2008-03-10_GENERAL DOCUMENTS - M1977247

Home Browse Search

ACZ Laboratories, Inc. August 10, 2007 Quality Assurance Plan Version 12 ' SOPAD018.08.07.12 Page 46 of 96 • RPD for a matrix duplicate is evaluated only if the observed concentration is greater than 10 times the MDL; otherwise each associated client sample must be reported ' with the appropriate qualifier, regardless of RPD. • In the absence of other contributing factors, a DUP failure for a solid or semi-solid ' matrix is attributed to non-homogeneity of the sample, and each associated client sample may be reported with the appropriate qualifier. • For an aqueous matrix, if the DUP fails then all associated samples and the DUP must be retested. If permitted by the instrument software the sample and DUP can be reanalyzed at the end of the analysis in lieu of retesting all associated samples. ' • For an aqueous matrix, if the MS/MSD RPD fails then the associated samples must be reanalyzed. If permitted by the instrument software the sample and MS/MSD can be reanalyzed at the end of the analysis in lieu of retesting all associated samples. , • If applicable, evaluate the LCS/LCSD if the RPD fails for a matrix duplicate or spike duplicate. Each associated client sample may be reported with the appropriate , qualifier if the LCS/LCSD meets the criteria stated in 11.1.2.2. • For a solid or semi-solid matrix, if both the LCSS and LCSSD recoveries pass but ' the RPD fails, then acceptable precision may be demonstrated by a passing RPD for the MS/MSD, and each associated client sample may be reported with the appropriate qualifier. , 11.2 Instrument Specific Controls All data must be associated with a passing instrument calibration and initial calibration verification. To the extent possible, all data must be associated with passing continuing calibration verification. If the initial calibration verification results (ICV/ICB) are outside of the acceptance criteria, then the source(s) of the failure must be identified, corrective action(s) performed if necessary, and the instrument recalibrated before ' proceeding with sample analysis. If the continuing calibration verification results (CCV/CCB) do not meet the acceptance criteria then the ' source(s) of the failure must be identified and corrective action(s) performed, including recalibration if necessary, before continuing with sample analysis. If reanalysis of any sample(s) associated with failing calibration verification is not possible then the associated data must be reported with the appropriate qualification. ' For instruments that permit the analysis of subsequent workgroups using the most recent calibration, two (2) consecutive attempts of the opening CCV/CCB are allowed. If both attempts fail to produce acceptable results ' then the source(s) of the failure must be identified and corrective action(s) performed, including recalibration if necessary, before commencing sample analysis. Unless stated otherwise by the test SOP, passing calibration verification must always bracket all batch quality , control samples, and results for additional instrument check standards, if applicable, must be within the acceptance criteria stated in the SOP. However, when the acceptance criteria for a CCV or CCB are exceeded (i.e. high bias) any associated client sample with a measured concentration less than the MDL may ' be accepted and reported with the appropriate qualification. 2773 Downhill Drive 970-879-6590 ' Steamboat Springs, CO 80487 www.acz.com