Back to Basics

[ click here for the Back to Basics Archive ]

Adding and Maintaining Bubble-Free Water in Ultrasonic Scanning Systems

by David L. Petricola*

 

All practitioners of ultrasonic nondestructive testing (NDT) understand the importance of maintaining bubble-free water in immersion scanning systems. Air bubbles in the tank water cause problems when they form on transducer or test object surfaces, and also when they exist as a suspension within the water. Bubbles on the surface cause complete reflection of the ultrasonic energy, preventing the propagation of the ultrasound or producing indications similar to discontinuities. When in suspension, the bubbles cause higher attenuation of the propagating ultrasound, especially at ultrasonic test frequencies above 20 MHz.

To eliminate air bubbles, general practice (often formally specified) requires a waiting period of 24 to 48 h after filling a test tank for the bubbles to dissipate. This system downtime is costly and encourages long periods between water changes resulting in dirty, smelly water.

The University of Dayton Research Institute, through a contract with the US Air Force, designed, built, evaluated and implemented an automated ultrasonic system for the testing of serviceable turbine engine components. The Turbine Engine Sustainment Initiative (TESI) sponsored this system development effort beginning in December 2001, with system implementation at the engine overhaul depot at Tinker Air Force Base in June 2005.

This technique vigorously agitates water
by passing it through an ordinary sprayer
similar to a common garden hose.

The TESI automated ultrasonic system shown in Figure 1 has many unique characteristics, including a water management system that fills and empties the immersion tank as needed during testing (Stubbs et al., 2005). For this system, a wait of 24 to 48 h to allow air bubbles to dissipate from the water is unacceptable; fill/empty cycles occur in times as short as 30 min. Thus, as part of the system design effort, a technique was developed to prevent the formation of air bubbles in the water and to hasten the dissipation of any existing bubbles. This article describes how this water management system works and describes qualitative observations of the presence and absence of air bubbles in water processed by the TESI ultrasonic water management subsystem.

Figure 1 - The Turbine Engine Sustainment Initiative's automated ultrasonic testing system.

A Quick Discussion of the Science: Temperature and Pressure

The amount of gasses dissolved in a liquid depends on parameters such as temperature, pressure, salinity and depth (Hitchman, 1978). It is beyond the scope of this article to go through the thermodynamics of explaining the dependency of the solubility of gasses on these parameters. Rather, an intuitive approach will be taken to explain the dependence of solubility on two significant parameters: temperature and pressure.

The effect of temperature on the solubility of gasses in water is commonly observed. Cold water in a pan initially contains no air bubbles on the bottom of the container, but as the water is heated, bubbles begin to form (the bubbles being less soluble in the warmer water). The solubility of gasses in water is inversely proportional to temperature, or

As for pressure, it is noted that cookbooks advise lowering the temperature for cooking at higher altitudes, where the atmospheric pressure is less. It is easier, for example, to boil water in Denver, Colorado, which is located well above sea level, than in the Mojave Desert, which is below sea level. Therefore, the lower the pressure, the lower the force holding the gasses in the liquid. Solubility, then, is directly proportional to pressure:

Combining the two proportions, we have

If we wish to remove as much gas as possible from the water in an ultrasonic testing tank, we need to lower the solubility by either increasing the temperature or decreasing the pressure on the water. Since the TESI system has an open testing tank and is designed to operate with the water temperature near room temperature, lowering the water's gas solubility by increasing the water temperature appreciably is impractical. The alternative approach of lowering the pressure in the water conditioning process was pursued.

Figure 2 - Formation of bubbles in unshaken jar B and shaken jar C after: (a) 17 min; (b) 24 h; (c) 72 h; (d) nine days.

The first design of the TESI water management system contained a 37.9 L (10 gal) vacuum tank. As tap water flowed into the vacuum tank, the lower pressure in the tank drew gasses out of the water. A pump then moved the water through a filter into a large storage tank. When the TESI ultrasonic system needed to fill the test tank, a 151 L/min (40 gal/min) pump transferred the de-gassed water from the storage tank to the test tank. Although the vacuum tank system successfully removed gas bubbles from the water, it added complexity to the TESI system and increased the probability of maintenance problems. Consequently, the investigation into techniques for de-gassing water was continued.

A Surprising Observation

A simple experiment, originally designed to "create" air bubbles in water, showed that allowing the water to remain dormant for an extended period was not the most efficient way to eliminate bubbles from a container of water. For example, Figure 2a shows a 960 mL (32 oz) specimen jar (marked B) that was filled with tap water and allowed to stand undisturbed. A second jar (C) was also filled with tap water, but in this case the jar was closed and shaken vigorously for about 1 min, with the desired goal being to create many bubbles in the jar. The lid was then removed and the jar was allowed to stand undisturbed. It was expected that many bubbles would quickly appear in the shaken jar C, but very few, if any, would be observed in the undisturbed jar B. However, within 15 min of filling jar B, it was observed that very minute, dust-sized bubbles appeared, whereas only a few stray bubbles were seen in jar C. The condition of the water in these two jars is shown in Figure 2a, taken 17 min after the jars were filled. The squares marked on the jars allowed the observer to count the number of bubbles in the marked areas on the wall of the jars. Figure 2b shows the jars after 24 h. During this 24 h period, the bubbles increased in size but showed no signs of dissipating. As shown in Figure 2c, after 72 h little had changed since the 24 h photograph was taken. Even after nine days, some bubbles remained (see the top half of jar B in Figure 2d). Of course, as would be expected, the number of bubbles in the tap water varied in repeat tries of this experiment, but the primary trend of fewer bubbles in the shaken jar remained.

Results of several of these air bubble experiments indicated that, rather than keeping the water still, agitating the water released bubbles more efficiently. Apparently, the mechanical agitation of the water that occurred when shaking the jars effectively lowers the pressure on the water and released gasses contained in the water. This result caused a reevaluation of the means for de-gassing the water in the TESI system; the vacuum system was discarded and various techniques for agitating the water were considered.

Pursuing the Unobvious

Removing air bubbles from the water in a "passive" fashion rather than "actively" removing them with a vacuum tank, motor driven stirrer or other device became the goal. It was suggested that the water be agitated by filling a storage tank from above with water from a spray nozzle. Questions arose such as "But isn't that how water is aerated?" and "Isn't that exactly what not to do?" As a garden hose nozzle and water are relatively inexpensive laboratory items, the objects were purchased and experiments quickly created.

Figure 3 - Traditional C-scan test tank being filled by hose and nozzle.

In the laboratory, a 1.1 by 1.2 by 0.8 m (45 by 47 by 32 in.) open ultrasonic immersion tank is typically filled by putting a garden hose without a nozzle in the bottom of the tank and turning on the faucet. Typically, this results in air bubbles on the walls of the tank and on the part to be tested — and at least a 24 h wait before starting the tests. Using a newly purchased garden hose nozzle, the tank was filled with the nozzle placed above the tank and set to spray into the tank (Figure 3). Although the observed outcome was now expected, it was still surprising to see that as the tank filled, the water was bubble-free and appeared ready to use immediately for C-scan ultrasonic tests.

This spray nozzle technique was integrated into the TESI system. To help eliminate air bubbles and make the water clean and free of odors, an air separator and charcoal impregnated filter were placed in the water source line before the nozzle. Figure 4 shows the inside of the 946 L (250 gal) storage tank with the nozzle in place. This nozzle, shown in Figure 5, is a full-cone brass spray nozzle for a 12.7 mm (0.5 in.) pipe. The diameter of its orifices is 5.8 mm (0.228 in.) and its spray angle is 120º. The water in the TESI system is constantly recirculated through the air separator, filter and nozzle when the test and calibration tanks are not being filled or emptied.

Figure 4 - Cut away view of 946 L (250 gal) storage tank with spray nozzle used to de-gas and store water for use, on demand, in the Turbine Engine Sustainment Initiative's automated ultrasonic system.

Figure 5 - Standard, commercially available nozzle assembly used in the Turbine Engine Sustainment Initiative's water storage tank.

Additional Experiments

To further study this technique of de-gassing water, a water conditioning system identical to that used in the TESI system was built. Tests were conducted to compare the sprayed water with water straight out of the tap and with water that had been recirculated through the spraying system.

Using the water conditioning test system, undisturbed tap water (the jar labeled A in Figure 6a) was compared with the water sprayed through the sprayer one time (jar D in Figure 6a) and with water that had been recirculated for an hour through the spray system (jar E). The water in these jars was observed over the course of several days. As was seen in previous experiments, very small bubbles appeared in undisturbed jar A just a few minutes after the jar was filled with tap water, and the bubbles remained attached to the jar wall throughout the observation period of nine days. The water in jar D contained a small number of bubbles immediately after filling, and the number decreased slowly over nine days. The recirculated water (jar E) contained no bubbles immediately after filling and remained bubble-free for nine days. The water in all three jars is shown immediately after filling in Figure 6a and after 24 h, 72 h, and nine days (Figures 6b, 6c and 6d, respectively). It is clearly seen that the simple spray technique does a good job of removing air bubbles from tap water after just one pass through the spraying system and completely removes bubbles when allowed to recirculate and pass multiple times through the sprayer.

Figure 6 - Formation of bubbles in unshaken jar A , nozzle sprayed jar D and sprayed and recirculated jar E after: (a) 90 min; (b) 24 h; (c) 72 h; (d) nine days.

Conclusion

An unobvious (at least to the author) technique has been developed for removing gasses from water used to fill ultrasonic testing systems. This technique vigorously agitates water by passing it through an ordinary sprayer similar to a common garden hose. The resultant sprayed and de-gassed water provides clean, clear, bubble-free water that is used, on demand, to fill and empty the TESI automated ultrasonic testing system implemented at Tinker Air Force Base. This technique of producing bubble-free water appears to eliminate the traditional need for a "hold time" of 24 to 48 h between the filling of an ultrasonic immersion system tank and beginning tests. As there are no moving parts and since only a low-cost sprayer head is used, the implementation of this technique is very robust and easy to maintain.

Acknowledgments

The author would like to acknowledge the contributions of Ron Cook, Dave Gasper, James Sebastian and David Stubbs, of the University of Dayton Research Institute, to this article. Photographs were taken by Senior Technician Greg Hartman, also of the University of Dayton Research Institute.

REFERENCES

Hitchman, M.L., Measurement of Dissolved Oxygen, New York, Wiley, 1978.

Stubbs, D., R. Cook, D. Erdahl, I. Fiscus, D. Gasper, J. Hoeffel, W. Hoppe, V. Kramb, S. Kulhman, R. Martin, R. Olding, D. Petricola, N. Powar and J. Sebastian, "An Automated Ultrasonic System for Inspection of Aircraft Turbine Engine Components," Insight, Vol. 47, No. 3, March 2005, pp. 157-162.

 

* University of Dayton Research Institute, 300 College Park, Dayton, OH 45469-0120; (937) 229-4607; e-mail david.petricola@notes.udayton.edu.

 

Copyright © 2007 by the American Society for Nondestructive Testing, Inc. All rights reserved.

[ Back to Materials Evaluation ]