The basic plan is that the CF will be the actual frame but there will be steel buried and bonded inside as a just in case kind of thing....just to be safe.
Ive done a lot of research into just this over the last few years. I intend spending the next 5-10yrs building an open wheel racer also built with a fully built(all)engine.
The main issues with a CF whatever you want to call it part of the chassis is testing.
One chassis could display different different torsional rigidity to another(if your making more than one)and safety aspect. This depends on so many things as its nearly Impossible to replicate exactly from one to the next without an F1 house, or aerospace composite department to hand.
But, that wont matter if your fixing the suspension points to a metal spaceframe as the CF can only help stiffness.
The spaceframe too will have some degree of torsional flex. If you run it all in autocad and apply loads, or make a wire frame, no matter what you do, you wont spread the loads evenly over every member and end up with a practical light design(In my opinion). So, if your bonding into a CF monocoque I think maybe you need to figure out what it is you want the carbon to do for you.
Be it to provide extra torsional resistance(careful of load concentrations between the two materials)
Or to provide the cabin, and also provide you with panels that slow down sharp objects that would otherwise pass through the spaceframe easily. (They use approx 20 layers of Zylon in an F1 tub for this)
Ive thought long and hard about all the above as its very much on my mind for my racer. Its a real tough one.
There is a few ways I think to go about it, that is manufacturing for someone that is handy with their hands.
It moreso applies to an open wheel racer in my case, as the geometry is a little simpler than what maybe you have in mind but Ill state what Im thinking anyway, sure why not.
Sadly the main brain bending issues Im having is that of the Carbonfibre construction, and if it would be good enough off the drawings - without crash/impact testing/torsional test.
For the 'total' CF build I was thinking of a two piece mould, striked off a foam core. Moulds would comprise of a top and bottom section(top side of tub, and bottom side)
Part line may not be straight, and would follow along the 0 deg draft point along the side of the tub.
The tub would be made in two parts, top and bottom, and then bonded together.
Layup into both independent moulds first with x amount of layers. Given that I have not, and no chance of access to an autoclave, it would have to be a wetlayup. Layers would be vacuumed to mould.
Then, as soon as the resin allowed, bond in precut(s) sections of foam filled honey comb such as nida-core polyurethane foam. The foam would add to stiffness, and also stop the cells from filling with resin(more weight) when laying up the rest of the layers. Doing this as soon as the first vac'd layers allowed would ensure both mechanical and chemical bonding between the honeycomb and the first layers since crosslinking would not yet have ceased.
Once I had a 'certain' amount of layers done onto the H.comb, both top and bottom would be bonded to eachother.
Before this was done, Id bond in any bulkheads needed for suspension points, etc. Im not sure if you could have these anodized and primed alloy parts lay onto the last layers applied, since they would be isolated from the first outer skin due to the honeycomb interlayer.
Any suspension cornering forces or other would then be transmitted to just the internal tub skin. Im not sure if this would be ok. Also, in a shear situation, I wouldn't want the suspension fixture bulkheads moving sideways, and shearing off my legs. So maybe a weak point in the wishbones would be needed.
Same goes for the engine mounting points, whether to use flat bonded in ferruled/bolted through plates, or to bond in alloy tubes the length of the car between layers. These would have solid tapped sections at the ends for engine fixture points.
Same applies to these as the bulkheads, to bond the to interior layer, or exterior layer(honeycomb separating them), or omit honeycomb in those areas and bond to both?
Another way to do it, is to use piano wire, bent at 90 deg at the end, spun in a drill, to clear out some of the honeycomb. This internal space is then filled with a chopped strand mix, and when dry, bored out to accept the bolt/ferrule.
Once the top and bottom were bonded(honeycomb longer at edges in bottom part, sliding into a section in the top part like a mortise and tenon), I was then thinking of filleting around the bulkheads at the bond line with sphere filled epoxy mix. This would allow a smooth low energy transition between the two. Another set of layers could be applied over the bulkheads, and engine tubes bonding them further to the interior tub layers. These could be vac'd on, or via an inflatable bladder arrangement.
Where both meet at the outsides of tub, the moulds would have been shaped in such a way(may require removable parts for mould release) that there would be say a 100mm high, 1mm depression, or low zone all along the join line, where I could vac on CF strips to complete external join(Inside could be done the same, no need for 1mm depression.
Whew.... thats what I was thinking for the CF tub construction.
The spaceframe tub is a little simpler, just slightly...
Make it in steel tube as we have all seen.
Bonding/fixing CF to steel tube is not easy if you want to work in flat CF sheets. You will also have to form each section which you will see when viewing the car from the outside.
Bonding these sections to a spaceframe will require them to have two geometry's. One for the outer aesthetic layer, and another rear edge layer/profile fitting the spaceframe tube geometries perfectly for bonding.
How you make the moulds for these panels is a big job.
If you change to square tube spaceframe things get a little easier, but of course square/rect sections are not as torsionally stiff as round ones.
You could concentrate on just bonding flat CF panels to the interior of frame, adding a lot of stiffness. You could also vac on CF to the joins and maybe get an ok looking interior pan and sides.
The outer mammoth shell may then only be required for aesthetics.
I dont know how carbon bonds to steel tube, or how you would prep it, Ive never studied it.
Still having said that, Id be more confident with bonding to aluminium since Ive studied, once its anodized, primed, and isolated from the CF with a layer of glass.
Aluminium has a finite life though, and is a bit more choosy to weld.
One last possibly way is cutting(compound shapes) Foam to go into all the spaces between the tubes or box sections. You could then apply CF over the lot, Internally and externally. But, In your case Mark, you would still need an aesthetic skin on the outside.
Doing it this way, would I feel provide nothing more than intrusion protection, since its only really touching a bare tangent of the tubes and would not offer much to torsional rigidity.
You could too I suppose encase the tubes with the carbon monocoque. Gluing both internal, and external skins together, and then filling the internal void around the tubes with an expanding foam of sorts.
With all the above in mind I traveled over to Horacio Pagani last month to ask him about the construction of his cars Since he uses a CF monocoque, I also wanted to see the construction up close, Ill put all that in another post as this one has run way too long as is.
A lot to think about for sure. The main points that I feel need to be closely addressed, is if you are intending the carbon to add rigidity, If not thought out correctly, all it may add is extra weight.
I really mean it when I say I cannot wait to hear your thoughts Mark, on how your going to go about it all.