How do i begin making a carbon fibre monocoque?

[kilcoo316]

To be honest were only a small institution and its very hard to find everyday books let alone magazines. Although i have visited the website i think i will get myself a subscription anyway. Do any of you recommend any back issues to order?

Ask about inter library loans then - it shouldn't be a problem.

[Ciro Pabón]

please read teh link i posted earlier it shows all of these the answers to all of these questions

Thanks, nice link!

[flynfrog]

it hasnt been updated in a while but solar car teams are probaly one of the first groups to build composite chassis


i can get some more pics of our molds and what not after i get a good rest ive been up 2 days strait now

[sion_williams]

Sorry about that flynfrog, i totally missed your post. Pictures speak 1000 words. Good link that!

Im currently working on the nose cone plug ready to make a mold for this years formula student. And ive also done alot of fibreglass work so hopefully, provided i design a chassis well, it should be a reletively easy task. "he sais confidently"

[speedrcr]

I'm assuming your question based more on the manufacturing side. But from my experience, the general procedure in the design portion is:
1.) Design. This is a partial list of information needed to start the design of your structure.
1.a) External Lofting. Usually done in UG or Catia is essential if proper molds are to be made and to insure that there is no surface waviness (surface going from positive to negative curvature)
1.b) Loads. With a shape defined, suspension/mass locations defined, you can start developing the suspension attachment reaction loads. To do this correctly, you will need good tire data (static and transient) and someone who knows how to do statics....which anyone in engineering should be able to do. Spring/dashpot loads will also need to be estimated with either shock dyno testing, or possibly real life driving with a load cell installed inline with the shock.
1.c) Materials. Not my area of expertiece, but look at several manufacturers of balanced cloth or unidirectional tape. Also consider the shape of your molds. Certain weaves have better "drape" then others so it should be a consideration. Core (Nomex, Aluminum, Carbon ect) if required will also need to be specified.

2.) Analysis. With shape, loads, and materials defined you will need to define the number and orientation of the plies. There are two ways to go about this. The first is:
2.a.) Classic Laminate Analysis. Start with the external loft of your chassis, and take several cross sections. Calculate the EI, JG, and QG section properties. You will need to make an initial "guess" on what the final laminate axial (E) and shear (G) modulii will be. With this step complete, the laminate running loads can be determined. Equipped with this information you can start on the laminate analysis. You will either need a lot of patience to go thru several iteration of classic laminate analysis procedures, or write a program to do it. I learned more about laminate analysis writing my own program then I did doing anything else. With this completed, you should now know what your lamina strains are. Now, the decision must be made as to which failure method you want to use. Tsai-Wu, Max Strain, Max Stress, Hill, Hoffman, and whether or not you want to look at first ply failure or do a progressive failure to determine the last ply failure. We typically only look at a first ply failure, which in general is conservative. I would also HIGHLY recommend doing small subcomponent testing for analysis validation.
2.b) FEA, I don't recommend doing this until you are comfortable with all of the hand analysis methods outlined previously. Without a good working knowledge of hand analysis, FEA can lead you down the wrong path. The old saying is, FEA WON'T make a good engineer a better engineer, but it WILL make a poor engineer a dangerous one. So assuming all the previous, import the surface loft you defined in step 1(a) define approximate property locations by dividing the surface into several smaller surfaces. Define material, property cards and mesh size. Try to use QUAD4 elements as much as possible as triangular elements tend to act as stress concentrations. Mesh density is up to you, but the more the better...within reason. Apply loads to the model, there are several options with this but remember that you want to represent the joint as closely as possible. Constraints think long and hard about which DOF you need to constrain to get a realistic load distribution, as this will have a HUGE effect on your results. Run mode, process results, add/remove plies/core as necessary.

This in no way is a complete list of everything required. But if you want to do a good job and not end up with a fancy, expensive carbon anchor, then this is the minimum required. I have seen quite a few carbon chassis's at the FSAE comp, which were either to flimsy or were about 2x heavier then was required.

Erik

[sion_williams]

Thanks for the info speedrcr, I didnt realis that this much work was required, alas its a very good learning tool knowing what is needed to be done. You have said to use CATIA, i am familiar with this programme but have never actually used it or even able to get a copy. I do have a programme called IDEAS, dont know if your familiar with this but its all i have.

Do you have and would be willing to share with me some examples of your work? I am very intrigued with this subject and would like to see some professional work

[AeroGT3]

I'm the aero lead for my school's FSAE team. we were one of the first schools to use a CF monocoque. unfortunately i can't really give you much information as i don't do anything chassis wise. i'd recommend http://www.fsae.com though.

[Jersey Tom]

How do you begin making a carbon fiber monocoque..

1) Justify doing it!! Is the heaps of extra time, money, and testing worth it? Will there be substantial weight savings? Is the extra stiffness necessary? How stiff is stiff enough in torsion and in bending? Is time better spent pulling weight out of something else? One FSAE school finally got their CF chassis squared away after 5 years of development, and their car still weighs over 500 lbs. Ours, with a steel tube chassis, is 460.

2) Understand all the rules of your race series with respect to how much of the chassis can be CF, if anything still needs to be steel, etc.

3) Become well-versed in mechanics and practicalities of composities. Weave vs uni, different weights, effect of ply orientation and size, effects of different resins (polyester vs epoxy vs whatever), ultimate strength, elastic modulus, etc. One FSAE school estimated a torsional rigidity of their drive case of something like 7000 ftlbs/degree, and it wound up being 700 ftlbs/degree. Ooops!

4) Make sure you can find carbon fiber. This past year there was a severe shortage to the point most composites places in the US could only get carbon for their top-priority defense contracts.

5) Determine a good, SOLID way to mount all your suspension etc pickups. Figure out the exact position of everything in advance. Not like a steel chassis you can weld on tabs as you go, its all gotta be set in stone in advance.

6) A good CAD program would be nice. CATIA's industrial licenses are $45,000 a piece, but it can do all sorts of different composities stuff with ply layout and orientation, etc.

7) Design the thing. You dont 'convert' tube chassis to carbon ones. Spaceframe geometry is laid out by completely different rules, triangulation and usage of space which is needed for rigidity. Carbon is a different ballgame. Its geometry driven.. figure out all your pickup points, for the seat, suspension, engine, etc. Flesh in from there.

Design molds.

9) Fabricate molds. TAKES A LONG TIME. If you can break it down into chunks and machine them out on a large 3- or 5- axis CNC mill, you will save yourself a lot of time.

10) Layup, and hope you did it right.

That's how I'd go about doing it. But I'm not a composites guy.. just another engineer.

[pauljohnmartin16]

I could tell you but alas this is an old post. I designed my first carbon mono in 1982. Made many 100's since. I even designed the McLaren F1 supercar mono back in 1991. There is a world of difference between a group of ex patternmakers and laminators who have set up their own composite facility and offer essentially a build to print service and someone who really understands not only the fundamentals but also what you really should be doing. No black art, no old wive's tales , no folklore, just sound honest engineering.If you want to design monos that is what you need and you will struggle to find that in a book.

[holycowman]

Sheesh pauljohnmartin16, I could use you in my back pocket for a current (not 2006) project : )