My nephew’s currently studying composite manufacturing techniques as a physics student at University. A good guy - 22 years of age and totally keen on getting into carbon fibre car design etc.
So we were discussing the resurrection of King Tut’s mummy last night and we got to talking about the cat scan which was used to do a 3d profile on the old bugger’s cranium etc. Very interesting stuff we agreed.
Anyways, we got to talking about doing similar “cat scan” 3d profiles of old classic car shapes and a debate ensued about which manufacturing technique would be superior to doing a Jurassic Park re-invention of classic car shapes of the late 60’s etc. It was an entertaining debate too I might add - fuelled by much laughter and fine wine.
I’d love your collective input on this… we agreed that 3D Laser profiling could be used to build the cadcam modelling of any classic car - from unibody to chassis or body panels. But we couldn’t agree which would make for a superior manfacturing technique. My nephew argued that if you could use the cad cam profiles to build the right moulds, you could theoretically mould an entire 67 Mustang Fastback unibody say, out of carbon fibre. It’d be as expensive as hell of course - but we weren’t really concerned about that. We were more concerned about the physics involved if budget was not an issue.
Conversely, I was arguing (based on some of the modern CNC polished ports in aluminum heads I’ve seen lately) that theoretically you could start out with two huge chunks of aluminum and then CNC either half of the same 67 Mustang Fastback unibody and weld the thing together down a north south plane and it would be an infinitely stronger, stiffer unibody than the original because you could strategically include extra “I-Beam” ribbing where necessary. My nephew argued the carbon fibre unibody would be stiffer. I agreed, however, the aluminum CNC unibody could be repaired and would survive a multiple impact crash with greater safety.
The question is, can CNC machines make such lovely curvaceous body lines? Or would it wiser to CNC a laft and right female mould, and then forge the aluminum into said moulds, and then CNC the insides of said forgings as part of post-machining process?
I know, I know, this is all hypothetical - but it’s good for my nephew - he’s thinking outside of the square already from a physics point of view and I like that. In closing, which option would be cheapest do ya think?
You’re forgetting about a lot of fundamentals of what vehicle structures are for. Machined (CNC’d) parts aren’t necessarily the best for any arbirtrary part of a vehicle structure. The way metal is formed has a huge, huge impact on the types of forces that that metal can withstand. I have to imagine that the same goes for composite materials like carbon fibers.
Okay, structural issues aside, and we’ll assume that the vehicle “skin” doesn’t affect structure (it does, though), then CNC’ing cast metal is still way, way more expensive than just stamping it from a die. There’s a lot of waste, and it takes lot of time. So, take your scans, and build a die from the scans. Maybe the die-makers use a CNC machine to cut the die. Even so, it’s still not as cut and dried. Die making is as much an art as a technology. Metal springs back. It draws. It does all kinds of things in the die press. You can’t just make a die in the mirror image of your measurements.
I have no idea how carbon fiber acts. It may be cheaper, then, to make a die for carbon fiber or vacuum forming or whatnot. In that case, it may be possible just to cut a die from the inverse of your measurements. Any vacuum forming or carbon fiber experts around?
The carbon fiber method would be the way to go. One of the problems you’re going to run into with machining aluminum is that to get the very thin profiles you’d need for the for body, you’d have problems with the parts vibrating so severely that it’d be almost impossible to pull off. (We’ll ignore the amount of noise that this would also generate.) At best, you’d end up with some severely irregularly sized parts, which would have to be reshaped by hand before you’d have any hope of welding the thing together. I wouldn’t vouch for how strong the body would be, because of all that flexing that had gone on.
Oh, and you have no idea how big those aluminum blocks would have to be. Jesse James got an idea to make some wheels for a bike out of solid aluminum blocks. The finished weight of the wheels was around 30 lbs or so (each). The blocks started out as 300 lbs of aluminum.
You know those executive toys that consist of a grid of pins in a frame - you push your hand/face/genitals against one side of the grid and the pins are pushed out in the same pattern, resulting in a temporary 3D image of the object you pushed in at the other side.
OK, make something like that, with pins that can be locked once positioned, or better still, a CNC machine that can assemble the two halves of a die by stacking a bunch of different-length rods into a frame. Another CNC machine grinds away the jaggies at the ends of the rods, then you can use the dies in a press to make the metal body part (or you could make a former for carbon fibre the same way).
I used to run CNC lathes and am familiar with CNC milling. Now I sell tooling for those machines. If you do decide to mill out a mold the size of a car out of a solid chunk please call me. I could retire after selling you the amount of cutting tools you would need.
Neat idea. Tuckerfan brought up some good points of the difficulties on how the material will react. Myself, I can’t recall seeing a michine big enough to do the job in one piece. I have no idea if you could even get a solid piece of aluminum big enough. That said, with graphic interpolation the right cutting tool can cut any manner of curves to give the classic lines of a car. I see some problems holding it. It would be a multi step operation with the workpiece being rotated or moved to get different axis cut.
I also know a fair bit about carbon fiber from pit crewing motorcycle racing. Far and away a better choice to make a skin out of. Lighter by magnitudes and a hell of a lot stronger than I would have thought the first time I started playing with it.
Hmmm… I think this is as close to a near perfect description of a prototype style die than I’ve ever heard – heard being the key word, as I’ve never actually seen one. I’ve always gotten the impression, though, that prototype dies work very much like these executive pin toys.
This ought to be large enough, but again, I’d think that it’d be such a PITA to do it out of aluminum that carbon fiber is lightyears ahead as the best choice.
Duplicating mills have been around long before computers. Think of it as a 3-D pantograph with a spinning cutter header like a router bit instead of a pen on the output side. The duplicating lathe is used to make long parts like rifle stocks. The part rotates slowly and a spinning cutterhead cuts the contours as it is controlled by a stylus tracing the contours on the original.
Even if you could mill a duplicate of a sheet metal auto body I don’t think you wouldn’t want to. Stresses and grain changes indruduced by stamping and forging are part of what gives those parts their strength. No reason a mechanical duplicating machine coldn’t be configured to make a die for the original part.
Except that the metal springs back after stamping, so a die isn’t a complete mirror image of most automotive body panels. Also one of the reasons car bodies have so many parts (aside from material differences) is that not every shape is even stampable. Think of y=f(x) – if there are two values of y for a single x, you don’t have a function, i.e., the part’s not able to be stamped.
It may serve as a good mould for composite materials, though, but I can’t attest to their material properties to say with any confidence.