The largest and most accurate flow calibration rig
in the world

Click to enlarge.

Up to 40,000 m3/h with calibration volume uncertainty <0.013%!
In Dordrecht, NL, KROHNE has commissioned a new large-scale calibration rig for flowmeters.  Fig. 1 conveys an idea of its size.

Calibration rig approved up to 40,000 m3 per hour
NMI, the Dutch calibration authority, has certified that the volume serving as the testing standard for this large-scale calibration rig is accurate to within ±0.013% when calibrating volume-flowmeters in the range from 18 to 40,000 m3 per hour.  This flow rate would be sufficient to fill a normal indoor swimming pool in less than one minute!
This test rig permits the precise calibration of flowmeters up to DIN 3000 (120") - which is big enough for a small goods vehicle to pass through - with an average flow rate of more than 1 m/sec (>3 ft/sec).

The principle of this calibration rig
The stainless-steel calibration tower is 42 m high and has a water volume of some 400 m3.  High-precision level switches at various heights mark a series of partial volumes. These partial volumes and the total volume were precisely calibrated by the Dutch calibration authority NMI. 
The test rig is certified by the NCO (the Netherlands Calibration Organisation, comparable to the DKD in Germany) under EN 45001, thus conforming to national and international standards.


How a calibration takes place
The specimen (a magnetic-inductive or ultrasonic flowmeter, for example) is installed in the outlet conduit of the tower.  The computer determines the required water volume as a function of the target flow and the duration of the calibration, and accordingly selects two level switches as the start and end points of the calibration. 
The regulating valve in the outlet is then closed and the tower filled with water.
The calibration process is commenced.  The valve opens, and the water begins to flow through the specimen under the influence of gravity.  The flow is automatically adjusted to the target value by a regulating valve. 
The measuring transducer of the specimen emits scaled volume pulses in proportion to the flow, e.g. one pulse per litre.  At this point the actual-value counter is still blocked.
When the water level passes the first high-precision level switch, the actual-value counter is released.  It totals the volume measured by the specimen.  Counting ceases when the second selected level switch is passed.
The volume counted by the actual-value counter is compared with the target value between the two level switches, which is known precisely, and the measurement error of the specimen determined.
For each calibration run the computer automatically produces a calibration report documenting flow values and measurement errors.

Every magnetic-inductive and ultrasound flowmeter from KROHNE has a calibration certificate!
Every flowmeter is wet-calibrated in a direct comparison with the volume calibrated by the Dutch calibration authority NMI.  Meters with small nominal bores are calibrated using either a smaller, but otherwise similar, tower test rig or piston-based devices (Piston Provers) with uncertainty less than or equal to 0.03%.
Each individual flowmeter is thus directly referenced to a volume-measurement standard, and hence complies with international standards.  Only in special cases are so-called Master Meters calibrated, and these in their turn are directly referenced to calibrated volumes, exhibiting uncertainty of only 0.1%.
Each flowmeter has its own calibration certificate giving details of the calibration method, its uncertainty and the accuracy values achieved.

The calibration method is 10 times more precise than the flowmeter under test!
The uncertainty of the calibration standard used is generally 10 times lower than the specified accuracy limit of the flowmeters that it is used to test.  This provides the certainty that the specified error limits of these devices are genuinely achieved under reference conditions.
This precludes the phenomenon that sometimes takes place in the market where flowmeters claimed to have measurement uncertainty of 0.15% are calibrated on test rigs with uncertainty of 0.2%.  In these cases there is no way of knowing whether the device sold as 'accurate to ± 0.15%' does not in fact have a measurement error of +0.35%.  It is as if a joiner with a gauge accurate to only ± 2 mm were to guarantee the length of boards to ± 1.5 mm.
The large difference between the uncertainty of KROHNE calibration equipment and the specified accuracies of its flowmeters guarantees true, verifiable data.