WO1992008961A1

WO1992008961A1 - Method for reducing the volume of vapors released during tank leakage testing

Info

Publication number: WO1992008961A1
Application number: PCT/US1991/008332
Authority: WO
Inventors: John Emerson Tuma
Original assignee: Tanknology Corporation International
Priority date: 1990-11-13
Filing date: 1991-11-07
Publication date: 1992-05-29
Also published as: AU8775291A; MX9101942A; NZ240564A

Abstract

A method for reducing the volume of gases released into the atmosphere during a test for detecting perforations in an underground fuel storage tank or a container for other chemical or volatile liquids using a method involving the reduction in the pressure in the ullage of the tank. The method involves the cooling of the evacuated gases as they are evacuated from the tank to condense a portion of the evacuated gases, thereby reducing the volume of the evacuated gases which is released to the atmosphere.

Description

METHOD FOR REDUCING THE VOLUME OF VAPORS RELEASED DURING TANK LEAKAGE TESTING

BACKGROUND OF THE INVENTION

Liquid chemicals, especially volatile hydrocarbons used in manufacturing, refining, distribution and other industries, gasoline, diesel fuel, and heating oil are widely used and often stored in tanks at refineries, industrial plants, retail service stations, bulk service stations, municipal garages, or by the end user. There is, therefore, a risk that these liquids may escape into the ground from perforations which develop in the walls of the storage tanks. Tank leakage problems are recognized by the petroleum marketing industry and by governmental and environmental agencies. Furthermore, a leaking tank is generally undis¬ covered until a flagrant appearance of hydrocarbons is traced to it. The tracing procedure can be long and costly because of factors such as unusual soil strata, a network of backfilled trenches, large and frequent variations in the height of the water table, or a dense concentration of underground tanks.

Perforations in tanks usually develop from corrosion; internal corrosion is less prevalent than external corro¬ sion, but does occur, especially in a narrow band along the tank bottom. This pattern is a result of the fact that water, which condenses in the air space at the top of the tank, is more dense than the hydrocarbon and sinks to form a layer at the bottom of the tank beneath the hydrocarbons. Internal corrosion is aggravated under the fill pipe because protective rust is removed by the flow of liquid against the tank bottom during filling or by the impact and movement of a gauge stick.

It has previously been proposed to detect perforations in a tank by a method involving reducing the pressure in the interior of the tank to cause the passage of air into the tank through any perforation which may be present due to a static pressure differential between the interior and the exterior of the tank. The method also involves acous¬ tically sensing the formation of the bubbles of air in the liquid in the tank, converting the acoustic information into an electrical signal, and processing the electrical signal to provide an indication of the existence of the perforation. This method is described in detail in United States Patent No. 4,462,249, hereby incorporated in its entirety by this specific reference thereto.

Although this method is an effective method for detecting leaks in storage tanks, in improperly controlled situations, a relatively high volume of vapors, which may be harmful if not released in accordance with applicable regulatory requirements, is released into the atmosphere during the test. The reduced pressure in the storage tank causes accelerated evaporation of the volatile liquid, especially chemical hydrocarbons having a low vapor pres¬ sure, and subsequent release of that vapor into the atmo- sphere as evacuated gas. The problem is exacerbated by the elevated temperature of the evacuated gases at the exhaust point of the vacuum system, caused by the heat produced by the vacuum pump and other factors, which keeps the evacuat¬ ed gas in the vapor state even though pressure is returned to ambient.

It is, therefore, an object of the present invention to reduce the volume of potentially harmful vapors released into the atmosphere during testing conducted in accordance with the method of U.S. Patent No. 4,462,249. It is another object of the present invention to reduce the exposure of the person(s) conducting a tank test in accordance with U.S. Patent No. 4,462,249 to potentially harmful vapors released during the test. It is another object of the present invention to prevent increased product loss from a chemical or gasoline storage tank as a result of the reduction of pressure in the tank during a test conducted in accordance with the method of U.S. Patent No. 4,462,249. Other objects, and the advantages, of the present invention will be made clear to those skilled in the art by the following description of a presently preferred embodi¬ ment thereof.

SUMMARY OF THE INVENTION These objects are met by providing a method of reduc¬ ing the volume of vapors released into the atmosphere during tank leakage testing comprising reducing the pres¬ sure in the ullage of a tank containing a volatile liquid to cause the formation of bubbles in the liquid therein for detection of the sounds produced by the bubbles. The temperature of the gases evacuated from the ullage of the storage tank is reduced to a temperature below that of the storage tank, causing condensation of the gases and a consequent reduction in the volume thereof. The gas condensate is collected and returned to the storage tank and any uncondensed gas, containing a reduced volume of vapor, is released into the atmosphere.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 illustrates a diagrammatic view of an appara- tus used for testing a storage tank for leaks having means connected thereto for accomplishing the method of the present invention.

Figure 2 shows a top view of the condensation unit connected to the apparatus of Fig. 1 having the top removed therefrom. Figure 3 shows a side view of the condensation unit of Fig. 2.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT As shown in FIG. 1, in a typical installation in which the leak testing method of U.S. Patent No. 4,462,249 is utilized, an underground fuel storage tank 10 is buried in the ground 11 and may be surrounded by backfill (not shown) . The tank 10 contains a body 14 of liquid hydro¬ carbon fuel, for example gasoline, or other chemical or hydrocarbon having a liquid surface 15.

Bubbles 17 are shown rising from a perforation 18 in the tank 10 and a body of water 19 is shown below the fuel 14, the water 19 either having entered through the perfora¬ tion 18 in the tank 10 which is located below the water table or collected by condensation. It will be appreciated that the liquid 14 will tend to leak outwardly of the tank through the perforation 18 since pressure in the tank 10 is greater than the outside, so that the bubbles 17 will not normally be present. Water 19 may collect at the bottom of the tank as a result of condensation at the top of the tank; the denser water sinks and collects at the bottom rather than passing through a perforation such as is shown at 18.

In conducting the tank leakage test, the pressure within the tank 10 is reduced below the pressure outside of the tank 10 to deliberately induce the inflow of air in the form of bubbles 17 in the liquid therein, all as described in more detail in the above-incorporated U.S. Patent No. 4,462,249. The apparatus for use in conducting that test includes a probe 22 which is immersed in the storage tank 10 and located near the bottom 10a thereof. The probe 22 is suspended within the tank 10 by a cable 24 passing through a fill pipe 25 forming a part of the tank 10. The cable 24 is suspended from a closure 26 which hermetically seals the fill pipe 25. The probe 22 includes a hydrophone 41 for detecting the acoustic signatures of the bubbles 17 produced by the ingress of air as described in U.S. Patent

No. 4,462,249, as well as other sensors, each of the sensors being connected through the cable 24 to a control and display unit 27. The control and display unit 27 is connected by cable 28 to an oscilloscope 29 and cable 30 to a pair of headphones 31.

The tank 10 also includes a vent pipe 33, to which a flexible hose 34 is connected by a coupling 32. The flexible hose 34 is connected to an evacuation control unit 35, which in turn is connected by a flexible hose 36 to a vacuum pump 37 which is driven by an electric motor 38. The evacuation control unit 35 is also connected by a flexible hose 40 to a gas source (not shown) containing nitrogen or other inert gas. Depending upon the chemical being tested, it is useful to use stainless steel fittings and stainless steel, teflon-lined vacuum hoses to reduce reactivity with chemicals.

A flexible hose 43 connects the vacuum pump 37 to a condensation unit 42 by a coupling 44. The evacuated gases pass through the condensation unit 42 as shown in Figs. 2 and 3 and out of exhaust pipe 50. The condensation unit 42 is connected by an inlet and outlet refrigerant tube 46 _L and 46₀, respectively, to a refrigeration unit 51 through which refrigerant is circulated to reduce the temperature inside the condensation unit 42. A heat transfer medium such as air, water or polyethylene glycol is contained within the condensation unit 42 to effectuate the cooling of the gases routed therethrough. By means of a coupling 45 and a tee-connector 54, the condensation unit 42 is attached to a removable condensation collection cell 49 at coupling 47 and to exhaust pipe 50 at coupling 48. Fig. 2 shows a top view of the condensation unit 42 with its top removed therefrom. The incoming flexible hose 43 is connected to a stainless steel tube 52 at coupling 44, routing the evacuated gas through the condensation unit 42. The refrigerant tube 46 is similarly routed through the interior of the condensation unit 42, which also contains the heat transfer medium 53. Fig. 3 shows a side view of the condensation unit 42. The stainless steel tube 52 is routed downwardly through the condensation unit 42 so that its low point is at coupling 45, allowing the gas condensate to flow into the collection cell 49.

The operation of the above-described apparatus is as follows. The probe 22 is inserted into the tank 10 through the fill pipe 25 and positioned near the bottom of the tank 10. The fill pipe 25 is hermetically sealed by the closure 26. The flexible hose 34 is connected to the vent pipe 33 by the coupling 32. Using the appropriate control valve settings on control unit 35, inert gas is routed from hose 40 into the ullage of tank 10 to pressurize tank 10 and reduce the likelihood of untoward events. The controls on control unit 35 are then adjusted and pump 37 is driven by the motor 38 to decrease the pressure in the ullage of the tank 10 above the fuel surface 15 by discrete amounts. The pressure in the tank ullage is reduced until the pressure difference at the tank bottom, between the interior and the exterior of the tank, is such that the minimum permissible perforation size can be detected. When the pressure in the ullage of tank 10 has been reduced to the point at which the external atmospheric pressure exceeds the combined pressure of the ullage gases and the pressure head of the liquid 14 above any perforation which may be present, air passes through the perforation, creating bubbles 17.

As the bubbles 17 break away from the internal wall surface of the tank 10, they emit the characteristic sounds or acoustic signatures which are detected by the hydrophone 41, which produces repetitive signals. These acoustic signatures are produced as the formed bubbles 17 change shape as they rise toward the surface. The pressure forces, buoyancy forces, and the surface tension of the liquid 14 in the tank 10 cause the bubble shape to deform as it rises, and these changes in shape or "volume pulsa- tions" emit acoustic waves which are detected by hydrophone

41. As noted above, this method for characterizing the acoustic signatures of bubbles within a storage tank to detect leaks is described in detail in U.S. Patent No.

4,462,249. When the pressure within the ullage of the tank 10 is reduced, the normal evaporative rate of the stored liquid, often a liquid such as an aromatic solvent or other chemi¬ cal having a low vapor pressure, increases. Consequently, controls (not shown) are preferably provided in control unit 35 such that the vacuum pump 37 is automatically triggered to evacuate the gas in the ullage of tank 10 to maintain a desired reduction in the pressure in tank 10. As a general rule, it is preferred to use a reduction in pressure which is as small as possible but which will still induce formation of the bubbles 17 in liquid- 14. A relatively small reduction in pressure decreases the amount of liquid 14 which vaporizes in the first place, and also prevents what is referred to as "champagning", e.g., formation of a continually multiplying stream of bubbles which effectively cover up the useful acoustic information emitted by bubbles 17. It is also preferred, although not required, to test a tank 10 which is at least about half filled with stored liquid 14 so that a smaller reduction in the pressure in the tank can be used during the method of the present invention.

The preferred reduction in pressure in the ullage of tank 10, which may be referred to as the pressure set point, also depends upon the depth and character of the liquid 14 stored therein. Generally, the pressure set point must be sufficient to overcome the pressure exerted on the bottom 10a of tank 10 by the liquid 14, e.g., the "pressure head", and differs depending upon whether the tank 10 is made of steel or fiberglass. By way of example, if tank 10 contains gasoline and the depth of the gasoline is 60 inches, the pressure head is 1.558 psi (e.g., the weight of a one square inch column of gasoline that is 60 inches high) . Consequently, the pressure set point preferred for use in tank 10 is a reduction in pressure of about 1.558 psi plus an additional reduction, which is referred to herein as the probe set point (because a pressure transducer for determining when the preferred pressure set point has been reached is conveniently mounted in the probe 22) of preferably about 0.5 psi (if tank 10 is fiberglass) or about 1.20 psi (if tank 10 is steel). The probe set point is chosen so as to be large enough to insure a vacuum in the tank 10 (note also that the gases in the ullage of tank 10 will contribute to the pressure at the bottom 10a of tank 10; although that contribution is sufficiently small that it can be ignored for the purpose of .calculating the pressure set point, it is an additional head pressure which must be overcome by the probe set point) so that air will enter the tank 10 through any perforations which may be present, but not so large as to cause champagning or damage to, for instance, an already weakened or damaged tank 10. Depending upon the vapor pressure of the liquid in tank 10, the depth of the liquid, and certain ambient conditions, the preferred probe set point may range from as little as zero to about 0.5 psi for fiberglass tanks and from zero to about 1.2 psi for steel tanks.

Another variable which affects the pressure reduction needed to effect proper testing of a tank and the recovery of the vapors evacuated therefrom, e.g., the pressure set point, is the water table depth. When the bottom 10a of tank 10 is below the water table, it is necessary to determine, preferably before initiation of the procedure, the decrease in the pressure set point that results from the head pressure of the water in which the tank 10 is

< positioned. This decrease in the pressure set point is

5 conveniently calculated by the following method. The diameter of tank 10 (designated by the variable A) and the distance between the top of tank 10 and the surface of the ground (variable B) are added together (variable C) , then the distance between the water table and the surface 10 (variable D) is subtracted from the tank bottom (C) to obtain (E) :

A + B = C [1]

C - D = E [2]

Water head pressure at the tank bottom is then obtained by

15 multiplying (E) by 0.036 psi to obtain (F) in pounds per square inch. The water head pressure is converted to the tank bottom pressure differential (G) by adding the probe set point (H) in pounds per square inch to the water head pressure (F) :

20 Ε x 0.36 psi = F psi [3]

^• F psi + H psi = G psi [4]

As noted above, the probe set point (H) differs depending upon whether the tank 10 under test is a steel or fiberglass tank: in the case of a steel tank, (H) is 25 preferably about 1.2 psi, for a fiberglass tank, (H) is preferably about 0.5 psi. In either case, the tank is preferably tested only when (G) is less than about 4.0 psi, and control unit 35 is preferably provided with a pop-off, or safety, valve (not shown) set to trip in the event the 30 tank bottom pressure differential (G) exceeds, for instance, 5.0 psi.

By way of illustration, if tank 10 is made of ξ fiberglass and is 72 inches in diameter, the top of the tank being 12 inches between the surface of the ground and

35 the water table being 48 inches below the surface of the ground, the water head pressure at the tank bottom (F) is obtained using equations [1] - [3] as follows:

72 + 12 = 84 84 - 48 = 36 36 X 0.036 = 1.296 psi

If it is assumed that the tank contains 60 inches of gasoline, the weight, or head pressure, of the gasoline is 1.558 psi, and when the water head pressure is subtracted from the head pressure of the gasoline, it can be seen that a pressure set point of only 0.762 psi (e.g., 1.558 psi - 1.296 psi + 0.5 psi = 0.762 psi) is needed to obtain the approximately 0.5 psi probe set point which is preferred for use in accordance with the method of the present invention. To complete the calculation of tank bottom pressure differential, equation [4] is used as follows: 1.296 psi + 0.5 psi = 1.796 psi and it can be seen that H is well below the preferred 4.0 psi safety limitation.

The temperature of the gases evacuated from storage tank 10 is reduced by routing the gases through the heat transfer medium 53 contained within condensation unit 42, the temperature of medium 53 having been cooled to a temperature below that of storage tank 10 by the refrigera¬ tion unit 51. The gas condensate resulting from that reduction in temperature is collected in collection cell 49, with a consequent reduction in the volume of the evacuated gases, and uncondensed evacuated gases are released to the atmosphere. Collected condensed evacuated gases are preferably returned to storage tank 10 upon completion of leak testing.

In an alternative embodiment of the present invention, rather than using the refrigeration unit 51 and condensa¬ tion unit 42, the gases evacuated from tank 10 are mixed with air or other gas cooled to a temperature below that of storage tank 10, with the result that a portion of the evacuated gases is condensed for collection in cell 49. Because that modification is preferably accomplished by mixing cool gases with the evacuated gases in control unit 35, no separate structure has been shown in the figures.

Although described in terms of the above preferred embodiments, those skilled in the art will recognize that charges in the method may be made without departing from the spirit of the invention. Such changes are intended to fall within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method of reducing the volume of vapors re¬ leased into the atmosphere during leakage testing of a tank containing a volatile liquid comprising: reducing the pressure in the ullage of a tank containing a volatile liquid to cause the formation of bubbles in the liquid for detection of the sounds produced by the bubbles in a test for leaks in the tank; reducing the temperature of the gases evacuated from the ullage of the tank; collecting the gas condensate resulting from the reduction in the temperature of the evacuated gases, thereby reducing the volume of harmful vapors; and releasing any uncondensed evacuated gases to the atmosphere.

2. The method of claim 1 wherein the temperature of the gases is reduced by routing the gases through a heat transfer medium cooled to a temperature below that of the tank.

3. The method of claim 2 wherein the temperature of the heat transfer medium is reduced by a refrigeration system.

4. The method of claim 1 wherein the temperature of the gases is reduced by mixing the gases with air cooled to a temperature below that of the tank.

5. The method of claim 1 additionally comprising returning the gas condensate to the tank.

6. The method of claim 1 wherein the pressure reduction in the ullage of the tank is greater than the difference between the pressure head of the liquid stored in the tank and the pressure head of any water in which the tank may be positioned.

7. The method of claim 1 additionally comprising the purging of the ullage of the tank with an inert gas before reducing the pressure therein.