Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Long-wavelength acoustic flowmeter

LWAF facility

Figure 1. 1:100 scale (10 cm OD pipe) LWAF facility. The fan forces air through a calibrated flow standard (0.2 % uncertainty with a 95 % confidence level), then through a swirl-generating turn-around section, and finally through the LWAF.

Conventional gas flow measurements conducted in large ducts (such as the smoke stack of a coal-burning power plant) have uncertainties of 5 % to 10 %. Consequently, the quantity of pollutants (sulfur dioxide, mercury, nitrogen oxides, and carbon dioxide) emitted by such ducts have equally large uncertainties. As part of NIST's Greenhouse Gas Measurements Program, we are testing long-wavelength acoustic flow meters (LWAFs) to reduce this uncertainty. LWAFs measure the average volume flow of flue gases in smoke stacks. The measured volume flow can be combined with measurements of the flue gas's pressure, temperature, and composition to determine the mass of pollutants emitted by the stack.


This animation illustrates the principle of the LWAF
Figure 2. In NIST’s model LWAF, loudspeakers generate low frequency sound waves that propagate past several axially-spaced microphones. The sound is partially reflected at the ends of the ducts and passes the microphones again. The data from the microphones determine the speed of sound and the average flow velocity. Because the sound’s wavelengths are longer than the duct’s diameter, the data determine the axial flow velocity, averaged over the entire cross section of the duct. In contrast, ultrasonic (short wavelength) acoustic flow meters determine the axial flow velocity averaged over a small fraction of the cross section.

LWAF graph
Figure 3. Flow rates measured by the LWAF compared to a standard. Error bars show one standard deviation of 20 measurements, each averaged for 60 seconds.

To test LWAFs, we constructed the calibrated, 1:100 scale model, test facility shown in Fig. 1. The model LWAF determined the speed of sound in ambient air with a standard uncertainty of 0.05 %. Within this uncertainty, the speed of sound agrees with the value calculated from NIST's REFPROP database.

The same LWAF determined the average flow velocity. It agreed, within ±1 %, with the velocity determined from a NIST-calibrated flow standard upstream from the LWAF. This good agreement was maintained for flows up to 24 m/s, even after a swirl-inducing turn-around section was inserted between the flow standard and the LWAF. (See Fig. 3.) Similar uncertainties were obtained with highly distorted flows generated by placing obstructions upstream of the LWAF.

To verify the scalability of the method, we will test LWAFs in NIST's 1:10 scale smoke stack simulator.


Created February 3, 2014, Updated November 15, 2019