Shot peening and surface finish 2

[inline6]

i want to get a small shaft for a piece of equipment shot peened for better fatigue resistance. is it beneficial to grind to a very fine almost polished surface finish before shot peening? o

 

[kumkumvijay]

Helli
Fatigue life is affected by surface finish. Shot peening increase the surface compressive stress and surface roughness.
I would suggest shot peening then roller burnishing or only roller burnishing.
Roller burnishing provide more deep surface compressive stress than shot peening and also good surface finish which is good for fatigue strength.
Surface compressive stress with good surface finish is good for good fatigue strength.

 

[CoryPad]

Usually not very beneficial to grind before peening, unless it is removing undesirable surface microstructures like oxidation, decarburization, etc.

 

[tbuelna]

inline6-

I agree with kumkumvijay's comments. There is benefit with regards to fatigue life by employing some type of surface modification process after shot peen, provided the modified surface has a texture profile where the valley bottoms are more smoothly blended (reduced profile slope) and thus create less of a local stress concentration. Making the surface "smoother" simply by removing the tops of the peaks, while leaving the valleys untouched, will not help the fatigue situation much.

The optimum amount of stock removal that is possible after shot peen depends on the process variables used for the shot peening. Factors such as shot size, hardness, intensity, coverage, impact angle, etc. affect the depth and degree of surface compression produced by the shot peening. If the surface being shot peened is very rough, has fillets with a radius similar to or smaller than the shot size, or has an orientation that the shot peen nozzle cannot be directed normal to, then the results of the shot peening in those areas will not be optimum.

Lastly, I personally don't think it would be necessary to both shot peen and roller burnish. Mechanical processes like roller burnishing, fillet rolling, or roll forming are much more effective than shot peening at producing a surface with high retained compressive stress. I would prefer roller burnishing over shot peening if the design of your shaft permits.

Good luck to you.
Terry

 

[Tmoose]

sounds like you have experienced some fatigue failures of this shaft. Can you provide more detail?

 

[inline6]

Thanks everyone, i have not seen a failure in the shaft at all.

I have an existing one where i need to modify it by reducing the diameter in one section hence i want to minimize the risk that fatigue occurs. I am lucky enough to be able to do this in a region away from peak bending moment but still in the area where I am modifying the nominal stress and SCF are highest risk of fatigue due to the reduced diameter.
I am not 100% sure on the load spectrum and material properties to have anything but qualitative evidence that things will be ok based on similar modifications done by others. Hence i want to get the best BFYB as far as improvement goes. It’s a one off thing so spending a bit of coin is ok.
I have access to someone who can shot peen and I can easily polish it myself but will look into roller burnishing

The other thing is should I use a ‘new’ shaft that has seen no cycles or will a used one be ok given the fact that I am machining away the outer surface where accumulated fatigue damage however small would have been highest? Diameter is reducing from 20mm to 15mm with transition radius each end

 

[Tmoose]

Is the loading steady state, or have some variability?
Regardless, I think the geometry where the smaller diameter turns into a larger diameter is at least as important as surface treatment. "Good" geometry and gentle finishing help during "static" loading (when failure by yield might be the concern) //AND// loading that tests endurance strength/fatigue limit.
I'm a fan of shot-peening but if the loading is ~ constant ( big torque at start up, with starts being infrequent, big unbalance, etc) versus constantly fluctuating (like reversed bending due to misalignment of coupled constantly rotating shafts, or jerky or pulsating processes) then shot peening or other fatigue enhancing processes don't have that much to offer. But good shaft geometry does.


Some basic shotpeening info with likely improvements in fatigue life here -

http://www.metalimprovement.com/pre.../MICShotPeeningApplicationsGuide-FullText.pdf

Withdrawn mil spec 13165 calls out shot size, intensity, and maximum post processing temperatures for various materials and section thicknesses.
I was also thinking it specified the maximum material that could be removed after peening ( 10% of Almen intensity??) but I could not find it in 13165C.

http://www.shotpeener.com/library/pdf/1989020.pdf

should you be considering possible excitation of torsional resonant frequencies?

 

[inline6]

basically its an idler shaft that carries bending loads created from tension of a drive belt, it rotates for many tens of millions of cycles in an engine with fully reversed stresses hence fatigue is important. there is no possibility of failure by static overload and torsional loads are not of concern in there area modified.

i have optimized geometry and much as i can with respect to lowering SCF and determined the maximum diameter that functionally will work. when i compare with a few similar modded shafts by others of slightly bigger diameter that create a sharp notch in a region of higher bending moment i should be a bit better off when the nominal stress x SCF is compared but the highly statistical nature of fatigue is a concern.

 

[Tmoose]

Is the belt sheave/pulley on an overhung portion of the shaft, or between support bearings?
Is the belt sheave mounted on the reduced diameter?

Unless the sheave mount is secured with an interference fit or wedging tapers ( taper lock hub, ringfedder, etc) I'd expect micromotions and fretting that will knock the stuffing ( and nearly a decimal) out of the material's fatigue strength.
Page 82 here-

http://www.timken.com/en-us/Knowled...s/Timken_Practical_Data_For_Metallurgists.pdf

It also will be trying persistently to loosen setscrews, eat keys and keyseats, and make a mess. Even with a tight fit the edges of hubs can bring severe micromotion problems on top of stress concentrations unless highly sculptured or other mitigating details are included.

http://www.me.metu.edu.tr/courses/me307/images/drawings/drwg13.jpg

http://www.me.metu.edu.tr/courses/me307/images/drawings/drwg14.jpg

If A keyseat emerges from under the edge of the hub life can be especially tough.

http://image.fourwheeler.com/f/9310...original_jeep_axles_old_school+break_axle.jpg

Shotpeening can help endurance in those fretting situations, but can only do so much.

 

[inline6]

pulley is on the overhung portion that isnt changing at all. the shaft is being reduced midway between the two bearings hence at point of half peak bending moment nothing is mounted in this area. bearings pulleys etc unchanged

 

[inline6]

what are peoples opinions on using a new shaft with zero cycles vs a used one? i ask because would not machining part of it away remove previous "fatigue damage" however little it is?

 

[metengr]

If you are going to re-use a shaft I would as a minimum perform a wet fluorescent MT or liquid penetrant test. If there are no linear indications, it should be suitable for use. As far as cumulative fatigue damage, this requires an understanding of service stresses in the shaft and using stress-life or strain-life fatigue data for the shaft material.

 

[tbuelna]

Unless the cost of using a new shaft presents a significant financial burden to your test effort, then it would make sense to use a new shaft. If I understand correctly, what you wish to determine with your test effort is the effect of modifying the profile and processing of your shaft design versus a baseline design, in terms of fatigue life. As with any testing effort, ideally you want to minimize the variables and use a test article with a known pedigree. A new shaft should present fewer unknowns than a used shaft.

It appears that you intend to conduct a test to validate the results of your shaft modification fatigue analysis. The highly statistical nature of fatigue life was discussed above, and fatigue testing can require quite a bit of time and expense. Obviously you want to make sure the test article and test set-up are well prepared so that testing goes smoothly.

 

[inline6]

no testing will be done only the real thing. i just want the longest life whether it be from a new shaft or i would get the same from an old one.

when i mentioned comparison i did hand calculations of my proposal against other modified ones ive seen which i feel are inferior due to having introduced sharp notches in a higher stressed portion of the shaft as a result of not transitioning with a nice radius, i didn't mean testing.

for the cost of a proper fatigue test i could make one from scratch out of much better steel