(1) Wind tunnel. The particle delivery system shall consist of a blower system and a wind tunnel having a test section of sufficiently large cross-sectional area such that the test sampler, or portion thereof, as installed in the test section for testing, blocks no more than 15 percent of the test section area. The wind tunnel blower system must be capable of maintaining uniform wind speeds at the 2 km/hr and 24 km/hr in the test section.

(2) Aerosol generation system. A vibrating orifice aerosol generator shall be used to produce monodisperse solid particles of ammonium fluorescein with equivalent aerodynamic diameters as specified in table F-2 of this subpart. The geometric standard deviation for each particle size generated shall not exceed 1.1 (for primary particles) and the proportion of multiplets (doublets and triplets) in all test particle atmosphere shall not exceed 10 percent of the particle population. The aerodynamic particle diameter, as established by the operating parameters of the vibrating orifice aerosol generator, shall be within the tolerance specified in table F-2 of this subpart.

(3) Particle size verification equipment. The size of the test particles shall be verified during this test by use of a suitable instrument (e.g., scanning electron microscope, optical particle sizer, time-of-flight apparatus). The instrument must be capable of measuring solid and liquid test particles with a size resolution of 0.1 µm or less. The accuracy of the particle size verification technique shall be 0.15 µm or better.

(4) Wind speed measurement. The wind speed in the wind tunnel shall be determined during the tests using an appropriate technique capable of a precision of 2 percent and an accuracy of 5 percent or better (e.g., hot-wire anemometry). For the wind speeds specified in table F-2 of this subpart, the wind speed shall be measured at a minimum of 12 test points in a cross-sectional area of the test section of the wind tunnel. The mean wind speed in the test section must be within ±10 percent of the value specified in table F-2 of this subpart, and the variation at any test point in the test section may not exceed 10 percent of the measured mean.

(5) Aerosol rake. The cross-sectional uniformity of the particle concentration in the sampling zone of the test section shall be established during the tests using an array of isokinetic samplers, referred to as a rake. Not less than five evenly spaced isokinetic samplers shall be used to determine the particle concentration spatial uniformity in the sampling zone. The sampling zone shall be a rectangular area having a horizontal dimension not less than 1.2 times the width of the test sampler at its inlet opening and a vertical dimension not less than 25 centimeters.

(6) Total aerosol isokinetic sampler. After cross-sectional uniformity has been confirmed, a single isokinetic sampler may be used in place of the array of isokinetic samplers for the determination of particle mass concentration used in the calculation of sampling effectiveness of the test sampler in paragraph (d)(5) of this section. In this case, the array of isokinetic samplers must be used to demonstrate particle concentration uniformity prior to the replicate measurements of sampling effectiveness.

(7) Fluorometer. A fluorometer used for quantifying extracted aerosol mass deposits shall be set up, maintained, and calibrated according to the manufacturer's instructions. A series of calibration standards shall be prepared to encompass the minimum and maximum concentrations measured during size-selective tests. Prior to each calibration and measurement, the fluorometer shall be zeroed using an aliquot of the same solvent used for extracting aerosol mass deposits.

(8) Sampler flow rate measurements. All flow rate measurements used to calculate the test atmosphere concentrations and the test results must be accurate to within ±2 percent, referenced to a NIST-traceable primary standard. Any necessary flow rate measurement corrections shall be clearly documented. All flow rate measurements shall be performed and reported in actual volumetric units.

(1) Establish and verify wind speed.

(2) Generate aerosol.

(3) Introduce particles into the wind tunnel. Introduce the generated particles into the wind tunnel and allow the particle concentration to stabilize.

(4) Verify the quality of the test aerosol.

(5) Aerosol uniformity and concentration measurement.

(ii) Determine the quantity of material collected with each isokinetic sampler in the array using a calibrated fluorometer. Calculate and record the mass concentration for each isokinetic sampler as:

Equation 7

where:

i = replicate number;

j = isokinetic sampler number;

M_iso = mass of material collected with the isokinetic sampler;

Q = isokinetic sampler volumetric flow rate; and

t = sampling time.

(iii) Calculate and record the mean mass concentration as:

Equation 8

where:

i = replicate number;

j = isokinetic sampler number; and

n = total number of isokinetic samplers.

(iv) Precision calculation.

(A) Calculate the coefficient of variation of the mass concentration measurements as:

Equation 9

where:

i = replicate number;

j = isokinetic sampler number; and

n = total number of isokinetic samplers.

(B) If the value of CV_iso(i) for any replicate exceeds 10 percent, the particle concentration uniformity is unacceptable and step 5 must be repeated. If adjustment of the vibrating orifice aerosol generator or changes in the particle delivery system are necessary to achieve uniformity, steps 1 through 5 must be repeated. When an acceptable aerosol spatial uniformity is achieved, remove the array of isokinetic samplers from the wind tunnel.

(6) Alternative measure of wind tunnel total concentration. If a single isokinetic sampler is used to determine the mean aerosol concentration in the wind tunnel, install the sampler in the wind tunnel with the sampler nozzle centered in the sampling zone (paragraph (c)(6) of this section).

(7) Measure the aerosol with the candidate sampler.

(iii) Determine the quantity of material collected with the test sampler using a calibrated fluorometer. Calculate and record the mass concentration for each replicate as:

Equation 10

where:

i = replicate number;

M_cand = mass of material collected with the candidate sampler;

Q = candidate sampler volumetric flow rate; and

t = sampling time.

(iv)

(A) Calculate and record the sampling effectiveness of the candidate sampler as:

Equation 11

where:

i = replicate number.

(B) If a single isokinetic sampler is used for the determination of particle mass concentration, replace C_iso(i) with C_iso.

(8) Replicate measurements and calculation of mean sampling effectiveness.

(ii) Calculate and record the average sampling effectiveness of the test sampler for the particle size as:

Equation 12

where:

i = replicate number; and

n = number of replicates.

(iii) Sampling effectiveness precision.

(A) Calculate and record the coefficient of variation for the replicate sampling effectiveness measurements of the test sampler as:

Equation 13

where:

i = replicate number, and

n = number of replicates.

(B) If the value of CV_E exceeds 10 percent, the test run (steps in paragraphs (d)(2) through (d)(8) of this section) must be repeated until an acceptable value is obtained.

(9) Repeat steps in paragraphs (d)(2) through (d)(8) of this section until the sampling effectiveness has been measured for all particle sizes specified in table F-2 of this subpart.

(10) Repeat steps in paragraphs (d)(1) through (d)(9) of this section until tests have been successfully conducted for both wind speeds of 2 km/hr and 24 km/hr.

(1) Graphical treatment of effectiveness data. For each wind speed given in table F-2 of this subpart, plot the particle average sampling effectiveness of the candidate sampler as a function of aerodynamic particle diameter (D_ae) on semi-logarithmic graph paper where the aerodynamic particle diameter is the particle size established by the parameters of the VOAG in conjunction with the known particle density. Construct a best-fit, smooth curve through the data by extrapolating the sampling effectiveness curve through 100 percent at an aerodynamic particle size of 0.5 µm and 0 percent at an aerodynamic particle size of 10 µm. Correction for the presence of multiplets shall be performed using the techniques presented by Marple, et al (1987). This multiplet-corrected effectiveness curve shall be used for all remaining calculations in this paragraph (e).

(2) Cutpoint determination. For each wind speed determine the sampler Dp₅₀ cutpoint defined as the aerodynamic particle size corresponding to 50 percent effectiveness from the multiplet corrected smooth curve.

(3) Expected mass concentration calculation. For each wind speed, calculate the estimated mass concentration measurement for the test sampler under each particle size distribution (Tables F-4, F-5, and F-6 of this subpart) and compare it to the mass concentration predicted for the reference sampler as follows:

(iv) Calculate the estimated mass concentration ratio between the candidate method and the reference method as:

Equation 14

where:

C_cand(est) = estimated mass concentration measurement for the test sampler, µg/m³; and

C_ref(est) = estimated mass concentration measurement for the reference sampler, µg/m³ (calculated for the reference sampler and specified at the bottom of column 7 of table F-4 of this subpart).