Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Modifying Sharp Edges for Fatigue

  1. Dec 22, 2016 #1
    Hello all,

    I was wondering, for modifying the sharp edges of a shaft, which one of the cases attached is better? Why? My guess would be the one with the radius since there are more materials remaining than the other but there should be a reason for the other case since that is also frequently done.

    Attached Files:

  2. jcsd
  3. Dec 23, 2016 #2


    User Avatar
    Gold Member

    Speaking from a baseline of pure ignorance of materials science, my guess would be the one with the smoothly curved radius rather than the slightly irregular one because smooth surfaces are less likely to serve as fracture points.
  4. Dec 23, 2016 #3
    I was not able to find a better representation for the second case, now there is a better one attached.

    Phinds, the shape is as described in the standard, the drawing attached belongs to it. My guess is the same as you, but there is probably a reason for it.

    Attached Files:

  5. Dec 23, 2016 #4


    User Avatar
    Gold Member

    I wonder if the second one is to create a slight indentation to better seat/retain an O-ring. Just a thought.
  6. Dec 23, 2016 #5


    User Avatar
    Science Advisor
    Gold Member

    Right sort of thinking . The undercut is to :

    (a) Allow the flat and cylindrical surfaces to be ground true easily

    (b) Allow assembly of the component into another component which has a true bore and face without problematic contact in the corner zone . Unwanted contact here prevents proper alignment and seating down and can be cause of fretting and other difficulties .

    (2) The undercut does of itself cause a stress concentration but in suitable applications there is an overall benefit in the undercut being there .

    (3) Better practice where possible is to have a normal fillet radius and chamfer the bore of the mating component .
    Last edited: Dec 23, 2016
  7. Dec 23, 2016 #6


    User Avatar
    Science Advisor
    Gold Member

    Undercuts can sometimes reduce stress concentration . An example of this is a shaft with a section of splines cut on the end . Undercutting the splines at the point where they meet the plain section of shaft reduces stress concentration at the spline roots .
  8. Dec 23, 2016 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    That was my thought as well. And if there is no such component, might as well go with the simple radius.
  9. Dec 23, 2016 #8

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Having spent many a cold night in a damp dark race car garage with a die grinder de-burring an engine block, the theory is to smooth all parting lines and any sharp edges to prevent stress risers. The engine can produce many harmonics ( vibrations at various RPM levels) that can cause cracks over time. A smooth blend is sought. Under cuts are to be avoided as well. Why??

    Lets review - Crankshaft manufacturers typically enlarge the journal radius to provide a smooth stress free transition at this point. This is to minimize the potential for cracking or breakage as might normally occur with a sharp transition or ninety degree angle. Unfortunately the larger and stronger fillet radius provides greater opportunity for the rod or the bearing insert to contact and ride directly on the fillet with catastrophic results. Racing rods and bearings are chamfered to accommodate larger crank fillets, but be careful. Its up to you to check each one to verify compatibility. Most rods are adequately chamfered, but some bearing inserts have insufficient chamfer for proper clearance. You modified a stock crank so always check for proper clearance.
    Fillet ride is a potential cause of rod bearing failure in Full race engines. This condition most often occurs with racing crankshafts that have a larger fillet radius on the transition from the journal surface to the side thrust face of the rod journal. Fillet ride can also cause issues with connecting rod side clearance as it prevents the rod from using the full amount of side clearance available between the journal shoulders.
    Always check your side clearance. Too little and no lubrication. Instant boat anchor as engine seizure occurs.
  10. Dec 23, 2016 #9


    User Avatar
    Gold Member

    Stress relief at a sharp corner can be accomplished by either adding material, or removing material, as seen in both of your pictures.

    From what I remember,
    From the first picture, an adequate fillet radius may leave not as much flat surface area for stress flow between mating parts, and there would be concentration of stress where the parts meet. An undercut can provide a larger fillet radius with a greater flat mating surface, and thus less stress concentration at both the corner and where the parts meet.

    Threaded connections can have an undercut, with no loss in strength as long as the undercut is not below the thread root.

    Gear teeth can have an undercut at the root, thus lowering stress concentration there also.

    Type of material - a ductile material can experience some plastic flow, and so can get by with a smaller radius at a corner. A brittle cannot.

    Internal properties of the material itself, such as Inclusions, can be instigators of internal cracks occurring before that at a fillet. Fillet radius and type may not be all that critical.

    To stop a noticed crack from advancing from a corner of say a flat plate, one can remove ( surprising ! ) material, by drilling a hole at the base of the crack. The new larger radius of the hole becomes a stress reliever.
  11. Dec 24, 2016 #10
    Perfect answers, thank you all! Really helpful.
  12. Dec 25, 2016 #11
    I'm thinking another possible reason for the undercut is a lubricant trap
  13. Dec 29, 2016 #12


    User Avatar
    Science Advisor

    The undercut is only there to accommodate the meshing tooth tips. The undercut of a gear tooth is bad practice as it significantly reduces the strength of the cantilevered tooth. It is better to increase the contact angle or to relieve the tooth tips and to polish the roots of the gear teeth where there is no contact.

    In general, rounded contacts do relieve stress concentration at corners. Polishing the rounded surfaces will improve the stress relief.

    Compressing or rolling a surface improves the relief effect because it replaces tension with compression in the surface layer. The ball peen hammer evolved in the hands of blacksmiths to relieve surface stress by compression. Once the metal was cold it could be used to work harden the surface.

    After drilling a round drill hole in a critical application, the chance of a crack starting at the hole can be reduced by reaming the hole slightly under-size. Then chamfer or radius the edges at the ends of the hole. Hone the hole to remove reamer marks. Then press a lubricated hard bearing ball all the way through the hole to compress the wall material. Repeat if needed with a slightly larger ball. Polish the inside of the hole and the end radii.
  14. Jan 1, 2017 #13


    User Avatar
    Science Advisor

    The slightly deeper cut of notch1 has a wider surface area with minimum shaft diameter, so stress reduction should be improved over the simple radius1. I think the notch1 example would be easier to “shot peen” than radius1.

    Application to; gear parts, camshafts, clutch springs, coil springs, connecting rods, crankshafts … crankshafts.

    Follow the link to https://en.wikipedia.org/wiki/Peening#History
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Threads for Modifying Sharp Edges
How can I modify this drawing to have fail safe