Alcoa Davenport aluminum rolling mill thickness variation

Tom's graduate student Julie investigates thickness variation on hot-rolled aluminum plate at Alcoa's Davenport, Iowa works. Diagnosis: the deliberately mismatched roll speeds (to give the plate slight upward curvature and keep it from cobbling into the mill bed) create a water-pooling cup at the slower end. Film-boiling heat transfer at the cooled end explains 80% of measured thickness variation. Proposed fix: a compressed-air "air knife" to blow standing water off the plate after each pass. Cost: ~$10,000 in piping; the compressed air infrastructure is already in place. Payback: one scrapped plate. Owen Richmond (head of metal forming, Alcoa Research) personally promises to see it implemented. A year later: not done. Tom uses this as his closing parable for "the way it's done in American industry."

They wanted the plate to hot-roll through the mill with a little upward curvature. It's a reversing mill, so the plate goes this way and that way. Sometimes you get Rocky Mountains on both sides, sometimes only one side. They actually ran one roll a little faster than the other, because you don't want the plate to cobble down into the bed of the mill — you want a slightly upward curvature. So one roll goes slower than the other, intentionally. That would create a little cup on the end where water would pool, and the water would cool it off while the operator was out there making his measurement, and then it goes through again. So one end was cooler than the other.

Tom's LFM (Leaders for Manufacturing) student Julie diagnosed the world's largest rolling mill's biggest reject category — plate thickness variation — by ultrasonic survey instead of single-point micrometer measurement. Pattern showed thickening at plate ends; cause traced to water pooling at one end because the top roll ran slightly slower than the bottom roll. **Central case of §8.**

I guess we have time, I can tell you one more story that goes along with that. In 1989 or 90 or so, I had a student in our manufacturing program. She was getting a master's degree in what you'd call Leaders for Global Operations around here now — it was Leaders for Manufacturing back then. She was working in the Davenport, Iowa works for Alcoa. This is where they roll the great big heavy plates — I saw the stretcher leveler for the four-inch or six-inch thick wing plates and stuff. What Julie was given for her master's thesis was to figure out why their biggest — if you did a Pareto on their plate, the thickness, the variations in thickness — they rejected more plate for being off thickness than any other reason. These are big heavy plates; this is the world's largest rolling mill.

Closing anecdote on the limits of deformation-process tolerance for thick aluminum plate (2–4 inch). Tom's LFM student "Julie" diagnosed a 5–10% thickness variation that was causing wing-plate scrap rejection; Alcoa never implemented the fix despite a one-day payback. Used to set up the larger theme: deformation processing can hold thousandth-of-an-inch tolerances, but only within geometric limits, and industrial uptake of academic findings is patchy.

If you're rolling something, and you have something that's really rigid, you can roll a piece of steel or a piece of aluminum to within 1/1000 of an inch. That's if it's 1/4 inch to 1 inch. You get to 2 or 3 inches thick or 4 inches thick, and you might have bigger problems. Maybe I'll tell you something about that the next time, of a student I had work at the world's largest rolling mill — Alcoa in Davenport, Iowa — where they make the aluminum plates for all the wings of all the aircraft. They were having like a 5 or 10% variation in thickness, and they were scrapping plates right and left because of this variation in thickness. Julie, the student who was doing an LFM thesis at the time, had to go out and figure out why. She did. She figured out why. And I'll tell you that Alcoa never did anything with her thesis, even though it had a one-day payback. That's just the way big companies work. They assume that people at a university are a bunch of academics that never came up with any practical idea in their life, and so they discount whatever we do. But that's another story.