Hollow sway bar- metallurgists question?

[TRE Cup]

Looking to make a pair of hollow sway bars for a non Porsche app. Since these will be bent, the questions come up on happens to the strength ? What is the best material that will keep the torsional strength? or perhaps heat treating is required ?

we want to maintain a vintage appearance, so the "right way" of a straight bar actuated by lever arms (bent to fit the mounting points on the control arms) are too much of a change

Thanks in advance

[Flieger]

You should send chris_seven a PM.

[Quicksilver]

Torsion bars are carefully heat treated to get the correct characteristics for a spring: extreme resistance to yield, crack resistance/toughness.

Steel has a basically uniform modulus of elasticity regardless of the alloy and heat treat. The difference that we see is mostly the difference in yield point and crack resistance or toughness. By changing the heat treatment these properties can be changed across an incredible range. Alloys can modify the range of these properties but the modulus of elasticity doesn't change any appreciable amount.

So back to the core question...
A spring can not be bent (besides to destruction) without getting heat involved. All commercial springs are formed and then heat treated. In theory you could start with a finished spring, heat it and form it, and then re-heat treat it but it would be less steps to just form the spring from rough stock and then heat treat it. From your description I'm not real clear on what you are doing. Is it literally a bent bar or are the ends to be formed to a shape that fits the pickup points?

First major point I see is: Why are you looking at a hollow bar?
Hollow bars are only done for a little bit of weight savings. The metal on the outside of the bar contributes more to the spring rate then the metal on the inside. Removing the inside metal removes the least effective weight of steel. It introduces a problem though. Cracks always start on the surface. (assuming the basic forming isn't a train wreck...) Hollow bars adds a new surface for cracks to start from. You can't shot peen the inside so you are reducing the options for treatment to straight heat or immersion.

Any manufacture of bent bar swaybars should be able to pop something like this out very easily. A swaybar is a torsion bar so they will have the equipment, the material, the heat treating equipment, and the know how to knock this out.
As an alternative you could check into a manufacture of conventional springs. (A coil spring is a torsion bar would into a spiral.)

You might call Glen Sander Engineering in Torrance as they are the maker of about the best 4340 hollow bars out there. I suspect they farm out portions (or most) of the manufacturing process but they have definitely figured out what needs to happen at every step to make it work.

The last wild idea I have is to talk to extreme hard core pre war restoration people. The stuff they are working on was hand formed to begin with and the only way they get new stuff is to find the expert on hand forming it to recreate it. I wonder if you called the Nethercutt Museum in Sylmar, if they would give you a name of who they farm spring work to. They have an amazing expertise in recreating a car from nothing.

[chris_seven]

I tend to agree with most of the above comments concerning spring manufacture but as with all engineering problems the devil is always in the detail.

I would almost certainly want to manufacture a very highly stressed sway bar from a good grade of Spring steel.

Most torsion springs for automotive applications are made from silicon - manganese spring steel, of which the main constituents are: 0.55 - 0.6% C, 1.8 - 2.1% Si and 0.8 - 1.0 Mn. Typical processing gives a nominal hardness of 49.5 Rc, equivalent to a UTS of 1600 to 1700 MPa. These springs are designed for a maximum stress under full bump conditions of 1100 MPa.

A steel of this type would certainly need to be in a resonably soft condition before bending and would then need to be heat treated to develop the correct properties.

These types of material are difficult to find in a tubular form and I would imagine 'gun drilling' is completely out of the question.

You could therefore select a 4340 tube which is commonly used for roll cages and depending on the bend radius may not need heat treatment prior to being used.

The bend region will certainly work harden and this will provide some strengthening and fatigue property improvement which should be a benefit.

If the bend readius is too tight then too much cold work could result and the tube will beome less reistant to fatigue and may fail. It really needs some calculations and knowledge of the specific design.

There have been some examples of sway bars being made from mild steel tubes providing that the surface shear stresses are less than 50% of the materials yeild strength they may work quite well and would be easy to make.

The life expectancy of the bar also needs some thinking about as 4340 and mild steel can suffer from 'softening' when fatigue cycled, so parts made from these materials may have a shorter life than one made from a spring steel but this may not be an issue depending on the duty of the sway bar.

With regard to the use a tube then it is clear that for a given torsional stiffness a tube will always be lighter than a solid bar. Again without the detail it is impossible to quantify.

I wouldn't worry about fatigue due to an extra surface as the torsional stress is proportional to the 4th power of diameter and the outside diameter will almost always be the initiation site for a fatigue crack unless there is a gross inner defect present and with a good quaility cold drawn tube this is unlikely.

As an alternative to a 'bent' bar could you install a blade style unit? I did this on my Lotus Seven and it has been very good and easy to adjust.

If you know the wheel movement and the arc that end f the bar will move through and can send an outline sketch I would be happy to carry out a few grubby calcs and suggest a material.

Just as a final note I have been looking at buying some 'soft' torsion bars for a 911 and I am not confident that 4340 is a good material choice for this application.

I would much rather use a 'proper' spring steel manufactured by a specialist such as Bekaert and I have decided to make my own tubular bars and have them corrctly scragged.

[Flieger]

Carbon fiber. Painted black to look like steel.