Looking for a Highly Fatigue Resistance Spring Material

[axleshox]

We're currently using a stainless 17-7 Condition C material.
The material needs to go through a 4 slide die for cold forming and cutting. Thus, we need it in a 2" width by 0.015" thick roll.

It is subjected to a stainless of about 0.005, 1/2% strain and must be resistant to fatigue for at least a 1 million cycles.

We've tried 304 and 316 alloys as well as nickel titanium alloys, i.e. Nitinol. The 304 and 316 were far less resistant to fatigue and came out slightly warped because of the heat treating proces. The Nitinol proved to be too difficult to manufacture and thus too expensive.

I'd appreciate any suggestions or leads.

Elmer Lee

 

[israelkk]

I am sorry but I could not understand your post. Are you designing a spring? If so, please explain the spring dimension, deflections, loads etc.

 

[axleshox]

Yes, we are designing a leaf spring, not a coil spring. There is no load requirement, only a deflection and fatigue requirement.

The question was not meant to elicit assistance in designing the spring itself. I simply was trying to get suggestions for materials.

The spring's max strain is about 0.005, which for most metals is pretty close to is fatigue limit, I think. I was hoping someone out there has work on applications requiring super elastic metals.

I hope that helps.

 

[israelkk]

First of all you can not differentiate deflection from force, a defection is usually a result of force, pressure etc.

However, at this strain the stress in the spring is approximately 103.7 kgf/mm2 or 147.5 ksi. According to AMS 5529B 17-7PH strips at CH900 condition has a minimum tensile stress of 240ksi. Therefore, the stress on the your leaf spring is far beyond the 0.4 to 0.5 of tensile stress or yield stress. This leads to the conclusion that it is beyond the metal fatigue limit even before we take into account surface defects etc.

There are not many materials stronger than 17-7PH but you can try to find Marage 300 or 350, Custom 475 from Carpenter or Elgiloy. However these materials are difficult to attain and they are expensive.

There is other things that you may do with the current 17-7PH but they fall into the design area and manufacture process.But, as you mentioned you do not need assistance in designing the spring itself.

 

[axleshox]

Thanks Israelkk,

We're using 17-7 condition C that goes thorough a 900F age hardening process. Would that be any better that 17-7PH. Appreciate your help and input.

Elmer

 

[israelkk]

The material comes in the C condtion to allow forming of the spring. After the heat treatment in 900F it becomes condition CH900 and the stength is going up to the 240ksi and as I mentioned in my last post.

You didn't mention how is the spring deflects is it defelects the 0.005 strain only to one direction all the time or both directions?

 

[axleshox]

It's a little envolved, but the gist of it is that the spring is essentially a cantilever beam. It's formed in the straight position and then the tip is raised and held at an upper position. During the lifespan, the tip is displaced downward upto 2x the amount it was originally displaced upwards.

The 0.005 strain number is the theoretical amount calculated using a FEM model and some fudging. It is achieved with the half amplitude oscillation, in other words, just by lifting the tip up by x, the max strain should be 0.005.

So, I believe the answer to your question is that it is subjected to 0.005 strain in both directions.

 

[israelkk]

If so then according to my calculations (based on minimum tensile strength of 240ksi) your spring should not last more than 20000 cycles. You may encounter larger results if the actual strip of 17-7PH will have higher tensile strength. However, one should design for the minimum the spec says because future batches could be in the minimum allowed properties.

 

[TVP]

axleshox,

Type 301 stainless steel is often used for small spring applications that require exceptional fatigue resistance. Type 301 develops its strength from strain hardening (cold working) instead of heat treatment, so warpage would not be a factor. Somers Thin Strip, a division of Olin Brass, offers Type 301 stainless steel strip in several different tempers, with the highest strength being 280 ksi ultimate tensile strength. I strongly recommend you contact them to discuss your application. Use the following link for more information:

http://www.olinbrass.com/somindex.html

 

[NickE]

Another stainless material with an endurance limit that is very high and also assured low inclusion and specifically designed for flat flexure is the Swedish Flapper valve Steels. One of these is Sandvik's proprietary grade of 420SS. It is called 7C27Mo2. There is a higher strength version called High-Flex.

Google: Sandvik Materials Technology.

(Also you will find that surface conditon has tons of effect on fatigue performance)

nick

 

[mattis]

continuing on the paranthesis in the previous post. Surface condition is extremly important here do not forget to shot peen the spring to make sure you have negative compressive stresses in the surface.
good luck

 

[NickE]

"by 0.015" thick roll."

Another note from the first job I had in the steel industry.

Toilet Paper comes in rolls. Steel comes in coils.

 

[axleshox]

I've tried shot peening. The material is far too thin. Rather than compressing the surface layer it deformes the actual spring shape.

My mistake NickE. I'll know what to use when talking to a toilet paper manufacturer now. You've saved me from the ultimate embarrassment.

 

[mattis]

About shot peening, I know realize that you have quite thin material. In my line of work 0,4 mm would be considered as foil... got confused by the imperial measures. Maybe try glass bead blasting which will give you residual stresses but not deform the surface too much. Residual stresses will probably deform the spring anyhow especially if you don’t get an even distribution on your part.

//ma

 

[israelkk]

axleshox

It seems we all grinding water. The solution is probably not by changing the material. I think you need to consider design change. If you can give more info on the spring shape and what it is supposed to do it may help. As you mentioned in your posts the deflection is what you are looking for. If so, you may be able to recieve the same delection with less strain using the same strip thickness or even thicker strip but with a different shape of spring.

 

[EdStainless]

No load requirement?
Have you looked at low modulus alloys? BeCu, BeNi or alpha/beta Ti? These would involve a lot less force for the same defection.

If you want to stay with the same stiffness tehn look at 21=6=9 (XM-11). It is available as re-melted coil product from Allegheny-Ludlum. It is used for aircraft hyrol lines for places where Ti isn't suitable due to temperature or damage tolerance requirements.
In tubing it is cold drawn to about 125ksi yield and 150ksi UTS, though it can be worked to higher values and still have good ductility.
I'll try to get an S-N curve for you.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[Compositepro]

Composites can have exceptional fatigue life. Fiberglass or carbon fiber can be combined with thermosets or thermoplastic resins for sping applications.

 

[NickE]

Another thought is that you might be able to utilize two thinner spings in parallel. IE: For flat bending at the same beam geometry and same P-P deflection a strip half the thickness sees far lower stresses. In my work I see this type of application every day. What helps here is that you say you have no load requirement. You may be able to parallel two half thickness strips and get roughly the same stiffness but far lower stresses.

Also you may want to look around at this website:

http://www.destacomanufacturing.com/technologies.htm

 

[Maui]

Elastic Design - Automobile Leaf Springs

How do you decide what material to use in designing a specific spring? Consider the leaf spring used for the suspension systems on older cars. These springs are essentially rectangular elastic beams loaded in bending. Suppose that such a beam of length L thickness t and width b can be modeled as simply supported at both ends, and loaded centrally with a force F. If the elastic modulus of the material is E, then the deflection D at the center of the beam is

D = (FL^3)/(4Ebt^3)

The maximum surface stress S is given by

S = (3FL)/(2bt^2)

If the spring is plastically deformed while under load, then it won’t be able to “spring back”. So we set the condition that the maximum stress must always be less than the yield stress Sy.

Sy > (3FL)/(2bt^2)

Substituting for F from the initial equation we find,

(Sy/E) > (6Dt)/(L^2)

For a spring in service to undergo a deflection D, the ratio of Sy/E must be high enough to prevent plastic deformation. So the best leaf springs are made from materials that have high values of Sy/E. For this reason spring materials tend to be heavily strengthened by work hardening, precipitation hardening, or solid solution strengthening. Note that small values of thickness t permit larger deflections D for the same value of Sy/E. This is why leaf springs are usually constructed of several beams stacked on top of one another.

I am uncertain of the specific geometry and loading conditions that you are using, but a similar analysis for your specific case should provide you with the combination of physical properties that will allow us to determine the appropriate materials for use in your design. If you do this analysis, get back to us and we may be able to provide you with specific materials that can satisfy your criteria.

Maui

 

[NickE]

Dang Maui you always manage to put the right equations down. Thanks for clarifying what I tried to say (I did badly I should add)