WM_S2014_22

Put on, I showed you welding electrodes before, this is the biggest electrode okay. They actually have machines they use about 2000 amps, and they basically repair big iron castings. So here's a guide for welding iron castings, and here's an ingot mold and some guys using a gas metal arc torch, but they also use these. If anybody wants to, anyway, you want me to pass it around? It's just a big hunk of steel with a flag on it but anyway, here it is.

The other thing — oops — I want, passes around a roll all over the — another thing I will pass around, since Mike talked about standard welding symbols, here's the AWS manual on standard welding symbols. And you will find them on CAD programs that when it gets to welding, they'll show you how to weld things. [Tom holds up the structural welding code.] This is the structural welding code, this thick okay, for plate, for bridges and buildings, I-beams. This is a structural welding code for sheet steel. You can tell, this is plate thickness, this is sheet thickness okay. Yeah, why is sheet metal welding not a big code? Well, who does the most sheet metal welding? People like General Electric, Appliance Park, where they're making refrigerators and washers, and the automotive companies. They write their own specifications. They don't need some national professional society. Whereas all those bridge builders and architects and smaller firms that are building, could be billion-dollar firms, they need kind of a national code. So the sheet metal welding code's not a very significant code okay. And that's 'cause each company basically develops their own specifications and procedures, so far as that goes.

The American Welding Society and a lot of other code right — I actually have a module on codes and standards. American Welding Society knows they could make a lot of money by selling codes, and so they have part of the structural welding code, the D14 series. The big structural welding code we've been talking about is D1 for steel — and D2 — D1.2, D1.1 is steel, D1.2 is aluminum, we'll talk about that later. D14 are specific industries, welded joints in machinery equipment, welding of presses and press components. When I talked about that great big 7,500-ton forging press, here's the code that applies to that. It's not very thick, that's partly because it calls out the ASME boiler and pressure vessel code, which is this thing okay. Welding really heavy steel — the structural welding code goes up to about one-and-a-half, two inches for bridges and buildings and stuff. Presses and heavy machinery and equipment can go four inches, eight inches, twelve inches, sixteen inches, and that's where you get into the boiler and pressure vessel code. It's got things for heavy sections. The boiler and pressure vessel code is not just steel, it's all materials, anything, composites. I mean it's got rules for — that's why it's so thick — rules for all kinds of materials.

Any questions about that? If you want to know more about codes and standards, I have a whole twelve lectures on codes and standards. But let me talk a little bit about the structural welding code. This is one of the codes that you might say got thinner. So here's the structural welding code, AWS D1.2-97, and here's a structural welding code 2003. And if you compare the two, the structural welding code got thinner. Now why would a code get thinner unless complicated? Well, because in 2000, the Aluminum Association came out with a little design manual which is this thick okay. So it got more complex, so my trend of complexity in codes holds true. The only reason this one used to have a lot of — has the design rules — this one refers you to this document for the design rules. So the whole chapter of a hundred pages in here on design of aluminum structures, welded aluminum structures, has been replaced by this. And there's actually a 2008 or something like that — you remember someone — yeah, I think the last time was 2008. In any case, there are structural welding codes for these various things.

I want to start talking about aluminum. Anybody having — quite — yes?

Student: [inaudible question, apparently about code update frequency]

It depends on the organization. ASME, every three years you get a new update. ASME code right now, to buy the full set of like twelve, thirteen volumes — maybe it's only eleven volumes — is $16,000 last I looked okay. And so people would get very upset if they had to buy a new one every year. For a while the structural welding code — the code writing organizations have the money, of many of which are supposed to be not-for-profits, they find they can make big profits on writing codes. So the structural welding code, which used to be updated every few years as they thought it was needed, they decided about ten years ago to update that every year. Because that way the law requires, if you're building a bridge in a building or a building, most jurisdictions, states or cities in the United States, will require this code in the law. Which means if you're a designer, this is part of your overhead. You got to pay $400 for this every year. Well, I think a few people screamed, and so now they update this every two years okay. So it's a balancing act of how much can the professional societies get away with in bringing out new codes.

The boiler and pressure vessel code, when you pay $16,000, you get semiannual updates. And they can be that thick for each volume, they can be thicker than the volume. But what do you get, six updates in three — you get six updates in three years plus the original code. And so now this thing is this thick becomes this thick, and every three years they condense it back okay. And so they have blue, pink, yellow you know, and you can tell what insert it is, because the contract people sign will say "you shall build it to the ASME boiler and pressure vessel code as of March 2014." Well, you better have the current code. You buy a whole set for $16,000, that's good for three years.

Okay — oh no — there are a lot of people who spend a lot of volunteer time writing these things. Is it all volunteer time? Some of it's, because of their company that they belong to. But if I look in the beginning of this, personnel, this is page after page of people who have contributed to this code if you start looking at it, and the names and stuff. Duane Miller is chair. To become chair of a code like this probably takes twenty years of going to these meetings twice a year okay. Well, many people are doing this as part of their job. Duane Miller started out at — what's the big prefab building guys in Kansas — Butler Buildings. Duane started out at Butler Buildings designing, you know, these prefab steel buildings okay. He then went to Lincoln Electric Company in Cleveland, and Lincoln of course wants him to attend these things and put things in the code, or control what's in the code. I mean whether he's at Butler Buildings designing prefab buildings, steel buildings, or whether he's at Lincoln Electric, you know, the company has an interest in his participating.

Who's the vice chair, Sindel? I don't know Sindel. Neiman, I don't know. Morales is a secretary, that's just an American Welding Society person. And you see here's another guy from Lincoln Electric. Has enough to do with — they're the world's largest welding company, they put multiple people on there so that when Duane Miller retires they got another guy coming up to be one of the leaders okay. And when Duane Miller can't attend one of the meetings, they'll have someone. Walt Disney World — they build things all the time, why is this guy — I don't know why Walt Disney World is paying for this guy to come there. But H.H. Campbell the Third, I'm sure that's Howitt Campbell, grandson. Howitt Campbell was a Harvard, and sometimes I tell some stories about Howitt who was an interesting guy, but anyway. There's all kinds of people. Shell International Engineering and something okay, probably shale oil. ConocoPhillips, American — you know, Canadian Welding Bureau. Anyway, there's lots of — you look at the names, there's Duane Miller again because this is a different thing. Hobart's the second largest welding company. US Army Corps of Engineers, Massachusetts Department of Transportation retired, and then another guy, Waukesha City Technical College, probably teaches a welding thing. Federal Highway Administration. Anyway, these people are donating their time to write these codes.

And the professional society which is managing the whole process sells these. They used to sell them for like eighty bucks, almost at cost. But about fifteen or twenty years ago they all decided this is a good business, and you should see their balance sheets okay. Since the American Welding Society started certifying welders and selling codes, you know — they probably can show to the federal government, to the IRS that they're doing it at real cost. But real cost is all their overhead, you know, they just built a forty-million-dollar building down in Miami Beach right okay. I bought a brick okay, you know, when they get donations I think you have to pay $500 for the brick anyway, whatever. So it's a business for the nonprofits.

But it is also collecting the historical knowledge okay, of all the failures that have gone before. And I mean I could — well, I do teach a section, you know, a section of this structural material sequence that we do on codes and standards. And how — actually the theme of that module is, are codes and standards good or bad. They carry with them all the corporate knowledge of all these previous failures we've had for a hundred years, and what are the best practices that we know of today. But they also stifle innovation. You know, you can't go against the code, it's written into law in many cases. And if you want to do something novel or different, you got to go get it approved by the code. That could take ten years and millions of dollars okay. So you know, it's on the one hand and on the other hand — and you know the story about hands, right? Harry Truman once said he wanted a one-armed economist okay, so he couldn't say "on the one hand and on the other hand" okay. So I think Harry Truman also said, "if you want a friend in Washington, get a dog." Anyway.

Okay any other questions? Yes?

Student: [inaudible question about how code changes are marked]

What they will do, is in the code they will have little lines in the margin typically, of what changed. So just like you might take a yellow highlighter and put a little line marker on that paragraph, in the margin they will put little black lines. I'll see if I can find one in here. Maybe they quit doing it in this code, but it doesn't change — nowadays it doesn't change that much. Coming up with the aluminum design manual, that was a pretty major change okay, and that occurs like every thirty years or forty years. But the other changes are five percent, ten percent. No, so far as that goes.

Are we — oh, you can get it online for a fee. You can buy the boiler pressure vessel code I think $13,000 for an online version. What a discount, right? You can also get it in the MIT libraries because they have to pay the $16,000 every three years. What do you think part of your tuition is going for okay. There's a place in Colorado that virtually will sell online, so you can download in the next five minutes if you have a credit card okay, virtually any code in the United States okay. They are sort of a clearinghouse for codes, that's all they do is maintain a web server, and you give them — you find the code, you order it online, you give them your credit card for $100,000 or $1,000 or whatever. I bought, had to buy codes that cost a thousand dollars that are less than a hundred pages. You know, ten bucks a page is pretty pricey, particularly when some of the pages are just a list of the people who participated. You know, hey, save me — heck, I don't need that page okay. But this, you know, it's become a racket.

And I actually spoke out at a meeting at NIST once and said, the National Institute of Standards and Technology used to be the National Bureau of Standards, and I said you guys should do something about this. They're basically stifling manufacturing or construction or whatever you want to call it in this country, because the overhead, the cost of codes is just going berserk. Go talk to the MIT libraries. They've been — not just the codes but the journals and everything else. Used to be journals were almost available for the cost of printing. Not anymore, you know, some journals can cost you thirty thousand dollars a year to subscribe to a very popular journal okay. So anyway, it's a racket okay. So — and that fell on deaf ears down in Washington okay. That's one of the reasons they don't call me back to meetings, which is a plus. Want your — no, I mean, open your mouth and say things that are not pleasant and people won't invite you back. Try it sometime.

Okay, this is just looking at a number of materials, steels, titanium, and aluminum, magnesium, and you can see that steels have tremendous strength compared to other materials. But if you do it as specific strength, you find everything becomes fairly uniform. In fact, titanium has some of the highest specific strength of any structural alloy system. Magnesium is right up there. Problem with magnesium, it's stress corrosion cracking.

The material of the future has been for the last fifty years, and probably will be for the next fifty, the aluminum alloys that we're going to talk about now. Go from the 1000 series — 1060 is electrical conductor. So if you see a high tension line, you know, carrying 300,000 volts of electricity across the landscape, the wire hanging down is made out of 1060 aluminum. And it might be in a work-hardened condition like an H18 temper, or a zero annealed condition. But in fact, it's not just that, it actually typically is on a steel core because the aluminum is so soft that aluminum has no strength. And when everybody's got their air conditioners running during the summer, one of the big problems with those big transmission lines is they sag, they elongate. So you got some big tower and another big tower, and in the winter the catenary looks like that, in the summer with all the electricity going through it looks like that, as it expands and the catenary drops. And it gets hot, and aluminum doesn't have much strength when it gets very hot, we're gonna find, because it melts at a much lower temperature. And if you have a little tree here okay, it will short out. And it doesn't take too much to short out when you got 300,000 volts okay. And there have been some major outages, particularly in California, where the catenary came down, hit a tree and shorted out, and you've now lost your 300,000-volt transmission line, which is quite a few homes okay. So people don't have their air-conditioning.

Worse, if you're having a drought in the area, some of the South Carol— so Southern California wildfires and stuff started by striking the ground or striking a tree, arcing. You got a drought, you're hitting a tree that, you know, a fir tree that hasn't had any water, it's really, it's a pretty interesting fireball okay. Starts big forest fire. Some of the utilities in California have paid hundreds of millions of dollars in damages for starting these major fires that you read about in the paper. Because they've got a thousand firefighters fighting for three or four weeks and homes are being destroyed and everything, well there have been some big settlements. And why do the utilities do this? Well, the utilities do it because they're gonna get sued, and when they go in front of a jury they're gonna lose, and they're gonna lose even more, so they settle for hundreds of millions of dollars. But it's okay, they just put it right back into the rate base. So everybody in California is gonna be paying for that. So these people, who obviously, if you lost your home you want to get your money back, but most people don't realize that's why California has some of the highest rates okay. They got attorneys, of course. You've heard the joke right — why does — what is — how's the joke go? Why does California have so many attorneys and New Jersey has so many toxic waste dumps? New Jersey got first choice. Anyway.

Someone's frowning out here like their father was an attorney, you know, or their mother. In any case, we have all these different materials, and it turns out in specific strength they are all fairly comparable. And it turns out aluminum is one of the best for lightweight materials.

Now the aluminum alloys fall in a number of different systems, and here are some of the alloy systems designations by the Aluminum Association in the United States. The raw alloys are 1000 series, so the 1060 is 99.5%, or actually the 1060 is 99.6% aluminum, the 1100 aluminum is 99%. 2000 series, it's copper-based alloys. Manganese is 3000 series. Silicon is 4000 series. And you can see the different alloying things. Eight is all the others. And so you've heard of aluminum-lithium alloys, they're 8000 series. They don't use the 9000 series yet. They weren't using the 8000 series much when I was a student like yourself okay. What they have started using them as, a catch-all.

A lot of the first aerospace — well, I'm going to show you something — the non-heat-treatable aluminum alloys, which are the 1000, 3000, 4000, 5000 series, have lower strengths. And the heat-treatable ones, which are the 2000, 6000, 7000 — aluminum-copper alloys — were the first aerospace alloys. And I think I mentioned that the Wright brothers, the housing for the engine of the Wright brothers aircraft, the first one to fly, was an aluminum-copper alloy, and that was sort of new back in 1910 or 1903 or whatever it was. The 6000 series are common high-strength aluminum alloys, general-purpose, sort of structural, architectural alloys, high-strength. The 7000 and 8000 series are most of the aviation, aircraft, Boeing, Airbus type alloys nowadays, and we'll talk about some of that okay.

The aluminum alloys, actually I can probably show you — well, let me back up and go below here. There are casting alloys for aluminum, and it has something similar. The 100 series are casting alloys, really pretty pure aluminum. The copper alloys, and then it starts to break down. 3000 series of the raw alloys was a manganese system, 4000 silicon, 5000 magnesium, 6000 is unused, 7000 — or 700 casting — is zinc base. So it's not exactly a one-to-one match, but it's reasonably close. And the welding of the heat-treatable and non-heat-treatable alloys turns out to be quite different. We're going to talk about that.

So here's the 1060 alloy, 99.6% minimum aluminum. 1350 is the 99.5%. Those are the big catenaries, the electrical conductors and whatnot. 1100 is garden-variety aluminum. 3003 and 3004 for sheet metal. 3004 is basically Coca-Cola cans or beer cans or whatever okay. It's like thirty or forty percent of all the aluminum made. And it's got one percent manganese and about one percent magnesium. It's fairly lightly alloyed but it has high strength, and it's got a lot of mechanical work in it, work hardening, for all you metallurgists out there. And then there are the — these are the non-heat-treatable alloys, the 5000 series are basically very weldable marine components. Naval ships that are made out of aluminum, sailboats that have aluminum — a lot of these things want to be lightweight, and so they use aluminum. But anyway, the 5000 series are readily weldable, and you can get without very much difficulty, you can usually get one hundred percent joint efficiency.

The non-heat-treatable alloys have basic temper designations. Zero is annealed, dead soft if you will. F is as-fabricated, right off, it comes off the rolling mill in the plant. H1, H2, and H3 are different levels of strain hardening, only partially annealed and stabilized. And you have to look over for each one what strength level is. The heat-treatable alloys have a T designation that tells you something about their heat treatment, so you can have solution-annealed heat treat — I'm going to tell you what that is if you don't know. You can have heat treated, because these are the heat-treatable alloys. The W condition is solution heat treated, it's basically in an unstable condition, it will precipitate naturally at room temperature, and we'll see that. T1 is cooled from the elevated temperature during the shaping operation and then naturally aged. T4 is solution heat treated, naturally aged.

So if I'm Werner Ladder Company, I'm extruding the rails for an aluminum ladder that you go buy at Home Depot or Lowe's or something. Turns out I'll extrude it, and I will set up my extrusion press such that the heat from extrusion will heat up that thing. It's going to be hot when it comes out from just the work, and it will be solution treated. It will be heated up so that all the alloying element goes into solution, and then as it cools down sitting there in the plant, as it's cooling down, it will naturally age to the precipitation condition. And so all they have to do is extrude it. The extrusion heat does the solutionizing heat treatment, and naturally ages as it sits there if you get the right type of alloy composition.

Other things might be T3, which is solution treated, cold worked, naturally aged. There's all kinds of combinations. There's T6, which is solution treated and artificially aged. You put it in a furnace at a certain temperature, you try to get the maximum strength. T7 is solution treated and over-aged. Why would you over-age, to not get the maximum strength? Because you get like triple the stress corrosion cracking resistance in T7 that you get in T6 okay. If you take some of my other modules, I'll talk about some of those things.

So if I look at the weldability of some of these heat-treated alloys, it varies dramatically. A lot of the aluminum-copper alloys don't have an A grade, they have C grade, which is pretty poor weldability. Essentially you make an arc weld on one of these things and you have cracks right behind it. Basically it's not completely solution — it's not completely homogenized, and you'll get aluminum-copper eutectic, and you'll get hot tear cracks within the solidification, we call it. So you actually form, in the heat-affected zone you'll actually melt some of the grain boundaries if you will, some of the inter-dendritic spaces, and you'll get liquid. And because aluminum has a 6% volume change on solidification, you get big internal stresses. Residual stresses and distortion are big problems in welding of aluminum okay. And you'll get cracking almost every time if you try to weld one of these alloys.

One alloy here in the aluminum-copper series, 2219, has very good weldability. That's what the space shuttle main tank was made out of for many years okay, because you had to weld it. You don't like leaks — they found out with space shuttle Challenger what happened with leaks of the o-ring. So they weld this — well, the tank — used to weld the tank when we had a space shuttle, and they used 2219 aluminum okay. If you get down to the 6000 series, you get a lot of A and B weldability. 6061 and 6063 — the difference, this is typically plate or sheet and this is typically extrusions — they have A-grade weldability. The largest tonnage of aluminum alloys is 6061 okay. Architectural aluminum, you know, door frames, window frames, you name it. You can get good strength in the base plate, but we're gonna see you only get about half the strength in the weld because as heat treated. The 7000 series — some of the 7000 series have very good weldability. Some of them, like 7075, which was the workhorse aviation alloy for, from let's say — wow — 2024 was the workhorse alloy from 1920 to 1950, and somewhere from 1960 to 1970, 7075 was the workhorse aviation aircraft alloy for things like Boeing and McDonnell Douglas and stuff. Since then, Alcoa has come up with other alloys that have better combinations of properties.

But all the aluminum alloys — not all the aluminum alloys, but almost all the aluminum alloys have a phase diagram. So if I'm looking at percent copper or percent magnesium or silicon or zinc, they have essentially a eutectic diagram. This is temperature, that's the percent of whatever the alloying element is, and here's your liquid — here's your solid alpha FCC aluminum, and you'll get some sort of intermetallic precipitate, whether it's Al₂Cu — I don't remember that — there's all kinds of intermetallics. I mean, dozens and dozens just in the aluminum-copper system. There's probably four or five different intermetallics with different ratios of aluminum to copper. Magnesium has several, and silicon has a bunch, and then you have ternaries where you have magnesium-copper and magnesium-silicon ternaries. And so if you want to be an aluminum metallurgist, you get to learn all about intermetallic precipitates that form.

But basically for welding, we have two regimes. We have kind of a regime right here, we have a regime over here, where we have a relatively small freezing range. We don't like to work in here, where the freezing range is large. Why? Because a big delta T in the mushy zone, and a 6% volume shrinkage, means lots of strain and lots of cracking. So you like to work with very highly alloyed aluminum, or very good — I'm sorry — very relatively pure aluminum, lightly alloyed, or very highly alloyed. You don't like the intermediate stuff in the middle. And so you have to be very careful when you've got an aluminum alloy, even more so than steel in some ways. You got to know what the composition of the aluminum alloy is before you select the filler metal okay.

If you look at the relative crack sensitivity, that's what we're plotting here, of different aluminum alloys, silicon, copper, magnesium, magnesium-silicon, you'll find this is composition of the weld of the alloying element. And all they're doing here is they're just plotting the freezing range or the strain okay. It's alpha-delta-T okay, how big is delta T. This is basically delta T for this system. The maximum here is at one or two percent, this is the maximum. Highly relatively pure — well, this pure aluminum, this melts at one temperature right. If it doesn't have much alloy content, the delta T is small. You get up here towards the eutectic and you see the crack sensitivity goes way down. So this is a plot of crack sensitivity. Where does it come from? It comes from the freezing range okay of the alloy. So you have to be careful okay.

If I look at the relative strength of some of these heat-treated alloys, this is 2024, the workhorse heat-treatable aluminum alloy from 1920 to 1950, and still used today. The annealed condition has a strength of 12 ksi. The heat-treated T3 condition, which is optimum strength, is over 50 ksi. T4 is a little, like five or ten percent less, but it has much better stress corrosion cracking resistance. Here's the — this is yield strength, this is ultimate tensile strength, and here's the elongation. The elongation is about the same in all cases. They have pretty good ductility. Aluminum in general doesn't have the brittleness that steel sees from hydrogen and things like that. Hydrogen creates other problems in a moment.

This is 7075 — again big increases from 16 ksi annealed, T6 conditions 75 ksi. Not as strong as a baseball bat, an aluminum baseball bat, which could be a hundred ksi, but still pretty good. And here's the elongation. We actually, when we start getting these types of strength, we do start losing some elongation in some of these alloys. Here is actually the aluminum-copper system, showing you the phase diagram. You have to solutionize up in this region, this is the all-alpha solid phase, here's the liquid, aluminum-plus-liquid, this is the phase diagram. So you heat up to 550 centigrade or so, and you can solutionize and put all that copper into solution. If you quench it, you can trap all that copper in solution. If you heat it up to intermediate temperatures around 200 degrees centigrade, you can precipitate out and get high strength. You can anneal it in the 300 to 400 degree range.

So if I look at the heat-treated condition for aluminum, here's the phase diagram for aluminum-copper. Solutionize up here, precipitation heat-treat here. The time-temperature diagram, temp — this is temperature versus composition, this is temperature versus time. Heated up for one hour, 940°F, which is 560 or whatever centigrade. Quench it in cold water, and then heated up to 340°F, which is like 170 centigrade or something, for ten hours, pretty long heat treatment time. These alloys, these heat-treated alloys are fairly expensive. I remember, it takes about — most furnaces, heat treatment furnaces are about a thousand dollars an hour okay. So whatever you're putting in there, if you're putting in ten thousand dollars worth of aluminum for ten hours, you're adding — or ten thousand — if you're putting in ten tons of aluminum, you're gonna end up adding a thousand dollars a ton of heat treatment cost because of this low temperature heat treatment and whatnot.

Now what about the heat treatment alloys, what gives them their precipitation hardening? This is the as-quenched hardness way down here. It's soft because it doesn't have any precipitates. The —

[Recording cuts off mid-sentence.]