Aging of 302 Springs 1

[Aero1207]

A vendor wound a large lot of 302 SST springs about two years ago. Springs were stress
relieved by heating to 450°F for 30 minutes, work hardened by cycling to solid height 100 times,
vendor then measured load at working ht and then placed them on his shelf.

Two years later, vendor says loads of the same springs have consistently increased by about 1 to 2 %.
Is it possible that spring load can increase over time for the unloaded springs given the scenario
described, i.e. forming, stress relieving, work hardening, sitting for two years?

 

[dgallup]

I rather doubt it. More likely his load tester calibration has changed 1 to 2%.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[EdStainless]

If you had said 1/2% I would say yes.
These are very highly stressed parts, and they weren't stress relieved after the final setting so they could change a little.
302 is metastable and 20k hours is a long time. But at least part of this is a calibration offest.


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[israelkk]

To my best experience and knowledge it is possible. This is a recover of the spring free length after heavy preset. We had even worst cases where the change was much greater. Those were very high preset spring that the spring free height was recovered after sitting on the shelf for a long time. We solved the problem by keeping the spring pressed to constant length even when on the shelf.

Regarding the 1 to 2% change in your case, if such a small change is detrimental to your system, to my experience you are expecting too much from a spring to be such accurate. It may suggest that such a system is not well designed. If such accuracy is needed, that suggests that an adjusting and calibrating means for the spring load should be incorporated into the system design.

 

[israelkk]

I would like to add that if such a small change is important, do not be surprised if the springs already installed in the systems for long time are no longer exert the same load they exerted when installed, due to long time relaxation. Unfortunately, this may be revealed when such a system is serviced or failed in operation where needed.

 

[Aero1207]

Thanks. System design is tenuous, no doubt, but cost for a more robust solution and space constraints
really lock us in.

This has been helpful so far and confirmed some of what I was thinking. One other item
that was brought to our attention from a heat treat vendor was that our temperature of 450° F may
be too low and should be scaled according to the rated tensile strength of the spring wire. We have
been using 450° for a number of years but I believe that was based on very small wire diameters, on
the order of .005-.010". The spring that is the subject of this thread is.059" in dia. Heat treat
vendor suggested heating to 600-700°F for 30 minutes. Would a higher temp help to further
reduce stress and subsequently improve the stability of the working ht. load over time?

 

[dgallup]

We specify 302 stress relieve at 700 F for 30 minutes, typical wire size is .020". Some spring vendors have said that is too high but it was developed when we actually had a metallurgist on staff & he insisted that was the correct temperature for 302.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[EdStainless]

600-650 is very typical.
I would go with that range.
You will get a little size change when you do it.

We used to measure the magnetic properties of the 3xx spring wire.
When you see those properties change then you know that you still have issues.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[tbuelna]

A tolerance of 2% is pretty tight for a force at height measurement. With a spring having a low rate and long length, it might be difficult to control the force at height characteristics to within 2%.

Also, are you sure the 302 wire was stressed sufficiently to work hardened it by repeatedly coil binding the spring? A compression spring with a .010" wire diameter and a coil OD of .059" likely has a very low spring rate. Depending on how many active coils it has the wire stresses at coil bind may vary quite a bit, and the amount of compression set it experiences would also vary.

 

[israelkk]

According to MIL-STD-29A SPRINGS, MECHANICAL, DRAWING REQUIREMENTS FOR

30. SPRING MANUFACTURE

30.1 GENERAL. Information on only a few operations is essential for design purposes.

1. STRESS RELIVING. The usual types of hardening and tempering ovens are used for stress relieving. Springs made from prehardened wire such as Music Wire, Oil Tempered & Hard Drawn, Corrosion Resisting 18-8 and similar materials are stress relieved by heating at low temperatures from 400 to 660°F to reduce the residual stress trapped
in the wire during the coiling operation.

In SPEC's SPRING DESIGN HANDBOOK Table 3.4 the range for austenitic stainless steels is 450 to 950°F

From: SPEC's SPRING DESIGN HANDBOOK

3. Heat Setting - various procedures can be employed to expose springs to stress and heat for varying times to prepare for subsequent exposures. Depending on the method used, the effect is to remove a usually large first-stage relaxation and/or to establish a residual stress system which will lessen relaxation influences. In some cases, the latter approach can be so effective that in application, compression springs may "grow" or exhibit negative relaxation. Increase in free length does not
usually exceed 1 to 2%.

 

[tbuelna]

Interesting. I just checked the Associated Spring catalog, and it shows a +/-10% tolerance for load at working height of their standard compression springs.

 

[israelkk]

tbuelna

This is the standard commercial tolerance. To ask for tighter tolerances means to test every spring from the manufactured lot and sort those that qualify. This is very costly process and it doesn't assures long term accuracy due to relaxation and other phenomena.

 

[tbuelna]

israelkk-

I agree with what you say. But the OP asked about a variation of just 1%-2% in force at height characteristics, which is a huge difference from +/-10%.

As an aside, I recently ordered some custom stainless steel (17-7) wave springs from Smalley. Just before they were delivered I got a call from a QA guy at Smalley asking if I would accept the lot of springs that measured just 0.10 lbf outside the force at height tolerance limits of what the purchase order required. I accepted the lot. But it was nice that the QA guy at Smalley made the effort to call about such a small discrepancy.

 

[Aero1207]

A big THANK YOU to all contributors. We have a better understanding of the issue and have established a path forward.
This is a DOD part that is currently getting a lot of use so each improvement is greatly appreciated.