PSD & Moisture Content (71) PROFICIENCY TESTING PROGRAM REPORT

Transcription

1 PSD & Moisture Content (71) PROFICIENCY TESTING PROGRAM REPORT Accredited for compliance with ISO/IEC Copyright: LabSmart Services Pty Ltd Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 1 of 47

2 Report This report is available on the LabSmart Services website. The issue of this proficiency report was authorised by Peter Young, Director, LabSmart Services Pty Ltd, June Contact Details Mobile: Fax: (03) Program Coordinator The program coordinator for this program was Peter Young, Director, LabSmart Services Pty Ltd. Contact Details Mobile: Fax: (03) Please note that any technical questions regarding this program are to be directed to the program coordinator. Z-scores Summary A z-scores summary for this program was issued in 18 April This technical report supersedes the z-sores summary. Accredited Proficiency Testing Provider LabSmart Services is accredited by NATA to ISO/IEC 17043, Conformity assessment General requirements for proficiency testing. Accreditation number The accreditation provides additional assurance to participants of the quality and importance we place on our proficiency testing programs. LabSmart Services Further information regarding LabSmart Services proficiency testing programs can be found on our website. Copyright This work is copyright. No part of this publication may be reproduced in any form, transmitted or stored in any repository (e.g. mechanical, digital, electronic or photographic) without prior written permission of LabSmart Services Pty Ltd. Please contact LabSmart Services should you wish to reproduce any part of this report. Amendment History Reports may be downloaded from the LabSmart Services website. Version 1 Issued 30 June 2017 Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 2 of 47

3 CONTENTS PAGE 1. Program Aim 4 2. Performance? 2.1 Identified Outliers 2.2 Overall Performance 3. Technical Comment 3.1 Material less than 75 µm (Washed) 3.2 Particle size distribution 3.3 Moisture content 4. Less than 75 µm (Washed) 4.1 Z-Scores & Graph 5. Particle Size Distribution (% Passing) 6. Moisture Content 5.1 PSD Z-scores (Combined) 5.2 Z-scores & graph for 2.36 mm sieve 5.3 Z-scores & graph for 1.18 mm sieve 5.4 Z-scores & graph for 600 μm sieve 5.5 Z-scores & graph for 425 μm sieve 5.6 Z-scores & graph for 300 μm sieve 5.7 Z-scores & graph for 150 μm sieve 5.8 Z-scores & graph for 75 μm sieve 6.1 Z-Scores & Graph 7. Program Information 7.1 Z-Score Summary 7.2 Program Design 7.3 Sample Preparation 7.4 Packaging and Instructions 7.5 Quarantine 7.6 Sample Despatch 7.7 Homogeneity Testing 7.8 Participation 7.9 Statistics Appendix A Instructions for testers Appendix B Results log Appendix C Participant results Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 3 of 47

4 1. Program Aim The proficiency program was conducted in February/March 2017 with participants throughout Australia. The program involved the performance of AS , Determination of the particle size distribution of a soil and AS , Determination of the moisture content of a soil Oven drying method. The program provides confidence to the construction materials testing industry regarding the competency of participants (and the industry) to perform these tests. Each participant s performance is statistically assessed and used as a measure of competency relative to all those who participated. Other measures of performance are also used. This report has been prepared using robust statistics. Information regarding the conduct and design of the program can be found in section 5. Comprehensive technical comment (section 3) is provided to assist participants improve the overall performance of these tests. In addition, test data has been reviewed for consistency and additional feedback regarding aspects of the test are provided. 2. How well did we perform? 2.1 Identified Outliers The results and statistical analysis of participant s; Material less than 75 μm (washed) results, PSD results and moisture content results are detailed in sections 4,5 and 6 respectively. There were 25 outliers identified across the tests performed. These were spread across 11 participants. This represented 24% of the 46 participants (1,2) who returned results in the proficiency program (Table 2.1A). Participant s test results are tabulated in section 4 along with the robust statistics and a z-score graph. The z-score indicates how far away a participant is from the program s median value. A z-score of zero indicates a strong consensus with respect to all other participants and represents a very good outcome. The z-score graph gives a quick visual indication of how a result compares to others in the program. Outliers are where a z-score value is greater than 3 or less than -3. It is recommended that participants with outliers investigate their performance of the test. Participants with outliers are detailed in table 2.1A. Those participants with z-scores greater than 2 or less than -2 may wish to review their testing methodology. Only those approaching a z-score of 3 (i.e. outside ± 2.75) have been specifically identified in table 2.1A as feedback. More detail on the robust statistics used can be found in section 5. Technical comment and feedback in section 3 is provided to assist participants investigate or review their results as well for those seeking to improve their testing performance. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 4 of 47

5 Sample Test Investigate Review Less than 75 µm (by washing) R6, L7 - A B Table 2.1A Particle size distribution (% Passing) Moisture content 2.36 mm K9, W8, L7, L6 S mm L7-600 μm L7-425 μm L7-300 μm R6, W8, L7-150 μm R6, W8, L7-75 μm R6, W8, L7 - R7, R6, P4, T5, A9, Z3, Z7 Participant codes where remedial action is recommended based on z-scores. - T4 2.2 Overall Performance Overall a satisfactory level of performance (1) was achieved by the majority (76%) of participants with 24% having one or more outliers (2). The performance by participants is very good overall and compares favourably with previous proficiency programs. For Sample A the spread in PSD results (NIQR) for many sieve fractions was higher than might have been expected. A large spread of results in each fraction may indicate inconsistent sieving practices. However, the bulk of participants were within one standard deviation as can easily been seen from the z-score graphs for each fraction. The variation in test results for PSD in this program are similar to those of the 2016 and 2015 programs. This is a good outcome. Competency of washing was assessed through the calculation of a washed material finer than 75 μm result. Although not a reportable result under the test method the mass of the washed material is still accountable within the test and competency of washing is an important aspect of the test. Most participants completed the washing satisfactorily. The spread in results for washed - material finer than 75 μm was larger than what might have been hoped for. Different soil samples will yield different variation for each fraction according to material quantity and type. The performance by participants is consistent with previous programs. There is an opportunity for laboratories to improve. There is greater variation in the performance of this test than might be expected which may lead to an inflated standard deviation. This means that there are fewer outliers than would be expected. The data indicates that there is considerable room for improvement in washing technique, use of check sums and the whole checking process employed by supervisors. Having 20 grams in the pan after washing a sample should ring some bells. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 5 of 47

6 The determination of moisture content (Sample B) by participants was very good with a very low standard derivation. It was slightly better than both the 2015 and 2016 moisture content variation. A very good outcome. Section 3 provides technical comment which may be useful in identifying possible areas for improvement as well as for investigating outliers. The proficiency program allows many of the laboratories to have greater confidence in their results while for others providing an opportunity to improve their competency with respect to the test in this program. The following is a summary of some of the statistics for this program. Sample Test Units Participants Median Normalised IQR Less than 75 µm (by washing) % A Particle size distribution (% Passing) 2.36 mm mm μm % μm μm μm μm B Moisture content % Table 2.2A Summary of test result statistics from the program. Results have been rounded. (1) Overall performance outcomes can vary from one program to another and should not be taken as either an improvement or deterioration in industry performance. Variation in program outcomes may be attributed to the difficulty of the material under test or where participants overall in one program may have more experience or greater skill levels than those in another program. Evaluation of industry performance endeavours to balance these issues. Industry outcomes and individual performance outcomes are detailed in section sections 3 through 6. (2) Statistics relating to the number of outliers or participation rates are intended as an overview only for the program. They are calculated based on the total number of participants, however not all participants perform each test. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 6 of 47

7 3 Technical comment Please note that the feedback in this section is aimed at providing information that can assist participants and laboratories to improve laboratory operations. All participants identified below are considered to have successfully completed the program unless otherwise identified in section 2. Each submission was assessed to provide additional feedback to participants. Results submitted by participants (Appendix C) were checked and in many cases recalculated. The data submitted on the program log sheet should agree with the results provided. In other words, the mass retained value should match the reported % passing value. Much of the technical comment deals with inconsistencies around the submitted data. It is apparent that many participants did not perform sufficient checks such as Check Sums. There is sufficient material provided for Sample A to perform the test more than once. In many instances checks, had they been performed, would have indicated that retesting may have been warranted. In checking participant s data, it is often difficult to determine exactly what may be incorrect. The following comments should be taken as guide towards reviewing submitted results. Incorrect results do not necessarily mean they will be an outlier. Many of the participants identified below may not have shown up as outliers but the results may still be incorrect. Those with outliers or those that are mentioned on more than one occasion below would benefit from reviewing their results. 3.1 Material less than 75 µm (Washed) Washing is important as it can have a large impact on accuracy of the results obtained. The more material under 75 µm the greater the influence. The material less than 75 µm -washed, % Passing the 75 µm and the pan data are all linked. Reviewing them together is one indicator of how well the sieving process has been performed. Washing It is important that laboratories can wash a sample thoroughly and not loose material. The particle size distribution test method incorporates both washed and unwashed samples. The test method does not specifically calculate the amount of material washed out of the sample. However, it is relevant to include the mass of the material obtained from the washing process in the check sum determination. Loss of material, incomplete washing or inaccurate drying will significantly affect the results obtained. For this proficiency program the calculation of a washed - less than 75 μm result has been used as a measure of competency for the washing process. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 7 of 47

8 It is expected that the material retained in the pan should be small and therefore the % Passing the 75um sieve amount should be very similar to the Material passing the 75µm sieve obtained from the PSD determination. Such comparison should be incorporated into the check sum process. Section 4 details the statistical analysis performed on participant s results. There were two outliers (R6, L7) that need to investigate their results. The standard deviation is around 4%. It is higher than desirable and could be improved by participants with z-scores above 2 or below -2 improving their technique. Overall 32 participants (73%) were within 1 standard deviation which is a good outcome. Calculation of material less than 75 μm - washed >> Comparing with % Passing the 75 µm The material less than 75 µm - washed and % Passing the 75 µm should almost agree. The % Passing the 75 µm would be expected to be higher by the amount the material had broken down during the sieving process. It is usually not possible to obtain a % Passing the 75 µm value that is less than the material less than 75 µm - washed. This may occur if none of the material broke down during the sieving operation. This is not the case for this program. Consequently, two participants (J8 & W8) should review their results. Several participants (23%) had the same result for both the material less than 75 µm - washed and % Passing the 75 µm. This indicates that the testing has not been performed to the proficiency testing requirements but does meet the test method as usually performed. One of these results is incorrect. Participants K6, E5, X7, M6, K9, P7, E6, A9, L6 and R4 should review the testing process. A difference of more than 1% between material less than 75 µm - washed and % Passing the 75 µm would influence the results obtained for this program. Participants that had large differences between these results (Y9, R6, F4, M9, B7, P4, Y5, L9, D5, W3 & L7) should review the results they obtained. With these participants, it also coincided with high pan weights. >> The calculation processes The material less than 75 µm - washed calculation from the washing stage (Section 4) was checked using the submitted mass values. The participants shown below had results that differed from those recalculated that were greater than 1% (up to 4%). The recalculated result was closer to the median value in each case. Either a transcription error has occurred or a calculation error. E5, X7, M6, K9, P7, B3, L6 Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 8 of 47

9 Pan The amount of material left in the pan after sieving can be used as an indication of how well the washing process has been performed. It is also an indication as to how much some of the material breaks down during sieving. The soil sample may contain both hard and soft particles. If a high pan weight is obtained, then how much of the material broke down needs to be considered and why. Did it break down easily because that was the type of material it was? Feeling the hardness of individual particles in larger fractions can help with this. It could also be due to too rough a washing process. If it was felt that the material did not break down sufficiently to account for the pan weight obtained, then incomplete washing may need to be considered. There was a large amount of fine material in this sample requiring careful washing. Based on the homogeneity data the material appeared to break down only slightly with pans weights typically below 2 grams in 400g (0.5%). Participants with z-scores between -1 and 1 performed the washing process well. Typically, their pan weights were under 4 grams (1%). Around six participants (13%) had pan weights above 10g (>2.5%). Washing may have been incomplete or the sieving action may have caused excess material to break down. It may be beneficial for these participants (J8, F4, M9, B7, D5, B3) to examine the washing process. Four participants (P4, Y5, W3, A9) had pan weights greater than 5 grams and should also review these results. Two participants reported no material retained in the pan (E5, R4). This would be highly unlikely for this material and may indicate incomplete sieving. Overall there were around 24 participants or 52% that could improve based on the submitted data covered under section 3.1. Participant improvement would help significantly reduce the variation observed in this program. 3.2 Particle size distribution Start Mass Participants were asked to use a 400g sample. Nominating a start mass helps ensure that participants are not disadvantaged using too little material or too much. Too small a subsample is less forgiving should there be any separation or loss in the material being sampled. The following participants used material that was either less than 350g or greater than 450g. Y9, R6, Y5, J5, T4, X7, S8, L6 Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 9 of 47

10 Sieving a split sub-sample The sieving method refers to AS for the appropriate sample size. In this case for a material with a maximum particle size of 2.36 mm the minimum sample size is 200g. As explained above this program required a 400g sample to be used. Participants using either the 300 or 200 mm diameter sieve would have needed to sieve the fine fraction in two possibly three passes to avoid overloading the sieves (300, 150 & 75 μm, depending on starting mass used). Alternatively, the sample could have been split. The test method does not indicate when it is appropriate to split a sample. Most laboratories used 200mm sieves (80%) and most likely split the sample as this is the most efficient sieving approach (based on time taken). It may not necessarily be the most accurate approach. Washed material is prone to segregation so quartering a sample down needs to be done carefully to obtain a representative sample. Splitting could under some situations decrease the overall accuracy if there has been a significant loss of material (or gain). In any sieving operation, the smaller the amount of material sieved the greater the effect any loss or gain has on the result. For example, if a subsample (split) was used, involving only the fractions less than 2.36, then the factor should have been calculated as follows: Factor = Dry Washed Mass Mass retained on 2.36mm sieve Mass of split subsample Some participants indicated that a 200mm diameter sieve was used but it was not split perhaps indicating that it was sieved over more than one pass. Further checking of these results was not possible. Calculation of % Passing All participants returned results for mass retained and % Passing as requested. This was very good. Cumulative results were not required but some participants provided these. The calculation of percent passing was checked by recalculating from the submitted mass retained data for each fraction. There were 33 participants (72%) whose results matched the recalculated value. Results that differed by more than 0.5 % are identified. The remaining participants (28%) had recalculated results that did not match the results shown. J8, U7, R2, J5, K6, Q3, E5, X7, W3, W8, P7, Z3, L6 The mass retained is primary data (i.e. not calculated) and hence participants should be able to transfer this data accurately to the proficiency testing log sheet. Participants need to understand the calculation process from primary Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 10 of 47

11 data to final result. The following participants omitted data and calculations could only be partially checked. These participants should review the reasons why data could not be supplied. Z3, S8, L6, K9, M8 % Passing the 75 μm sieve There were three outliers identified for % Passing the 75 μm sieve. R6, W8, L7 See also comments under section 3.1. Check Sums AS , Clause 5.4(b) (i) Note 3 indicates that the sieved fractions plus washed material and pan should equal the starting mass to within ± 1 % retained. The mass retained results were used to check how well participants met this requirement. It is important to noted that Check Sums can be (and should be) applied to other aspects of the calculation process. Incomplete sieving, lost material or excessive sieving Not easy to identify based on a single result. Obtaining nothing in the pan may be an indication. A proficiency program provides some feedback assuming the result has been calculated correctly. If the sieving for a fraction is incomplete there would be more retained on a sieve and hence a larger mass retained value. Less passes so a lower value for % Passing is likely to be obtained compared to the median value shown for that fraction. This corresponds to a negative z-score. A positive z-score indicates that less was retained on the sieve (due to more complete sieving, too vigorous sieving or material was lost etc.) giving a higher % Passing value compared to the median. The above is an indication only as natural variation of each fraction within the proficiency sample supplied occurs. Very high or low z-scores are less affected by sample variation. This can provide a reasonable indication for % Retained results but less effective for % Passing as gains and loss can accumulate. Organic matter The proficiency testing Sample A had some organic material in the sample. Small amounts do not pose a problem unless blocking of sieves is an issue. The amounts present were determined to not significantly affect the results obtained. Around 0.2 g of organic matter was estimated to be present or 0.05%. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 11 of 47

12 Mechanical Sieving Of the 46 participants who performed the particle size distribution, 27 participants (59%) used a mechanical shaker. The time used varied considerable from 5 to 30 minutes. It should be noted that excessively long periods in the shaker may lead to unacceptable levels of material break down. There appeared no difference in results compared to those who did not use a shaker. This is expected as all samples must be finished by hand. Decimal places used The decimal places used indicates broadly the accuracy and is related to the LOP. Understanding what is required and reporting on the log sheet correctly indicates a reasonable understanding of the test method. Most participants performed well on this aspect. A small number of participants had difficulty with the concept as indicated by the following: Did not putting anything on the log sheet LOP not matching the decimal places used Entering balance resolution instead of LOP The proficiency program requested results to be recorded to two decimal places (nearest 0.01) on the log sheet where possible. If you have a balance that reads only to 0.1g then the nearest 0.1g is reasonable. The proficiency program requests results to 0.01 where it is possible as it helps improve the quality of the feedback that can be given. There was a random change in the number of decimal places reported within some of the participant s set of results. The most common were: Start mass to 1 decimal place the rest of the weights to 2 decimal places Weights to 1 decimal place while % passing was to 2 decimal places Dropped numbers, most likely zeros at end of a number None of these are likely to have a significant impact on the results but it is preferable that participants understand the correct number of decimal places to use. The LOP of the balance indicates how many decimal places can reasonably be reported. The test method indicates the minimum requirements. Laboratories of course can use more accurate balances than this. Test method Max LOP Implied Resolution Typical balance calibration LOP Course/Intermediate ±5g 1g ±2.4g Fine ±0.5g 0.1g ±0.37g More accurate ±0.05g 0.01g ±0.046g For this program, the LOP was requested. Either the maximum LOP as per the standard or the actual balance calibration LOP was acceptable. Several participants would benefit from reviewing this aspect of the test. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 12 of 47

13 3.3 Moisture content Sample B was double bagged and each bag heat sealed. All participants indicated that the bag was sealed and satisfactory. Any loss of moisture during transit was therefore very unlikely. The sample was prepared with the moisture content at ± < 0.01 %. A moisture content value less than this would be expected due to evaporation during weighing and moisture adhering to the sample bag. Participants could elect to test the whole sample or a subsample. The majority selected to test the whole sample with only two (L9, Z4) using a sub sample. Participants were requested to report the wet mass of the sample. Participants M3, T4, K9, N6 and L7 report weights more than the sample sent indicating that the wet mass probably contained the weight of the pan used as well. Depending on the weight of the sample used the test method allows a range of balance LOPs to be used. Many opted for a balance with a smaller LOP then required by the test method. It is recognised that most laboratories have an assigned balance and associated LOP allocated to this test. Sample A had a MC of around 3% so participants could not accidently test sample A. There were seven outliers identified (R6, R7, P4, T5, A9, Z3, Z7) for moisture content (15%) plus one close to becoming an outlier (T4). Outliers need to be investigated and T4 should review the result obtained as well. Participants with high moisture content results should review results for transcription and calculation errors. These participants used all the sample so incorrect sampling would not have been an issue. Participants with low moisture content results should review results for transcription and calculation errors. Incomplete drying to constant mass may also contribute to a low result. A lengthy delay between opening the sample bag and recording the weight could lead to loss of moisture and a low moisture content result. All other participants had z-scores well under 2 standard deviations with the median very close to the prepared moisture content value of 12.00%. This was a very good outcome. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 13 of 47

14 4.1 Sample A - Material finer than 75 µm (Washed): Z - Scores Code Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W R # P F B M A R E D R B V P L # Y Z3 L S8 T L M Z L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 44 Median Median MU 0.73 First Quartile Third Quartile IQR 5.26 Normalised IQR 3.90 CV (%) 15.3 Minimum (4.25) Maximum (31.57) Range (27.32) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 14 of 47

15 4.1 Sample A - Material finer than 75 µm (Washed): Z - Score Graph M8 Z4 Q7 R4 S5 M6 E6 B3 X7 M3 K9 N6 E5 Q3 T4 T5 W8 S6 Z7 L9 K6 J5 J8 W3 L8 V8 D6 R2 R5 P4 R7 U7 M2 L6 D5 A9 Y9 P7 Y5 B7 F4 M9 R6 L Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 15 of 47

16 5.1 Particle Size Distribution - Sieving : Z - scores Sieve Size Code 2.36 mm 1.18 mm 600 μm 425 μm 300 μm 150 μm 75 μm % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score J U J3 Y K4 M R C9 R # # # F M R D B P Y L T M L D M R Z J T K Q E X W S Q M K # Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 16 of 47

17 Sieve Size Code 2.36 mm 1.18 mm 600 μm 425 μm 300 μm 150 μm 75 μm % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score % Passing Z Score % Passing N W # # # # P B A E R V L # # # # # # # Z S L # R R Z R S Z Score % Passing Z Score Statistic Sieve Size 2.36 mm 1.18 mm 600 μm 425 μm 300 μm 150 μm 75 μm Number of results Median First Quartile Third Quartile IQR Normalised IQR 3 x Normalised IQR CV (%) Minimum Maximum Range Minimum Δ Maximum Δ Range Δ Note:1 Note:2 # Denotes an outlier result. Δ Calculated with outliers removed. R = Result incorrect and removed from data set. Particpants who did not return results are shown as blank entries. Results in green have been calculted for the particpant. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 17 of 47

18 5.2 Sample A - % Passing 2.36 mm : Z - Scores Code Test Result % Z Score J X U W J3 S Y Q K4 M M K # R N C9 W # R P F B M A R E D R B V P L # Y Z L S T L # M Z L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 46 Median Median MU 0.07 First Quartile Third Quartile IQR 0.52 Normalised IQR 0.39 CV (%) 0.4 Minimum (95.10) Maximum (100.00) Range 2.02 (4.90) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 18 of 47

19 5.2 Sample A - % Passing 2.36 mm : Z - Score Graph L8 J8 R2 K6 M2 L9 V8 N6 B7 R4 T5 E6 Z3 Y9 P4 P7 A9 M8 Z7 R6 T4 K9 L6 W8 L7 S8 M6 S5 D5 Q3 M3 W3 D6 M9 U7 R7 Z4 S6 E5 J5 B3 Y5 Q7 R5 F4 X Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 19 of 47

20 5.3 Sample A - % Passing 1.18 mm : Z - Scores Code Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W R P F B M A R E D R B V P L # Y Z L S T L6 M Z L R D S M8 NR Z J T K Q E Statistic Code Value Test Result % Number of results 44 Median Median MU 0.15 First Quartile Third Quartile IQR 1.10 Normalised IQR 0.81 CV (%) 0.9 Minimum (92.60) Maximum (99.00) Range 4.11 (6.40) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 20 of 47

21 5.3 Sample A - % Passing 1.18 mm : Z - Score Graph L7 M6 S8 S5 Z4 M3 D5 Q7 Q3 W3 B3 D6 R4 M9 X7 U7 R7 J5 S6 L9 E6 Y5 Z3 E5 J8 R2 T4 L8 V8 K6 M2 F4 R5 N6 Y9 Z7 W8 T5 B7 R6 P7 A9 K9 P Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 21 of 47

22 Code 5.4 Sample A - % Passing 600 µm : Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W R P F B M A R E D R B V P L # Y Z L S T L6 M Z L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 45 Median Median MU 0.32 First Quartile Third Quartile IQR 2.31 Normalised IQR 1.71 CV (%) 1.9 Minimum (86.74) Maximum (98.00) Range 5.87 (11.26) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 22 of 47

23 5.4 Sample A - % Passing 600 µm : Z - Score Graph E5 K6 J8 M8 Z3 V8 Y5 R5 N6 M2 K9 Z7 M9 Y9 F4 A9 T5 W8 P4 B7 P7 R6 M6 S5 S8 Z4 Q7 M3 Q3 B3 R4 U7 S6 X7 T4 E6 D5 W3 L9 D6 R7 R2 J5 L8 L Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 23 of 47

24 Code 5.5 Sample A - % Passing 425 µm : Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W R P F B M A R E D R B V P L # Y Z L S T L6 R M Z L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 45 Median Median MU 0.47 First Quartile Third Quartile IQR 3.39 Normalised IQR 2.51 CV (%) 3.1 Minimum (72.50) Maximum (95.00) Range (22.50) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. An "R" indicates a result that has been rejected as not being compatiable with the program. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 24 of 47

25 5.5 Sample A - % Passing 425 µm : Z - Score Graph E5 K6 J8 M8 Z3 V8 Y5 R5 N6 M2 K9 Z7 M9 Y9 F4 A9 T5 W8 P4 B7 P7 R6 M6 S5 S8 Z4 Q7 M3 Q3 B3 R4 U7 S6 X7 T4 E6 D5 W3 L9 D6 R7 R2 J5 L8 L Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 25 of 47

26 Code 5.6 Sample A - % Passing 300 µm : Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W # R # P F B M A R E D R B V P L # Y Z L S T L6 M Z L R D S M8 Z J T K Q E Statistic Code Value Test Result % Number of results 44 Median Median MU 0.53 First Quartile Third Quartile IQR 3.77 Normalised IQR 2.79 CV (%) 4.5 Minimum (48.60) Maximum (91.00) Range 9.80 (42.40) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 26 of 47

27 5.6 Sample A - % Passing 300 µm : Z - Score Graph K9 T5 J5 K6 L8 V8 R7 Z3 R5 P4 Y5 M2 J8 M9 P7 A9 S6 Y9 B7 F4 R6 W8 L7 Z4 S5 M6 Q7 S8 B3 R4 X7 M3 T4 W3 D6 U7 D5 E6 R2 L9 E5 Q3 N6 Z Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 27 of 47

28 Code 5.7 Sample A - % Passing 150 µm : Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W # R # P F B M A R E D R B V P L # Y Z L S T L6 R M Z L R D S M8 R Z J T K Q E Statistic Code Value Test Result % Number of results 44 Median Median MU 0.54 First Quartile Third Quartile IQR 3.89 Normalised IQR 2.88 CV (%) 7.9 Minimum (20.39) Maximum (86.00) Range (65.61) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. An "R" indicates a result that has been rejected as not being compatiable with the program. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 28 of 47

29 5.7 Sample A - % Passing 150 µm : Z - Score Graph D5 Z7 D6 K6 L8 R7 P4 R2 V8 U7 R5 A9 Y5 Y9 P7 M2 M9 Z3 J8 B7 F4 R6 L7 W8 R4 S8 Z4 Q7 S5 M6 B3 M3 X7 E6 Q3 W3 L9 E5 T4 K9 T5 S6 J5 N Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 29 of 47

30 Code 5.8 Sample A - % Passing 75 µm : Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R N C9 W # R # P F B M A R E D R B V P L # Y Z L S T L M Z L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 46 Median Median MU 0.75 First Quartile Third Quartile IQR 5.50 Normalised IQR 4.08 CV (%) 15.6 Minimum (8.53) Maximum (83.00) Range (74.47) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 30 of 47

31 5.8 Sample A - % Passing 75 µm : Z - Score Graph D6 D5 L8 R2 K6 V8 R5 P4 U7 R7 M2 Y9 L6 Y5 A9 Z3 P7 M9 B7 F4 J8 W8 R6 L7 M8 Z4 Q7 S5 R4 S8 M6 E6 M3 B3 X7 N6 K9 E5 Q3 T5 T4 W3 S6 L9 J5 Z Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 31 of 47

32 Code 6.1 Sample B - Moisture Content: Z - Scores Test Result % Z Score J X U W J3 S Y Q K4 M M K R # N C9 W R # P F B M A # R E D R B V P # L Y Z # L S T # L M Z # L R D S M Z J T K Q E Statistic Code Value Test Result % Number of results 46 Median Median MU First Quartile Third Quartile IQR Normalised IQR CV (%) 0.8 Minimum (10.60) Maximum (21.73) Range 0.43 (11.13) Z Score Note: A # indicates an outlier where the z-score obtained is either greater then 3 or less than -3. Codes for all participates are shown. The results column shows a blank entry for those participants that did not submit a result for this test. Minimum, Maximum and Range are calculated with outliers excluded, those in brackets include outliers. Particpants results that have been corrected are shown in green. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 32 of 47

33 6.1 Sample B - Moisture Content: Z - Score Graph E6 Y5 Q3 D5 N6 W8 P7 L6 S6 Y9 L8 B7 L7 L9 R2 T4 R6 A9 Z7 R7 T5 P4 Z3 M6 S8 E5 M3 K6 W3 D6 J5 U7 M2 J8 Z4 B3 R4 M9 Q7 X7 S5 V8 F4 M8 K9 R Z-score Review Weak Consensus Strong Consensus Weak Consensus Review Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 33 of 47

34 7. Program Information 7.1 Z-score Summary The proficiency program was conducted over February and March A Zscore Summary summary was issued on the 18 April 2017 and posted on the LabSmart Services web site. A summary was ed to participants. The summary is intended as an early indicator of participant performance. The proficiency testing program report supersedes the z score summary. Further information can be found in section 7.9 Statistics. 7.2 Program Design Design It is expected that the level of experience/skill need to perform these tests will present a reasonable assessment of the overall competency of the tester and industry performance. Sample A - PSD Part of the design of each program involves determining what information needs to be requested to allow for the correct analysis of the data collected. This allows the best possible feedback to be offered to enable participants to improve in the performance of this test. The retained mass for PSD is used for this purpose. In designing a proficiency program, it is sometimes necessary to minimise the effect of some inherent test method variability. Sufficient sample was provided to allow a larger sample size than required by the test method. In addition, the sample size to be used was stipulated to also reduce the variability associated with using variable sample size. Unaccounted material losses or gains (lost material, binding, material break down etc) have a greater effect the smaller the sample size. There was sufficient material provided to allow retesting should the participant need to do so. Laboratories performance is based on the % Passing results. Sample B Moisture Content Calculations were undertaken to ensure that an accurate moisture content was achieved across all samples prepared. The accuracy needed to be sufficiently small as to not have only a minor influence on the testing performed by participants i.e. equal to or less than 0.01%. The program was designed to provide technical feedback regarding performance as well as possible improvements in performance. Further considerations involving the design of the program are detailed below. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 34 of 47

35 7.2.2 Selection of material used in the program Sample A - PSD A soil sample was selected for the PSD test (sample A) with a large amount of material passing the 0.75 µm sieve. The material breaks down only slightly under test. The sample also had particles covering all sieve sizes requiring a reasonable level of skill to test. The sieving determination has been designed to allow accurate comparisons between laboratories to be made. The sample used contains sufficient material at each fraction size to produce a result suitable for comparative purposes. The number of significant places that results are required to be submitted has been increased to facilitate comparison. The test method allows for a moisture correction to be performed. For this program, it was desirable to focus as much as possible on the sieving outcome. Consequently, the material was dried prior to commencement by each participant to eliminate any variability due to a moisture correction. The sieving standard does not cover material passing the 75 μm sieve as it is covered by the hydrometer test. It is however desirable to record this information. It allows check sums to be performed on the sieving analysis and as a measure as to how well the washing operation may have been performed. A sample was selected to ensure there was overloading of some sieves to simulate typical sieving operations and to fully cover the PSD process. The sample size was restricted to a set starting mass to reduce the variability associated with variable sample size. Unaccounted material losses or gains (lost material, binding, material break down etc.) have a greater effect the smaller the sample size. Sample B Moisture Content Moisture content is a common test performed by soil laboratories. Sand was used for the moisture content test to reduce variability in testing. Samples of accurately known moisture were provided to ensure that homogeneity would be satisfactory for the program. The samples were prepared so that the moisture contend would be %. Homogeneity samples were tested to ensure integrity of the samples sent and validate the integrity of the preparation process Role of proficiency testing The determination of outliers is an important task of this proficiency program. A secondary function is to provide feedback that can help those with outliers identify possible areas to investigate as well as assist all participants to improve. In addition to the statistics, proficiency programs often obtain other information that is not normally available to a laboratory. It allows for a better understanding of the testing and can provide information that can lead to improvements in the testing process or test method. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 35 of 47

36 Proficiency testing enables participants to measure competency against others. It is also a measure of staff performance and the equipment used. Apart from measurement uncertainty it is the most useful tool a laboratory has in better understanding the performance of a test Participant assessment Assessment of each participant is based on a z-score that is related to the program consensus value (median). This is used to determine any statistical outliers. Compliance to proficiency program requirements including the correct calculation of results and adherence to program and test method requirements may also be used as part of the assessment process. Participants may also be asked to investigate any discrepancies detected with the paperwork submitted Reporting of results - Significant figures The number of decimal places (significant figures) reported for a test has a bearing on the statistical analysis and therefore the interpretation of the results. There is a need to strike a balance between what is desirable from a statistical viewpoint while recognising how the results are used in practice. Too few decimal places (e.g. due to rounding) can cause an increase in the observed spread of results. Increasing the number of decimal places (with respect to normal reporting) can distort the observed spread of results compared to that encountered in actual practice. Large numbers of similar, rounded results can also cause a distortion in the analysis. For example, rounding to 0.5 % means that any number between and will be 11.0%. If the largest value is in a set of results it is pushed out to 11.0 through rounding. Rounded results are useful from an end user perspective but are not as useful when considering laboratory performance. The test method acknowledges additional decimal places may be used for statistical purposes. For this program, it was decided that the benefits of using additional decimal places would complement the aim of the proficiency program. Participants results were analysed as received regardless of whether there were more or less significant figures than the number requested by the program Additional information requested This program requested additional information as detailed in Appendix C not usually reported. The additional information is however consistent with the performance of the test and the records the test method requires laboratories to maintain. The additional information is used to interpret participant s performance and assist with providing technical comment including feedback on outliers and possible participant improvement. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 36 of 47

37 5.2.7 PSD data checks The determination of outliers is an important function of this proficiency program. A secondary function is to provide feedback that can help those with outliers identify possible areas to investigate as well as assist all participants to improve. This information also helps with identifying any random or systematic errors associated with the test methodology. As observed in other proficiency programs operator errors can often creep into the result calculation process. Assessment of participant s data was incorporated into this program to ensure data was comparable. Every participant s PSD results are recalculated. Any inconsistencies identified during this process do not need to be investigated (as do outliers) but are identified as possible feedback for participant improvement Role of % Retained The sieving component of this proficiency program is based on % Passing results as normally reported by laboratories. The % Passing involves a cumulative calculation which can at times give rise to misleading outliers, particularly on smaller aperture sieves. In such cases an outlier may not necessarily be attributed to the sieve size on which the outlier occurred. Participants need to be aware of this should they need to undertake any investigation. To provide feedback Mass Retained is normally requested for each participant (Appendix C). Increasing the number of significant numbers that results are reported also aids accurate analysis and feedback. The calculated % Retained results may be provided as an appendix to the report where relevant. It should be noted that if the mass retained results submitted are themselves not correct then this will most likely show as z-scores greater than 3 or less than -3. This may be the case even if no outlier was obtained for the % Passing results. To perform a comparison there needs to be a one for one test result i.e. a Mass Retained / % Passing correspondence for the analysis to be statistically valid. That is the accuracy of the analysis is dependent on all of participants suppling accurate mass retained results. 7.3 Sample Preparation Sample A - PSD Sample A consisted of approximately 1.7 kg of a dark moist soil. Samples for the program were drawn and packaged from a single, well mixed, lot. Ten samples were drawn evenly spaced throughout the lot for homogeneity testing. Each participant received randomly drawn samples from the remainder. A unique participation code was assigned to each sample. Each sample was placed in a plastic bag and heat sealed. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 37 of 47

38 Sample B Moisture Content Sample B consisted of approximately 1 kg of a light coloured sandy soil. A single lot of bulk sample was oven dried and well mixed. A known weight of soil was added to a sample bag. A known amount of water was added to each sample. Bags were immediately heat sealed placed into a second bag and heat sealed again. Ten samples were drawn evenly spaced throughout the lot for homogeneity testing. Each participant received randomly drawn samples from the remainder. A unique participation code was assigned to each sample. Sample A and B were packed into a sturdy box prior to dispatch. 7.4 Packaging and Instructions Each sample was sealed in a plastic bag, labelled with the program name. Sample A and B were packed into a sturdy box. Participants were instructed to test according to the nominated test method and report to the accuracy indicated on the results log sheet. See Appendix A for a copy of the instructions issued to participants and Appendix B for the log sheet used. A set of instructions and log sheet were placed in the box prior to sealing and despatch. 7.5 Quarantine Samples sent to Western Australia (WA) are subject to quarantine regulations that require treatment of the soil prior to importation into WA. Samples sent to WA are heat treated and compliance certificates enclosed with samples. Where necessary additional information regarding handling and preparation of the sample are included. 7.6 Sample Dispatch Samples were dispatched to participants on the 10 February 2017 using Toll Priority. Dispatched samples are tracked from despatch to delivery to each participant by LabSmart Services. 7.7 Homogeneity Testing Samples for homogeneity testing were packed in the same way as those for participants. Ten samples were selected at approximately equal intervals throughout the set of samples. The same instructions were given to the laboratory performing the homogeneity testing. Analysis of the homogeneity testing results indicated that the variability associated with the proficiency samples was satisfactory (Table 7.7A). The average value for each homogeneity test lies within 1 s.d of the participant s median value. The assessment of the homogeneity provides confidence that any outliers identified in the program represent statistically valid outliers. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 38 of 47

39 Sample Test Units % Passing Average s.d Coefficient of Variation H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 (%) Less than 75 μm by washing % mm % mm % A PSD 600 μm % μm % μm % μm % μm % B Moisture Content % Table 7.7A Homogeneity results for sample A & B Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 39 of 47

40 7.8 Participation Forty-nine participants entered the program. The nominated date for participants to return their results was 10 March There were 3 participants (6%) who were unable to return their results in time for inclusion in the final report. 7.9 Statistics Z-Scores were calculated for each test and used to assess the variability of each participant relative to the consensus median. A corresponding z-score graph was produced for each test. The use of median and quartiles reduces the effect that outliers have on the statistics and other influences. Consequently, z-scores provide a more realistic or robust method of assessment. Some results were reported by participants to more decimal places than requested as part of the proficiency program and by others to fewer decimal places. In all instances test results have been used as submitted by participants. A z-score is one way of measuring the degree of consensus with respect to the grouped test results. The z-scores in this report approximate standard deviations. For each test, a z-score graph is shown. Use the graph to visually check statistically how you compare to other participants. The following bar (Figure 7.9A) is shown at the bottom of each graph. This helps to quickly visualize where each participant s result falls. Review Weak Consensus Strong Consensus Weak Consensus Review Figure 7.9A Z-score interpretation bar For example: A strong consensus (i.e. agreement) means that your test result is close i.e. within 1 standard deviation of the median. A weak consensus means that your test result is satisfactory and is within 2 standard deviations of the median. If you have obtained a test result that is outside 2 standard deviations, then it may be worth reviewing your testing processes to ensure that all aspects are satisfactory. Only those obtaining a z-score approaching 3 (I.e. outside 2.75 range) have been highlighted in the report for review. If you have obtained a test result that is outside 3 standard deviations then you will need to investigate your testing processes to ensure that all aspects are satisfactory. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 40 of 47

41 For further details on the statistics used in this proficiency program can be obtained from LabSmart Services or download the Participant Guide from the LabSmart Services website Z-score summary A Z-Scores Summary is issued soon after most results are received. It gives participants early feedback as to any program outliers. The summary is usually available on the LabSmart Services website up until the final report is issued. The final report supersedes the z-score summary. The final report contains detailed technical feedback regarding the performance of tests and revised z-scores. The inclusion of late results or corrections are at the discretion of the program coordinator. In some instances, this may change some of the z-scores slightly but generally the performance outcome remains the same. If there is any impact it will be discussed within section 7.1 of the report Comparing statistics from one program to another The statistics generated from one proficiency program are not usually comparable to those from another proficiency testing program. Only very general comparisons may be possible. The reason statistics from one program may not be compared to another is due to the range of variables that differ from one proficiency program to another. These variables include: Type of material selected Variability of the sample The number of participants Experience of participants Test methodology variations Equipment used Test methods used Experience of supervisors Range of organisations involved Program design Type of statistics employed The program outcome represents a snap shot of the competency within the industry and hence provides an overview of the industry. The more participants involved in the program then the more representative the overview Measurement uncertainty The statistics detailed in this program do not replace the need for laboratories to separately calculated measurement uncertainties associated with each test when required by the client or NATA. The proficiency program does give information useful for calculating the MU and bench marking the MU calculated Metrological traceability The assigned median value used in this proficiency testing program is derived from participant performance and is not metrologically traceable. Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 41 of 47

42 LabSmart Services Appendix A Soil Grading & Moisture Content Proficiency Program 2017 (71) INSTRUCTIONS FOR TESTERS 1. Please check that the package you have received contains the following: Instructions (for testers) Results Log Sample A PSD sample (Dark), approximately 1.7 kg Sample B Moisture content sample (White), approximately 1 kg Contact LabSmart Services if the bag is damaged or any item is missing. Phone Please do not discuss aspects of this program with other organisations or other staff within your own organisation who may also be testing a sample from this program. Confidentiality is important in order to ensure statistically valid measures of participant performance. 3. Before testing please read and examine the results log sheet. Follow these instructions carefully. 4. Use AS and AS test methods unless you are unable to do so. You may perform a test even if you are not NATA accredited for the test. Sample A - PSD 5. Mix the contents of the bag marked Sample A PSD. Ensure the whole sample is thoroughly mixed and breakup any clumps that may have formed during transit. 6. Screen the sub-sample using the 4.75mm sieve. 7. Split to obtain a sub-sample of approximately 400g. 8. Dry the sub sample (of around 400g) as per the standard. Testing 9. Record the mass obtained in step 8 above as the start mass for the PSD. 10. Wash the sample and dry as per AS Record the washed dry mass. 11. Calculate the % passing 75 um using the mass recorded under step 9 and Perform the remaining grading and calculate the % passing using the start mass recorded under step Some sieves may become overloaded. If needed there are two columns on the log sheet to record the mass of each split portion. 14. It is recommended that the entire sample following testing be retained until the technical report has been issued. Instructions continued over page Page 1 of 2 PT Instructions V3 Copyright LabSmart Services Pty Ltd Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 42 of 47

43 Sample B Moisture Content 15. Do not open the bag marked Sample B Moisture Content. 16. The moisture in the sample will be unevenly spread throughout the sample during transit due to vibration, gravity and differences in temperature. 17. The sample should be kneaded while in the plastic bag to help spread the moisture. This needs to be performed several times a day for at least two days. Alternating how the bag is left on the bench to cure will also help promote an even spread of moisture. Curing at around 20 0 C will reduce condensation and promote even dispersion of the moisture. 18. The whole sample may be used or a subsample as per the test method AS to determine the moisture content. If whole sample is used then no need to perform step Record the wet mass of the sample used to determine the moisture content. Make sure it is the wet mass and not the wet mass plus container. 20. Determine the moisture content. 21. Record all information and calculations as per the proficiency testing results log sheet and to the accuracy shown on the results log sheet if possible. In many cases a greater reporting accuracy is required compared to that nominated by the standard. 22. Have a query? Contact Peter at LabSmart Services. Phone Please fax or the Results Log to LabSmart Services by 10 March 2017 Fax: (03) OR Petery@labsmartservices.com.au 24. Please retain the completed Results Log as this contains your participation code that will identify your results in the technical report covering the proficiency testing program. It is also recommended that a copy of completed worksheets be kept with the results log in your proficiency file. Thank you for participating in this proficiency testing program. Page 2 of 2 PT Instructions V3 Copyright LabSmart Services Pty Ltd Copyright: LabSmart Services Pty Ltd Issued - 30 June 2017 Page 43 of 47