Vibratory vs Thermal stress relieve processes 1

[swaneyg]

I have a large fabrication that I call out for thermal stress relieve. It is fabricated from W10 X 100 ASTM A572 grade 50 steel. It's approximate overall size is 14W X 14L X 6H and has several wear plate bosses machined on it after stress releive. One of the vendors has taken an exception to thermal stress relieve and is offering vibratory. I am not familiar with this process and do not know if it will actually do what I need it to do. The tolerances for the machined surfaces or pretty close and portions of this &quot;frame&quot; will see ambient temperatures >900°F for short durations of <60 sec. I do not want the overal parallelism in the finished asembly to move around and cause any frictional interfearences or binding with the mechanism that operates within this fabrication.
Any help would be appreciated.
Glenn Swaney

 

[Carburize]

Do a search on this site - this subject was discussed in some detail a few months ago.

 

[bklauba]

We have a report describing an 18 m L milling machine gantry that was VSR Processed before and after rough machining. Resulting TIR was + or - 0.08mm, monitored over several weeks time thru assembly. Weight was ~ 17 tons.

But vibratory stress relief equipment varies greatly in terms of its performance, and the set-up is critical. The report supplies details and concepts. Post your e-mail address, and I can forward it.

 

[unclesyd]

If it is going to see 900°F for short durations I would at least take it to at least 900°F for 1 Hr. Not much stress relief but this exposure will uncover any unforseen movement due thermal Stresses imposed by this periodic exposure to 900°F. This is especially true if the 900°F exposure isn’t uniform.

 

[kenvlach]

See
Weld Stress Relief with Subharmonics
thread330-61212

 

[bklauba]

A more effective means than the subharmonic approach is using resonant vibration. See the website for the Center for Environmental & Energy Reserach at Alfred University,

http://ceer.alfred.edu/research/vibratory.html

for one of several papers that have made this point.

If a simulation of the 900 F. is going to be done (I wonder if the relatively uniform raising and lowering is a realistic simulation of 900 F. localized subjectation?), a good preparation for it, so as to minimize distortion is a properly performed VSR treatment. Manufacturers of a wide variety of precision components, from machine tools, paper mill machinery, mining equipment, and other large parts use this process.

Since the workpiece is relatively square, 3 load cushions should be placed at one of the 14 (feet?) long midpoints, and two placed only a few feet apart on the opposite side. This will allow the torsional mode of vibrtion to be minimally dampened, which will be (more than likely) dominant over the simple bend mode, and also highly effective in distributing load throughout the part. This load cushion arrangement will also allow other modes of vibration to be undergone, with minimal damping.

You can expect a growth and shifting of the resonance peak(s) (growth of ~ 20 - 40%, shifting [negative] of ~ 1 - 3%) as a result of dwelling on the peaks. A superimposed before and after slow scan through the vibrator speed range will allow documentation of this change in resonance pattern.

Once the resonance pattern has stabilized, the dimensional character of the workpiece will also be stabilized.

The equipment used to perform this process should have its accuracy and repeatability authenticated; vibrator speed regulation and accuracy should be significantly less than 1% (say 0.2% at most) and the accelerometer amplifier should also be linear within similar bounds.

 

[swaneyg]

Thanks all for your replies. I have about 20 projects going and an outage scheduled for the first week of December. I haven't had much time to read through all of this but I will as soon as I can get some loose ends tied up.

The contractor fabricating these items won't be ready for several months for stress relieving. That does give me a little time to read the posts and do some research.

Again, Thanks,
Glenn

 

[bklauba]

If you would like me to review the workpiece so as to recommend proper setup for vibratory stress relief, please send a sketch or drawing of it to me at bklauba@airmatic.com

I will also then attach to a reply materials on various large precision components than have been stabilized using this technique. Included will be a case study on the 18 meter (59.5 foot) long milling machine gantry (one of two that were stabilized) which remains straight + or - 0.08 mm (about 0.003&quot over the full length, according to the data included in the study that was supplied by the milling machine manufacturer's Q/A department. Charts showing the changes in the workpiece's response to vibration will also be included.

 

[BTIGUY]

Glen,

A good vsr process (subharmonics, as referenced by someone else) should work every bit as good as the TSR only if tempering is NOT needed. Often ASME calls out for stress relief when they really need stress tempering. Check it out. You mentioned that the goal is for machine stability and stability in service. That being the case, no problems foreseen with sub-harmonic vsr.

The CEER report referenced from Alfred Univ was a student's thesis. A review of this report reveals that this was theory only, no vsr was done to any test bar, only &quot;mechanical induced&quot; stress was applied not thermal, no cross reference to field performance, and no reference to over 150 articles written about subharmonic vsr processing including a previous report one from Alfred Univ. In this former report, 1994, for the US Army at Watervliet Arsenal, the report stated that the sub-resonant [sub-harmonic] vsr process was applied to cannnon barrels with a 99.7% confidence that they would produce successful results. This same report stated that these barrels experienced a 67% scrap rate using TSR only! I'm puzzled why the current CEER report from AU is available to anyone. Some people are going to have a hay-day with that report. Very misleading, most unfortunate.

Did you know that you can use subharmonic vsr during welding? This prevents 50-90% of normal weld distortion plus stress relieved the weldment as you weld.

Best of Luck
BTIGuy

 

[bklauba]

Dear BTIGuy,

Agreed, the CEER / Alfred Univ. Report was a graduate student's thesis. Most university published research starts in this form. Later, depending upon review of department heads (in this case, the Chairman of the Mech. Eng. Dept.) and peer review, i.e., an effort to vet work in a meritous manner, it gathers increasing support within its immediate community.

This, obviously, is what happened at Alfred Univ. Otherwise, they would not have taken the time and trouble (and bannered the work with CEER / Alfred Univ.'s name) to post it upon their website. This distinguishes it from the vast majority of work, both at AU and other universities and research institutions.

Putting that aside for the moment, the researchers at AU have joined many others in expressing the following point: On a fundamental basis, since more energy, more flexure and greater force levels occur within the material during resonance, (all of which appear to be deeply woven into the modus operandi of the vibratory stress relief process) how can an off-resonance approach be of any advantage??

There are several examiners of this issue. One notable one, since the material examined is quite similar to that mentioned at the beginning of this thread, is the work of Dr. S. Shankar, then at the Oregon Graduate Center (which has since been absorbed by the Univ. of Oregon). Quoting from the abstract of his work: Vibratory Stress Relief of Mild Steel Weldments, Feb. 1983 (ISSN: 0419-4217, MA Number 83-312652),

&quot;The influence of resonant and subresonant frequency vibration of the longitudinal residual stresses in A-36 mild steel weldments has been studied. Residual stress analysis was carried out using sectioning, X-ray, and blind-hole drilling techniques. . .The resonant frequency vibration had more pronounced stress redistribution as compared to the subresonant frequency vibration. Transmission electron microscopy studies indicated local plastic deformation as the mechanism by which the stress reduction occurred. Constant amplitude axial fatigue experiments on samples machined from regions adjacent to the weld showed that both vibratory techniques did not induce any fatigue damage.&quot;

So it would appear that both methods can relieve stress, but resonance has shown, not only in theory, but in practice, to be more effective.

Perhaps the subresonance approach is easier on the vibrators used to perform the process?? At the risk of anthropomophizing (or is it personification, I forget), electric motors hate to be vibrated, esp. in the motor brush/commutator area, or if the rotor or armature is high-inertia. I would think that tuning one upon the resonances of a structure it was mounted upon could be a pretty rough ride. A vibrator design involving very rugged, yet compact components, plus a rather tight servo-drive might help. Lacking some of these, subresonance more than likely would extend vibrator longevity.

 

[BTIGUY]

Glenn,

The explanation presented by bklauba sounds genuine. That's what makes it deceptive or perhaps he doesn't know the details of this report. The thesis from the Oregon Graduate Center that was referenced was even more out of touch than the thesis from Alfred Univ. In examining the &quot;resonant&quot; peak, according to Klilad Khan, who also worked on the OGC's report, &quot;They stopped increasing frequency when the part became unruly.&quot; Since they didn't want to work with an unruly part they stopped just short of the violent behavior yet called that [subharmonic] frequency &quot;resonance.&quot; Then for sub-resonance frequency they reduced the frequency down to 25%, 50% and 75% of the resonant FREQUENCY, not amplitude like what they were supposed to do. Of course they were way too low to affect a decent change, besides their original change in identifying the true harmonic peak was in error.

As far as resonance vs. sub-harmonic (sub-resonance) processing keep in mind that the sub-harmonic process has been &quot;specified&quot; by name by such organizations as NASA, US Army and Navy, General Motors, Grumman Aerospace and Boeing AFTER first trying and giving up with the resonant approach. These companies continue to use subharmonic processing frequently on numerous high-tech parts in place of heat treat stress relief. There must be a good reason.

Glenn, best of luck treating your weldment. Skip vsr if the weldment needs to be tempered at all. If machine stability, long-term stability, or reducing the possibility of premature fatigue are a concern sub-harmonic vibration processing should prove to be an acceptable option.