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.
81 CERAMIC CURVE, ALTON, RICHARDS BAY 3900 l TEL: +27(0)35 751 2446 l CEL: +27(0)83 6500151 l FAX: 0862605793 E-MAIL: [email protected]
FLOAT AND SINK PROFICIENCY TESTING
REPORT SEVENTEEN
Revision: 00
Final report
FEBRUARY 2020
R BABOOLAL (SCHEME MANAGER)
PARTICIPANT:
Page 2 of 15
EXECUTIVE SUMMARY
1. Sixteen samples were sent out to participants with 15 result submissions.
2. Since most of the results do not follow a Normal distribution, non-parametric statistics were used. This is
a straightforward statistical calculation and can be found on the link below
https://www.youtube.com/watch?v=qkI-HeMiKzQ
3. Observations at the differing densities
DENSITY COMMENTS
1.30 Three laboratories obtained yields at this density. The data obtained for these is not sufficient to carry out reliable statistical analysis. The typical ash for this coal round had an ash of around 30%, hence not much yield was expected
1.40 All participants reported yields at this density, however the results obtained were quite varied. Three outliers were detected on the yields and two on the ash determination. The standard deviation on the yields was low, after the exclusion of outliers. Standard deviation on the ash was high at 2.43.
1.50 The yields were varied, but no outliers detected. The sample preparation and ash determination were also quite varied 2.79
1.60 There was one outlier each, on the yields and the ash determination with the standard deviations still high on both
1.70 The yields had a high standard deviation whereas the ash standard deviation was reduced, indicating greater accuracy on the ash determined at this yield
1.80 At this density as well, the yields had a high standard deviation whereas the ash standard deviation was reduced, indicating greater accuracy on the ash determined at this yield
Sink 1.80 No outliers detected; however varied results obtained. Generally, particularly good results all round considering that this was not a high-grade coal.
COAL CONCEPTS - PROFICIENCY TESTING - FEBRUARY 2020
ANALYTICAL PARAMETER : YIELDS AT F1.80
LAB ID MASS (Kg) %YIELD
3f 1.81 18.50
5f 2.06 21.10
6f 1.66 16.70
7f 2.01 20.40
8f 2.43 24.87
11f 1.92 24.71
14f 2.86 29.43
15f 1.80 18.10
18f 1.48 15.60
19f 1.80 18.10
20f 1.36 18.90
21f 1.73 17.90
23f 2.25 23.01
24f 0.79 18.20 NUMBER OF RESULTS - 14 14 OUTLIERS - 0 AVERAGE - - 20.39 STD DEVIATION - - 3.84 MEDIAN - - 18.70
COAL CONCEPTS - PROFICIENCY TESTING - FEBRUARY 2020
ANALYTICAL PARAMETER : ISO ASH (%) (Density at 1.80)
LAB ID MOISTURE IN ANALYSIS SAMPLE (%) AIR DRY DRY BASE
3f 2.70 33.10 34.02
5f 2.70 35.20 36.18
6f* 4.00 37.00 38.54
7f 3.10 33.10 34.16
8f 4.40 35.36 36.99
11f 3.80 33.90 35.24
14f 2.83 32.60 33.55
15f 2.40 33.70 34.53
18f 1.60 34.50 35.06
19f 2.40 33.90 34.73
20f 2.20 32.80 33.54
21f 2.20 34.70 35.48
23f 2.91 33.71 34.72
24 1.20 32.90 33.30
NUMBER OF RESULTS - 14 14 14
OUTLIERS - - 1 1
AVERAGE - 2.86 33.81 34.73
STD DEVIATION - - 0.91 1.07
MEDIAN - - 33.71 34.72
Non parametric stats on yield results Non parametric stats on DB results
Quartile 1 18.10 Quartile 1 34.02
Quartile 2 18.70 Quartile 2 34.73
Quartile 3 23.01 Quartile 3 35.48
INTERQUARTILE RANGE (IQR) 4.91
INTERQUARTILE RANGE (IQR) 1.46
1.5 X IQR 7.37 1.5 X IQR 2.19
Acceptable lower limit 10.74
Acceptable lower limit 31.83
Acceptable upper limit 30.38
Acceptable upper limit 37.67
Results outside acceptable limits 0
Results outside acceptable limits 6f
Page 13 of 15
3.7 YIELDS AND ASH RESULTS AT DENSITY OF SINK 1.80
COAL CONCEPTS - PROFICIENCY TESTING - FEBRUARY 2020
ANALYTICAL PARAMETER : YIELDS AT Si 1.80
LAB ID MASS (Kg) %YIELD
3f 2.83 28.90
5f 2.44 25.00
6f 2.40 24.10
7f 2.56 26.10
8f 2.07 21.20
11f 1.92 24.71
14f 3.34 34.34
15f 3.11 31.20
18f 2.73 28.90
19f 3.21 32.20
20f 2.21 30.60
21f 2.87 29.60
23f 2.44 24.95
24 0.75 17.30 NUMBER OF RESULTS - 14 14 OUTLIERS - - 0 AVERAGE - - 27.08 STD DEVIATION - - 4.61 MEDIAN - - 27.50
COAL CONCEPTS - PROFICIENCY TESTING - FEBRUARY 2020
ANALYTICAL PARAMETER : ISO ASH (%) (Sink portion after density1.80)
LAB ID MOISTURE IN ANALYSIS SAMPLE (%) AIR DRY DRY BASE
3f 2.70 45.40 46.66
5f 2.20 47.60 48.67
6f 1.50 48.10 48.83
7f 2.60 46.60 47.84
8f 3.50 47.78 49.51
11f 2.90 47.20 48.61
14f 2.65 43.70 44.89
15f 2.40 44.40 45.49
18f 2.20 44.70 45.71
19f 2.70 44.60 45.84
20f 2.10 44.10 45.05
21f 2.20 44.50 45.50
23f 2.41 45.76 46.89
24f 1.90 42.00 42.81
NUMBER OF RESULTS - 14 14 14
OUTLIERS - - - 0
AVERAGE - 2.54 45.46 46.59
STD DEVIATION - - 1.79 1.90
MEDIAN - - 45.05 46.25
Non parametric stats on yield results Non parametric stats on DB results
Quartile 1 24.71 Quartile 1 45.49
Quartile 2 27.50 Quartile 2 46.25
Quartile 3 30.60 Quartile 3 48.61
INTERQUARTILE RANGE (IQR) 5.89
INTERQUARTILE RANGE (IQR) 3.12
1.5 X IQR 8.84 1.5 X IQR 4.68
Acceptable lower limit 15.88
Acceptable lower limit 40.81
Acceptable upper limit 39.44
Acceptable upper limit 53.29
Results outside acceptable limits
0
Results outside acceptable limits 0
Page 14 of 15
4. HOMOGENEITY CHECK
SAMPLE NO.
TEST PORTION
1
TEST PORTION
2 sample av (Xt)
range (Wt)
range sqd
1 30.46 30.18 30.32 0.28 0.0784
2 30.86 30.72 30.79 0.14 0.0196
3 30.17 30.17 30.17 0.00 0.0000
4 30.3 30.38 30.34 0.08 0.0064
5 30.39 30.76 30.58 0.37 0.1369
6 30.04 30.11 30.08 0.07 0.0049
7 30.8 30.89 30.85 0.09 0.0081
8 30.35 30.82 30.59 0.47 0.2209
9 30.16 30.29 30.23 0.13 0.0169
10 30.37 30.11 30.24 0.26 0.0676
GENERAL AVERAGE 30.42 STANDARD DEVIATION 0.266 WITHIN SAMPLE STANDARD DEVIATION 0.167 BETWEEN SAMPLE STANDARD DEVIATION 0.238
5. STABILITY CHECK
SAMPLE NO.
TEST PORTION
1
TEST PORTION
2 sample av (Xt)
range (Wt)
range sqd
1 30.09 30.66 30.38 0.57 0.3249
2 30.19 30.01 30.10 0.18 0.0324
3 29.98 30.11 30.05 0.13 0.0169
4 30.03 29.94 29.99 0.09 0.0081
5 30.5 30.02 30.26 0.48 0.2304
6 30.36 29.88 30.12 0.48 0.2304
7 30.3 29.9 30.10 0.40 0.1600
8 30.16 30.6 30.38 0.44 0.1936
9 30.56 30.07 30.32 0.49 0.2401
10 30.5 30.35 30.43 0.15 0.0225
GENERAL AVERAGE 30.21 STANDARD DEVIATION 0.158 WITHIN SAMPLE STANDARD DEVIATION 0.270 BETWEEN SAMPLE STANDARD DEVIATION 0.107
The Float and sink samples were confirmed to be sufficiently homogenaous and stable, for analysis to be
done
Page 15 of 15
6. CONCLUSION Generally acceptable/comparable results were obtained.
The ash results were very good considering this is obtained from a washed fraction (sampling). (If the compounding contributions
from sample preparation and testing errors are taken into account as well.)
The use of non-parametric statistics had been applied
COAL CONCEPTS: Terms and Conditions
Return of results: Laboratories participate in proficiency testing programs on the understanding that they will be sharing their results and information anonymously with other laboratories performing the same analysis. No return of results compromises the spirit of the programs, and reports will not be sent to laboratories unless they return results. Payment in full is required from all laboratories enrolling whether they return results or not. Errors in Participant Proficiency Testing Results: Proficiency testing reports should reflect the level of accuracy that a regular testing client would receive. If a participant finds an error in their proficiency testing results, they may notify us in writing and change their submission PRIOR to the due date for return. Changes after this time will not be accepted. Coal Concepts’ reports results as submitted by participants. On occasion, it seems as though participants have mixed up the samples or not processed the samples according to the instructions. Coal Concepts cannot make assumptions of this nature and change results 'to suit'. We also cannot compromise the integrity of the programs by suggesting to some participants that they should review their results prior to the due date. (This is unfair to other participants) It is the responsibility of the participants to check all aspects of the program, including sample identification, preparation, testing instructions, calculations and reporting of the results prior to results submission. If samples are not in good condition on arrival to the participant laboratory, Coal Concepts must be notified in writing IMMEDIATELY, as often samples can be replaced in good time. Claims about samples received in bad condition will not be accepted after the report has been issued. Late Enrolments and Late Results: Late enrolment requests cannot always be accommodated, as sample manufacture must be scheduled well in advance to the shipping date of the program to allow all necessary quality assurance activities to be carried out. Shipping of PT materials and evaluating test results from PTPs out of cycle with the mainstream programs is considerably time consuming and therefore costly. In order not to disadvantage participants able to comply with time frames, Coal Concepts may charge a late fee in the following circumstances: Requests that Coal concepts staff enters results on behalf of participants Requests to record results after the due date Requests for PTP participation that is out of cycle with the scheduled dates Shipping fees and Customs clearance: Costs incurred for shipping samples and clearance of same through customs are the responsibility of the participating laboratory unless otherwise indicated Non-payment of fees: Coal Concepts retains the right to withhold reports and/or test materials and services when invoices are outstanding. Confidentiality of results: All data and information received by Coal Concepts from its clients are considered confidential unless the client has given express permission to pass on information. Definitions: The dictionary definitions of “collusion” and “falsification” are as follows. · Collusion: A secret agreement or cooperation for a fraudulent or deceitful purpose. · Falsification: Deliberately changing something to be false. In proficiency testing terms, collusion is comparing data (and perhaps changing data) to fit in with a believed “correct” result. This is contrary to the spirit of proficiency testing programs, which are issued with the intention of providing an objective comparison of a laboratory’s performance with others. Coal Concepts tries to minimise the occurrence of collusion by being aware that laboratories should be objective when they report their results, and should therefore not know the intended results at the time they are reporting to us. Answers are not provided to clients until results have been submitted. To prevent collusion and falsification our advice to clients is: DON’T confer with others about PT samples or results. DO accept the fact that everyone makes errors. DON’T average the results or opinions of every person in the laboratory before selecting the answer to be submitted. Instead, use one of the answers AS SUBMITTED to you and take advantage of the Coal Concepts internal QA services and submit all answers generated by the technicians. DO have confidence in your own results. Proficiency Testing (PT) is a compulsory part of laboratory accreditation, but it is also an important tool for giving you confidence in your results. “Enhancing” your PT results with assistance from another participant cannot increase confidence in your laboratory’s performance. Coal concepts’ testing staff are not told what the expected results are, nor what we are expecting. We subject ALL results to analysis, even if they are different. The staff have the right to check that the results we enter on their behalf are correctly transcribed. Clients are always welcome to contact Coal Concepts to seek advice or information about collusion or falsification of data. Policy for Participant Appeal of PT Performance Assessment: If participants disagree with their performance assessment in a proficiency report, they should inform Coal Concepts in writing. The response will include Coal Concepts interpretation of the outcome of the reassessment and an explanation of that outcome. (For example, explanation of a calculation, or the rationale for the outcome of the evaluation.) If a mistake has been made by Coal Concepts, it will be dealt with via Coal Concepts’ non-conformance system. Liability In no event shall a party's liability to the other party for direct damages exceed an amount equal to the value of the amount for the PT Programme, under that specific month