Calculating Torsion In A Chassis? 2

[willwoll100]

Hello guys and gals I'm brand new to the forum and have question, here goes;

I've done a torsion test on a car chassis with one end secured and the other end attached to a device which has a bar connected perpendicular to the chassis which rests on an arced section of steel on the floor. Now the arced section was placed centrally to the chassis and a load applied on the end of the bar. There was 6 DTI's set up at various points of the chassis (3 either side) and the deflection was taken. Now this was repeated when the shear plates were removed and only the top one in place then only the bottom. So the data that I've got is the load applied, angle of twist, radius of twist, distance from fixed point to load applied and deflection. Now what I was wondering is, does anyone know how to calculate the torque of the chassis per degree, I know how to calculate it for simple round bars/tubes but am confused on how to do it for a chassis. Any help would be very much appreciated.

Will

 

[GregLocock]

Torque applied= load*radius

Doesn't get much easier than that does it?

Your degrees of twist are just the angular motion of your rocker system.

Cheers

Greg Locock

 

[willwoll100]

I don't think that is right for what I need as this would give me the same torsional stiffness for the chassis when it had the shear panels on and off? Or are you saying to incorporate the angle of twist as well?

Will

 

[GregLocock]

I don't understand where your shear plates are.

Cheers

Greg Locock

 

[willwoll100]

Sorry I should have said that the chassis is a ladder frame style with shear plates on the top and the bottom.

Will

 

[GregLocock]

So why does their presence or absence affect how you calculate the torional stiffness?

perhaps you'd be better to post an example of your calculation, this is going nowhere fast.

Cheers

Greg Locock

 

[willwoll100]

It doesn't effect how I calculate it its just that the chassis should be around 2 - 2.5 times stiffer than with them fitted.

Sample Calc:

(mass*gravity*deflection)/angle of twist

This will give you N.m/degree, now I'm not sure if this is the right calculation to do?
Thanks for your patience and help

Will

 

[mloew]

Will,

The formula you are looking for as follows:

stiffness = torque (applied) / angular deflection (measured)
Torque applied = load applied * radius of force application

The mass of your structure only comes into calculations if you are trying to come up with a specific stiffness measurement.

Best regards,

Matthew Ian Loew
"Luck is the residue of design."
Branch Rickey

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

 

[willwoll100]

MLeow thnks alot maybe I should go to bed now as its the early hours and sorry to Greg for battering your head but thanks again.

Will

 

[sean01]

WillWoll100,

Although (Like Greg) I am still a little unclear about just exactly what you are trying to achieve - Mleows formula is the one I use for this sort of calculation. Two things I would add though.

(1) Make sure you work in radians !!!
(2) You may find that it is appropriate to add a torsional stiffness factor K and warping stiffness factor Cw into this calulation based not only on the geometry of your chassis - but also the way in which the fixed end is secured and the load applied.

Formulae for the calculation of Cw and K can be found in most stress formulae handbook - I personally use Roarks - but there are others which people can recommend.

The deflections measured by the DTIs should capture the detail you require to assess how effective your shear plates are in their respective positions. Be aware though that the effect of transverse shear stress can significantly decrease the torsional stiffness of the structure. This effectively sets a maximum limit for K as J - the polar moment of inertia (J = Ix + Iy)

For the structures I have experience of (Automotive chassis) this figure is generally in the 22 - 29 NM/deg.

Sean

 

[GregLocock]

Sean wrote:

For the structures I have experience of (Automotive chassis) this figure is generally in the 22 - 29 NM/deg.

Greg writes:

For the structures I have experience of (Automotive chassis) this figure is generally in the 1.5-30 kNm/deg.

(ie, you got the units wrong, fair enough, and the range is MUCH bigger than you might think, once you go back in history or consider some of the more extreme examples)

The lower end is/was things like Austin 7 replicas and other horrible ladder chassis. The top end is any decent German sports sedan. 10-20 is when the magic happens, above that you are getting worthwhile improvements, but it is no longer night and day.

If you have a 'normal' sedan the clue is that your first and second modes start to look like pure bending and pure torsion, whereas below the 'magic' stiffness you tend to have a pair of mixed torsion and bending modes.





Cheers

Greg Locock

 

[sacem1]

Try reading the "Super Seven chassis design thread where this topic has been discussed before.

A lot of good advice and ideas in there.

SACEM1