Lessons in carbon fiber laying: Boeing 787

[Ciro Pabón]

Burkhard Domke, head of Airbus Engineering Inteligence, "published" a PDF document on lessons learnt from new Boeing 787 first batch of airplanes.

It's interesting how the document, leaked at FlightBlogger made its way into the public domain and, moreover, how informed are modern companies about their competitors.

However, beyond that, it contains a number of graphics on the technologies used by Boeing to lay carbon fiber. Before I copy/paste one that called my attention, allow me a brief introduction.

The 787 is a last generation airplane, with a "cigar" made of carbon fiber, instead of aluminium.

The size of the airplane and the goals that Boeing has for 100 airplanes per month (!), stretches the technology of carbon fiber to its limits and beyond, not from the point of view of structural design, but from the point of view of manufacturing.

First, I find remarkable that Airbus proclaims Boeing having problems with their assembly line due to "low-wage, trained-on-the-job workers that had no previous aerospace experience" who work for Boeing's suppliers. It does not make me comfortable about the quality of the airplane... Nor I am tranquil about Boeing having issues with the FAA because heat is transferred through the carbon fiber fuselage. I think (based only on my very limited readings about the material) that could bring forward structural problems in the long term.

Second, and final, here you have the more important slide, from my point of view:

From 0.5 lb/hour of manual carbon-fiber lay down rate (in the 80's) to 30 lb/hour in actual production, a 15 times increase in actual carbon-fiber manufacturing speed. Still, that is way beyond optimistic rates of 500 pounds/hour predicted. At this last rate, an F1 car could be produced (if round, I imagine) in 2 hours.


Any comments by people with experience are appreciated, starting by how this huge use of carbon fiber could either starve the market, or make carbon fiber so cheap that it becomes available to the mass market.

The names of the companies involved in carbon-fiber manufacturing are Alenia, Spirit, Hawker de Havilland, Kawasaki, Vought and Global Aeronautica. I don't have the time, but it would be interesting to watch a picture of a multiple-head carbon fiber laying machine for an airplane the size of a 787 (I think they have one at Spirit). I posted here, ages ago, a picture of the machine that Rutan used for building Spaceship One, the first private space rocket.

Bob Rutan, master of composites, inside Spaceship Two

[WhiteBlue]

I my view it is often more intelligent to use energy in a one time investment instead of continuing expenditure. That goes for the use of polymer foams for insulating buildings instead of paying for heating an cooling them. It also applies to KERS, HERS and light weighting aircraft and cars.

The fiber laying issues are going to be solved by machine engineering. Of that I am quite sure. It is a straightforward problem of coping with upscaling requirements of unprecedented dimensions.

The more difficult question is the cost curve of carbon fiber vs the world production rate of the material. To make substantial gains fundamental improvements to the physical/chemical production process must be found. The market will sort out the margin fluctuations that result from demand jumps in a relative short time like 2-3 years. So the question of cost vs experience gained by rising quantity is the decisive issue. What I keep hearing from people involved in the industry it will still take a long time until carbon fiber composites will be able to compete wit metals in mainstream automotive bodywork.

[xpensive]

Having friends working in the manufacturing of the Swedish "Gripen" jet-fighter, I understand that carbon-fibre has not had the impact everybody expected some 25 years ago.
Cost of material is possibly not such a major issue in that industy, but the time-consuming manufacturing method, as carbon-fibre is not an isotropic material, why the direction of the individual layers have to be carefully engineered and applied.

In this context, it is imortant to undertand the difference between "Carbon-fibre" and "Graphite". In what we call carbon-fibre, the long fibres are oriented in certain directions within a matrix to achive specific properties in different directions.

Grahite however, such as in a tennis-racket, are chopped fibers mixed up in the matrix without any specific direction, basically producing an isotropic material.

[West]

I work on the technical side of the 787 (documents) and all I can say is... wow...

[timbo]

I heard that Renault F1 yeam was somehow involved in carbon-fiber technology for dream-liner, is that true?
On a side note, I wonder what resins does Boeing use. It must be something with very unusual properties as aircraft is exposed to temperatures well below -30 oC at its cruise altitude. Very few polymer materials are capable to withstand low temperatures and heating/cooling cycles like this for a long periods of time.

[flynfrog]

My former employer was building a composite jet and prop plane. The money saved by going to carbon is not from the matl. cost but from the speed and cheapness you can tool up for it. Carbon will cut assembly time way down. As far as the 787 and the A320 starving the market they did. For a while it was really hard to get carbon unless you were military. This has subsided a little bit as the carbon companies have started to catch up. Some of the other pluses of going to carbon are that they can run more humidity in the fuse since they don't have to worry about corrosion. the HVAC system on the 787 is unlike anything we have seen.

[Jersey Tom]

That does not look like the kind of document that should have made it into the public domain...

[Carlos]

Just some offhand remarks, advances in the processing speed of laying up large carbon fiber structures is accelerating, a company in partnership with an Australian university has developed a totally automated process, almost like a large stamping machine, large carbon fiber sheets, in multiple layers, heated resin is injected, ultrasonic transducers remove bubbles and voids, dies compress the piece under pressure, Toyota bought one of 2 machines built, let me confide the documenting links and articles have been lost from my collection, there was also a PBS documentary about the US strike fighter competition, one of the companies, it might have been Boeing, were laying up a great big, one piece delta wing and having a lot of trouble due to its size, there is a world wide shortage of carbon fiber at this time,the Japanese government has partnered with industry with a 100M + grant to accelerate the production and lowering the price point of CF, yes carbon fiber has lost its lustre as the end all of strength and fabrication, a small US aero company is producing, light jets from aluminum using friction welding which is said to be cheaper and stronger than other aluminum methods, almost a microscopically one piece fusalage, almost. Ciro and one other member posted about a multilayer resin aluminum composite material thats almost as strong as CF panels, there's also work being done on natural fibers that is looking promising, in the late 1800's the inner rims of railway car wheels were commonly made by stamping machines,with a paper based composite, must have been very strong, for such applications. A lot of reseach with reprocessing natural fiber on a nano level to rival Carbon fiber is ongoing.
Just a few more OT remarks, long term members know how I prefer to offer research, and that brings to mind some long term memories of F1T, Tom a prolific poster, 2000+, who we haven't heard from in a very long time, in his very last post he was headed to the Caribbean with his University team to race in a rally, we never hears from him again, since Ciro is nearest to the vicinity, could you look him up? Bsck to carbon fiber. I had mentioned some university/industry research on CF used for intake valves and although I haven't read a post originated by checkered here on the forum, in a very long time, he passed along that CF is used commercially to manufacture ICE pistons, very good to hear from him.

[Ciro Pabón]

No, it doesn't, J-Tom. Isn't the Internet a wonderful thing? For example, if West is "wowed", I'm even more by being able to reach it from my house in the middle of the Andes.

Flyn is right on the mark, as usual. In that document there are a lot of diapositives devoted to the weight/number of passengers history of the plane. After all, that's what specifies the target market of the plane.

I'm reposting because I finally found an interesting article on Alenia.

Besides, here is the machine that intrigued me: the multiple head carbon fiber laying thingie. I imagine that the red arm is the working head. According to the previously linked article, it lays down the fiber in the right direction, with polymer and all. The molds (not visible in the picture) collapse under the cigar after the fiber has cured enough. In the article they compare it to laying down a wax coat on a ballon and then collapsing the balloon to release the cold wax.



Yes, the red thingie is the head. After writing the previous paragraph I find this closeup of the carbon fiber laying arm and head. Pretty interesting.



I think it follows Carlos's post ideas (hi, Carlos, long time no see, I miss your PMs! ), but I'm not sure. The carbon is spun and "sized" before being mixed with the polymer and layed, if I'm not mistaken. I think it's kind of an inkjet printer, but with carbon fiber instead of ink.

BTW, sizing means that the carbon fibers are covered with an extra hard polymer (330 MPa!) while being spun to improve the interphase bonding between the fibers and the polymer matrix, which, I read, is crucial for a better composite final strength. You can read about it here: http://scholar.lib.vt.edu/theses/availa ... apter1.pdf

Carbon fiber sizing. I had no idea it was done


This is how the fuselage appears. Isn't it a thing of beauty? (As poet Bhuwan says: "I've never seen beauty dancing with eternity")



This is the machine used for machining (duh) the carbon fiber, to trim it and make holes in it after curing.



This is the largest autoclave I've ever seen:



BTW, after reading Carlos's post, I read elsewhere that the bubbles are "fought to death" by Alenia. The building where the fuselage is built has the largest clean room in Europe (they compare it with a 3,000 rooms hotel), temperature controlled to avoid the "bubbling".

This is the truck (for civil engineers, hi, you guys!) that will carry the finished fuselage parts. It weighs 375.000 pounds on 32 tyres. I'd love to build the road for it!



Well, thanks for all the comments, specially the clever ones of WB about capital cost vs maintenance and the delay foreseen by xpensive friends in the time to mass market (and the CF-graphite length of fibers comment, I did not know).

Timbo, the polymers use a compound with a name I cannot pronounce, much less remember, optimized for temperature changes, you can read about it in the PDF I linked about CF "sizing". After all, timbo, Rutan's ships go to the stratosphere and then they do an aerodynamic braking to return, so I imagine they have temperature changes under control. Yes, Rutan ships go up and go back in little time, but...

[WhiteBlue]

The red arm seems to be a huge custom made servo electric industrial 4 axis robot on a linear portal. The last rotating hand is build as a hollow ring. through the ring prepregs (preimpregnated fabrics) are fed. I asume we are talking tapes here and not rovings of fibers. The resin they typically use is a thermosetting compound that has the crosslinking inhibited at room temperature. From the design of the robot I would think that the whole fuselage rotates as an additional axis which is driven by the other end of the machine that looks like a lathe. Multiple instances of such a head would obviously speed the prozess up dramatically but would probably also need very sturdy or additional portals for mounting.

On the robot hand you can see the roller that presses the impregnated tape or rovings to the collapsible core. When the layer is finished the chemical reaction is kicked off by heat and/or radiation in the autoclave.

Doors and windows can be machined by the same robot by exchanging the tape laying hand für a milling head.

[flynfrog]

Doors and windows can be machined by the same robot by exchanging the tape laying hand für a milling head.

Could be but there is no way you would allow a dirty operation in a clean room.

There was a good article in one of my trade magazines a few months ago on the dream liner and these machines we are talking about. I think it was composites world but Im not sure. The really only make sense on very large scale operations such as this. The rest of the world still hand lays. We implemented a laser system to help ply placement. We cu our scrap 75% and sped up layup almost 80% before they were measuring each ply by hand

[WhiteBlue]

Good point. They probably use a dedicated similar machine in a different location.

[xpensive]

Conventional machining, such as sawing, drilling or milling, of Carbon finre is difficult due to the very abrasive nature of the fibre itself.

As far as I am updated, high-speed water-jet is still the prefered method.
Besides, the cut becomes so mush cleaner than with mechanical cutting.

[rjsa]

Ciro, thank you very much for this!

[flynfrog]

Conventional machining, such as sawing, drilling or milling, of Carbon finre is difficult due to the very abrasive nature of the fibre itself.

As far as I am updated, high-speed water-jet is still the prefered method.
Besides, the cut becomes so mush cleaner than with mechanical cutting.

water jetting is far from the preferred method. Water jetting can result in alot of edge delamination and full on blow outs. most places still use carbide and diamond cutters. Water jetting on composites needs a backer usually water but something as large as a 787 would need a insanely huge water tank