OCR Text |
Show , The applicable Method 301 acceptance criteria for correction factors is the range 0.7 to 1.3. Using these criteria from Method 301, the FfIR measurement method for all three aldehydes meet the criteria. A draft, revised Method 301 was received from EPA for evaluation with the aldehyde data. The aldehydes validation data meet the draft, revised method validation criteria as well. Although the bias for formaldehyde and acetaldehyde were calculated to be statistically significant, the correction factors were well within the 0.7 to 1.3 acceptance limits. Validation Results for NOX' COl' CO, and Moisture. The EPA Method 4 moisture values were used to correct all FfIR results, measured on a wet basis, to a dry basis for comparison to the EPA reference methods. Total NOx from the exhaust was determined by adding the FTIR concentrations for NO and N02• Table 5 shows the comparison of the concentration data from the EPA reference methods to the FfIR data. These data were used in the calculations described in Method 301, Section 6.2, Comparison with a Validated Method. To determine method precision acceptability, the F-statistic was calculated as follows: SD 2 F -statistic = p SD 2 y (4) where SDp and SDy are the standard deviation of the proposed method (FfIR) and the validated method measurements, respectively. The significance of the method bias was tested using the t-statistic: .. I ~ I t-statIstIc = ~) (5) where dm is the absolute value of the mean of the paired sample differences and n is the number of replicates. The correction factor, calculated where method bias is observed, was determined as follows: (6) where V m is the mean of the validated method values. 13 |