Interlaboratory Reproducibility of Contour Method Data Analysis and Residual Stress Calculation

Journal Article · Wed Jun 03 00:00:00 EDT 2020 · Experimental Mechanics

DOI:https://doi.org/10.1007/s11340-020-00599-0· OSTI ID:1671104

D’Elia, C. R. ^[1]; Carlson, S. S. ^[2]; Stanfield, M. L. ^[2]; Prime, M. B. ^[3]; Araújo de Oliveira, J. ^[4]; Spradlin, T. J. ^[5]; Lévesque, J. B. ^[6]; Hill, M. R. ^[1]

---

1. Univ. of California, Davis, CA (United States)
2. Southwest Research Inst. (SwRI), San Antonio, TX (United States)
3. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
4. Open Univ., Milton Keynes (United Kingdom). StressMap
5. Air Force Research Lab. (AFRL), Wright-Patterson AFB, OH (United States)
6. Hydro-Quebec Research Inst., Varennes, QC (Canada)

While the contour method for residual stress assessment has developed rapidly, no published study documents its interlaboratory reproducibility. Objective: Here we report an initial reproducibility experiment focused on contour method data analysis and residual stress calculation. The experiment uses surface topography data from a physical process simulation of elastic-plastic beam bending. The simulation provides surface topography, for input to the contour method data analysis, as well as a known residual stress field with 130 MPa peak magnitude. Additionally, to increase realism, noise and specific artifacts are added to the topography data. A group of participants received the topography data (without the known residual stress), independently analyzed the data, and submitted results as a two-dimensional residual stress field. Analysis of submissions provides a group average residual stress field and the spatial distribution of reproducibility standard deviation. The group average residual stress agrees with the known stress in magnitude and spatial trend. The reproducibility standard deviation ranges from 2 to 54 MPa over the measurement plane, with an average of 5.4 MPa. Reproducibility standard deviation is smaller in the cross-section interior (≤ 5 MPa), modest near local extrema in the stress field (5 to 10 MPa), and larger near the cross-section boundaries (10 to 30 MPa). Overall, the largest values of reproducibility standard deviation (up to 54 MPa) occur in limited areas where artifacts had been added to the topography data; while some participants identified and removed these artifacts, some did not, leading to systematic differences that elevated the standard deviation.

• Research Organization: Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
• Sponsoring Organization: USDOE
• Grant/Contract Number: 89233218CNA000001
• OSTI ID: 1671104
• Report Number(s): LA-UR--20-24098
• Journal Information: Experimental Mechanics, Journal Name: Experimental Mechanics Journal Issue: 6 Vol. 60; ISSN 0014-4851
• Publisher: Springer
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
contour method
precision
reproducibility
residual stress
uncertainty