Sway Bar Design 4

[MGMech]

Hi,
We are designing new Cabin Say Bar (Stabilizer bar or Anti-roll bar). I am searching information for material selection for good fatigue properties and also good weldability, proper heat treatment, optimum welding parameters. It seems quite big job to me.

We will use hollow tube, it is welded to forged steel at both ends. material alterantives seem 34MnB5, 26MnB5, AISI 4140(42CrMo4), AISI 4130 (25CrMo4). I haven't seen any other alternatives than these ones. Austenize temperature and quenching time can not be optimized because our supplier always make the same process. Austizeing is done between 900-1000 Celciues degree, quenching is done at oil. We can only optimize tempering temperature. I want to test weldability and fatigue performance of sway tubes those are tempered at three different tempering temperatures. I do not know if it will worth to check different tempering temperatures.

I am not sure for Forged metal at ends. I think welding to forge metal is better than welding to cast iron but I could not find good information to provide. And I think it would be better if sway tube material matches with forged steel. So, I am close to select 4140 forged steel.

I need to prepare welding and fatigue properties with GMAW (MIG) Welding. For serial production, Magnetic Impelled Arc Welding is best however machines are very expensive like 1 million dollar. So it is also possible to make GMAW at serial production.

It seems like opmization problem between welding and best fatigue performance. I will be happy if you can help me.

 

[CoryPad]

Why are you welding? Spring steels are not easily welded without cracking, and the welding will reduce the strength and fatigue life. Most stabilizer bars use bolted connections.

 

[metman]

4140 is not a good choice for welding. 4130 is more weldable and and don't know of any reason not to forge.

6150 is a commomonly used spring material but not sure about forgability.

Design for RELIABILITY, manufacturability, and maintainability

 

[tbuelna]

For a torsion member you'll want to use drawn/seamless tube, which can be procured in 4130 or 4140. For your end fittings both 4130 and 4140 can be die forged hot.

MIG welded 4130 might give satisfactory results provided the weld is carefully designed and the weldment receives proper post weld heat treatment. Regarding fatigue properties at the MIG weld joint, you may need to adjust the local section thickness to compensate for any knock down applied to the weld HAZ.

4140 would give better mechanical properties, but it is not easy to fusion weld. However, you might look at modifying your design to allow friction welding the end fitting to the tube. Friction welding is a solid state joining process that produces a weld with excellent metallurgy, so a friction weld would be much better than a fusion weld in terms of fatigue performance. While friction welding will likely be a bit more expensive than MIG welding (even with large numbers of parts), you'll have to weigh added cost against the improved fatigue performance.

Lastly, for an automotive sway bar torsion member shot peening is a cost effective way to improve fatigue performance.

Hope that helps.
Terry

 

[FennLane]

Welding to any cast iron would a risky process and unlikely to provide very good results.

Forgings would be much better and if you used Friction Welding steels of slightly different composition would not present any real issues.

Even with Friction Welding you will need to consider joint design if you use any steel with a Carbon Equivalent of greater than 0.4.

The squareness of the ends prior to welding is quite important and if you can profile the forging so you have a tube/tube type joint you will obtain better results than trying to weld a tube to a bar. If you weld a tube to a bar the weld interface will tend to misalign and you could have problems with the end of the tube becoming significantly misaligned. the cooling rate would also mean that the HAZ in the bar/block of steel may become too high and some cracking could occur.

If you use a typical Continuous Drive Friction Welder then you will have a TIR between the two parts of around 10 thou for each joint at the best abut this depends on how well you can fixture the forging.

If the ends of the sway bar need to be aligned one to the other you will need to find a friction welder with two drive heads that are geared together. This type of machine is quite rear and was normally used for the manufacture of propshafts.

You could also consider Flash Butt Welding or Magnetic Arc Welding.

If very high levels of alignment are required then Electron Beam Welding would be worth considering.

 

[MGMech]

Thanks for your answers.

(CoryPad) At Cabin Sway Bars all competitors are also welding. Actually if no one was welding sway bar, I would not be the first one, however that is cheap and simple design. Beacuse of weldin process, we do not use spring steel like 55CrV4. Instead, 26MnB5, 30MnB5, 4130, 4140 are candidates as I've mentioned.

(metman) AISI 6150, in other words 55CrV4 is very risky to weld. I agree with you that 4130 is better choice for weldability than 4140. As I know they are both good for forging. One important issue is as I see at benchmark, our main competitor used 4140. I'm trying to find out the logic why they haven't used 4130 instead of 4140. Only reason that I can think could be convenience to find 4140 at market.

(tbuelna) Thanks for advices. We design the ends as just you've mentioned. Thickness of hollow sway bar is thickened to twice at ends to reduce stress. I'm studying on the issue that do I have to make post weld heat treatment. Because end section thickness and also diameter is more than bar, stress becomes half of the bar.
Friction welding gives good results however I could not find supplier at my country that has capability to make friction welding for both sides at the same time. And also while we know that we can make this welding with MIG cheaper, making friction welding would be luxury choice for us.
I also examine shot peening process to maka hollow tube before welding. After welding, just cooling at air slowly could be sufficient.

(FennLane) Beside welding to cast iron is risky, I see also that forging end parts would be cheaper, so it's win win for me. For aligning, we consider pin inside bar. Hollow tube is fitted to pin, welding groove at the ends of hollow tube is filled with welding material. I think slow cooling is required. I will try to achieve withoud post welding treatment however I may also prepare one prototype with post heat treatment after welding and see the fatigue test results.
I still look for two drive head friction welding machine. I talked with propshaft supplier however they did not interested with sway bars.
I was not aware of flash butt welding. I will try to communicate with suppliers tomorrow, thanks.
For Magmetic Arc welding, that is great however machine cost is too high (app. 1 million €) also maintance and control of parameters are required every year.
I will also try to learn details of Electron Beam Welding, however at first sight it seems expensive and could be over engineering.

Thanks you all.

 

[Tmoose]

Could you provide a picture of one of these welded sway bars?
Preferably one that has been successful for a few years of service.

I'm wondering if the ends are tapered somewhat larger than the center section, so the poor geometry and metallurgy of the welded end are in a relatively low stress region.

 

[tbuelna]

Not too long ago it was common to see TIG welded 4130 roll bars used on race cars. But I think the configuration is a bit different than what MGMech is considering.

 

[MGMech]

Some samples each are used at OEMs (not after market) can be seen at this site.

http://www.kongsbergautomotive.com/...iveline/vehicle-dynamics/cabin-anti-roll-bar/

Ends are forged to the diameter is bigger for all examples. "Poor geometry and metallurgy of the welded end are in a relatively low stress region", that's right.

As I've mentioned, strong material candidates are 26MnB5, 30MnB5, 34MnB5. These are Boron Steels. I can not find sufficient information for these steels, like mechanical properties at different annealed conditions, fatigue properties, TT diagrams. I will be very happy if you can give me a link for Boron Steel properties.

Thanks

 

[MGMech]

I talked with Flash Butt Welding (FBW) Machine supplier. He stated that Hollow Tubes can not be welded with FBW. Also diameter of tube, bar or thickness of beam should not be more than 7mm. So, FBW is not choice for my case. I'm not sure if this information is related only for their machine or general requirement though.

 

[TVP]

MGMech,

You need to investigate Friction Welding or Resistance Welding (specifically, Resistance Projection Welding using an annular projection feature). Annular projection welding is also called Ring Projection Welding. Resistance welding can use either mid-frequency DC or capacitor discharge power supplies. Here are some links for more information:

http://www.thompson-friction-welding.com/

http://www.twi-global.com/capabilities/joining-technologies/friction-processes/

http://www.twi-global.com/technical...of-annular-projections-in-resistance-welding/

http://mercury.kau.ac.kr/welding/We...ction Welding and Resistance Seam Welding.pdf

http://peco-us.com/wp-content/uploads/2014/10/Prod_Weld_combined_REV812.pdf

 

[FennLane]

In another life I was the Chief Metallurgist at Thompson Friction Welding from 1972 until 1979 when I left to take a research post with regard to Elasto-Plastic Fracture Mechanics.

We used Friction welding to replace many flash butt welding machines used to make tubular components including trailer axles and Grade E Drill pipe so I an surprised to hear that tubes can't be welded using this technique.

 

[MGMech]

Thanks for information FennLane. I will check welding suppliers.
For my case, it seems like we will continue with MIG welding. I will prepare prototypes with MIG welding. It is the most economic solution.
As I read at research papares, heat treatment (so hardness and microstructure) is not effective for weldability. That is good new for me, my prototype variation will be reduced.
If heat treatment will be applied to reduce residual stresses, welding different materials will be problem because heat treatment requirements are different. For example welding 26MnB5 to 4140. So, I will prepare prototypes for welding 30MnB5 to 30MnB5 and also 4140 to 4140.
To evaluate best welding parameters, we will make statical mechanical tests like tensile test, squeze test and checking hardness of welding location. After specifying best welding parameters, we will make torsional fatigue test. I can not find any reference that evaluates torsional fatigue test via flexural or tensile fatigue test. It would be more easier for me to make other test than torsional fatigue test, unfortanately it seems that not make sense.

 

[tbuelna]

I could see where linear friction welding might be difficult with a thin wall tubular cross section at the weld interface, since the surface overlap around the interface is not uniform in relation to the direction of motion used during welding. Rotary friction welding would work if there were some way to index the parts when the weld is completed. Another thing that might work would be to swage the tube ends prior to welding so that they have a much thicker local wall, which would allow linear friction welding to be used.

BTW, I took a look at the sway bars shown on your company's website. They are some very nice looking products.