Stiffening a Roadster 1

[scrimshaw]

I am looking at ways to make a roadster chassis stiffer, I know it'sa common problem!. Chassis has not been built yet so I have no problems making adjustments. Here is as much relevant info I can think of.

Overall the car is similar to a 20s/30s Ford roadster. Two seater but wheelbase has been stretched to 140”. Body tub is FRP using sandwich panels for stiffness and wood laminate/foam at high stress points. Minimal steel at hinge points etc. Tub is built to support itself, doors, trunk but at the moment is not intended to contribute much to overall chassis stiffness.

The chassis is ladder configuration using 2 main rails originally sized at 2x4 rectangle tube at 11 gauge. Front suspension is independent with full crossmember, rear is live axle with parallel leaf.

Chassis is not low so I am hoping to run the propshaft under the chassis so I can use full 2x4 crossmembers just aft of bellhousing and before rear kickup effectively making a box directly under the tub. Maybe diagonal crossmembers as well. (Too much?)

I am aware of the inherent problems of roadsters and that torsional rigidity is always lacking because of the door openings. I am concerned about this twist popping the doors open especially with this long wheelbase. Here are the suggestions/thoughts I have so far.

a) Enlarge main rails to 2x6 or even 2x8, something like this.

http://www.artmorrison.com/2006cat/17.pdf

This is my favourite.
b) Change configuration of main rails to 4x4.
c) Construct a fairly elaborate framework of 1” tube under the cowl around the firewall, doors and behind the rear bulkhead, and bolt to the chassis. This is favoured by many FRP constructors though their bodies are very thin and flexible. My thoughts are that this will only minimally stiffen the chassis and is more for body support, door hinges, windshield etc.
something like this

http://www.wescottsauto.com/32-ford-roadster/p-32-ford-roadster-2.htm

There will be no racing or standing quarter miles, engine will probably be Ford 302 V8 relatively stock. Estimated weight is very roughly 3500lbs. (Or less I hope!)
Raising the door sill is not possible. I realize there are many unknowns so I am not looking for numbers just your thoughts on what would be the most efficient way to stiffen everything up if necessary. I would welcome any completely different ideas that don't require advanced machinery/construction techniques.
Thanks for your time.

 

[GregLocock]

The cage in the last link will do nothing for the torsional rigidity of the vehicle.

Improving the stiffness of ladder chassis is not incredibly difficult - but they are not very easy to analyse by hand. I use a beam element FEA model and an optimiser that tries changing each beam section, to give the best mixture of weight and stiffness. Using this I've designed a ladder chassis that is 3 times as stiff as the original (55 Tbird), for a 20% weight increase. The FEA agrees with physical tests on the original chassis to better than 15% error.

General principles-

1 triangulate to the opposing rail wherever the main rails are bent.

2 make the main rails as large as possible

3 square sections are better than rectangles, but a larger rectangle b x a is better than a square section a x a , where b>a . The ratio of b/a shouldn't be too extreme (not a problem in my application)

4 Joint efficiency is crucial. Take the x members right through the rails and weld at both intersections.

5 Round tube is often recommended for efficiency reasons, I think this doesn't apply in space limited applications. My calculations say a square tube a x a x t is better than a tube of diameter a and the same weight - because I am interested in the bending stiffness as well as torsionals.

I don't like the sound of 8x2 rails, I think 6x4 or 8x4 would be better. I've got one bit that is 8 x7.

Running the propshaft under the chassis will make designing it much easier.


Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[scrimshaw]

Greg

Thanks for the reply and confirming my thoughts on the my third option.

You dont mention the thickness of the material. If you are talking about 11 gauge it gets pretty heavy at 8x4.

Thanks for the info in your first 3 points, that helped me a lot.

I don't quite understand your 4th point. Do you mean to take the x member through TO the rail or are you talking about someother type of joint that goes through the rail?

I will probably not be using any round tube.

Thanks again for your time.

 

[GregLocock]

The T bird chassis, where it can, uses boxes that are as large as possible that are still thick enough to be welded.

The x member needs to go right through the main rail, and is welded to both sides of the main rail, and some additional stiffening may be advisable.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[scrimshaw]

Thanks for the info Greg.