The flow metering element design shall be of the differential pressure producing type for insertion within the pipe line. The inlet shall consist of a conic convergent section for flow conditioning. The throat inlet section included angle shall not exceed 22 degrees and shall have a minimum length of 0.5 times the throat diameter. The throat section shall be cylindrical with a length at least 0.5 times the throat diameter and with sharp intersections to support a protected and stable throat pressure profile. Meters with blended radii will not be considered. Devices amplifying the differential pressure by causing change in the direction of the flow stream cross sections within the throat section shall not be considered. The outlet cone shall have a divergent included angle of 10 degrees. The outlet cone length may be selected to minimize laying length or to maximize pressure recovery.
The entire flowmeter inlet section, throat section, outlet cone and integral flange shall be fabricated of thermosetting epoxy vinyl ester resin, reinforced with not less than 30% Fiberglass by weight. Pressure connections shall be 316 stainless steel. The pressure connecting passage shall be within the integral mounting flange. The device shall be designed for mounting between ANSI flanges and include flange gaskets. The metering element shall not have debris collecting annular chambers or cavities and shall have a single pressure connection for high pressure and a single pressure connection for low pressure sensing of the differential pressure. The manufacturer shall furnish a flow versus differential curve to the design engineer for approval.
The device performance and metering similitude shall be substantiated by the manufacturers traceable independent testing. Performance substantiation shall include coefficient value, constancy with Reynolds number predictability, effects from installation, and line pressure loss from hydraulic shape. The tube coefficient shall be independent of line size and be constant and predictable to low Reynolds numbers.
The uncalibrated accuracy shall be at least 0.75% of flow based on a 2X Standard Deviation value calculated from a significant sample size. The device shall be a Lambda Model 2300 or approved equal. Headloss shall be no more than 8% of the measured differential pre