Twisted anti-roll bar

[cgscgs]

Due to a gentle brush with the wall my rear ARB is twisted about 3 degrees. There is enough adjustment to dial out the pre-load. Is this something I can live with or does it need to be un-twisted? Thanks.

[billtebbutt]

My view is that you throw it out and replace with new. Once they start to bend/yield, they do not stop - they have given up whatever consistency they were built with

cheers
BT

[Purple Frog]

That's something I would PM Richard Pare about. He or Steve Lathrop.

[Lotus7]

Once steel has yielded into the plastic range, it no longer behaves linearly;
sadly you own a paperweight, urge you to buy a new one.

[R. Pare]

It will continue to twist yield, but at lower stress levels, though I cannot tell you just how much lower. Get a new one made, hopefully with a better alloy.

[Robins Ken]

Once steel has yielded into the plastic range, it no longer behaves linearly;
sadly you own a paperweight, urge you to buy a new one.

Can anyone explain in simple (that's me) terms how the Van Diemen bars function? ie: where is the 'twist'.

[jchracer]

I would only be concerned if you yielded (permanent twist) it far enough to initiate a crack in the bar.....highly unlikely....If you are really worried, an engine shop could magneflux it. Re-adjust and move on. You may have slightly reduced it's fatigue life but the stiffness should not be significantly altered.

[JHerscher]

Presuming you mean later bars, from the vertical bar. The torsion is from the twist imparted by the crossbar.

[BeerBudgetRacing]

Can anyone explain in simple (that's me) terms how the Van Diemen bars function? ie: where is the 'twist'.

On the chop-stick bars I believe it's the portion where the 2 chop-sticks are mounted but mostly the bend of the chop-sticks themselves.

Had/have the T style myself.

[Robins Ken]

Presuming you mean later bars, from the vertical bar. The torsion is from the twist imparted by the crossbar.

I meant the earlier ones circa 1990. Thanks Ken.

[provamo]

or so I thought
.
Is was the frame of our Lola T-204

way back when, lol

[problemchild]

Sorry, but I will disagree with the experts once again.

The only operational difference with a bent ARB is that you cannot adjust it on the fly. If an ARB is totally symmetrical, then you can slide the links on each arm and have reasonable expectation of maintaining your crossweights without putting the car on a pad or flat surface. Once you have twisted it, you need to check cornerweights (after adjustment) to ensure you have not preloaded it.

The simple solution is to disconnect one side, put a stiff hollow bar over the ARB arm, and tweek it until it matches the other side. Use a level if you don't trust your eye. Hook it up and go racing. Five minutes spent and zero cost.

For those that like spending money buying parts, buy a new one!

FWIW, the ARBs on Mygales bend all the time. That can be from mild wheelbanging or just driving over curbs. Mygale teams straighten them all the time. Because my cars have blades on each ARB at each end of the car, I have tooling that I can replace the blades with when I straighten them. Straightening ARBs saves $1000s, money that I don't need to collect from my customers or absorb as the team. More money for racing!

For those that question my advice, my cars win a lot of races at the highest level of competition. I would much rather have ARBs that bend than break. That costs us money! That applies equally to having a Tiga CF or a Mygale FF.

[old 59]

I meant the earlier ones circa 1990. Thanks Ken.

Front or rear?

[Lotus7]

Greg, your reply leads me to expand on my previous one - considering pure metallurgy vs the practicalities of racing....

I agree that a yielded bar can be bent straight and reused, and will function in an anti-roll context...
will it be linear? maybe not
will driver notice? very probably not
will it save him many dollars? very much so :-)

[Robins Ken]

Front or rear?

Rear.

[DaveW]

...The only operational difference with a bent ARB is that you cannot adjust it on the fly. If an ARB is totally symmetrical, then you can slide the links on each arm and have reasonable expectation of maintaining your crossweights without putting the car on a pad or flat surface. Once you have twisted it, you need to check cornerweights (after adjustment) to ensure you have not preloaded it.

The simple solution is to disconnect one side, put a stiff hollow bar over the ARB arm, and tweek it until it matches the other side. Use a level if you don't trust your eye. Hook it up and go racing. Five minutes spent and zero cost....

Straightening ARBs saves $1000s, money that I don't need to collect from my customers or absorb as the team. More money for racing!...

Greg,

As usual, spot on. I can't remember how many times I've straightened and continued to use twisted/bent ARB's. It's gotta be in the multiple dozens. And, once straightened, they always functioned as if they were new.

The key to all of this is that steel modulus/stiffness does not change significantly even when it has yielded. The only caveat is that the deflection to yield after deformation may be different due to internal stresses the initial and subsequent yields caused. In practice, this will not cause issues as long as one inspects the bar after some use and restraightens it if it took another set.

[Lyngengr]

The key to all of this is that steel modulus/stiffness does not change even when it has yielded. The only caveat is that the deflection to yield after deformation may be different due to internal stresses the initial and subsequent yields caused. In practice, this will not cause issues as long as one inspects the bar after some use and restraightens it if it took another set.

As a mechanical engineer, let me explain what is happening when a typical metal is yielded. In essence, you now have a fatigue life problem. If you bend the material enough to get a permanent set, there is a zone of plastic deformation, which on a stress-strain curve does not return to a state of zero strain or stress. Where it returns depends on how much stress was applied. Depending on where this new setpoint is, a subsequent application of load may take more or less stress to yield again. You need to know very accurately how much initial yield stress was applied to determine the response to subsequent loads.

So straightening a bent roll bar is basically introducing two low-cycle fatigue events to the material. Will it take additional stress cycles below yield? Sure, but remember the yield point has now changed, probably lowered. Keep in mind that to be produced in a "U" shape, the material would have to be yielded at least once already. The key is to keep any additional loads below the new yield point of the material, after all the bending and straightening. Actually, to be safe, any additional loads should be less than half the new ultimate strength to avoid any high-cycle fatigue problems.

The difference between low-cycle and high-cycle fatigue is that low-cycle involves plastic deformation, think of straightening a paper clip. How many times can you do that before the clip breaks? That is, in essence, what you are doing when you straighten a bent roll bar.

[Rick Kirchner]

To reply to Ken Robbin's question:

the "chopstick" rear and "u" front bars bend both in the uprights and in the crossbar. If you make a custom one you can simplify the design process a bit by making either the crossbar very large and the uprights thin (which might lead to some rate variance side to side) or make the uprights very stiff and the crossbar thin. If the crossbar is short I believe that can lend itself to yield problems. I'm not an ME, I only impersonate one occasionally.

I was trying to simplify the back of my car - eliminate hanging the bar off the wing to make gear changes easier, and also to eliminate the T-blade bar I had which requires slotting the bodywork for the pushrods.

Alana Spurling and Art Smith helped me with this folded rear bar design where we tried to make the uprights and what I would call the "transfer tubes" - the sleeves that allow you to position the uprights while actuating the far ends of the crossbar - as stiff as possible to take them out of the design process.

I always left the blades on my rear bar on full soft, so Art tried to calculate the T bar rate, then we designed the new bar to match that as closely as possible. I don't have the numbers handy, but it came out pretty close (as close as my lead ass can feel it).

[DaveW]

...The difference between low-cycle and high-cycle fatigue is that low-cycle involves plastic deformation, think of straightening a paper clip. How many times can you do that before the clip breaks? That is, in essence, what you are doing when you straighten a bent roll bar.

Except that the plastic-strain level in a twisted swaybar is much less than what you induce in paper clip when you repeatedly bend it far enough to break it. As Greg said, experience teaches... And I can say for a fact that I never had a straightened swaybar break due to straightening it from a reasonable amount of bending.

And, BTW, I was also a mechanical engineer by profession.

[problemchild]

Putting a new ARB on a 45 year old car has to change the handling of the car. Might be better, but certainly different, and I am assuming, would need to be tuned to the car. One more factor in the process to be considered.

[LJennings]

A few assumptions here

• Material 4130 tubing
• 90 deg bends both ends
• Heat treated to less than 40 RC
• Only slightly bent - no kinks
• No cracks

should be okay to bend back and you may end up with a .01% stiffer bar

[Robins Ken]

To reply to Ken Robbin's question:

the "chopstick" rear and "u" front bars bend both in the uprights and in the crossbar. If you make a custom one you can simplify the design process a bit by making either the crossbar very large and the uprights thin (which might lead to some rate variance side to side) or make the uprights very stiff and the crossbar thin. If the crossbar is short I believe that can lend itself to yield problems. I'm not an ME, I only impersonate one occasionally.

I was trying to simplify the back of my car - eliminate hanging the bar off the wing to make gear changes easier, and also to eliminate the T-blade bar I had which requires slotting the bodywork for the pushrods.

Alana Spurling and Art Smith helped me with this folded rear bar design where we tried to make the uprights and what I would call the "transfer tubes" - the sleeves that allow you to position the uprights while actuating the far ends of the crossbar - as stiff as possible to take them out of the design process.

I always left the blades on my rear bar on full soft, so Art tried to calculate the T bar rate, then we designed the new bar to match that as closely as possible. I don't have the numbers handy, but it came out pretty close (as close as my lead ass can feel it).

That's stout ! Nicely done. More thoughts on a 90's VD rear. Do the long drop links flex as they make their way over the bell housing and gearbox?

[Rick Kirchner]

I wouldn't doubt it. As you know, you can remove them off of a T-blade while the car is on the ground. Once on blocks, they don't come off easily, but I could always bend them out of the way.

[ejsprint30]

These parts are perfect candidates for the cryo process...especially after any welding or straightening operations to eliminate any newly introduced stresses.

[ejsprint30]

When I got my car, the front arb was bent right at the welded tab area. It seemed to easy to straighten, and it bent again on my next session...and yes, after it bends you can not adjust on the fly. I ended up buying a new one and decided that I wasn't going to weld the adjustment tab on it, as I was unsure if the original one lost some toughness after the welding operation. I made a clamp on, and so far it is working flawlessly.

[t walgamuth]

That's stout ! Nicely done. More thoughts on a 90's VD rear. Do the long drop links flex as they make their way over the bell housing and gearbox?

There is always flex over a load point....the question is is it enough for you measure it?

[t walgamuth]

In thoughts of the OP.....I was wondering if the actual bar bent or something else....a link or the frame? My first reaction was adjust for it and run the car. If there is a difference in the handling you can always replace it later.

[Rick Iverson]

Kirchner, how tall are your blades?

To reply to Ken Robbin's question:

the "chopstick" rear and "u" front bars bend both in the uprights and in the crossbar. If you make a custom one you can simplify the design process a bit by making either the crossbar very large and the uprights thin (which might lead to some rate variance side to side) or make the uprights very stiff and the crossbar thin. If the crossbar is short I believe that can lend itself to yield problems. I'm not an ME, I only impersonate one occasionally.

I was trying to simplify the back of my car - eliminate hanging the bar off the wing to make gear changes easier, and also to eliminate the T-blade bar I had which requires slotting the bodywork for the pushrods.

Alana Spurling and Art Smith helped me with this folded rear bar design where we tried to make the uprights and what I would call the "transfer tubes" - the sleeves that allow you to position the uprights while actuating the far ends of the crossbar - as stiff as possible to take them out of the design process.

I always left the blades on my rear bar on full soft, so Art tried to calculate the T bar rate, then we designed the new bar to match that as closely as possible. I don't have the numbers handy, but it came out pretty close (as close as my lead ass can feel it).

[Rick Iverson]

Kirchner;

The reason I asked on the lengths is they look shorter than mine.

To reply to Ken Robbin's question:

the "chopstick" rear and "u" front bars bend both in the uprights and in the crossbar. If you make a custom one you can simplify the design process a bit by making either the crossbar very large and the uprights thin (which might lead to some rate variance side to side) or make the uprights very stiff and the crossbar thin. If the crossbar is short I believe that can lend itself to yield problems. I'm not an ME, I only impersonate one occasionally.

I was trying to simplify the back of my car - eliminate hanging the bar off the wing to make gear changes easier, and also to eliminate the T-blade bar I had which requires slotting the bodywork for the pushrods.

Alana Spurling and Art Smith helped me with this folded rear bar design where we tried to make the uprights and what I would call the "transfer tubes" - the sleeves that allow you to position the uprights while actuating the far ends of the crossbar - as stiff as possible to take them out of the design process.

I always left the blades on my rear bar on full soft, so Art tried to calculate the T bar rate, then we designed the new bar to match that as closely as possible. I don't have the numbers handy, but it came out pretty close (as close as my lead ass can feel it).

[Rick Kirchner]

Don't know for sure. As you can see, the pushrods just barely miss the cross brace. The reason I made this bar was because at the time I couldn't afford that zippy blade adjustable bar. My bearing mounts for instance, were made from heavy aluminum L with a sawzall and a snag grinder. I also never used the adjuster, and wanted to eliminate it.