By John Wright, Aaron Johnson, Gina Kline, Chris Crowley, Jodie Pope, Vern Bean, Kevin John, Jeremy Latsko, David Pfeffer, James Winchester, Michael Case, Gary Mahoney, Gary Quigley, Jason Bellavance, Jon Fitzgerald, Bradley Ramsey, Wes England, Russell Kauffman, Michael Vickers, Dennis Page, Juan Garcia, William Monette, Curtis Van Diepen, Timothy Tippins, Jeffrey Fetzer, Michael Wusterbarth, Andrew Yee, Mark Evans, Rusty Gavagan, Jeff Morrissey
During 2010 and 2011, NIST piloted a 12-laboratory comparison of hydrocarbon liquid flow calibration standards spanning the range 3.8 L/min to 38 L/min. The laboratories were in mutual agreement within the expected 0.3% uncertainty, which is approximately half as large as the differences measured in a similar 1988 comparison. The transfer standard (a pair of turbine flow meters in series) introduced an uncertainty of 0.17% into the comparison. The comparison protocol used methods that were developed during international comparisons including: using uncertainty weighting to generate a best fit comparison reference curve, using statistical criteria to remove discrepant results from the fit, assessing and including in the data analysis the uncertainty contributed by the transfer standard, and reporting a standardized degree of equivalence between the participants. Several laboratories used mixtures of propylene glycol and water (PG + W) instead of Stoddard solvent (the commonly used surrogate for jet fuel) because the PG+W mixtures are safer and cheaper to manage environmentally. This comparison and other studies show that there is no significant difference in the calibration results between Stoddard solvent and a PG + W mixture with the same kinematic viscosity. Therefore, NIST is changing its calibration fluid to PG + W and encourages other laboratories to do the same. Read the Full Article (PDF)
