Page:NIOSH Manual of Analytical Methods - Chapter N.pdf/3

 the sample size necessary to achieve an acceptable failure rate for an entire batch. For example, to achieve a one percent “average outgoing quality” (average failure rate of one percent) for a batch of 51 to 90 cassettes, a sample of 13 should be tested and produce no failures. If one or more of the 13 cassettes fail, the entire batch of cassettes must be tested or the batch discarded. The leak test results in a small quantity of background aerosol (typically from less than 3 L of air) being collected on the filter. Field blanks should also be leak tested in the same manner to correct for this small amount of background material. The particle count leak test should be performed prior to sampling. It was found that soot particles can clog the leak and prevent detection of particles downstream even though loss of material from the filter occurred during sampling. Thus, a negative leak test after sampling is not a guarantee that leakage did not occur. Another approach sometimes used to check for leakage after sampling is to observe the sealing area at the edge of the filter for improper compression or for particle deposits. These post-sampling tests can indicate when samples should be rejected, but are not generally recommended because they are subject to false negative results. 2.

FIGURES

Figure 1. Picture of a 37-mm and a 25-mm cassette cut in half. The cut cassette surfaces are colored black for easier visualization. Bypass leakage can occur around the filter edge where the seal is formed [2].

3/15/03

181

NIOSH Manual of Analytical Methods