David Spitzer, process engineer from Spitzer and Boyes, describes how gas pressure variations affect flow measurement.

A flow control loop consisting of a flowmeter, controller and control valve is used to control the flow of a gas to a process.

Measurements from a pressure transmitter installed some time ago indicate the incoming air pressure is typically two bar, but that it can vary from one to 2.5 bar because of process upsets that occur upstream.

How will these pressure variations affect the flow measurement, assuming the flowmeter is calibrated to operate at two bar? First, it is the absolute pressure of the gas that is important when measuring flow.

The pressure of a standard atmosphere is 1.01325 bar absolute, but to make the mathematics simpler, an atmosphere will be approximated to be one bar absolute.

Making this approximation, the typical operating pressure is three bar absolute with variations that can reach from two to 3.5 bar absolute.

Second, many gas flow measurements are inferential mass measurements in units expressed as a standard cubic volume per unit time.

However, regardless of the gas pressure, raw flowmeter measurements reflect the mass, volume, velocity or velocity head of the gas.

If the flowmeter measures mass flow, there will (in theory) be no effect on the flow measurement.

That is, a mass meter should not be affected by process pressure variations.

However, if the flowmeter measures volume or velocity, the effects are inversely proportional to the variation in absolute pressure.

For example, if the gas was operating at a pressure of 3.5 bar absolute, its volume would be approximately 3/3.5, or 85.7 per cent of what it would be if the operating pressure was three bar absolute.

As a result, the flowmeter would measure approximately 14.3 per cent lower than it would if the same mass of gas was flowing at three bar absolute.

The output of flowmeters that infer mass flow by measuring the differential pressure across a restriction is also inversely affected by variations in operating pressure.

However, the relationship is not one-to-one as in the case of flowmeters that measure volume or velocity, but rather approximately 0.5-to-one.

In the above example, the effect of operating at 3.5 bar absolute is approximately 0.5*14.3, or 7.2 per cent.

A more accurate approximation can be calculated as one-sqrt (3/3.5), or 7.4 per cent.