You are on page 1of 13

McNeese et al.

Proceedings of Meetings on Acoustics


Volume 9, 2010 http://asa.aip.org

159th Meeting Acoustical Society of America/NOISE-CON 2010


Baltimore, Maryland 19 - 23 April 2010

Session 4aAO: Acoustical Oceanography

4aAO8. An investigation of the combustive sound source


Andrew R. McNeese*, Jason D. Sagers, Preston S. Wilson and David P. Knobles *Corresponding authors address: Mechanical Engineering Department and The Applied Research Laboratories, The University of Texas at Austin, 1 University Station C2200, Austin, TX 78712-0292, mcneese@arlut.utexas.edu The Combustive Sound Source (CSS) is a versatile impulsive underwater sound source with an adjustable bandwidth and output amplitude. Unlike typical impulsive acoustic sources, CSS can maintain a wide bandwidth at low-amplitude, hence it is a more environmentally-friendly impulsive source for ocean acoustics experiments and surveys. The CSS consists of a submersible combustion chamber, open to the water, which is filled with a combustible fuel/oxidizer mixture. The mixture is ignited and a combustion wave propagates through the mixture. During this process the ensuing bubble expands due to an increase in temperature and can collapse to a smaller volume than before ignition. This bubble activity radiates acoustic pulses. The CSS can be used as a source for low frequency sediment characterization and TL measurements, and when deployed on the bottom, can produce seismic interface waves. In addition to stationary deployments in the water column, CSS can be deployed in a tow body and as an array. Discussion will focus on the latest CSS design including functionality, acoustic output, long-term operational stability and future development plans. Published by the Acoustical Society of America through the American Institute of Physics

2010 Acoustical Society of America [DOI: 10.1121/1.3543876] Received 19 Nov 2010; published 30 Dec 2010 Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 1

McNeese et al.

1 Motivation

The motivation behind this research is to create an alternative to underwater explosive charges for use in underwater acoustic experiments. It is desired that an alternative source provide an adjustable source level while maintaining a wide bandwidth, provide a repeatable signal in both magnitude and phase, and eliminate the need to store and transport explosive charges. Although several alternatives to underwater explosive charges are currently available, most do not have the ability to maintain a constant spectral shape throughout amplitude adjustment. The present apparatus was only intended to be used to study the acoustic output and was not intended for eld deployment.
2 Description of Apparatus

There are three essential hardware components of a repeatable, reusable Combustive Sound Source (CSS): a gas delivery system, a submersible chamber to entrap gas below the surface of the water, and an ignition source. A repeatable gas delivery system is needed to allow users to accurately deliver the desired amounts of fuel and oxidizer into the submersible chamber before ring the CSS. The chamber must be capable of withstanding the forces and shock loading applied by the combustion process and subsequent bubble activity at a range of depths. Failure of the chamber to withstand high levels of shock loading will quickly result in cracks and leaks. An ignition source within the chamber is needed to initiate the combustion process. This ignition source must be repeatable to ensure that the combustion event is initiated in the same manner over long periods of use. A schematic of these components along with the electronic controls and a recording system is presented in Fig. 1. As seen in Fig. 2, the gas delivery and ignition system are primarily comprised of a three tiered tower bolted on top of a combustion chamber. The bottom tier of the tower consists of a combustion chamber head and ange held directly above the chamber, which allows spark plugs and tubing to be mounted into the combustion chamber. The gas delivery system utilizes bottled gas to produce an air-hydrogen mixture or an electrolytic cell to produce an oxygen-hydrogen mixture. Various mass ow controllers and valves assist in regulating the delivery of each gas through the bottom tier and into the combustion chamber. The ignition system is comprised of a capacitor bank, automotive stick coils, and spark plugs mounted in the bottom tier which repeatably ignite the gases in the chamber. Each subsystem plays a vital role in ring the CSS.
3 Basic Operation

The basic operation of the device is presented in Fig. 3. Initially, the chamber is lled with a combustive mixture. The spark plug then initiates the combustion process, turning the gaseous mixture into high temperature combustion byproducts. The bubble expands due to the increase in temperature and ultimately takes the shape

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 2

McNeese et al.

"  

 

"  


# 


   

  

   

 

     

! !



  

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 3

McNeese et al.

Figure 2: Representation of the Combustive Sound Source.

Figure 3: Descrition of bubble growth and collapse.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 4

McNeese et al.

Gas Mixture

D V
hydrophones

Figure 4: A general schematic of the test set up where D is the depth of the CSS and hydrophones, V is the volume of gas in the combustion chamber, and d is the base diameter of the chamber.
growth, followed by a second peak in pressure corresponding to the bubble collapse. The time between these two peaks is referred to as the duration of bubble collapse. The inverse of the duration of bubble collapse can be seen in Fig. 6, as it is the frequency at which the maximum spectral value occurs. Similar waveforms and spectra were found for various other CSS events containing the air-hydrogen mixture.

Eect of Increasing Volume


The measured energy source level (ESL) is shown in Fig. 7 for events comprised of the air-hydrogen mixture ignited at 5 m and 10 m for various volumes (liters at STP) of gas. The trend of increasing ESL was expected since an increase in gas volume directly relates to an increase in fuel, which yields an increase in radiated acoustic energy. However, it is seen that the ESL increases logarithmically with respect to an increase in total gas volume. The ESL increases at a high rate for small gas volumes, whereas the rate of ESL increase is reduced at larger volumes. This implies that the eciency of energy transfer from potential to acoustic energy is reduced for large volumes of gas. Doubling the volume of gas does not necessarily correspond

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 5

McNeese et al.

150 100 50 0 50 0 0.05

Duration of Bubble Collapse

Pressure (kPa)

0.1

0.15 0.2 Time (s)

0.25

0.3

0.35

Figure 5: Example time series produced by CSS event (18 liters of an air-hydrogen mixture red at a depth of 10 meters).

200 ESD (dB re 1Pa s / Hz @ 1m) 180 160 140 120 100 80 1 10 100 Frequency (Hz) 1000 10000

Figure 6: Example spectra produced by CSS event (18 liters of an air-hydrogen mixture red at a depth of 10 meters).

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 6

McNeese et al.

210 ESL (dB re 1 Pa2s @ 1m)

205

200 5m 10 m 0 50 100 150 200 Total Gas Volume (Liters @ STP) 250

195

Figure 7: Energy Source Levels for total gas volumes at STP of the air-hydrogen mixture at 5 m and 10 m.
to doubling the output acoustic energy. It can also be seen that the data points for events at 5 m and 10 m follow the same trend and are of comparable magnitude when the volume of gas is calculated in terms of volume at STP. This implies that the ESL depends more on the volume (liters at STP) of gas rather than the depth at which the combustive mixture is red.

Oxidizer Comparison
The dierences in the acoustic output for various gas chemistry are apparent in Figs. 8 and 9, which show time series data and the energy spectral density (ESD) for the above mentioned events. Although each event in Fig. 8 has a similar duration of bubble collapse, the peak pressures exhibited by each event and the bubble oscillations after the initial bubble collapse are signicantly dierent. The oxygen-hydrogen mixture is characterized by a sharp peak during bubble collapse, corresponding to the peak pressure of the entire signal. The air-hydrogen mixture is characterized by a much lower pressure during bubble collapse and bubble oscillation for a longer period of time after the initial bubble collapse. Figure 9 also shows that the spectral content of an event is aected by the gas chemistry, hence the use of various gas mixtures allow users to alter the output spectrum.

Consistency of Bandwidth
A unique feature of the CSS is its ability to maintain bandwidth through source level (SL) variations. Figure 7 shows the ability to vary the SL of CSS events, and Fig. 9 shows that each oxidizing gas provides a distinct spectral shape. It can also be seen that the acoustic output can keep its shape while being adjusted in amplitude.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 7

McNeese et al.

300 Pressure (kPa) 200 100 0 100 0 300 Pressure (kPa) 200 100 0 100 0 0.05 0.1 0.15 0.2 0.25 Time (s) 0.3 0.35 0.4 OxygenHydrogen 0.05 0.1 0.15 0.2 0.25 Time (s) 0.3 0.35 0.4 AirHydrogen

Figure 8: Comparison of time series produced by equivalent volumes of Air-Hydrogen and OxygenHydrogen mixtures.

200 dB re 1Pa2 s / Hz @ 1m

airhydrogen O2hydrogen

180

160

140

120

10

100 Frequency (Hz)

1000

Figure 9: Comparison of average octave band energy levels produced by equivalent volumes of Air-Hydrogen and Oxygen-Hydrogen mixtures.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 8

McNeese et al.

190 ESD (dB re 1Pa2 s / Hz @ 1m) 180 170 160 150 140 130 120 1 10 100 Frequency (Hz) 1000 9 L (STP), 5 m 28 L (STP), 10 m

Figure 10: ESD of two CSS events comprised of the air-hydrogen mixture. The blue curve is the result of 9 liters (STP) ignited at a depth of 5 meters, and the red curve is the result of 28 liters (STP) ignited at a depth of 10 meters. The ESD is increased by approximately 10 dB across the spectrum while keeping its shape.

Figure 10 shows the ESD for two CSS events comprised of the air-hydrogen mixture. The two events shown dier in SL by approximately 10 dB, but it is seen that the spectrum keeps its shape and bandwidth is maintained throughout the change in SL, although depth must be varied to achieve this. To further show the relation between the shape of the two events, Fig. 11 shows the two curves given in Fig. 10 with an amplitude shift to overlay the spectral shapes. Here the curves are plotted on an arbitrary vertical axis in order to more easily compare the spectral shapes regardless of the absolute amplitude of each curve. Figure 11 shows that the two events have nearly identical spectral shapes. Figure 12 shows the ESD for two CSS events comprised of the oxygen-hydrogen mixture. The two events shown dier in SL by approximately 19 dB, but it is seen that the spectrum keeps its same basic shape and bandwidth is maintained throughout SL variations. To further show the relation between the shape of the two events, Fig. 13 overlays the spectral shape of the two curves given in Fig. 12 on an arbitrary vertical axis in order to compare the spectral shape regardless of the amplitude of each curve. Figure 13 shows that the two events have similar spectral shapes with only minor dierences. These dierences could be due to the fact that the 6 L (STP) event ignited at 25 m was recorded during the Shallow Water '06 experiment (SW06), which was in a dierent location with a dierent chamber, than the 150 L (STP) event ignited at 10 m. The experimental dierences could be the cause of the minor dierences in spectral shape seen in Fig. 13.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 9

McNeese et al.

0 9 L (STP), 5 m 28 L (STP), 10 m dB (relative) 20

40

60 1 10 100 Frequency (Hz) 1000

Figure 11: Spectral shape of two CSS events shown in Fig. 10. The two events are given here on an arbitrary vertical axis to compare the spectral shape without regard to the individual amplitudes of each event. In other words, the red curve has been shifted down to overlay the blue curve, for comparison purposes.

Repeatability

Throughout testing it was seen that the acoustic output of the CSS is repeatable, especially at low frequencies. Figure 14 shows three dierent time series produced by 80 liters of an air-hydrogen mixture ignited at 5 meters. The repeatability of these waveforms is such that the waveforms lay directly on top of one another making it dicult to distinguish each of the three curves. Similarly, the repeatability of the spectrum produced by each of these events can be seen in Fig. 15. Each of these events diered in source level by less than 1 dB.

Explosive Charge Comparison

Figure 16 shows the ESD for a explosive charge (1.8 lb TNT) detonated at 23.5 m [2] along with CSS events produced from an air-hydrogen mixture and an oxygen-hydrogen mixture. It can be seen that the oxygen-hydrogen mixture produces a spectrum similar to that of the explosive charge with reduced amplitude. The CSS events do not exactly replicate the spectral shape of an explosive charge; however, it is seen that oxygen-hydrogen CSS events do produce a spectral shape similar to that of an explosive charge and have the advantage of easily being adjusted in SL.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 10

McNeese et al.

ESD (dB re 1Pa s / Hz @ 1m)

190 180 170 160 150 140 1 10 100 Frequency (Hz)

6 L (STP), 25 m 150 L (STP), 10 m

1000

Figure 12: ESD of two CSS events comprised of the air-hydrogen mixture. The blue curve is the result of 6 liters (STP) ignited at a depth of 25 meters recorded during SW06, and the red curve is the result of 150 liters (STP) ignited at a depth of 10 meters. The ESD is increased by approximately 19 dB across the spectrum while keeping its same basic shape.

6 L (STP), 25 m 150 L (STP), 10 m

dB (relative)

10

20

30 1 10 100 Frequency (Hz) 1000

Figure 13: Spectral shape of two CSS events shown in Fig. 12. The two events are given here on an arbitrary vertical axis to compare the spectral shape without regard to the individual amplitudes of each event. In other words, the red curve has been shifted down to overlay the blue curve, for comparison purposes.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 11

McNeese et al.

150

100 Pressure (kPa)

50

50 0

0.1

0.2

0.3 Time (s)

0.4

0.5

0.6

Figure 14: Three time series produced by 80 liters of an air-hydrogen mixture ignited at 5 meters.

200 dB re 1Pa2 s / Hz @ 1m

180

160

140

120

10

100 Frequency (Hz)

1000

10000

Figure 15: Spectra produced by the three time series in Fig. 14.

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 12

McNeese et al.

220 dB re 1Pa2 s / Hz @ 1m 200 180 160 140 0 10

TNT

O2H2

AirH2

100 Frequency (Hz)

1000

Figure 16: ESD of CSS event described by 150 liters (STP) of an oxygen-hydrogen mixture ignited at 10 meters (shown in blue), ESD of CSS event described by 240 liters (STP) of an air-hydrogen mixture ignited at 5 meters (shown in red), and ESD of 1.8 lbs of TNT detonated at 23.5 meters [2] (shown in black).

5 Conclusions

The CSS exhibits the behavior of a versatile source. It has the ability to vary source level, octave band energy levels, and repeatably maintain bandwidth through each of these variations. It also has the unique ability for users to tune the output to a desired spectrum through use of various gas mixtures, and the use of gas mixtures eliminates the need to store and transport explosive charges. This work shows the potential for the CSS to become a viable high amplitude, impulsive underwater sound source.
References
[1] P. Wilson, J. Ellzey, and T. Muir, Experimental investigation of the combustive sound source, IEEE Journal of Oceanic Engineering 20, 311 320 (1995). [2] N. R. Chapman, Source levels of shallow explosive charges, J. Acoust. Soc. Am. 84, 697702 (1988).

Proceedings of Meetings on Acoustics, Vol. 9, 005002 (2010)

Page 13