Give a summary of the test report, explaining the scope of the tests and the main findings. Include the sampler performance and restrictions on its use. Describe any practical difficulties in the routine use of the sampler that are known to exist.
Annex A (normative)
Procedure for a workplace comparison of a candidate sampler and a validated sampler in order to obtain a correction factor
A.1 General
This Annex is intended for the user, rather than the manufacturer, of sampling samplers. It describes a recommended method for establishing the equivalence of a candidate sampler and a validated sampler in a specific workplace. The purpose of establishing the equivalence of the two samplers is to enable the user of the samplers to carry out screening measurements or periodic measurements of dust concentration with samplers that have not been evaluated in laboratory tests according to this standard. The validated sampler and candidate sampler shall both be either personal samplers or static samplers. The methods described in this Annex shall not be applied to comparisons of personal samplers with static samplers or vice versa.
Analysis of the field comparison data is carried out to obtain a correction function that relates the aerosol concentrations measured by the candidate sampler to those measured by the validated sampler. In subsequent use in conditions to which the correction function applies, the results from the candidate sampler are scaled by application of the correction function. The correction function will depend on both the properties of the aerosol present during the test, i.e. its composition, concentration, size distribution, shape, charge, etc., and on the environmental conditions existing at the time of the test, i.e. wind speed, temperature, pressure, humidity, etc. The correction function will be specific to the workplace activities included in the comparison exercise and cannot be assumed to apply to different circumstances.
This annex recommends criteria for deciding whether the transformed candidate sampler results show an adequate degree of equivalence to the validated sampler results. If the equivalence or performance is poor, this will probably be due to either that the inherent variability at the workplace is exceedingly large, or by including too wide a range of workplace activities in the comparison, rather than to poor performance by the samplers. In the first case the large variability can only be reduced by averaging over several samplers, which can be difficult with personal samplers. In the second case the comparison exercise shall be repeated for a more narrowly-defined group of exposed persons or workplace activities, until an adequate degree of equivalence is established. In some workplaces the nature or organisation of the work can make this impossible.
A.2 Procedure for field comparison of candidate sampler with validated sampler A.2.1 General
Pairs of measurements are obtained with both validated sampler(s) and candidate sampler(s) (preferably more than one individual for both the validated sampler and the candidate sampler), exposed to the same aerosol. The number of pairs of measurements (experimental runs) obtained shall be as large as possible and never less than 10. The measurements shall cover the range of aerosol properties, concentrations and environmental conditions occurring at the sampling sites, and be obtained over a period of at least two days (although a larger number of days is preferred)5).
Both validated sampler and candidate sampler shall be operated in accordance with the instructions given in the manufacturer's instruction manual for the sampler types. Any deviations from these instructions shall be
5) For examples of published field comparisons of aerosol sampling instruments, see Bibliography, references [7] to [9].
documented in the comparison report. Only those samples obtained in accordance with documented operating procedures may be regarded as valid and included in the data analysis.
A comparison test that demonstrates equivalence (see A.3.6) between a candidate sampler and validated sampler does not entail that the performance of the candidate sampler is as required (see EN 13205-1:2014, 5.1).
A.2.2 Comparison of two types of personal samplers for the inhalable aerosol fraction See 6.2.
A.2.3 Comparison of two types of static samplers See 6.3.
A.2.4 Comparison of two types of personal samplers for the respirable or thoracic aerosol fractions
See 6.4.
A.3 Calculation methods A.3.1 General
Designate the concentrations measured with the candidate sampler in experimental run r using candidate sampler individual s as Xrs . Designate the concentrations measured with the validated sampler in experimental run r using validated sampler individual s as Yrs.
A.3.2 Estimation of the correction function
Take logarithms of the measured concentrations of the candidate sampler, Xrs[mg/m3], and (the average of) the measured concentrations of the validated sampler, averYr [mg/m3], for each experimental run r and plot the pairsln(Xrs), ln(averYr) graphically. The functional relationship y = f(x) between the concentrations of (the average of) the candidate samplers and (the average of) the validated samplers shall be computed from the logarithmic data using a recognised statistical procedure, documented in the report. An example of a suitable technique is curvilinear regression by the least-squares method.6) The derived correction function shall be monotonic in the range in which it will later be used. The correction function shall be plotted onto the graph showing all data. The correction function’s parameter values and their uncertainties shall be documented.
A.3.3 Calculation of ratio of corrected sampler concentration to workplace (test) aerosol concentration, as determined with the validated sampler
For each determined workplace (test) aerosol concentration, calculate the corrected individual candidate sampler concentration s for experimental run r, Yrs∗, from Formula (A.1):
Yrs∗=expf(ln( )Xrs ) (A.1)
where
Xrs is the measured candidate sampler concentration for candidate sampler individual s, for run (workplace (test) aerosol concentration) r, [mg/m3]; and
Yrs∗ is the corrected concentration measured by the candidate sampler individual s, for the run experiment (workplace (test) aerosol concentration) r, [mg/m3].
For each determined workplace (test) aerosol concentration by candidate sampler individual s in experimental run r, calculate the ratio of the corrected individual candidate sampler concentration to (the average) validated sampler concentration from Formula (A.2).
Rrs= Yrs*
averYr (A.2)
where
Rrs is the ratio of the candidate sampler individual s concentration in experimental run r to the (average) workplace (test) aerosol concentration in experimental run r, [-];
Yrs* is the corrected concentration measured by the candidate sampler individual s, for the run experiment (workplace (test) aerosol concentration) r, [mg/m3]; and
averYr is the average of the measured validated sampler concentration for the run (workplace (test) aerosol concentration), r, [mg/m3].
A.3.4 Exclusion of outliers
Exclude comparison pairs (of averages) for which ratios Rrs > 2,5 or Rrs < 1/2,5 from further analysis, provided a) no more than one pair of results is excluded for every 10 pairs of data points; and b) there are at least 10 unexcluded points.
A.3.5 Residual uncertainty after transformation by the correction function Calculate the geometric mean of the ratios Rrs, from Formula (A.3), i.e.
lnRg = 1
NRun lnRrs
r=1 NRun
∑ (A.3)
where
NRun is the number of experimental runs, pairs of (average) validated sampler and candidate sampler concentrations;
Rg is the geometric mean of the Rrs values; and
Rrs is the ratio of the candidate sampler individual s concentration in experimental run r to the (average) workplace (test) aerosol concentration in experimental run r.
Calculate the geometric standard deviation of the ratios Rrs from Formula (A.4), i.e.
lnsR= 1
NRun−1 (lnRrs−lnRg)2
r=1 NRun
∑
1 2
(A.4)
where
NRun is the number of experimental runs, pairs of (average) validated sampler and candidate sampler concentrations;
Rg is the geometric mean of the Rrs values;
Rrs is the ratio of the candidate sampler individual s concentration in experimental run r to the (average) workplace (test) aerosol concentration in experimental run r; and
sR is the geometric standard deviation of the Rrs values.
If any outlying points have been excluded as specified in A.3.4, adjust NRun accordingly.
A.3.6 Equivalence
Provided that the expanded standard uncertainty of the candidate sampler is not calculated according to EN 13205-5, it is suggested that the two methods have an acceptable degree of equivalence if:
— sR≤1,3 for subsequent measurements at concentrations in the range 0,5×COEL to2×COEL;
— sR≤1,5 for subsequent measurements at concentrations less than0,5×COEL; where
COEL is the appropriate occupational exposure limit (OEL) value [mg/m3] applying to the substances being measured.
That a candidate sampler is found to be equivalent to a validated sampler at a specific workplace does not imply that the equivalent sampler also will be a validated sampler (type C) at that workplace. To determine whether an equivalent sampler also is a validated sampler at the specific workplace requires a test according to the main part of this standard.
A.4 Periodic validation
The equivalence of the two methods shall be checked periodically by comparing a small number of new candidates and validated sampler results.
If the results of such an exercise indicate that the correction function no longer applies, or in any case when circumstances in the workplace change, the full comparison exercise shall be repeated. Where new comparison data are described adequately by the previous correction function, the function shall be updated periodically by combining new and existing data.
A.5 Documentation A.5.1 General
The comparison exercise should be documented in sections as described as follows.
State type of test (comparison) and personnel carrying out the tests and date(s) of the experimental work.
A.5.2 Description of the candidate sampler and validated sampler See 9.2.
A.5.3 Critical review of sampling process See 9.3.
A.5.4 Circumstances of field comparison
— Description of the site at which the comparison was carried out, and the activities included;
— description of the properties of the aerosol present, i.e. composition, concentration, size distribution, shape, colour, refractive index, charge, etc;
— description of the environmental conditions at the test site, i.e. wind speed, temperature, pressure, relative humidity, etc.
— description of the experimental arrangement; position and orientation of samplers;
— details of sampler flow measurement;
— details of any external sampling pumps used;
— cleaning procedures.
A.5.5 Details of experimental design
Give a table that shows the design of the experiment in terms of the locations of the samplers during sampling, the sampling times and dates, the number of pairs of results (experimental runs) obtained and the number of validated samplers and candidate samplers per experimental run. The methods used for obtaining aerosol concentrations from both the candidate and validated samplers shall be described in detail.
A.5.6 Data analysis
Present the data in graphical form, showing the correction function plotted onto the data. Explain the procedure for estimating the correction function.
A.5.7 Equivalence
— state the correction function parameters and their uncertainties;
— state the geometric mean and geometric standard deviation of the ratios Rrs, Rg and sR;
— state the concentration range over which the candidate sampler and validated sampler can be regarded as equivalent.
Bibliography
[1] Mark, D., R. J. Aitken, O. Witschger, W. Koch, G. Lidén and H. Kromhout, Development of a novel calibration tool for workplace aerosol samplers. Final report on EU FP4 project SMT4-CT98-2254.
Available from the authors, 2006
[2] WITSCHGER, O., GRINSHPUN, S. A., FAUVEL,S. and BASSO, G., Performance of Personal Inhalable Aerosol Samplers in Very Slowly Moving Air When Facing the Aerosol Source. Ann. Occup. Hyg.
2004, 48 (4) pp. 351–368
[3] DE VOCHT F., HUIZER D., PRAUSE M., JAKOBSSON K., PEPLONSKA B., STRAIF K. et al. Field Comparison of Inhalable Aerosol Samplers Applied in the European Rubber Manufacturing Industry. Int. Arch.
Occup. Environ. Health. 2006, 79 (8) pp. 621–629
[4] SU W.-C., VINCENT J.H. Towards a general semi-empirical model for the aspiration of aerosol samplers in perfectly calm air. J. Aerosol Sci. 2004, 35 (9) pp. 1119–1134
[5] SU W.-C., VINCENT J.H. Corrigendum to “Towards a general semi-empirical model for the aspiration of aerosol samplers in perfectly calm air” [Journal of Aerosol Science 35 (9) (2004) 1119–1134]. J.
Aerosol Sci. 2005, 36 (12) p. 1468
[6] DRAPER N., SMITH H. Applied Regression Analysis. 2nd ed. Wiley Series in Probability and
Mathematical Statistics, Hrsg. W.A. Shewhart und S.S. Wilks. John Wiley & Sons, New York (NY), USA, 1981
[7] VAUGHAN N.P. Investigation of Size Fractionating Sampler Characteristics Using a Real-time Aerodynamic Particle Size Analyser. Internal Report IR/L/FD/83/9, Health and Safety Executive, London, 1983
[8] VERMA D.K., SEBESTYEN A., JULIAN J.A., MUIR D.C.F. Field Comparison of Respirable Dust Samplers.
Ann. Occup. Hyg. 1992, 36 (1) pp. 23–34
[9] WERNER M.A., SPEAR T.M., VINCENT J.H. Investigation into the Impact of Introducing Workplace Aerosol Standards based on the Inhalable Fraction. Analyst (Lond.). 1996, 121 (9) pp. 1207–1214 [10] CEN/TR 15547:2007, Workplace atmospheres - Calculation of the health-related aerosol fraction
concentration from the concentration measured by a sampler with known performance characteristics [11] CEN/TR 13205-3:2014, Workplace exposure — Assessment of sampler performance for
measurement of airborne particle concentrations — Part 3: Analysis of sampling efficiency data [12] EN 13205-6:2014, Workplace exposure — Assessment of sampler performance for measurement of
airborne particle concentrations — Part 6: Transport and handling tests
[13] EN 481, Workplace atmospheres - Size fraction definitions for measurement of airborne particles [14] EN 482, Workplace exposure - General requirements for the performance of procedures for the
measurement of chemical agents