Turbine flow meters are widely used to measure the flow of valuable fluids such as natural gas and petroleum products. Therefore is important to have primary calibration standards with low uncertainty. NIST has several hydrocarbon liquid flow standards. Historically the fluid used in these standards, and other primary and secondary standards throughout the world, is Stoddard solvent (MIL-PRF-7024E Type II). Stoddard solvent is a light mineral oil (having kinematic viscosity (ν) ≈ 1.2 centistokes at 20 °C) that is less flammable and toxic than jet fuel. Because of the toxicity associated with Stoddard solvent [1] it is highly desirable to rid the existing standards of it.
Propylene glycol and water (PG+W) mixtures are biologically and environmentally benign fluids [2] and a 5 % mixture by volume of PG+W matches the kinematic viscosity of jet fuel. Pure PG has a kinematic viscosity of approximately 50 centistokes that is in the middle of the range of hydraulic oils at 21 °C. Calibration laboratories are interested in utilizing PG+W mixtures to 1) decrease the cost of maintaining multiple standards with fluids of various kinematic viscosities, 2) reduce pollution and provide a biologically benign environment for workers, 3) reduce the danger associated with maintaining multiple standards that require the involvement of the fire department during the use of flammable fuels, and 4) decrease the expense and risk of disposal of the harmful substances.
Arnold Air Force Base calibration laboratories and NIST have began using PG+W mixtures. NIST has successfully changed the fluid in the 20−L hydrocarbon liquid flow standard. Upgrades are in progress for the 2−L hydrocarbon liquid flow standard to also be compatible with PG+W. To have a successful transition away from Stoddard solvent there are three key obstacles to overcome: 1) water compatibility of the standards, 2) water compatibility of turbine meters calibrated on the standards and 3) convincing the calibration community that PG+W gives the same result as Stoddard solvent regardless of the kinematic viscosity. This is true as long as the meter is used in its low-uncertainty, usable range (usually 10:1).
This web page is committed to educating the hydrocarbon calibration community on: 1) how NIST performs turbine meter calibrations, 2) NIST's progress on changing to PG+W, 3) how important bearing materials are when using PG+W and 4) studies and publications that support the move to PG+W. This is an ongoing process and we encourage feedback from the hydrocarbon calibration community.