Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Vibratory Stress Relief for 4140 Shaft Machining Stresses

Status
Not open for further replies.

IdealME

Mechanical
Aug 10, 2005
2
US
We’ve been using VSR for over 15 years to remove thermal stresses from large, A36 plate fabrications of welded components with reasonable results. We are trying to apply it toward removing residual stresses after shaft machining. I've been researching VSR for use in manufacturing 800 lb, AISI 4140, induction motor shaft assemblies.

Rotor construction has the motor core biased toward one end of the shaft. The core end has a duplex ball bearing set close to the core. At the opposite end, the shaft (and equipment) are overhung (cantilevered) almost half the shaft length past a journal bearing that is closest to the core.

The general shaft diameter is relatively uniform over its length; with the greater diametrical stock reductions located at the ends. The shaft size reductions are as great as 4¼" diametrical (stock removed) at the ends. We are using VSR on the shaft machining to relieve residual machining stresses. It is suspected that there are also additional mechanical stresses associated with shrink fitting of the laminated core to the shaft that are not being addressed.

It is preferred to use VSR over thermal stress relief to avoid any reduction in strength, unnecessary distortion and straightening, and surface oxidation of the rotor assembly. VSR also allows use to have in-house control of the stress relief process.

Modal analysis and critical speed predictions indicate that the mechanical design is stabile. When functioning purely as a motor, the unit operates as predicted. However, when operated in its integrated application, the shaft has been noted to develop as much as .004" eccentricity at the far end of the overhang. Considerable testing suggests that functional loading is probably not having an influence on stability.

To the best of our knowledge there is no known, proven evidence of what science is behind the VSR process. We are trying to better understand the mechanics and metallurgy behind VSR and what it can successfully accomplish. Also, we need to know what measurable (NDT) indicators there are to demonstrate that the rotors are effectively stress relieved. We are currently using TIR to demonstrate compliance.

Is there any suggests on how we can be successful with this method?
 
Replies continue below

Recommended for you

The only "VSR" process of which I'm aware is application of a peening gun to straighten large weldments. I would not characterize the result as 'removing stress', but as adding stress in such a way as to balance whatever stress is already locked in. Machine away either stressed area, and the remainder distorts the part again.

If you're talking about something completely different, ignore this message.





Mike Halloran
Pembroke Pines, FL, USA
 
The most repeatable means of monitoring and documenting the effects of vibratory stress relief is to monitor, compare, and track the progress of resonance pattern change. This change accompanies effective vibration treatment, and manifests itself in two ways:

1. Growth of the resonance peak(s) to higher amplitudes. This change is due to increased compliance, i.e., reduced rigidity of the workpiece as treatment progresses.

2. Shift of the resonance peak(s) in the direction of lower frequency. This change is NOT DUE to the stress relief, but rather due to the growth of the resonance peaks. (This change is also more subtle, being typically 1 - 3% of the amount of growth.

These changes can be seen if the speed regulation of the vibration is tight, so that the resonance peaks can be accurately gauged as treatment continues, and the peaks grow and shift. Acceleration is the preferred parameter to measure amplitude, being proporational to the load the workpiece undergoes.

Setup should minimize the damping of the workpiece. For rectangular stuctures, using a 3 point placement of rubber isolation cushions, placed far from the corners, achieves this goal. (Placement near the corners might allow the bend mode resonances, but severely dampens the torsional modes.)

For shafts, most successful arrangements employ two V-blocks located lengthwise close to the center of the shaft. Normally these are separated by 20% of the shaft's full length.

Keep in mind that there will be metallurgical effects from the vibration treatment, only lowering of residual stresses that are likely to threaten the dimensional stability of the workpiece.

You can see an abstract of a paper recently published by ASM International that presents this technology at:
The paper also includes an extensive list of references that show a steady progress of understanding of vibratory stress relief, plus a survey of companies that use the method. That is not to say that there are unanswered questions. Several universities have submitted, or are ccurrently crafting, proposals to study vibratory stress relief. As the use of vibratory stress relief continues, some answers to these questions will emerge. It is clear, however, that for at least large precision components, it is not only faster and more practical than PWHT, but often results in more stable, predictable components.

One case study that is presented in the paper attests to this point. The workpiece was an 18 meter long gantry for a milling machine. It remains stable and accurate to this day (two years after install) + / - 0.45 mm, full length, as does a twin to it that was also stress relieved using vibration. Vibration treatment was done both before and after rough machining.

When delivered and erected, it was found that the machine had some mild out-of-true problems. Components that had been PWHT'd had distorted while in transit. The 18 meter gantry had not, so it was used as a reference to align the machine.


BK
 
OOPS, try 0.045 mm!!

Amazing how much a zero is worth!

BK
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top