Spool Piece Ultrasonic Flowmeter QWUS-SP2000 DN15-DN600 0.5% Accuracy for Chilled Water Metering

Flow measurement problems in closed loops rarely arrive as a bang. They arrive as drift. The energy billing figure stops reconciling with the chiller log, the BMS trend turns jagged at night when pumps run at 40 %, and eventually someone cuts the pipe open to find a turbine rotor scuffed flat by grit that has been circulating since the first flush. By then the site has already paid twice: once for a meter that under-read, and once for the commercial dispute the under-reading created.

Three patterns push a buyer from mechanical toward transit-time ultrasonic measurement:

• Load swings are wide. Variable-speed pumps have pushed closed loops from a constant 2 m/s design point down to 0.3 m/s for most of the night. A turbine that is accurate at design flow can lose most of its usable turndown below 15 % of range.
• The water is not clean. Glycol top-ups, magnetite from old steel pipe and construction debris all attack bearings. Every maintenance intervention on a mechanical meter means a shutdown, a gasket set and a re-commissioning record.
• Pressure budget is spent elsewhere. In a loop already tight on differential head, a meter that costs 10–30 kPa of permanent pressure loss is a meter that forces a larger pump.

The measurement has nothing to do with the mass of the water and everything to do with the sound velocity of it. Two piezoelectric transducers sit in the spool at a known angle and a known path length apart. Each fires an acoustic pulse through the water. A pulse travelling with the flow arrives marginally sooner than a pulse travelling against it, and the resulting time difference is proportional to the average velocity along the acoustic path.

That single physical relationship is what makes the device so durable in service:

• Velocity is derived from time, not from motion. There is no rotor to stall, no bearing to clog and no reed switch to fatigue — the sensor only has to survive, not to move.
• Density, pressure and viscosity do not enter the calculation. A glycol concentration change alters the sound speed and is corrected internally, so the volumetric reading stays inside specification across the season.
• Multi-path averaging flattens the profile. The SP2000 uses a reflected double-path arrangement that averages across the diameter, which is why a compact upstream run of 5 D is sufficient instead of the 10–20 D demanded by insertion probes.
• Reverse flow is measured, not ignored. Both directions are counted with sign, so a circulating loop that reverses during valve changeover does not lose volume or corrupt the total.

Inside the spool the acoustic path sits in a machined seat with a defined path length and angle, and that is the decisive difference from clamp-on hardware: the geometry is fixed at the factory rather than established by a field installer with a tape measure. Consequently the stated 0.5 % of reading applies to the meter as it leaves the workshop, with no site-dependent derating.

Buyers who compare a spool body with a clamp-on set are usually weighing install effort against purchase price. The honest engineering answer is that the spool pays back in the first commissioning window, because it removes the widest source of error in ultrasonic flow measurement: variable transducer geometry.

Typical duty covers chilled and hot water in commercial buildings, district heating substations, boiler feed and make-up lines, process cooling water and CIP circuits. The sizing rule that keeps accuracy predictable is to land the design flow between 1.0 and 3.0 m/s so the night-time turn-down still sits above 0.1 m/s, and to step down one line size rather than accept a meter running permanently in the bottom fifth of its range.