|
||||||||||||||||||||||||||||
|
|
Tools and TemplatesThese winzipped tools were written for internal use but are general enough that CLAS makes them available to interested parties on the understanding that they have been validated against published examples and commercial software. NRC is not to be held liable in the unlikely event of loss from incorrect results. Uncertainty Budget TemplateExcel template for preparing simple uncertainty budgets. The spreadsheet includes a completed example for the calibration of a micrometer using gauge blocks. DOWNLOAD Calibration Control ChartsControl charts for analyzing quality control data obtained from period checks of the measurement process. It is presumed that the user has a rudimentary understanding of statistical process control (SPC). For information on SPC, consult any of a large number of publications written on the subject. CLAS provides a bibliography of representative publications on SPC. In general, control charts are used to detect variations in calibration processes which are not random in nature and which, therefore, can be assigned to a cause. The information is used to determine whether or not the calibration process is working properly (i.e., in control). This is analogous to the use of the Student's T-Test for deciding whether or not a single observation comes from a distribution (i.e., process) with known mean and unknown standard deviation. Like any tool, control charts can be used in many different ways to provide different types of information on the calibration processes. CLAS provides sample spreadsheets for a number of different types of control charts. It remains the responsibility of the user to select the most appropriate type of chart for the given application and to interpret the results accordingly. It is also the responsibility of the user to collect the measurements over a sufficiently wide range of operating conditions to ensure a correct characterization of the distribution and over a sufficiently long period of time to ensure independence. In some cases, 25 measurements over several months is sufficient. InstructionsTo use the spreadsheets, simply click on the Data tab and overwrite the sample values of Date and Observations with your own data. The Upper and Lower 2warning limits (UWL & LWL) and 3control limits (UCL & LCL) and other appropriate statistics will be calculated automatically. To view the control charts, click on the Charts tab.
To unzip the spreadsheets, evaluation copies of Winzip software are available for download at the following location: http://www.winzip.com/download.cgi. |
|||||||
|