Repairing a broken ARB

[alangbaker]

Some time in my race on Saturday (I honestly don't know when despite my "incredible" driver's sensitivity! ) the rear anti-roll bar on my RF98 broke at the weld between the bottom pivoting tube and the 1/2" vertical torsion bar.

Can this be welded back together successfully, and if so, what steps do I need to take to make sure it goes well?

[S Lathrop]

Someone may have much better advise than I can give but here is what I would recommend.

First grind the crack enough to assure that when you weld it you get full penetration.

Second have it TIG welded. I would check with a welding supply supplier as to what welding rod they would recommend. Most likely the sway bar is 4130 or a similar alloy.

Third post heat the welded area with a gas torch to just about glowing, and let it cool slowly. This is to normalize the welded area.

[S Lathrop]

Someone may have much better advise than I can give but here is what I would recomment.

First grind the crack enough to assure that when you weld it you get full penetration.

Second have it TIG welded. I would check with a welding supply supplier as to what welding rod they would recommend. Most likely the sway bar is 4130 or a similar alloy.

Third heat the welded area with a gas torch to just about glowing, and let it cool slowly. This is to normalize the welded area.

[mark elder]

I saw 4 broken ones this season alone in F1600 in New Zealand (one was mine). I have seen breaks mid shaft, down the bottom near the pivot and at the pivot (always just the shaft, never the actual pivot). Some were really old and some new (ie no consistent pattern).

The only repairs that seemed to last had extra plate partially wrapped around and welded across the break (TIG). But this concerns me that it increases the stiffness by an unknown amount.

I have half thought of trying to measure the ARB stiffness but keep getting distracted with other stuff (like getting to the track and racing; 7 two day meetings in the last 3 months all over the country). Given that the twisting resistance is a torque this must be easy enough to do. Odd thoughts include (on the car) disconnecting one bar and pulling on the end of it with fishing scales until the two bar ends are, say, an inch apart (yes you could use degrees of angle between the bars but that seems too hard and further from the real world). The force would then be some sort of measure.

Clearly, I have not given this much thought. Can someone put me out of my misery? Derek Sewards Race Car Design book tries to cover it but I get a little lost (my reports used to read 'could try harder in class', 'gets distracted easily')


I do have a dog in this race, I put a softer bar on when it rains and if it is pouring down (think aquaplaning), I disconnect it completely. But this is completely empirical (ie it works). Numbers would be nice. And it would show that a welded repair was (or was not) much stiffer than original.

[S Lathrop]

I saw 4 broken ones this season alone in F1600 in New Zealand (one was mine). I have seen breaks mid shaft, down the bottom near the pivot and at the pivot (always just the shaft, never the actual pivot). Some were really old and some new (ie no consistent pattern).

The only repairs that seemed to last had extra plate partially wrapped around and welded across the break (TIG). But this concerns me that it increases the stiffness by an unknown amount.

I have half thought of trying to measure the ARB stiffness but keep getting distracted with other stuff (like getting to the track and racing; 7 two day meetings in the last 3 months all over the country). Given that the twisting resistance is a torque this must be easy enough to do. Odd thoughts include (on the car) disconnecting one bar and pulling on the end of it with fishing scales until the two bar ends are, say, an inch apart (yes you could use degrees of angle between the bars but that seems too hard and further from the real world). The force would then be some sort of measure.

Clearly, I have not given this much thought. Can someone put me out of my misery? Derek Sewards Race Car Design book tries to cover it but I get a little lost (my reports used to read 'could try harder in class', 'gets distracted easily')


I do have a dog in this race, I put a softer bar on when it rains and if it is pouring down (think aquaplaning), I disconnect it completely. But this is completely empirical (ie it works). Numbers would be nice. And it would show that a welded repair was (or was not) much stiffer than original.

The math equations for a sway bar is not that difficult to make. Many of the books on race car suspension will give the formulas.

When I test the sway bar stiffness, I do it by placing the car on stands. Remove the wheel on one side. I remove the shock on the same side and leave the sway bar hooked up. I measure the distance from the upright to the ground. Then I hang a weight on the upright and take a new measurement. Simply divide the weight by the distance the upright moved and you have the spring rate of the sway bar at the wheel. This is the spring rate for a single wheel movement. The spring rate for the sway bar in roll, 2 wheels moving in the opposite direction is double the single wheel rate.

If you really want to know how stiff your car is in roll, you need one more number: the torsional stiffness of the chassis. Then you can calculate the the actual roll stiffness for the suspension. This is the number the tires actually see when you are cornering. For any car that I am working with, I actually run a test for this number. I have measured numbers from 1200 ft. lbs. per degree to 18,000. Doing a torsion test is an excellent way to check the condition of the chassis and chase down handling problems if they start to appear.

[problemchild]

With the current trends of very soft rear springs and RARBs, harder spec tires, and aggressive curb crashing, it is common to bend or break the RARB regularly. Many of us have lengthened ARB arms or relocated link ends to help this situation, and others have gone to the chopstick rear bar. I try to re-zero my RARB on the pad as often as I can, perhaps every session at some tracks.

I have repaired bars for emergency spare use but considering the situation, don't think that is proper for continual use. I suggest switching to a configuration with similar RARB rate but more linear application that will be more reliable but also make for an easier to drive car.