms of carbon equivalent. We're going to go through the American Welding Society code and show you the [formula] department.

But what they are, the fractions is representing, like one of our carbon, 1/30, part silicon, or without of attraction. Silicon's 1/30th is effective at hardening steel as carbon, in person ison. Silicon will add, hardens steel similar to carbon, but at one-and-one-thirtieth percent is effective, okay. So you can get great depth of hardening, which is hardenability. But this is a measure of carbon equivalent, how much hardness and you get. But you also get hardness when you have these other alloying elements, they don't just get depth, they also get an increasing, okay. But it so they're all careful formulas, everybody wants to own their own empirical formula.

This is what I really wanted to get to, as we haven't really talked about it: this is hydrogen-free samples. This is the notch tensile strength, so it's like ripping the piece of paper, that will not, you put your steel as a function of temperature. Turns out right here at room temperature is the greatest embrittlement of hydrogen. I mentioned before that if you wanted to measure the hydrogen in the steel in the laboratory, you'd quench it, put it in liquid nitrogen, because it takes forever, you're slowing this D is now down to, finalized at minus, in liquid nitrogen, and you can run one for three or four days before you do your chemical analysis and still measure your hydrogen, okay. Because you've got three or four orders of magnitude slower diffusion. At higher temperatures you're going to find the stuff diffuses out more quickly, okay. At 250 degrees, I got the data here somewhere, I showed that, you simply that before, but it might be 10⁻⁴ at 250 degrees with a preheat. And so the hydrogen's diffusing out more rapidly. Turns out, nature has it that room temperature is absolutely the worst, okay. You're losing — that one — okay, so that's just the way it is and we have to live with it.

Student: Why in this graph at higher temperature, it looks like increasing the strength?

Yeah, well the strength — this is a brittleness test, and so hydrogen-free samples are not, otherwise increasing the temperature, that's a weak Petro of it. I don't know if there's a really good data for that, okay. He may have had that, but he may have actually had a little bit of hydrogen. Here, free samples or whatever, there's scatter around this thing, and I bet the scatter is great as the slope, okay. But there is definitely a difference here, I mean there's the order banking dues, you can lose 75 percent of your strength through the hydrogen room, okay. So all of a sudden your design which was good for fifty percent of strength is two times overloaded in terms of embrittlement. So we do have to worry about it.

So now let me give you an example, this, the last elements we have, of this Venn diagram. This is one of my earliest. This is microstructure. [Tom draws or points to the Venn diagram.] This is stress. This is how you cope with that. So I was like 28 or 29 years old, I was a young faculty member, could the firm feed my family, remember what it's like. And you know remember, I would an innocent but I can't afford to feed my family, was born he ate, okay, now the salary of an assistant professor.

And so I got a phone call from what used to be the boss Navy [Boston Navy Yard] — or honor to have such a thing, but is, that I guess the Constitution of it. And it wasn't there, the Constitution, of Charleston exactly, downtown somewhere else. And they had a destroyer that was supposed to go out in one week, and for six weeks they had been trying to weld chromoly tubing in the steam system to stainless steel. And the Navy spec said you must use an Incanel [Inconel] base color mode. And they had been trying that, had people come up, Philadelphia Navy Yard, some Navy welders, they tried, and they kept on getting cracks, okay.

This is just a little one-inch tube of stainless steel and they had to join — [Tom draws the joint geometry] — like this, machine below my back, and they had to make it in the vertical position because that's the way the boots were in the destroyer. And they had to put in a root pass, and whenever they tried to do it, you know, they got a crack. They found they could do it with stainless steel — this was stainless steel to chromoly steel going, for a problem, mate, legalism, a high carbon — high chromoly steel for pressure systems, stainless steel is even higher than that too — but this is just an transition joint. Every time they tried to make it crack, they tried preheating, they tried all kinds of things, and they couldn't get it to work. So they call MIT and I was the only one, and I needed the money, so they said, we want you to come down and tell the Navy why we should be able to do this with stainless steel. And I, nicolau away, and I didn't know if this would work.

So I say a little bit lately, try to get home at five o'clock, the little kids, until they went to bed, and I stay late, I go to the library, and I find an article in journal ten years later [earlier], if you weld it with stainless steel it's going to crack on you, because the carbon will diffuse around and form some carbides with some of the things that are in the stainless steel, and you'll get something that would hold them — new name for that — but it's basically a long-term, when you're at temperature, shift over time the carbon diffuses, and you'll get cracking. So I knew I couldn't come and tell the Navy, hey just get respect, it's okay to do you stainless steel and they'll get, we did it out of the yard.

So I come down and I have a copy of the paper with me, and I meet with someone in a conference room, and they didn't tell me you all to be over, but I said, well, you can't, you're gonna have to think about, just like the spec says. But let's talk about why you're having a problem. So they showed me, and I asked some questions about all these things, hydrogen, check the composition of moisture in the shielding gas, they had, couldn't do much about the microstructure. This, you know, this was the chromoly steel, the stainless, with the filler metal in between, and that's what the Navy had.

The only thing I could think of in this half-hour, 45-minute discussion, as I was quizzing them and they were quizzing me — they weren't real happy because I wasn't just wonder like, letter, maybe say oh it's okay problem, porque estoy — I said, well I'll tell you what. Why don't you try, instead of the joint groove that looks like that, why don't you machine one that looks like this, so that when you weld it you can put your root pass in here rather than putting a root pass in here and getting a big deal right here. And that will reduce your stress because you have some flexibility, buh-buh-bah, and we'll try to mess it. Goodbye, let me know how it works out. They said, no, we'll do it while you're here.

And so we go out and they take a couple pieces of pipe, they put them on the lathe, achete it, guys we don't got enough of a length here, ethics just a little more. So then they go, they do it, TIG weld, that's what you do, fight typically particular group past, may be slow, it's great that, my body. Well, they do the TIG weld and then they have to let it slow cool, they wrap it in insulation to let it slow cool, little, do about hydrogen to come out, so maybe spec. And so we're standing around for 45 minutes while it's cooling down, and the whole time the foreman did the machining and welding, and saying, this is not going to work, it's not going to work. And this guy's got 40 years experience, I got four years experience, who do you think's going to be right? I was just thinking, look at my watch, singing, and there's somewhere I have to be.

So then he takes the insulation off and cools down, and he takes the first of all, I guess clean it with the dye penetrant, solvent, sprays that on, and then a bit calm and steady sprays on the penetrant, you got to let that penetrant for five minutes, okay. So I'm not going to work. And then he finally holds it up and sprays the white dog [developer], developer turning it around. I walked out of there. [Tom mimes walking off.] Oh, okay. I don't know if they got the chip out of that, I've, I just got my for the doctor's aware, but I could have walked out, and I had no idea except no, that's on the bells, and I knew, I could, they do it as good a job, I quizzed him about the whole getting rid of the hydrogen, no breeze so there'll moisture, you're shielding, and they checked all those things. About when the microstructure was going to be whatever it was, and this is how we had used, the only thing I could think of is how to lower the stress. It was sort of a shot in the dark, and it works.

That's been it, stress, of a structure, of mother, right. And then actually once you got the root pass, I think the spec allows you to use something else afterwards or whatever, but they knew that, they put in final test it was just that root pass they couldn't get crack-free, okay. And by just changing the joint job, so this actually gets to the point of, is the only tool you have as a hammer, you see every problem is a nail. The people that came up from Philadelphia, hydrogen down and, but then everybody thought of it as the welding metallurgy problem. Well it gets solved as welding metallurgy, while the solution is a welding residual stress problem, okay. So anyway, so one of my early successes that no idea how, sometimes, okay, thanks.