Report No. 768 Soils Proficiency Testing Round 14 July 2012capeyork.myownserver.net/~ptaasnau/documents/768.pdf · Plastic Limit ... This report summarises the results of a proficiency
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
SD 2.12.04
Report No. 768
Soils Proficiency Testing
Round 14
July 2012
Acknowledgments
PTA wishes to gratefully acknowledge the technical assistance provided for this program by Mr G Hooper, Golder Associates Pty Ltd. Also our thanks go to The Nielson Group for the supply of the samples.
Results Sheet: California Bearing Ratio.......................................................................... …C4
1
SD 2.12.04
1. FOREWORD
This report summarises the results of a proficiency testing program on the determination of Classification and California Bearing Ratio (CBR) on a soil sample. It constitutes the fourteenth round of an ongoing series of programs. The program was conducted in May 2012 by Proficiency Testing Australia (PTA). The aim of the program was to assess laboratories’ abilities to competently perform the prescribed analyses. The Program Coordinator was Ms L Galbraith and the Technical Advisor was Mr G Hooper, Golder Associates Pty Ltd. This report was authorised by Dr M Bunt, PTA Statistician.
2. FEATURES OF THE PROGRAM (a) Participants were provided with 1 labelled soil sample containing a clayey
gravel mix. (b) A total of 63 laboratories received samples, comprising:
- 62 Australian participants; and
- 1 Papua New Guinean participant.
Of these 63 laboratories, 7 were unable to submit results by the due date. (c) Laboratories were provided with the Instructions to Participants and Results
Sheet (see Appendix C). Laboratories were requested to perform the tests according to their routine methods and to record their results on the Results Sheet.
(d) Prior to sample distribution, a number of randomly selected samples were
analysed for homogeneity. Based on the results of this testing (see Appendix B), the homogeneity of the samples was established.
(e) Each laboratory was randomly allocated a unique code number for the
program to ensure confidentiality of results. Reference to each laboratory in this report is by code number only. Please note that a number of laboratories reported more than one set of results and, therefore, their code numbers (with letter) could appear several times in the same data set.
(f) Results (as reported by participants) with corresponding summary statistics
(i.e. number of results, median, uncertainty of the median, normalised interquartile range, robust coefficient of variation, minimum, maximum and range) are presented in Appendix A (for each sample and for each of the analyses performed.
2
SD 2.12.04
(g) A robust statistical approach, using z-scores, was utilised to assess laboratories’ testing performance (see Section 3). Robust z-scores and z-score charts relevant to each test are presented in Appendix A.
(h) The document entitled Guide to Proficiency Testing Australia, 2011 (reference
[1]) defines the statistical terms and details the statistical procedures referred to in this report.
(i) A tabulated listing of laboratories (by code number) identified as having outlier
results can be found on page 8.
3. FORMAT OF THE APPENDICES
(a) Appendix A contains the analysis of results reported by laboratories for the
samples. This section contains the following for each determinant, where appropriate: - a table of results and calculated z-scores;
- a list of summary statistics; and
- ordered z-score charts.
(b) Appendix B contains details of the homogeneity testing. (c) Appendix C contains copies of the Instructions to Participants and Results
Sheet.
4. STATISTICAL DESIGN OF THE PROGRAM (a) Outlier Results and Z-scores In order to assess laboratories’ testing performance, a robust statistical
approach, using z-scores, was utilised. Z-scores give a measure of how far a result is from the consensus value (i.e. the median), and gives a "score" to each result relative to the other results in the group.
A z-score close to zero indicates that the result agrees well with those from
other laboratories. Whereas, a z-score with an absolute value greater than or
equal to 3.0 is considered to be an outlier and is marked by the symbol “§”. (b) Results Tables and Summary Statistics Each of these tables contains the results returned by each laboratory,
including the code number for the method used, and the robust z-score calculated for each result.
3
SD 2.12.04
Results have been entered exactly as reported by participants. That is, laboratories which did not report results to the precision (i.e. number of
significant figures) requested on the Results Sheet have not been rounded to the requested precision before being included in the statistical analysis.
A list of summary statistics appears at the bottom of each of the tables of results and consists of:
- the number of results for that test/sample (No. of Results);
- the median of these results, i.e. the middle value (Median);
- the uncertainty of the median;
- the normalised interquartile range of the results (Normalised IQR);
- the robust coefficient of variation, expressed as a percentage (Robust CV) - i.e. 100 x Normalised IQR / Median;
- the minimum and maximum laboratory results; and
- the range (Maximum - Minimum).
The median is a measure of the centre of the data. The normalised IQR is a measure of the spread of the results. It is calculated by multiplying the interquartile range (IQR) by a correction factor which converts the IQR to an estimate of the standard deviation. The IQR is the difference between the upper and lower quartiles (i.e. the values above and below which a quarter of the results lie, respectively).
Please see reference [1] for further details on these robust summary statistics.
(c) Ordered Z-Score Charts On these charts each laboratory's robust z-score is shown, in order of
magnitude, and is marked with its code number. From these charts, each laboratory can readily compare its performance relative to the other laboratories.
These charts contain solid lines at +3.0 and -3.0, so that outliers are clearly
identifiable as those laboratories whose "bar" extends beyond these "cut-off" lines. The y-axis of these charts has been limited, so very large z-scores appear to extend beyond the chart boundary.
5. PTA AND TECHNICAL ADVISOR’S COMMENTS Analysis of Results by Method Groups Due to differences in the methods used by participants, results were grouped and analysed separately as Method 1 or Method 2. Method 1 includes the AS methods (relevant sections) and equivalent methods and Method 2 includes Q methods (relevant sections).
4
SD 2.12.04
No other methods were so prevalent in these groups that they could be analysed according to the method. The following table summarises the results submitted by participants for the program.
TABLE A: SUMMARY STATISTICS METHOD 1
Test No. of Results Median Normalised IQR
Liquid Limit (nearest 1%) 39 40.0 3.0
Linear Shrinkage (0.5%) 39 8.00 2.04
Plasticity Index (nearest 1%) 39 17.0 4.8
TABLE B: SUMMARY STATISTICS METHOD 2
Test No. of Results Median Normalised IQR
Liquid Limit (nearest 1%) 15 40.0 1.9
Linear Shrinkage (0.5%) 15 8.00 0.56
Plasticity Index (nearest 1%) 15 15.0 3.3
For the remaining tests as part of Soils Round 14, methods were grouped together. It has been found that there are little differences between methods and thus could be analysed as part of one consensus. The following table summarises the results submitted by participants for the program.
5
SD 2.12.04
TABLE C: SUMMARY STATISTICS ALL METHODS
Test No. of Results Median Normalised IQR
Apparent Particle Density (0.01 g/cm
3)
23 2.780 0.080
Plastic Limit (nearest 1%) 54 23.5 3.0
California Bearing Ratio @ 2.5mm (1%)
51 10.0 6.7
California Bearing Ratio @ 5.0mm (1%)
51 13.0 7.0
Achieved Dry Density before soaking (0.001 t/m
3)
51 1.8540 0.0056
Achieved Moisture Content (0.1%)
51 16.80 0.37
Grouped analysis was performed for methods appearing greater than or equal to 10 times in each test. The results are tabulated below.
TABLE D: ANALYSIS OF GROUPED METHODS
Test Grouped
method
No. of results Median Normalised
IQR Apparent
Particle Density (0.01 g/cm
3)
AS1289.3.5.1 13 2.720 0.193
Plastic Limit (nearest 1%)
AS1289.3.2.1
Q105
37
11
23.0
24.0
3.0
1.9 California
Bearing Ratio @ 2.5mm (1%)
AS1289.6.1.1 32 9.5 6.9
California Bearing Ratio @
5.0mm (1%) AS1289.6.1.1 28 13.0 6.9
Achieved Dry Density before
soaking (0.001 t/m
3)
AS1289.6.1.1 26 1.8555 0.0044
Achieved Moisture
Content (0.1%)
AS1289.6.1.1
AS1289.2.1.1
16
11
16.85
16.80
0.46
0.11
6
SD 2.12.04
Results returned for Soils Round 14 Proficiency Testing Program included Australian Standards (AS) and another predominate method (Q method). Due to the differences in each method the AS test method and Q test method were split into two separate consensus groups for tests relating to Sample 1: Classification. For Sample 2: California Bearing Ratio (CBR), both AS and the Q method were analysed together in the same consensus group.
5.1 Apparent Particle Density
Three outliers were reported for this test. Laboratory code 78 is encouraged to review their procedures.
5.2 Liquid Limit, Plastic Limit and Plasticity Index
Method one received one outlier for Liquid Limit and one outlier for Plasticity Index. Method two received one outlier for Liquid Limit. No outliers were reported for Plastic Limit. Laboratory Code 130 is encouraged to review their procedures for Plasticity Index. Plasticity Index is the calculation of the difference between the Liquid Limit (LL) and the Plastic Limit (PL) i.e. LL-PL. Thus Laboratory Code 130 may need to review its procedures for Liquid Limit in particular.
5.3 Linear Shrinkage and Linear Shrinkage Mould Length
Laboratory Code 61 and 122 received an outlier for Linear Shrinkage and was the only two outliers received in both method groups. No statistical analysis was performed on Linear Shrinkage Mould Length as it was for informational purposes only.
5.4 California Bearing Ratio at 2.5mm and 5.0mm
The spread of results received for CBR at 2.5 and 5.0 in Round 14 Soils Proficiency Testing Program is large. Although, the statistical analysis does not result in any outliers a smaller spread of results would be expected. As there is no industry standard for accepting a CBR result, a result within +/- 5 of the median would be expected. Laboratories reporting results outside this range are strongly encouraged to review their procedures. An analysis of laboratories’ uncorrected results was attempted to determine if it would reduce the spread, however, not all results were received and therefore the analysis could not be performed.
7
SD 2.12.04
Table E shows the Comparison of CBR results from previous rounds.
TABLE E: California Bearing Ratio By Round Comparison
Round Median (%)
2.5/5.0mm
Robust CV (%)
2.5/5.0mm
Range
2.5/5.0mm
No. Results
2.5/5.0mm
14 10.0/13.0 66.7/54.2 23/27 51/51
12 8.0/7.5 28.8/28.0 20.3/21 49/48
11 113.0/142.0 9.6/7.0 157/178 43/43
8
50.0/70.0 34.0/24.3 99/111 58/58
NATA No. 457
(July ’04)
140.0/185.0 37.1/30.1 260/210 38/30
5.5 Achieved Dry Density (ADD) Before Soaking and Achieved Moisture
Content (AMC)
Results reported for ADD and AMC were of a high standard with a CV of only 0.3% and 2.2% respectively. There were four outliers reported for ADD which comprised 7.8% of the total number of results reported. Two outliers were reported for AMC which comprised 3.9% of the total number of results reported.
5.6 Conclusion Overall a good performance was noted for classification tests, ADD and AMC. However a high CV was determined for CBR at 2.5mm and 5.0mm. All laboratories should be self examining their processes. Processes to examine include;
compaction patterns when moulding sample,
seating load at time of penetration
calculation, which includes careful scrutiny / examination of application to any correction which is applied to the penetration graph.
Laboratories must take greater care when reporting the method used in a test. Laboratory 31 reported an aggregate test method whilst laboratory 65 reported a method that does not allow corrections greater than 0.5. Also, laboratory 92B reported a method which is not a 1 point CBR. Laboratories returning outlier results and those returning results for CBR outside +/-5 of the median are strongly encouraged to review their processes.
8
SD 2.12.04
Metrological Traceability and Measurement Uncertainty of Assigned Values Consensus values (median) derived from participants’ results are used in this program. These values are not metrologically traceable to an external reference. As the assigned value for this program is the median of the results submitted by the participants, the uncertainty of the median has been calculated and is presented in the corresponding tables of Appendix A.
6. OUTLIER RESULTS
Laboratories reporting outlier results are listed in the following table:
Results Sheet: California Bearing Ratio.……………………………………………….C4
C1
SD 2.12.04
PROFICIENCY TESTING AUSTRALIA
Proficiency Testing Program
Soils Round Part 14A: Classification
INSTRUCTIONS TO PARTICIPANTS
Please read instructions carefully BEFORE commencing testing. To ensure that the results of this program can be analysed properly, participants are asked to carefully note the following: 1. The following tests are to be conducted:
Apparent Particle Density Emersion Class
Liquid Limit Linear Shrinkage
Plastic Limit Linear Shrinkage Mould Length
Plasticity Index
2. These tests are based on AS1289, however other methods may be used.
Apparent Particle Density AS1289.3.5.1 Determination of the soil particle density of a soil – standard method. The -2.36mm fraction only to be tested.
Liquid Limit AS1289.3.1.1. Determination of the liquid limit of a soil – Four point Casagrande Method.
Plastic Limit AS1289.3.2.1. Determination of the plastic limit of a soil – standard method.
Plasticity Index AS1289.3.3.1. Calculation of the plasticity index of a soil.
Emerson Class AS1289.3.8.1. Methods of testing soils for engineering purposes.
Linear Shrinkage AS1289.3.4.1. Determination of the linear shrinkage of a soil – standard method.
3. For each test on the sample, the result is to be reported on the results sheet to the
accuracy and reporting basis indicated. 4. Testing may commence as soon as the sample is received. All laboratories must return
Please read instructions carefully BEFORE commencing testing. To ensure that the results of this program can be analysed properly, participants are asked to carefully note the following:
1. Adjust the sample moisture content to 16.8% and allow to cure for a minimum period of 24 hours prior to remoulding (compaction)
2. Perform AS 1289.6.1.1 (1998) - preferable, or equivalent:
Determination of the California Bearing Ratio of a soil, following the criteria set out below;
- Standard laboratory method for a remoulded specimen soaked/
- Standard Compactive Effort to be used.
- Sample to be remoulded in 3 layers to a 1.854 t/m3 dry density.
- The surcharge mass to be applied = 4.5 kg.
- 4 day soaked Test
-Swell Measurements to be recorded and Swell % calculated
3. Report results (to the accuracy indicated) on the provided “Results Sheet”. Results
are to include an attached plot of the penetration data.
Laboratories are to perform tests according to their accredited method.
After all testing has been completed, laboratories are asked to seal and retain any of the remaining sample for at least 3 months.
Testing may commence as soon as the sample and documentation are received.
Please return results no later than 14 May 2012 to: