roping in uncertainty measuring the tensile strength of steel wire...
TRANSCRIPT
-
Roping in uncertainty – measuring the tensile strength of steel wire ropes
Riaan Bergh
27 September 2016
-
2
Presentation outline
The test environment
Why test?
The detail
The outcome
The process
-
3
Why do we test new ropes?
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
Theory vs. Practice
Rope strength depends on:
• Material properties of wires
• Wire uniformity
• Rope construction
mailto:[email protected]
-
4
Why do we test ropes in service?
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
5
High force, high stakes environment
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
• MFL Tensile tester
o 15 000 kN capacity testing from 750 kN (5% of FS)
o 2 m stroke length, 8 m daylight
• Laboratory space 1200m2, not climate controlled
• Calibration duration, cost
• No CRM, no PT schemes
• No re-test of the same UUT
mailto:[email protected]
-
6
Enter uncertainty…
? Instrumentation
? Environment e.g. temperature and humidity
? Variability
? Resolution of force measurement
? Electromagnetic noise
? Test method
? Personnel
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
7
The estimation process
The GUM method
• Model the measurement
• Identify and quantify all sources of error
• Evaluate all sources of error to obtain the uncertainty
contribution
• Combine uncertainty contributions
• Calculate the degrees of freedom for the combined
uncertainty
• Choose level of confidence
• Calculate the expanded uncertainty, U
• Report the result
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
8
The detail – Mathematical model
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
C
C
mailto:[email protected]
-
9
Sources of error
• Accuracy of machine force measurement system
• Accuracy of reference load cell – uncertainty quoted by
calibration lab
• Resolution of force measurement
• Repeatability
• Reproducibility
• Effect of temperature
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
10
Uncertainty budget
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
11
Temperature effect
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
-1.00%
-0.75%
-0.50%
-0.25%
0.00%
0.25%
0.50%
0.75%
1.00%
-3 2 7 12
Seas
on
al d
iff
fro
m m
ean
ro
pe
BF
Difference from mean BF (n = 12)
Winter diff frommean [%]
Summer diff frommean [%]
Summer: +0.3%Winter: -0.3%
ΔT = 20°C (12 months to date)
mailto:[email protected]
-
12© CSIR 2013
The outcome – Example 1 at 852 kN working point
-
13© CSIR 2013
The outcome – Example 2 at 1965 kN working point
-
14
Room for improvement
• Temperature effect
• Control lab temperature
• Condition specimens prior to testing
• Apply a temperature correction to test result
• Accuracy of machine force measurement
• Apply correction
• Adjust machine force calculation polynomial coefficients
• Uncertainty of reference load cell
• Load cell?
• Calibration laboratory
Riaan Bergh - September 2016 - [email protected] - 011-482 1300
mailto:[email protected]
-
Thank you.