GB2236180A - Detection of bubbles in flowing liquids - Google Patents

Abstract

Bubbles in a liquid such as beer flowing through a duct (5) are detected by means of a beam of light (17) or similar radiation passing transversely through it to a sensor (20). To achieve maximum fluctuation in the output of the sensor, the beam normally grazes an opaque barrier (19) but strikes the barrier when the beam is diverted by the presence of a bubble. The output of the sensor is monitored through successive observation periods, and the proportion of each observation period during which the output exceeds a predetermined value is calculated. if the proportion exceeds a predetermined value a signal is generated e.g. to close a cut-off valve. The device can distinguish between a few small bubbles and a large bubble or swarm of bubbles, particularly the bubbles that precede fob in beer, thus enabling flow to be quickly stopped. <IMAGE>

Description

The Detection of Bubbles in Flowing Liquids This invention relates to the detection of bubbles in flowing liquids. The invention has been developed in connection with the detection of gas bubbles in beer and similar beverages containing carbon dioxide but it is to be understood that the invention is applicable to the detection of bubbles in other liquids as well.

It is common practice to store beer in a keg and to dispense the beer from a dispense tap to which the beer has been pumped through a length of tube. During the discharge of beer from a keg it is normal for the dispense tap to be opened and closed intermittently and for the pump to be switched on and off intermittently.

Consequently there are intermittent and repeated changes in pressure in the beer and those changes sometimes lead to minor quantities of carbon dioxide coming out of solution and forming minute bubbles in the beer. Such bubbles are unimportant and have no deleterious effect on the dispensing process. During the final stages of discharging beer from a keg, however, larger quantities of carbon dioxide may be sucked into the tubing from the interior of the keg with the result that froth or "fob" passes along the tubing. Hitherto the presence of the fob has not been detected until it has passed the pump and reached sensors, downstream of the pump, that serve to detect the presence of liquid and the flow of liquid; those sensors yield signals that cause the pump to be switched off when liquid is absent or the flow of liquid is not detected.When the presence of fob has led to the pump ceasing to operate, it has been necessary for the beer inlet to be connected to a replacement keg, containing a fresh supply of beer, and for the dispense system to be operated until all of the fob has been discharged from the system by way of the dispense tap. That process has been wasteful of beer, and the present invention enables an improved and less wasteful practice to be adopted.

Now it has been observed that when fob is formed in beer and flows along tubing, beer preceding the fob contains numerous small bubbles, the concentration and size of such bubbles increasing progressively with a reduction in distance from the fob. This phenomenon can be used to advantage in the control of dispensing operations for beer and other beverages as it enables the presence of fob to be detected at an earlier stage than hitherto.

From a first aspect the present invention consists in a method of detecting bubbles in a flowing liquid comprising the steps of causing the liquid to flow through a duct, passing a beam of light or other similar radiation transversely through the duct and the liquid flowing through it, the radiation being such that it is at least partially diverted from its normal path on encountering a bubble or bubbles in the liquid, providing sensing means for sensing the radiation, the sensing means being so located that when the path of the radiation is diverted from its normal path on encountering a bubble or bubbles in the liquid the response of the the sensing means varies from its normal response, analysing such variations in response of the sensing means, determining whether1 during at least a predetermined proportion of an observation period, the response varies to an extent greater than a predetermined extent, and generating a characteristic signal if the response does so vary.

From a second aspect the present invention consists in apparatus for carrying out a method in accordance with the first aspect of the present invention and comprising a duct through which a liquid can be caused to flow, a source of light or other similar radiation operative to generate a beam of such radiation which can pass transversely through the duct when a liquid transparent to the radiation is flowing through the duct, sensing means operative to sense the radiation and so located that in use when the path of radiation is at least partially diverted from its normal path on encountering a bubble or bubbles in the liquid the response of the sensing means varies from its normal response, analysing means operative to determine whether, during at least a predetermined proportion of an observation period, the response varies to an extent greater than a predetermined extent, and signalling means operative to generate a characteristic signal if the response does so vary.

It is, of course, necessary for the duct and the liquid to be transparent to the radiation. While it is envisaged that the radiation would usually comprise visible light, it would be possible to employ electromagnetic radiation of another frequency or other frequencies such as radiation in the infra-red or ultraviolet range.

In order to determine whether, during at least a predetermined portion of an observation period, the response of the sensing means varies to an extent greater than a predetermined extent, it is necessary to add together or integrate those individual parts of the observation period during which the variation in response equals or exceeds that predetermined variation in extent.This addition or integration may be effected by analogue means but is preferably effected digitally, employing a method in which each observation period is sub-divided into a plurality of relatively short sub-periods of time, the number of such sub-periods of each observation period during which the variation in response of the sensing means equals or exceeds the predetermined variation is counted, and a characteristic signal is generated if the total number of sub-periods so counted equals or exceeds a predetermined number. The sub-periods would normally be of uniform duration.

The manner in which a beam of radiation is diverted when it encounters a bubble or bubbles in the beverage or other liquid is relatively complex. The diversion of the beam may result from refraction, reflection or other phenomena or from a mixture of such phenomena. In carrying out a method in accordance with the first aspect of the present invention it is preferred to adopt an arrangement which relies wholly, or at least principally, on refraction.To this end use is made of the fact that a beam of radiation passing an interface between the interior of the duct and the wall of the duct, at an angle inclined to the normal, is refracted at an angle which is a function of the refractive index of the contents of the duct; if the duct is full of liquid that is entirely free from bubbles the beam will then follow one path but if the duct has no liquid in it the beam will follow a different path. Now it has been observed that with an arrangement of that kind, the inclusion in the liquid of bubbles in increasing numbers or of increasing size usually results in an increase in the deflection of the beam.

It will be appreciated that the maximum extent to which the path of the beam is diverted - that maximum diversion occuring when there is no liquid at all in the duct - may well be a relatively small diversion, and that the extent of the diversion that occurs when relatively small bubbles are present in the liquid in the duct may well be still smaller. Nevertheless, it is these small variations in the path of the beam that must be detected. It is therefore preferred to adopt one or more strategies such that the variations in beam-path are rendered as large and effective as possible. To this end the beam is preferably arranged to cross the interfaces between the interior of the duct and wall of the duct at angles inclined at approximately 450 to the normal.If desired the duct may be of square-shaped cross-section and the beam may be directed against one of the sides of the wall at 450 so that it emerges from an adjacent side at much the same angle. Alternatively the duct may be of circular cross-section and the beam may be directed along a path spaced away from the axis of the duct.

The whole of the emerging beam may be allowed to fall on the sensing means. If that is the case, variation in the path of the beam can lead to a variation in the response of the sensing means only if the beam shifts to a more sensitive or a less sensitive part of the sensing means. The position of the sensing means relative to the possible paths of the emerging beam is therefore preferably adjusted to maximise this change in response. Alternatively, or in addition, there may be provided an opaque (or partially opaque) shield means so positioned that, as the path of the beam changes, a portion of the beam is cut off, or is reduced in intensity, by the shield means so that the intensity of the beam striking the sensing means is reduced and the response of the sensing means consequently changes.Two or more of these strategies are preferably employed together in order that the apparatus should give as large a variation as possible in the response of the sensing means.

As indicated above, the invention is particularly applicable to the dispensing of beer and other beverages.

By suitable selection of said proportion of the observation period and by suitable selection of said predetermined extent of the variation in the response of the sensing means, it is possible to provide a method and apparatus such that when used with beer, or other beverage containing carbon dioxide, in the process of being dispensed it is possible reliably to distinguish between the presence in the beverage of a few small and unimportant or insignificant bubbles and the presence in the beverage of any particularly large bubble or any swarm of smaller bubbles of the kind that normally precedes a quantity of fob.

Accordingly, in a preferred method in accordance with the first aspect of the present invention the liquid comprises beer or other beverage containing carbon dioxide and is pumped intermittently along ducting extending from a beverage container to a dispense point from which the beverage is intermittently dispensed, said duct constituting part of said ducting, and generation of the characteristic signal is operative to cause the flow of beverage along that ducting to cease.

Generation of the characteristic signal is preferably operative to close a valve situated in said ducting.

Preferably the beer or other beverage is pumped by pumping means situated in said ducting, and said duct constitutes a part of said ducting that is upstream of the pumping means.

From a third aspect the present invention consists in a method of dispensing beer or other beverage containing carbon dioxide in which method beverage is pumped by pumping means from a beverage container to a dispense point from which the beverage is dispensed, and bubble-detecting means is operative to detect the presence of bubbles in the beverage and to cause the flow of beverage to cease if the size or concentration of bubbles in the beverage exceeds a predetermined level.

From a fourth aspect the present invention consists in apparatus for dispensing beer or other beverage containing carbon dioxide comprising pumping means operative to pump beverage from a beverage container to a dispense point, and bubble-detecting means operative to detect the presence of bubbles in the beverage and to cause the flow of beverage to cease if the size or concentration of the bubbles in the beverage exceeds a predetermined level.

The bubble-detecting means is preferably disposed upstream of the pumping means.

The flow of beverage is preferably caused to cease by closure of a valve, the valve preferably being downstream of the pumping means, and the pumping means then preferably being caused to cease operation by the detection of an increase in pressure of the beverage between the output of the pumping means and said valve.

After the flow of beverage has been caused to cease, any gaseous carbon dioxide collecting upstream of the pumping means is preferably bled from the system, and to this end there is preferably provided a separator disposed upstream of the pumping means, the separator comprising a vessel through which the beverage passes on its way from the container to the pumping means and a bleed valve through which carbon dioxide collecting in an upper part of the vessel can be bled from the system.

In the accompanying drawings: Figure 1 is a diagrammatic layout of a beer dispensing system embodying the present invention, Figure 2 is a diagrammatic cross-section of apparatus embodying the present invention, and Figures 3 and 4 are graphs illustrating one particular method embodying the present invention.

Figure 1 shows diagrammatically the layout of a beer dispensing system such as may be used in a public house. Beer is stored in a keg 1 of the usual kind.

Carbon dioxide under 'pressure is fed from a cylinder 2 by way of a tube 3 to the headspace of the keg. Beer is withdrawn from the keg through a spear (not shown) with an opening closely adjacent to the bottom of the keg and flows through a pipe 4. The beer then flows through a short duct 5 (Figure 2) which constitutes part of bubble-detecting means 6. From that detecting means 6 beer flows to a separator 7 in which any gas entrapped in the beer can be separated from the beer and allowed to escape through a manually operable bleed valve 8 at its upper end. The separator comprises an upright vessel and is provided with an internal barrier 9 which extends upwards from the bottom of the vessel to a horizontal upper edge adjacent to the top of the vessel.At its bottom end the vessel has a beer inlet which enters the vessel to one side of the barrier 9 and a beer outlet which communicates with the other side of the barrier 9.

The outlet of the vessel is connected to the inlet of an electrically operated centrifugal pump 10 of a standard kind normally employed in beer dispensing systems. As is customary, the pump at its outlet has a non-return valve and a pressure-switch assembly 11 followed by sensing means 12 comprising a float-operated device operative to detect the presence of liquid in the system and a device comprising a bobbin which drops when there is no liquid flow through the device but rises in response to a flow of liquid and operates a switch, thereby enabling the occurrence of liquid flow to be detected. Immediately upstream of the sensing means 12 is a cut-off valve 13. All of the parts described so far are disposed adjacent to each other and may for example be disposed in a cellar.

After flowing through the normally open cut-off valve 13, the beer is fed through piping 14 to a dispense tap 15 at a dispense station, which is normally located in a bar at a level considerably higher than that of the cellar or other place where the keg, pump and other parts are disposed. The tap 15 is manually operable and incorporates an electric switch. The complete path followed by beer travelling from the keg 1 to the dispense tap 15 constitutes ducting of the kind referred to above.

When the system illustrated in Figure 1 is in use and beer is to be dispensed, the tap 15 is opened manually thereby operating the associated electrical switch. Operation of the switch temporarily overrides the sensing means 12 and the pressure switch in the assembly 11 and switches on the pump 10 so that beer is pumped from the keg through the ducting of the system to the dispense tap 15. When the tap is closed pressure builds up in the piping 14 so that the pressure-switch assembly 11 operates to switch off the pump.

If, during operation, a few small bubbles occur in the beer these may be detected by the bubble-detecting means 6, but provided their size or concentration is relatively small or low, the system will continue to operate. If, however, the size or concentration of bubbles increases above a predetermined value, as normally occurs shortly in advance of a quantity of fob passing from a nearly-empty keg, the bubble-detecting means produces a characteristic signal which causes immediate closure of the cut-off valve 13. The resultant increase in pressure between the pump 10 and the cut-off valve causes the pressure-switch 11 to operate and switch off the pump.

The nearly-empty keg can then be disconnected and replaced with a fully-charged one. Any fob in the ducting tends to be confined to that part of the ducting upstream of the barrier 9 in the separator 7.

Gas from the fob tends to rise and collect in an upper part of the separator vessel and can be discharged by temporarily opening the valve 8.

The system can be re-started by manual depression of a starting button (not shown). When the system is re-started the cut-off valve 13 is reopened and, on opening of the dispense tap 15, beer is pumped from the separator and fresh beer is drawn over the barrier 9, which may act as a weir if some gas remains in the separator or enters the separator from the spear and the tube 3. Any such gas collects in an upper part of the separator 7 and can be discharged through the valve 8. The process of discharging gas can be repeated as many times as necessary but two or three times is usually enough to get rid of the gas or substantially all of the gas. In this way fob is prevented from reaching the pump and there is no need to adopt the method of clearing the system previously adopted, that is to pump the fob through the piping 14 and discharge it from the tap 15.

The bubble-detecting means 6 comprises apparatus in accordance with the present invention. It is illustrated in Figure 2 which is a cross-section through a short length of duct 5, of circular cross-section, made of a transparent plastics material. In normal use, beer 16 fills the duct and flows through it when the dispense tap 15 is opened. A beam of light 17 from an LED 18 mounted outside the duct 5 is directed through the duct 5 and through the beer 16 flowing through the duct. The beam is directed along a path spaced away from the axis of the duct so that at each interface between the interior of the duct and the wall of the duct the beam is at a markedly inclined angle to the normal, that angle being 0 approximately 45 .An opaque radiation barrier or shield 19 is fixed inside the duct and extends diametrically across the duct from a bottom part of the wall of the duct to a location relatively close to a top part of the wall. The arrangement is such that normally, that is when the duct is full of beer, the beam 17 grazes the upper edge of the barrier 19 or is partially blocked by the barrier, being able to leave the duct and fall on a photo-optic or photo-diode 20 which acts as a photo-electric sensor. In passing through the beer the beam 17 preferably follows a path normal to the barrier 19.

In use, when the duct is full of beer and there are no bubbles in the beer, the beam follows its normal path and grazes the barrier 19, the emerging beam falls on the most sensitive area of the photo-diode 20. When there is no beer at all in the duct, the beam follows a fully diverted path so that the emerging beam falls on a less sensitive area of the photo-diode and the response of the photo-diode is significantly different. The position of the photo-diode is adjusted until this occurs. When there is beer in the duct but bubbles are also present, the path of the beam is also diverted and may be somewhere between the normal path and the fully diverted path. Nevertheless the arrangement is such that the presence of bubbles in the beer (unless they are very small in size) normally results in a relatively large variation in the response of the photo-dioide.

Thus it will be appreciated that the apparatus illustrated in Figure 2 operates in the manner described above and uses the strategies described above to maximise the sensitivity of the device.

Consequently, if a graph is drawn to represent the output of the photo-diode during an extended period of time, the graph shows a uniform output when no bubbles are present but shows a reduction in the signal as bubbles pass through the beam. When occasional bubbles pass there may be one or more isolated dips in the graph of the signal but when the concentration of bubbles increases there tends to be formed an irregular pattern of dips clustered together or overlapping and with different minima.

In order to derive useful information from such signals, the output of the photo-diode is fed to an analyser 21 which serves to perform a plurality of functions. Conveniently the analyser is centred on a digital processor although it would be possible to provide an analyser comprising electronic components acting in an analogue manner. The analyser 21 operates to analyse the signal during a predetermined period of time or observation period and on completing its analysis operates to repeat the process without delay.

This cycle continues indefinitely until or unless a characteristic signal is generated as described below.

Each observation period is subdivided into a sequence of relatively short time sub-periods of uniform duration each of which follows the previous one without a break. Each sub-period is categorised by the analyser, and in particular if during any sub-period the valve of the signal falls below a predetermined value, whether once or more than once, that sub-period is categorised as being a low-signal sub-period. The number of low-signal sub-periods is counted by the analyser and if the total number of low-signal sub-periods that occurs in any one observation period equals or exceeds a predetermined number, a part of the analyser that constitutes signalling means given rise to a characteristic signal. Generation of characteristic signal of that kind by the analyser 21 causes the cut-off valve 13 to close, with the consequences described above.

In a typical form of apparatus the rate of flow of beer and the duration of each observation period are such that between the beginning and the end of the observation period about a column of beer about 6 mm long has passed the beam 17. The number of sub-periods in each observation period may be selected as required but is typically sixteen. Each sub-period may have a duration of some 4 milliseconds. By varying the number of low-signal sub-periods necessary to bring about the generation of a characteristic signal, the sensitivity of the apparatus can be varied. In a typical arrangement a characteristic signal is generated if the number of low-signal sub-periods equals or exceeds three out of an observation period of sixteen sub-periods.

The analyser may be adjustable to enable any one or more of a number of settings to be adjusted. For example the minimum number of low-signal sub-periods necessary to cause the generation of a characteristic signal may be adjustable; said predetermined extent beyond which the response of the sensing means must vary before a sub-period is categorised as a low-signal sub-period may also be adjustable; the duration of the observation period may be adjustable; and the number of sub-periods in each observation period may be adjustable.

In a modification, the observation periods are not subdivided into sub-periods - each observation period is treated as being continuous. The analyser acts continuously to monitor the incoming signal from the sensor 20. Whenever that incoming signal is at its maximum or is above the predetermined level the analyser categorises it as a maximum signal and whenever that incoming signal is equal to or below the predetermined level the analyser categorises it as a minimum signal. Thus the incoming signal is converted into square-wave form. During each successive observation period the analyser integrates the total time during which there is a minimum signal and calculates the proportion of the observation period during which the signal is a minimum signal. When that proportion exceeds a predetermined proportion the analyser generates a characteristic signal.

The analyser may be adjustable to enable any one or more of a number of settings to be adjusted.

Examples of such settings are: said predetermined proportion of any observation period; said predetermined level to which or below which the incoming signal must fall before being categorised as a minimum signal; the duration of each observation period.

Operation of this modified form of analyser is illustrated in Figures 3 and 4. Figure 3 is a graph in which the horizontal axis represents time and the vertical axis represents the signal strength emanating from the sensor 20. Vertical lines 25 denote the boundaries of successive observation periods 26, 27 and 28 of equal duration. During the first observation period, 26, the signal 29 is substantially constant.

During the second observation period, 27, however, the signal on two occasions falls for a short while below a predetermined level 30. During the third observation period, 28, the signal falls on three occasions below the level 30.

Figure 4 is similar to Figure 3 but shows a signal 31 which results from analysis of the signal 29 by the analyser. Whenever the signal 29 is above the predetermined level 30 the signal 31 is at a maximum but whenever the signal 29 is below the level 30 the signal 31 is at a minimum. The signal 31 is thus of a square-wave form. The analyser totals or integrates the total duration of the minimum signal(s) during each observation period and compares it with a predetermined value. In the example illustrated the proportion of the second period 27 during which the signal 31 is at a minimum is too low to give rise to the characteristic signal, but the third period 28 differs in that the proportion of the period during which the signal 31 is a minimum exceeds a predetermined proportion. As a consequence, a characteristic signal is generated by the analyser 21, causing closure of the valve 13.

While the invention has been described as it applies to beer it is to be understood thst it is equally applicable to other beverages containing carbon dioxide. Moreover, the apparatus shown in Figure 2, together with a suitable analyser may be used to detect abnormal bubbles in any transparent liquid.

Claims (19)

1. A method of detecting bubbles in a flowing liquid comprising the steps of causing the liquid to flow through a duct, passing a beam of light or other similar radiation transversely through the duct and the liquid flowing through it, the radiation being such that it is at least partially diverted from its normal path on encountering a bubble or bubbles in the liquid, providing sensing means for sensing the radiation, the sensing means being so located that when the path of the radiation is diverted from its normal path on encountering a bubble or bubbles in the liquid the response of the the sensing means varies from its normal response, analysing such variations in response of the sensing means, determining whether, during at least a predetermined proportion of an observation period, the response varies to an extent greater than a predetermined extent, and generating a characteristic signal if the response does so vary.

2. A method according to claim 1 in which each observation period is sub-divided into a plurality of relatively short sub-periods of time, the number of such sub-periods of each observation period during which the variation in response of the sensing means equals or exceeds the predetermined variation is counted, and a characteristic signal is generated if the total number of sub-periods so counted equals or exceeds a predetermined number.

3. A method according to either of claims 1 and 2 in which the beam of radiation is caused to pass an interface between the interior of the duct and the wall of the duct at an angle inclined to the normal and is refracted at an angle which is a function of the refractive index of the contents of the duct so that the angle when no bubbles are present in the path of the beam differs from the angle when bubbles are present in the path of the beam.

4. A method according to any one of the preceding claims in which the sensitivity of the sensing means to the radiation varies from part to part thereof and the arrangement is such that in use, the diversion of the beam from its normal path on encountering a bubble or bubbles in the liquid results in the beam striking a part or parts of the sensing means either more or less sensitive than the part or parts that it strikes when following its normal path.

5. A method according to any one of the preceding claims in which there is shield means so positioned that in use, the diversion of the beam from its normal path on encountering a bubble or bubbles in the liquid results in at least a portion of the beam being cut off or reduced in intensity by the shield means so that the intensity of the beam (if any) striking the sensing means is reduced and the response of the sensing means consequently changes.

6. A method according to any one of the preceding claims in which the liquid comprises beer or other beverage containing carbon dioxide and is pumped intermittently along ducting extending from a beverage container to a dispense point from which the beverage is intermittently dispensed, said duct constituting part of said ducting, and generation of the characteristic signal is operative to cause the flow of beverage along that ducting to cease.

7. A method according to claim 6 in which the characteristic signal is operative to close a valve situated in said ducting.

8. A method according to either of claims 6 and 7 in which the beer or other beverage is pumped by pumping means situated in said ducting, and said duct constitutes a part of said ducting that is upstream of the pumping means.

9. A method according to claim 1 and substantially as hereinbefore described with reference to Figures 1 and 2 of the accompanying drawings.

10. A method according to claim 9 and substantially as hereinbefore described with reference to Figures 3 and 4 of the accompanying drawings.

11. Apparatus for carrying out a method in accordance with any one of the preceding claims and comprising a duct through which a liquid can be caused to flow, a source of light or other similar radiation operative to generate a beam of such radiation which can pass transversely through the duct when a liquid transparent to the radiation is flowing through the duct, sensing means operative to sense the radiation and so located that in use when the path of radiation is at least partially diverted from its normal path on encountering a bubble or bubbles in the liquid the response of the sensing means varies from its normal response, analysing means operative to determine whether, during at least a predetermined proportion of an observation period, the response varies to an extent greater than a predetermined extent, and signalling means operative to generate a characteristic signal if the response does so vary.

12. Apparatus according to claim 11 in which the beam of radiation is caused to pass an interface between the interior of the duct and the wall of the duct at an angle inclined to the normal and is refracted at an angle which is a function of the refractive index of the contents of the duct so that the angle when no bubbles are present in the path of the beam differs from the angle when bubbles are present in the path of the beam.

13. Apparatus according to either of claims 11 and 12 in which the sensitivity of the sensing means to the radiation varies from part to part thereof and the arrangement is such that in use, the diversion of the beam from its normal path on encountering a bubble or bubbles in the liquid results in the beam striking a part or parts of the sensing means either more or less sensitive than the part or parts that it strikes when following its normal path.

14. Apparatus according to any one of claims 11 to 13 in which there is shield means so positioned that in use, the diversion of the beam from its normal path on encountering a bubble or bubbles in the liquid results in at least a portion of the beam being cut off or reduced in intensity. by the shield means so that the intensity of the beam (if any) striking the sensing means is reduced and the response of the sensing means consequently changes.

15. Apparatus according to any one of claims 11 to 14 in which the duct constitutes part of ducting connected or connectable at one end to a container for beer or other beverage containing carbon dioxide and at the other end to a dispense point from which the beverage can be intermittently dispensed, the ducting including pumping means operative to pump the beverage from the container to the dispense point, generation of the characteristic signal being operative to cause the flow of beverage along that ducting to cease.

16. Apparatus according to claim 15 in which the ducting includes a valve which is closed in response to the generation of the characteristic signal.

17. Apparatus according to claim 16 in which the valve is downstream of the pumping means.

18. Apparatus according to any one of claims 15 to 17 in which said duct is upstream of the pumping means.

19. Apparatus according to claim 11 and substantially as hereinbefore described with reference to and as shown in Figures 1 and 2 of the accompanying drawings.