MK IV tubular spaceframe drawing

mark b.

Supporter
The perimeter triangulation does nothing for torsional stiffness. Looks heavy. I would design a center tunnel running the length between the seats, the larger the better, and use sheer panels where needed instead of all the triangulation. At the rear I suspect the engine/tranny is going to be a stressed member? The rearward weight transfer on acceleration goes into the rear suspension, with the top of the shocks at that height, the only weight that transfers down onto the suspension is the weight above the top of the shock mount, not much. All weight below that point is transferred to the rear and 'locks' the suspension, does nothing for weight transfer. Course I'm not an engineer and this is worth about as much as you paid:). Good luck.
 

H.J.Bayer

Supporter
I've taken some of your suggestions on board now.
I also had a good conversation with a car designer.
It was hard to make the foot space as big as necessary despite the long R8 wishbones.
we have also tried to consider the crash zone issue.
Everything not so simple....
Now I'll see what the designer means.
In the appendix you can find a 3D PDF, you can hide the sheets at the layers.
The whole thing is not light but hopefully stiff and safe :)
The process continues ....:cool:
 

Attachments

The perimeter triangulation does nothing for torsional stiffness. Looks heavy. I would design a center tunnel running the length between the seats, the larger the better, and use sheer panels where needed instead of all the triangulation. At the rear I suspect the engine/tranny is going to be a stressed member? The rearward weight transfer on acceleration goes into the rear suspension, with the top of the shocks at that height, the only weight that transfers down onto the suspension is the weight above the top of the shock mount, not much. All weight below that point is transferred to the rear and 'locks' the suspension, does nothing for weight transfer. Course I'm not an engineer and this is worth about as much as you paid:). Good luck.
I take issue with this. If the side structure/perimeter is properly built/configured, including stiff bulkheads, it will more efficiently carry torsional load than a center tunnel since they are further from the neutral axis. They also have the virtue of providing side crash protection. I might also look at a diagonal between the two rear stays. Your comment on shear panels is right, but in order to really provide shear strength, they have to be rigidly fastened to the structure. Just using pop rivets won't provide enough integrity. In this case, triangulation is a belt and suspenders approach to it, at the cost of some weight. Since this car leaves so much on the table from lack of aero/downforce etc, a little extra weight is a drop in the bucket.
 
The biggest weakness (torsionally) that I see is through the area near the front of the side "pods". The structure needs something to tie the two pods together like a large transverse box well connected to the side structures.
 

H.J.Bayer

Supporter
Thank you for your answer!
I am also of the opinion that a central tube does not bring so much.
But it causes a lot of problems regarding crash safety.

@Bob
I'm not quite sure what you mean, can you draw a line on the picture to indicate which area you mean?
Thanks a lot!
 

Mike Pass

Supporter
Your space frame is lacking in triangulation which is important in preventing "lozenging" and contributes a lot to stiffness. I have attached a pic of a Van Diemen Formula Ford chassis to illustrate.
A central spine chassis is fine for front to rear axle stiffness. Pics of Lotus Elan and Europa chassis to illustrate. However the critical draw back to these chassis is the total lack of side impact protection.
These 3 examples are well known for excellent handling due to good torsional stiffness with lightness.
Cheers
Mike
 

Attachments

Central spine chassis may be OK for lower powered cars like the Elan, but think of V8 powered examples and descriptions of wierd, spooky and terrifying handling come to the fore e.g. Lotus 30, De Tomaso Mangusta.
 
HJ, assuming your program does not do stress/strain analysis, it might be a good idea to build a model out of balsa wood. That will help you find where the problems are. As Bob noted, stiffness of the bulk heads is crucial, as that's where the loads get fed through the chassis, front to back and vice versa. Super stiff side pods won't keep the chassis stiff if the bulkheads lozenge. The rear bulk head is pretty easy to build stiff since it doesn't have an aperture, but the front is more difficult.
 

H.J.Bayer

Supporter
Hi,
I've redrawn everything now, starting from a skeleton model.
All lines are connected at nodes.
I want to use this model in a static analysis program.
Almost everything is made of triangles
As a test and for illustration I also created the profiles.
Looks pretty heavy. There are about 500 parts
"If a Little is good, More is better, and Too Much is just enough..." :)

Better so...?
 

Attachments

Last edited:

Neil

Supporter
H.J.;

I've built chassis with round tubing and with square/rectangular tubing. Believe me, round is far more difficult.
 
Top