Torsional Stiffness

[EgoFG]

When a car is released there is often a claim of say 15% greater 'torsional stiffness'.
The press often uses this expression or 'torsional rigidity'

Noticeably at time the FG release reviews the press said the VE had greater torsional stiffness.

What I would like to know is, since it is quoted in % gains it must be measurable - not just based on the driver impression.

Does anyone know of a source of data for 'torsional Stiffness' or 'rigidity'.
Some kind of competitive comparison.

thanks

[geckoGT]

Do not know of a source giving hard data that can be used as a comparison between manufacturers. It is more of a subjective observation that can be made in the handling characteristics of a vehicle.

A dead give away is opening a door when one wheel is jacked off the ground. Do this on a VN commodore and the door will not close again without misalignment causing it to strike on the opening. Do this on a BF and it closes easy, showing much less distortion of the body structure.

They do torsional rigidity testing in the design process which involves placing a measured amount of torsional force on the body and measuring the amount of deformation. Done with hydraulic rams and very sensitive measuring equipment.

[XR Martin]

About the only real 'comparison' I have seen.

http://www.youtube.com/watch?v=zRfE_XAk2mE

[geckoGT]

Awesome demonstration of the concept, thanks

[MAD]

Nice comparison, but that guy has sweet FA of an idea of what he is talking about.
I hate when sales people interview an engineer, have no idea what they are talking about, then try to replicate the information to the public.

Torsional stiffnes is a measurable unit, but it's effects on handling and are not so much. Many many factors come in to play when 'handling' is 'measured'.

[geckoGT]

Originally Posted by MAD

Nice comparison, but that guy has sweet FA of an idea of what he is talking about.
I hate when sales people interview an engineer, have no idea what they are talking about, then try to replicate the information to the public.

What were the faults?

[MAD]

Originally Posted by geckoGT

What were the faults?

Just some of the words he was using seemed in the wrong context. When he was talking about the frame being boxed, a line he used "the deep section was able to provide a good section and react the load". It sort of sounds right, but not at the same time, unless "react the load" is a term used in America.
I never heard that reference or anything like it while at uni, so it sounds out of place to me. Maybe I judged too quickly, maybe he was not very used to public speaking and got some of his words tied.

The whole experiment seemed like a very good example of what torsional stiffness is and how it can be measured.

[KPOJ]

The way we do it on race cars is to attach a 10+ foot pole and start putting you back in.

Its still a rough measurement but you know which car is stiffer because one is being twisted with the longer pole.

[geckoGT]

Originally Posted by MAD

Just some of the words he was using seemed in the wrong context. When he was talking about the frame being boxed, a line he used "the deep section was able to provide a good section and react the load". It sort of sounds right, but not at the same time, unless "react the load" is a term used in America.
I never heard that reference or anything like it while at uni, so it sounds out of place to me. Maybe I judged too quickly, maybe he was not very used to public speaking and got some of his words tied.

The whole experiment seemed like a very good example of what torsional stiffness is and how it can be measured.

Yes his public speaking was off and obviously not his forte. A lot of his theory when (you get past the terminology) was quite sound and the demonstration was good.

[Silver Ghia]

The demonstration and explanations of closed (boxed) section chassis vs. open (C) section chassis was quite sound, as geckoGT said.

Put simply, open (C) sectioned beams will twist (deflect) much more under torsion than similar dimensioned closed section beams. Get a short square section hollow tube in a vice at one end and a spanner at the other, and it will be quite stiff under torsion. Cut a slit along its length then see how much easier it is to twist it.

In the chassis demonstration, due to the cross members there is bending involved as well as torsion. The deeper section will be much stiffer in bending, especially so also being a box section. The torsional load applied at the rear is reacted by the various chassis members and finally at the anchor points (restraints). This is what is termed a 'load path'.

[madmelon]

Torsional stiffness is measured in Nm/degree. The aim is pretty much always for infinite but in reality it's a bit different. I think the alloy/carbon chassis Astons are somewhere around 30,000Nm/degree. Formula SAE racing cars (university students design and build) generally aim for about 1500Nm/degree. There are a large number of factors which determine how stiff you can make a chassis, including if you can put structural members through the cockpit (ie, roll cage/hoops/bars/etc).

[Silver Ghia]

Originally Posted by geckoGT

They do torsional rigidity testing in the design process which involves placing a measured amount of torsional force on the body and measuring the amount of deformation. Done with hydraulic rams and very sensitive measuring equipment.

They can actually check the torsional rigidity with the finite element (computer) model during the initial design phase, before the first article is made and tested. With the finite element model, they apply the loads on the model and see what the torsional rigidity is, and whether it meets their targets. They can then optimise the structure in the model by moving members around, changing thickness and shapes, add gussets and other members etc, to improve the torsional rigidity.

[Watto_Cobra]

Here's a video that really shows how sloppy the Toyota frame is next to the F-truck.

Youtube - Silver Creek

[Wally]

Originally Posted by EgoFG

When a car is released there is often a claim of say 15% greater 'torsional stiffness'.
The press often uses this expression or 'torsional rigidity'

Noticeably at time the FG release reviews the press said the VE had greater torsional stiffness.

What I would like to know is, since it is quoted in % gains it must be measurable - not just based on the driver impression.

Does anyone know of a source of data for 'torsional Stiffness' or 'rigidity'.
Some kind of competitive comparison.

thanks

There is static and dynamic deflections that take place. It's quite simple to simply measure compliance by shoving weights at various points along the frame and work out a cumulative result in radians per Nm, but it doesn't tell the whole story.

Just like when you size valve springs, you should find the natural resonant frequency to figure out best shape, mass, dampening, etc . To do this on structural frames you attach a shaker; on a car I suspect the frame is put in a jig of some sought with a shaker attached and some sought of accelerometer measures the natural resonant frequency at various points. I would guess the modal frequency where the maxiumum compliance occurs, becomes a yardstick, the higher the frequency the less dynamically compliant.

If you added up the side to side, end to end, diagonal to diagonal, etc compliances at various frequencies you'd get an overall cumulative compliance figure in radians per Nm, which that little video XR6 Martin's doesn't demonstrate . The mean or median frequency where the majority of cumulative compliance occurs would become another yardstick.

[KPOJ]

Originally Posted by madmelon

Torsional stiffness is measured in Nm/degree. The aim is pretty much always for infinite but in reality it's a bit different. I think the alloy/carbon chassis Astons are somewhere around 30,000Nm/degree. Formula SAE racing cars (university students design and build) generally aim for about 1500Nm/degree. There are a large number of factors which determine how stiff you can make a chassis, including if you can put structural members through the cockpit (ie, roll cage/hoops/bars/etc).

The astons torsional stiffness is amazing... is that for a race or street aston?

On the other hand the formula sae is quiet low, but i guess they don't have much weight behind them

My hyundai's TS is about 20,000-23,000Nm

[ltd]

Originally Posted by Watto_Cobra

Here's a video that really shows how sloppy the Toyota frame is next to the F-truck.

Youtube - Silver Creek

That's one of the coolest things I've ever seen. Thanks for posting that.

[EgoFG]

This is all good stuff.

The press has said VE has better torsional stiffness - how do we know they are correct.

These are the best comments I have been able to dredge up.

http://www.goauto.com.au/mellor/mell...1?OpenDocument

Originally Posted by goauto

Visibility from the cavernous new interior, which now accommodates even the tallest of occupants and then some, is not only better than before, but better than in the Commodore, which has an unusually thick A-pillar that encroaches on forward vision. Ford says it has strict A-pillar width dimensions globally, and that it achieved its chassis rigidity targets without widening its A-pillars to the same degree (or fitting a fixed rear seat).

That said, Ford executives privately admit the 92 per cent new FG body is not as rigid as the VE's and that it does not deliver the same structural rigidity gains as the BA Falcon did in 2002, when Ford applied a $500 million facelift to the slow-selling AU that lifted body strength by almost 60 per cent.

With gains of up to 20 per cent, the BA-FG model change was never going to feel as significant as the structural integrity improvements wrought by the VE, which replaced VT-VZ Commodore lineage that dated back to at least 1997. In short, the FG comes off a much higher base than the VE and is therefore a victim of the law of diminishing returns.

and this may seem to contradict the previous one
http://www.drive.com.au/Editorial/Ar...rticleID=48306

Originally Posted by drive

The FG Falcon’s structure – also known as the body-in-white – is stronger in comparison with the BF Falcon’s, though it is also 21.3kg heavier.

So If both are correct FG has lost some torsional rigidity over the BF but gained strength in the "Body in White", and the FG does not have as good torsional strenght as the VE.

But this is opinion - not something with figures that we can see and take into account.

[MAD]

I think they are saying the BA-FG stiffness gains were not as large as the 60% improvement seen in the AU-BA upgrade.

[JPFS1]

Ego

FG hasn't gone backwards in terms of Rigidity when compared to BF. They are just saying that the gains in rigidity from AU to BA in % terms is greater than BF-FG. The FG is still more rigid than BF, just not as much an improvement in % terms. The structure is pretty optimised with current materials.

And the statement about the VE being more rigid is correct. Partly because its BIW is all new and it doesn't have to contend with the vulnerable split fold rear seat.

The numbers themselves are kept very close to their chest though for obvious reasons.

[ZA-289]

Originally Posted by XR6 Martin

About the only real 'comparison' I have seen.

http://www.youtube.com/watch?v=zRfE_XAk2mE

What a dumb video, of course a C section chassis is going to bend more than a boxed chassis!

I want my 5 minutes back.....

But as said, yes jacking up one corner of a vehicle and seeing which doors open and which don't is an easy way to tell.

then again, torsional ridgity dosn't mean alot when subframe mounts are soft as buggery (VE commodore)

[KPOJ]

Gave my hyundai race car the torsional stiffness test tonight, from what i worked out the x brace in the roof ties the whole car together. Most important bars in the cage after seeing all the affects of forces in different areas.

Had a problem at first, the 500kg chassis jig was lifting with the chassis. chassis had 3 points of connection so it can twist. few bolts into the shed floor fixed that, then my bar wasnt long enough to twist it in the end i had a 10ft bar with 100kg's on it, faintest sight of movement was visible. once i weld the roof bars in it will be perfect.

: : : :

[madmelon]

I have to correct myself on what I posted before. The Formula SAE team I'm involved with aims for about 5000Nm/degree not 1500. Gets close too.