MEASUREMENT OF POWER PLANT EXHAUST FLOWRATES USING LONG-WAVELENGTH ACOUSTIC FLOWMETRY

Lee J. Gorny, Keith A. Gillis

As a part of NISTís program to standardize measurement of greenhouse gas (GHG) emissions from power generating facilities, a long-wavelength acoustic flowmeter (LWAF) is being developed to accurately and economically assess plant exhaust.  Flowrate, in conjunction with composition, plays a pivotal role in developing a platform of technologies that will enhance the regulatory effort of the EPA and other such organizations worldwide.  Currently flowmeters, due to their large scale and the non-uniform, unsteady, swirling profile of exhaust gas have a large associated uncertainty.  No standard exists to calibrate and evaluate flowmeters on this scale in this type of flow.

Potzick and Robertson first investigated LWAF techniques at NIST in the mid 1970ís. In its simplest embodiment, a LWAF is a device that measures flowrate by tracking propagation of acoustic plane waves in a duct.  Potzickís flowmeter was effective in complex, changing flows, largely because of the robustness of plane wave propagation to flow distortions on a first order approximation.  The purpose of this research is to develop the LWAF technique for GHG monitoring in power plant stacks and potentially as a flow reference standard. Difficulties arise when scaling up Potzick and Robertsonís approach because of the high Reynolds numbers in a stack that result in greater blower and turbulence noise (Re8).  To overcome this, either a substantial low-frequency source is necessary or another approach must be developed.

In our work, a 1:100 scale flow facility was built in Building 221 Room A108 where alternative LWAF concepts are being developed and characterized.  Successful techniques will be verified on a 1:10 scale flow facility (currently being fabricated by the Fluid Metrology Group) and on the National Fire Resistance Labís (NFRLís) exhaust stack to further stress their capabilities, and to better evaluate their applicability to a full-scale stack. Ultimately, successful approaches will be evaluated at an operational power plant.  This year, preliminary data were collected at a coal burning power plant as a sanity check for the LWAF approach to this measurement.  The geometry and vibration spectra were measured at points along the duct liner, its inlet and on top of the scrubber.  These data aid in locating and predicting tonal noise sources.  Direct, flow noise measurements in the stack were taken through a fixed port and simultaneously at two axial locations using a tether.  Long-wavelength, time-correlated acoustic propagations were observed at two frequencies in these data, and the baseline acoustic pressure was quantified. 

Using the LWAF 1:100 scale facility, three alternative LWAF techniques have been evaluated.  The first uses a swept sine acoustic source, and measures the phase difference of the signal between axially spaced microphones.  A second method measures time of flight of a pulsed source that propagates, flowrate is obtained by tracking the delay between peaks and valleys at subsequent mic locations.  The third method uses averaging to isolate dominant outgoing acoustic propagations from the fan and other structural sources.  Averaged propagations are obtained at several locations, from which flowrate can be determined by phase comparison.