Results 1 to 8 of 8

Thread: Help needed: tube fatigue failure stress at clamp point calculation needed

  1. #1

    Help needed: tube fatigue failure stress at clamp point calculation needed

    Hi

    I am experiencing a fatigue failure of grease tubes at bracket points. The brackets are steel and crush the pipe, which is on a vehicle so experiences high levels of vibration. Table shows main info, attached image shows stress I am searching for (bracket redesign probable, but I also need to find this value beforehand). Thank you for your help.

    Tube OD 12mm
    Tube ID 10mm
    Tensile Strength 480 MPa
    Yield Strength 325 Mpa
    Youngs Mod (E) 200GPa
    Measured Vibration 6g
    Mass tube and grease 0.342kg/m
    Clamping force 515MPa
    Deformation with clamping 8% OD
    Max dist clamp - clamp 1100mm
    Bolt torque (M10 bolt) 54000Nmm
    *Clamping force (from above) 36072 N
    *Clamping pressure (assuming 1mm contact of tube against clamp due to crushing) 515MPa
    * Moment of inertia of uncrushed tube * 5.27E-10 m^4
    Attached Images Attached Images

  2. #2
    Principle Engineer
    Join Date
    May 2015
    Posts
    216
    Other issues not addressed by your table might be chassis flex and the stress riser created at the pinched area. We might not be able to support the assumption that there are "no additional stress risers". Is the grease supplied under high pressure?

    Are you avoiding a commercial clamp for cost or other reasons?

    Attachment 2142

  3. #3
    Quote Originally Posted by Hudson View Post
    Other issues not addressed by your table might be chassis flex and the stress riser created at the pinched area. We might not be able to support the assumption that there are "no additional stress risers". Is the grease supplied under high pressure?

    Are you avoiding a commercial clamp for cost or other reasons?

    Attachment 2142
    Thanks for getting back to me. Regardless of what the new clamp is like, I need to prove the existing clamp is ineffective. I agree that there will be a stress raiser at the edge of the clamp, which is crushing the tube, but what I meant by 'no additional...' was to assume that the corner of the clamp is 90degrees and there are no raised edges to increase stress and no fillets etc. to reduce stress.

    I am aware that the approach I am taking overs-simplifies the problem but I am just after a way of calculating the stress at that point using the limited information available.

    Thanks

    Rosie

  4. #4
    Principle Engineer
    Join Date
    May 2015
    Posts
    216
    You have vibration induced fatigue failures and need to prove the clamp is ineffective? In addition, you seem reluctant to accept the proof that the customers are generating despite the evidence.

    From what you have told us, you exceeded the yield strength of the tube in clamping and created a stress concentration between zero and three. To this and what ever residual stress is in the part there is bending due to vibration which seems to exceed the fatigue limit of the part. If a manual grease gun is used to fill the tube, pressure can become a significant percentage of the yield strength. Corrosion can be a factor.

    In short all we know is that you exceeded the endurance limit, based upon the physical evidence.

  5. #5
    Lead Engineer
    Join Date
    Aug 2013
    Location
    Houston TX USA
    Posts
    421
    Since you have already stressed the tube(s) at those points beyond their yield stress by clamping then you must assume that their residual stress at those points remains at the material's min yield level even without any additional loadings. Ironically, the vibrations probably initially do some degree of stress relief to an indeterminate static stress below that yield stress point; but, without knowing that residual stress value it is impossible to calculate a potential fatigue life of the tube(s).

    At the same time, I agree with the prior post that if the customer is having cracking and failure at those points' then it doesn't matter what your calculation result is, the clamp designs are insufficient for their intended service. Working in new product development for industrial valves for 20+ years has taught me one thing above all else: "Regardless of the amount of inhouse product testing or its results, the final true test of a product will be performed in service by the customers."

    Anyone that is demanding any "proof" calculation(s) is both wasting your time, delaying an obviously required clamp redesign and insuring there will be more unnecessary customer problems.

    Adding more clamps on a closer spacing may help the issue; but, in the long run using clamps with the appropriate internal elastomeric cushioning will be a better and more all around economic solution.

  6. #6
    Quote Originally Posted by JAlberts View Post
    Since you have already stressed the tube(s) at those points beyond their yield stress by clamping then you must assume that their residual stress at those points remains at the material's min yield level even without any additional loadings. Ironically, the vibrations probably initially do some degree of stress relief to an indeterminate static stress below that yield stress point; but, without knowing that residual stress value it is impossible to calculate a potential fatigue life of the tube(s).

    At the same time, I agree with the prior post that if the customer is having cracking and failure at those points' then it doesn't matter what your calculation result is, the clamp designs are insufficient for their intended service. Working in new product development for industrial valves for 20+ years has taught me one thing above all else: "Regardless of the amount of inhouse product testing or its results, the final true test of a product will be performed in service by the customers."

    Anyone that is demanding any "proof" calculation(s) is both wasting your time, delaying an obviously required clamp redesign and insuring there will be more unnecessary customer problems.

    Adding more clamps on a closer spacing may help the issue; but, in the long run using clamps with the appropriate internal elastomeric cushioning will be a better and more all around economic solution.
    Again, thank you for your reply. However, I am in a situation where mathematical evidence is needed. Are you aware of how I may be able to work that stress out, even if assumptions need to be made?

    Thanks

    Rosie

  7. #7
    Principle Engineer Cragyon's Avatar
    Join Date
    Feb 2013
    Location
    Newark, NJ
    Posts
    299
    Here's a paper discussing the math behind Fatigue Life Calculations. You're going to need to find or develop a S-N Curve or Wohler Diagram for the shape/material.

    fatigue-calculations.pdf

  8. #8
    Lead Engineer
    Join Date
    Aug 2013
    Location
    Houston TX USA
    Posts
    421
    Let's say you are able to perform a fatigue analysis, that analysis will be of no value without having direct and verifiable data on the vibration frequency and amplitude of your tubing under your customers' service conditions.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •