Ipsofactoj.com: International Cases [2001] Part 8 Case 8 [Ch.D]


CHANCERY DIVISION (PATENTS COURT)

Coram

Micromatic A/S

- vs -

Dispense Systems

International BV

MR JUSTICE NEUBERGER

9 MAY 2001


Judgment

Neuberger J

INTRODUCTION

  1. The claimant, Micromatic A/S, a Danish company, is the registered proprietor of UK Patent GB 2192621B ("the Patent"), which claims an invention whose title is "A Valve Coupling for Securing a Valve in a Pressure Container". The Patent claims a priority date of 18th June 1986, the application having been filed a year later. The claimant contends that the first defendant, Dispense Systems International BV, a Dutch company, infringes. Though there are seven other defendants, only one of them is still relevant, namely Dispense Systems International UK ("DSUK"). The identity of the entity trading under that name is somewhat obscure, but it appears, as its name suggests, to be some sort of UK representative or associate of the first defendant. I shall refer to the first defendant, Dispense Systems International BV as "the defendant", and to the defendant and DSUK as "the defendants".

  2. The defendants deny infringement, but alternatively contend that if they infringe, then the Patent is invalid on the grounds of obviousness or lack of novelty. The essential piece of prior art relied on by the defendants to support their case of lack of novelty or obviousness is a valve coupling used by the claimant and known as the Charles Wells valve, or, in the claimant's documentation, the RS-SOS valve, which is what I shall call it.

  3. The specific type of device in issue in the present case is valve for use in a beer keg. The essence of the alleged invention is a valve which can be dismantled safely. This aspect of safety is particularly important in the context of unauthorised opening of beer kegs, as such kegs are pressurised vessels. Breweries have standard procedures and special tools which are supposed to be used to remove the valves from the kegs. There is normally no reason why the valve unit in a beer keg should be taken out when the keg is in a public house. However some proprietors or employees in public houses try and open kegs, for various, often nefarious, reasons. Such people will not have the benefit of the breweries' procedures and tooling and are unlikely to know how the valve is supposed to work.

  4. The claimed invention is a valve coupling which is safe to extract and will not, according to the Patent, be "suddenly shot out of the neck ring". This is achieved by organising the geometry of the body of the valve and of the neck ring (which is welded into the keg) so as to make sure that pressure is released before the dismantling process is finished. The Claimant submits that the prior art did not achieve this result because the rubber gasket under the underside of the valve might automatically self seal or reseal by re-adhering to the inside of the neck ring. However, it is right to add that the cited prior use RS-SOS valve system had a lower pin which, the two defendants contend, was intended to address such resealing.

  5. The issue of whether the defendant's product infringes the Patent turns essentially on the proper construction of the Patent, and, in particular, whether it is limited to a two piece valve, as the defendants contend, or whether, while incorporating a two piece valve, it also extends to a single piece valve, as the claimant contends. If the defendants are right on this issue, then there is no infringement, and they do not challenge the validity of the Patent. If the defendants are wrong on this issue, then particularly if they infringe, they maintain their attack on the Patent on the grounds of obviousness and lack of novelty. In addition, the claimant seeks to amend the Patent.

  6. I propose to begin by describing the three valves which have been provided during the hearing including the prior use RS-SOS valve relied on by the defendant and the allegedly infringing product of the defendant. I shall then set out the relevant parts of the Patent, including those parts which the claimants seek to delete or add by way of amendment. I shall then deal with the main issue of construction of the Patent, in its current form, and then the construction of the Patent as proposed to be amended ("the amended Patent"). I shall then deal briefly with the Patent's alleged contribution to the art. I shall then discuss infringement. Next, I shall turn to a set of experiments carried out by the claimants. I shall then discuss the issues of novelty and obviousness. Finally, I will deal with the question of amendment.

    THE VALVES

  7. During the course of the hearing, I was provided with three valves.

    • The first was the defendants' allegedly infringing product, known as the DSI Valve.

    • The second was the RS-SOS Valve, is said by the two defendants to anticipate the Patent, or, alternatively, over which they say the Patent is obvious.

    • The third valve was a two piece valve manufactured according to the teaching of Claim 1 of the Patent in its unamended form, according to the defendants' construction.

    I was also provided with two types of neck ring, into which each of the valves fitted.

    • The first type of neck ring was a metal neck ring similar to the sort of neck ring which would have been welded into a keg, but with a section removed to enable one to see how a valve actually engages with the neck ring.

    • The second neck ring was made of transparent plastic, which enabled one to see how a valve engages without the need for the cutting out of the section.

    Each of the two neck rings had the two pins in the neck. It was very helpful to have these items. In a case such as this, a model is worth more than a thousand words. As the drafting of Claim 1 of the Patent, even in its improved form as proposed to be amended, demonstrates, however skilful the drafting, one's understanding is immeasurably improved by seeing the thing in action.

  8. Mr Burkill helpfully provided four sets of photographs comparing RS-SOS and DSI valves in four different positions. They were each taken with the metal neck ring with the section removed, rather than with the transparent neck ring. I attach the four sets of photographs together with the descriptions as provided by Mr Burkill:

    As the two pins are on the inside of the neck ring, they cannot be seen, but one can identify their location quite easily by reference to the two indentations either side of the removed section of the neck ring: the upper pin is to the right, and the lower pin is to the left, of that removed section.

  9. If one imagines the neck ring as stationary, the valve (which is attached to a spear which extends into the keg) is lowered into the neck ring. As one lowers the valve into the neck ring, the valve will normally stop as the upper pin in the neck ring hits the bottom of the neck of the valve. However, if one turns the valve at that point, there is a groove in the bottom of the valve neck many times wider than the pin. As a result of this groove ("the lower vertical groove"), the valve can be moved further down into the neck, but only to a limited extent, because the vertical groove is short. However, the groove continues in a latitudinal direction (i.e. along the outer circumference of the neck of the valve). Accordingly, the valve can be turned in the neck, with the upper pin remaining in this latitudinal groove. After a rotation of some 300 degree, the latitudinal groove stops, and accordingly one can rotate the valve no further because the pin stops rotation. However, at that point there is a further vertical groove in the valve neck ("the upper vertical groove") so that the pin no longer prevents the valve from going further down into the neck ring. What next stops the valve travelling further down is not the pin but the bottom neck of the valve hitting the underside of the neck ring. Although not included in the photographs, a rubber gasket separates the bottom of the neck of the valve from the bottom of the neck ring.

  10. Once the valve reaches the point, this top of the valve is lower than the top of the neck ring, and the neck ring is so designed that a circular clip (or circlip) can be inserted into the top of neck ring so as to keep the valve in place despite the pressure in the keg pushing it out. The bayonet fitting at the top of the valve can then be used to attach a tap which so engages with the valve that, when the tap is turned on, beer can be obtained from the keg.

  11. If one subsequently wishes to obtain access into the keg, one must, of course, first remove the tap. One then removes the circlip. Once that happens, the pressure in the keg will serve to push the valve out of the neck ring. At that point, the valve is pressured out of the neck ring to the extent permitted by the upper vertical groove in the valve moving up along the upper pin in the neck ring until the upper pin reaches the bottom of that groove.

  12. The valve can then be turned so that the upper pin travels along the latitudinal groove (albeit that it is the valve, and therefore the groove, that is moving and the pin, together with the neck ring and the keg, which is stationary). However, shortly after this movement commences, the upper pin, which, if there is any pressure in the keg, will be very much at the bottom end of the latitudinal groove, will pass into a "pressure release hole", or a safety detent as I shall call it, to which I have not so far referred. This is a hole cut into the bottom end of the latitudinal groove and into which the upper pin will naturally fall if the valve is being turned in the neck ring under pressure. It is a safety device, whose purpose is to minimise the risk of the valve being able to disengage from the neck ring while there is still significant pressure in the keg. In effect, the upper pin should remain in the safety detent so long as there is any significant pressure in the keg. Once there is no significant pressure, the upper pin can be lifted out of the detent (or, more accurately, the valve can be pressed down into the neck ring so as to take the upper pin out of the detent) and the upper pin can then continue moving along the latitudinal groove (or more accurately the groove can continue turning on the pin).

  13. This part of the dismantling exercise effectively ends when the valve has turned to such an extent that the upper pin is at or near the end of the latitudinal groove into which it obtained through the lower vertical groove. At that point the valve can be taken out of the neck ring because the upper pin, being in the lower vertical groove no longer prevents the valve being lifted out of the neck ring, in the same way as the lower vertical groove had enabled the valve to be inserted into the neck ring not withstanding the presence of the upper pin.

  14. It will be noted that, in the above, regrettably rather lugubrious, explanation as to how the valves works, I have not differentiated between RS-SOS and the DSI valves. That is because, at least in terms of the physical description, they work in entirely the same way, albeit with one distinction which I have not so far explained. That distinction is that the groove which I have described as latitudinal (i.e. the groove which enables the valve to be turned in the neck ring because it effectively "contains" the upper pin) is strictly horizontal in the case of DSI, but it is slightly slanted or helical to the horizontal in the case of RS-SOS. In the RS-SOS device, the slant in the longitudinal groove is such that, when the valve is being inserted into the neck ring at the rotation stage, the upper pin travels a little distance up the valve (or, more accurately, the valve travels down the neck ring) at an angle of 3 degree up from the horizontal. By the same token, when being released from the neck ring, the valve, at that rotation stage, comes out of the neck ring to a small extent, because the latitudinal groove is at 3 degree down from the horizontal. On the other hand, in DSI, the latitudinal groove is strictly horizontal, and accordingly, during the rotation stage, the valve neither goes into, nor comes out, of the neck ring (whether one is considering the insertion of the valve or its removal).

  15. The RS-SOS and DSI valves each involve the neck ring having two pins. They each involve the valve having a first vertical groove, through which the upper pin passes, a latitudinal groove through which the upper pin can pass during rotation of the valve, at the end of which there is a second vertical groove through which the upper pin passes so as to fix the valve more firmly in the neck ring. They each have the same method of attachment (namely a circlip) and the same features, including the safety detent. The only difference between them is that the latitudinal groove is strictly horizontal throughout in the case of DSI, and almost, but significantly not, horizontal in the case of RS-SOS.

  16. I must now refer to the working of the two piece valve made up by the defendants to reflect what they say is the only correct reading of the Patent. When dismantled, it is in two pieces. The first and main piece contains the valve itself. It includes, close to the top of its neck, a ridge with a comparatively small vertical groove, which is the lower vertical groove. The second piece is an upper ring which is of the same width as the ridged part of the main piece; it also has a vertical groove, which is the upper vertical groove. If one places the main piece into a neck ring with two pins, the valve will only go fully into the neck ring if the upper pin passes through the lower vertical groove in the top ridge of the piece, and the valve is then turned so that the lower pin can also pass into the vertical groove, whereupon the underside of the neck of the valve will hit the underside of the neck ring and be fixed. One can then place the upper piece of the valve into the neck ring with the upper vertical groove in the upper piece of the valve fitting over the upper pin of the neck ring. The effect of including the upper piece of the valve is to create a horizontal groove under the upper vertical groove in the upper piece (in which the upper pin of the neck ring is resting) and above the lower vertical groove in the lower piece (in which the lower pin of the neck ring is resting). Accordingly, when inserted and assembled, this two piece valve looks pretty similar to the DSI valve except that

    1. the latitudinal groove goes all the way around the neck of the assembled valve;

    2. the lower vertical groove is much narrower, being only a little wider than either of the two pins; and

    3. there is no pressure relief hole or detent.

  17. As Mr Burkill, who appears for the defendants, argues, and, indeed, in accordance with the evidence, particularly of Mr Hendrik Doelman, the expert called on behalf of the defendants, the two piece valve, such as that assembled for the purpose of these proceedings by the defendants, work somewhat differently from a single piece valve, such as RS-SOS and DSI valves.

  18. Thus, in the case of a two piece valve, the upper piece or cylinder, and the lower piece or cylinder, each require some mechanism whereby they are prevented from rotating. In the case of the upper cylinder, an undesirable degree of rotation could lead to difficulties with attaching the dispenser heads to the coupling means (shown in figures 1 and 2 of the Patent as a bayonet fitting) in the top cylinder. The upper pin in the neck ring, by engaging with the slot in the upper piece or cylinder, achieves this. As to the lower piece or cylinder, it needs to be prevented from rotating, because otherwise, if it becomes aligned with the top pin, it would shoot straight out with the top as soon as the circlip is removed. Accordingly, the lower pin must not be vertically below the upper pin, and it must engage with a slot in the bottom cylinder. In the case of a single piece, the two parts - or cylinders - must move, and in particular rotate, as one: if the upper cylinder is prevented from rotating by the top pin, then the whole valve (as it is a single piece) is prevented from rotating. There is therefore no need to make any separate provision to stop the lower cylinder from rotating.

    THE PATENT

  19. The specification of the Patent begins by explaining what the alleged invention concerns, namely a valve inserted into a neck ring in a container with "at least one gasket to form a pressure-fluid-tight connection between the valve and the neck ring" (page 1, lines 9 to 11). The specification goes on to explain the risk of valve being "suddenly shot out of the neck ring" if the pressure has not been released in the keg before the valve becomes capable of such ejection. The aim of the invention is described as "to provide a valve .... which does not involve [that] danger, and which is .... simpler and cheaper .... than known before" (page 1, lines 28 to 31).

  20. The specification goes on to explain the claimed invention, concentrating on the two pins in the neck ring and explaining how they are to function. There is nothing in the first two pages explaining that the function of the lower pin is to prevent resealing, and the application to amend seeks to add the following words:

    in the first dismantling step the pressure / fluid tight seal provided by the gasket is released and the lower prevents the gasket from sealing again so that in the area of the lower pin there will be formed a sufficiently large opening for the pressure in the container to be relieved.

  21. On page 3 of the specification, there is the following passage at lines 4 to 19, and the application to amend involves all but the first sentence being deleted:

    Preferably, the open-ended guide groove extend parallel to the axis of the valve coupling. Preferably, the part of the valve receiving area comprising the upper cylinder part may be a separate loose ring. This has the advantage in terms of production that it is easier to manufacture the coupling means for coupling the valve with a dispenser head for dispensing the contents of the container.

    Additionally, it is preferable that the part of the valve receiving area comprising the upper cylinder part, the recess and the lower cylinder area consist of one or more separate loose rings. This makes the structure particularly simple and inexpensive to manufacture.

  22. At the bottom of the same page, there is this:

    It is preferable that, in order to prevent the pins in the neck ring from working loose in use, they may be integrated in the neck ring, said pins being formed by radial impression of a part of the cylinder wall of the neck ring.

  23. On page 4 of the Patent, reference is made to 5 figures. The reference is in these terms:

    fig.1

    is an exploded view [of an embodiment] of a valve coupling in which the upper cylinder has a separate base ring and is therefore not in accordance with the invention [according to the invention],

    fig.2

    is an enlarged top view of a valve and a neck ring for a pressure container, fully coupled,

    fig.3

    shows the same, but seen from the side,

    fig.4

    is a lateral, partially sectional view of the coupling between the valve and the neck ring shown in fig. 3,

    fig.5

    shows the same, seen in a first dismantling step,

    fig.6

    shows the same, seen in a second dismantling step, and

    fig.7

    shows the same, in a final dismantling step.

    In relation to figure 1, the words in square brackets are included in the Patent, but are sought to be excluded in the amended Patent, and the underlined words are not in the Patent but are sought to be included in the amended Patent.

  24. It should be explained that figure 1 shows four pieces. The first is the circlip or returning ring. The second and third pieces are the two pieces which constitute the valve and are, in effect, a drawing of what is the two piece valve which the defendants have had made up and which I have described above. The fourth piece shown in figure 1 is the neck ring with the two pins. I do not think I need comment in detail on the other six figures, at any rate at this stage.

  25. On pages 5 to 8 of the specification, the draftsman of the Patent explains in a little detail what the figures represent. After explaining again the risk of the valve being shot out of the keg if there is nothing to ensure that the pressure in the keg is released before it can shoot out, the Patent continues (page 5, line 33 to page 6, line 6):

    To eliminate this risk, in the valve coupling of the invention, the valve is moved out of the neck ring in at least two steps during dismantling, and in the first step of these the valve is only allowed an axial movement which is smaller than the height of the neck ring, but sufficiently great to cause relief of the pressure in the container. When the last dismantling step is initiated, the container is no longer under pressure, so that the valve may finally be removed without any danger ....

  26. The specification then goes on to explain the nature of the valve in a little more detail. On page 7 (lines 11 to 22), there is this:

    In the first dismantling step the gasket .... is no longer fixed between [the underside of the neck of the valve and the top of the underside of the neck ring] and it may therefore be moved upwardly and partly outwardly, thus opening a pressure passage from the container via the guide grooves [sc. the upper and lower vertical grooves] and a gap between the cylinder parts on the receiving area of the valve. This gap .... which keeps the valve safely guided in the neck ring while the air passage will be sufficiently large to ensure rapid relief of the pressure in the container.

  27. Importantly, the specification immediately goes on (at page 7, lines 24 to 32):

    The pressure in the container frequently urges the gasket to a new position where it seals against the pressure again, but this will be prevented in the area at the lower pin in that the gasket is kept spaced from the downwardly facing breast of the valve by the lower pin, so that, at least in this area, there will be formed an opening sufficiently large for the pressure in the container to be relieved.

  28. The specification then explains that, thereafter, the second dismantling step involves the valve being rotated, with the upper pin in the latitudinal groove moving away from the upper vertical groove, until it reaches the lower vertical groove, at which point the third dismantling step can be taken. This involves effectively pulling the valve out of the neck ring, which is not precluded by the upper pin, because it passes through the lower vertical groove.

  29. I should also refer to a passage on page 8 of the Patent (lines 16 to 33) which is in these terms:

    In the above embodiment of the valve coupling of the invention, the receiving area 3 of the valve 2 is integrally formed with the rest of the valve. In [the embodiment of] fig. 1, however, the upper cylinder 11 consists of a separate loose ring 5 in which the coupling means 10 for screwing on the dispenser head are provided. [This embodiment facilitates the production of these coupling means 10. In other advantageous embodiment, the receiving area 3 of the entire valve 2 consists of one or more separate loose rings so that the structure is particularly simple and inexpensive to manufacture.

    Again, the claimant seeks in its amendment to the Patent to delete the passages in square brackets.

  30. Claim 1, particularly as proposed to be amended, is of some length, and the claimant's expert, Mr Eric Partington, has helpfully divided it up into 19 integers. So divided up, it is in the following terms:

    1. A valve coupling for securing a valve in a pressure container, such as a keg of beer,

    2. comprising a neck ring secured to the container, a valve having a receiving area adapted to be received in an opening in the neck ring,

    3. the receiving area being formed integrally with the rest of the valve,

    4. locking means to secure the valve in the neck ring, and at least one gasket to form a pressure- and fluid-tight connection between the valve and the neck ring, and where said valve and neck ring have mutually co-operating coupling means respectively,

    5. which, when the receiving area is removed from the neck ring, are adapted to ensure that said removal can take place only through at least two removal steps,

    6. and to ensure relief of the pressure in the container at least prior to the initiation of the last one of these steps,

    7. wherein the valve, in its receiving area, has an upper and a lower cylinder part spaced by a recess

    8. and each formed with a groove to provide respective upper and lower open-ended guide grooves extending longitudinally of the coupling

    9. and being displaced an arc distance with respect to each other along the circumference,

    10. and that the neck ring along its inner side having an upper and a lower pin

    11. mateable with said guide grooves

    12. and each having a height and length which is smaller than the width and depth, respectively, of the recess,

    13. the upper pin detachably engaging with the upper guide groove (in the mounted position of the valve,)

    14. and the lower pin detachably engaging with the lower guide groove in the mounted position of the valve,

    15. said pins being moved, upon mounting and dismantling, through the guide grooves in two steps with an intermediate relative angular rotation

    16. corresponding to said arc distance,

    17. the upper pin sliding in the recess between the upper and lower cylinder part

    18. and the lower pin along the underside of the lower cylinder part during said rotation.

    19. In the first dismantling step the pressure/fluid-tight seal provided by the gasket is released and the lower pin prevents the gasket from sealing again so that in the area at the lower pin there will be formed a sufficiently large opening for the pressure in the container to be relieved.

    The claimant seeks to add the underlined integers by amendment.

  31. I need not refer to Claims 2, 5, 7 and 8. The claimant seeks to delete Claims 3 and 4. I should set out Claim 6, which the claimant also contends is infringed by the defendants:

    6.

    A valve coupling according to any preceding claim, including a cylindrical neck ring, whereas the pins of the neck ring adapted to engage the guide grooves of the valve are formed by radial impression of a part of the cylindrical wall of the neck ring.

    DOES THE PATENT EXTEND TO A SINGLE PIECE VALVE?

  32. This turns on the proper construction of the Patent. In this connection, there does not seem to be a dispute between the parties as to the identity of the notional addressee of the Patent. He would be a person with a general engineering background. Although I do not think it is disputed by the claimant, it is right to emphasise that I accept the submission on behalf of the defendants that the addressee would be familiar with the basic designs of valves in use as at the priority date, particularly valves used in beer kegs. Although, as pointed out by Mr Birss, who appears on behalf of the claimant, the Patent is not specifically limited to valves used in beer kegs, I agree with him that the addressee would be a person interested in valves in connection with beer kegs.

  33. Although one or two other issues of construction were canvassed during argument, the central, and not entirely easy, issue between the parties is whether the Patent is limited, as the defendants contend, to a valve in two pieces, namely (a) the main part of the valve and (b) a separate top cylinder or ring containing the coupling means for the dispenser head, or whether, as the claimant contends, although the Patent does extend to such a two piece device, it also extends to a single piece device, where the upper ring (with the coupling means) is physically part of the same piece as the main, lower, part. So as to avoid confusion, I should emphasise that, whoever is right on this issue, two other relevant and separate pieces (apart from the keg itself) would have to be involved before the claimed invention would perform its function. The first is the neck ring, (normally secured by welding into the keg) which contains the two pins, and into which the valve is placed. The other separate piece is the returning ring or circlip which effectively holds the valve down into the neck ring, and whose release (whether authorised or not) is the first necessary step for the removal of the valve from the neck ring. For ease of reference, I shall refer to a "two piece valve" as being the type of valve to which, according to the defendants, the Patent is limited, and to a "one piece valve" as being the type of valve to which the claimant contends the Patent also extends.

  34. There is no doubt that some of the figures in the Patent are consistent only with a two piece valve. This is as plain as could be in figure 1. Figures 2 and 3 are consistent with both a one piece valve and a two piece valve. To the untrained eye, figures 4 to 7 inclusive also appear to be neutral; however, to the more informed eye, and in particular to the eye of the notional addressee, it seems clear that figures 4 to 7 are consistent only with a two piece valve. I reach that conclusion because of the convention in engineering drawing which identifies physically separate pieces by a straight line going across a hatched area in which, in the case of a single piece valve, a straight line would be inappropriate.

  35. However, I consider that the description of the figures in the Patent calls into question the view that the Patent is limited to a two piece valve. The description of figure 1, "an exploded view of an embodiment of a valve coupling according to the invention", appear to me to indicate that the draftsman of the Patent had in mind other possible embodiments. The description of figure 2 (and therefore of figure 3 and the ensuing figures) is not so clear in its wording, but it does refer to "a valve and a neck ring". These references are on page 4 of the Patent.

  36. Earlier, on page 3 of the Patent, there are the words, on lines 6 to 8:

    Preferably, the part of the valve receiving area comprising the upper cylinder part may be a separate loose ring.

    This appears to me to emphasise that a two piece valve is preferable in practice, an assertion which, ironically, is pretty clearly wrong. However, of more relevance, the words appear to me to suggest, if anything, that the draftsman does not have in mind a requirement for a two piece, as opposed to a single piece, valve.

  37. On page 8 of the Patent, after referring in some detail to the way in which the claimed invention works, there is this passage (lines 17 to 23):

    In the above embodiment of the valve coupling of the invention, the receiving area 3 of the valve 2 is integrally formed with the rest of the valve. In the embodiment of Fig 1, however, the upper cylinder 11 consists of a separate loose ring 5 in which the coupling means 10 for screwing on the dispenser head are provided.

    Once again, the draftsman goes on to explain the advantage of having more than one piece.

  38. I include the numbering (which is meant to be a reference to the numbering on the figures) in the extract just quoted from the Patent, because it is somewhat confusing when one compares it to the figures. Nonetheless, it appears to me clear from that extract, and indeed not inconsistent with the figures, that the draftsman is indicating that he is contemplating a single piece valve, albeit that a two piece valve (indeed a valve consisting of more than two pieces) is comprehended in the invention and, indeed, may be preferable. Mr Burkill argues that the concept of the upper ring being "integrally formed" with the rest of the valve does not indicate, or at least does not unequivocally indicate, that there is a single piece. I do not agree. First, as a matter of ordinary language, it seems to me that if two components are "integrally formed", then physically there is a single piece containing both components. Secondly, that interpretation of the words "integrally formed" appears to be supported in the passage I have quoted in the second sentence. Thirdly, as Mr Birss points out, in a passage also on page 8 of the Patent, a "particularly advantageous embodiment" of the claimed invention is described, namely one involving the two pins being "integrally formed with the neck ring .... by radial impression of a part of the cylinder wall" with the consequence that the "structure is very strong and prevents the hinge from working loose in use" (lines 30 to 34). Particularly when the passages are so close to each other in the same document, it seems to me that there is considerable force in the point that the same expression, "integrally formed" should be given the same meaning in each passage.

  39. In the end, of course, the central issue is the proper construction of the claim, and not of the specification. Nonetheless, it would be wrong to determine an issue of construction relating to a claim in a patent without having regard to the contents of the specification. The words of the specification cannot be invoked to alter the effect of the plain and unambiguous words of the claim, unless, perhaps, the natural meaning of the claim is irrational or extraordinary, and cries out not to be given its natural meaning. However, although, save perhaps in exceptional circumstances, the specification cannot be invoked to alter the plain and unambiguous meaning of the claim, the contents of the specification are relevant when construing a claim. This is particularly true where a claim is badly or unclearly expressed, or is capable of having more than one meaning. In such a case, the contents of the specification will frequently play a part, sometimes a really crucial part, in determining the meaning of the claim.

  40. Turning to Claim 1, I do not need to discuss integers (a) to (f), (i), (l), (o), or (p), because Mr Burkill accepts that they are consistent with both a two piece valve and a single piece valve. However, I must consider the other integers, because it is contended on behalf of the defendants that, in one way or another, they are inconsistent with the notion of the Patent extending to a single piece valve.

  41. First, integer (g). As a matter of language, I do not see why a stipulation that a valve "in its receiving area has an upper and lower cylinder part spaced by a recess" requires the two pieces or cylinder parts to be separate, in the sense of not being part of, or welded together in, a single piece. If anything, this integer, read on its own, appears to me to be more suggestive of a single piece valve than a valve where the two cylinder parts referred to are separate pieces. However, that is not to suggest that the wording of this integer is anything like near enough to exclude a two piece valve from the ambit of the claim, particularly in light of what is said in the specification.

  42. The defendants are on somewhat stronger ground when one turns to integer (h), which must be considered together with integers (j) and (k). Reading these three integers together, Mr Burkill argues with obvious force that the two "guide grooves" (i.e. the upper and lower vertical grooves) mentioned in integer (h) are intended to be "mateable with" respectively, the upper pin and the lower pin described in integer (j). He says that this makes perfect sense in relation to a two piece valve: the purpose of each guide groove is to "fix" each of the two pins so as to prevent each of the two cylinders from rotating when the valve coupling is assembled, and this ties in with the adverb "longitudinally" in integer (h). However, in the case of a single piece valve, argues Mr Burkill, there is no need for any "guide groove", and, indeed, he argues that in such a case there is no such guide groove, and, indeed, the lower pin could not mate with the guide groove.

  43. A similar argument arises in relation to integer (n) which should be read with integer (m). The point made on behalf of the defendants is that the word "engaging" to be found in integers (m) and (n) is a functional one, and the function in question is to prevent the upper cylinder - in the case of integer (m) - and the lower cylinder - in the case of integer (n) - from rotating. Both those integers are appropriate for a two piece valve, because each of the two pieces needs a pin in a guide groove (or vertical groove) to prevent it rotating. However, Mr Burkill says, in the case of a single piece valve there is no need for any such engagement, and indeed there is no such engagement, so far as the lower pin is concerned.

  44. Much the same point arises in relation to integer (r), which is conveniently considered with integer (q). In either type of valve, the upper pin will slide in the recess (i.e. the latitudinal groove) between the two pieces, but only in the case of a two piece valve, runs Mr Burkill's argument, will the lower pin "[slide] along the underside of the lower cylinder part during .... rotation". As he puts it, in the case of a single piece valve "the top pin does all the work .... and the lower pin .... does not need to touch the lower cylinder at all during this rotation".

  45. In summary, the defendants say, in light of integers (h), (k), (n) and (r), Claim 1, read in its unamended form in a sensible way through the eyes of the notional addressee, envisages a two piece valve and not, contrary to the claimant's contention, a single piece valve as well. While acknowledging the force of Mr Burkill's argument on this issue, I have reached the conclusion that Claim 1 in its unamended form extends to a one piece valve, as well as to a two piece valve. If one confines oneself to the wording of Claim 1 in its unamended form, I would accept that, read as a whole, and in particular in light of the integers fixed on by Mr Burkill, it could well be understood by the notional reader as being limited to a two piece valve. However, even if one limits oneself to the wording of Claim 1 in its unamended form, I think that the notional addressee could well have understood it to extend in addition to a single piece valve.

  46. First, I am by no means convinced that expressions such as "guide grooves", "mateable with" and "detachably engaging with" have to be given the rather strict meaning which the defendants' argument ascribes to them. I accept that a "guide groove" for a pin can be envisaged as being not much wider than the pin itself and that a guide groove which is designed to be "mateable" with the pin is similarly to be only slightly wider than the pin itself. Nonetheless, given that neither of those expressions is said to have a special technical meaning, it does not appear to me to be an unduly wide reading of those terms if the "guide groove" is substantially wider than the pin. It is true that a groove can "guide" a pin in the sense of forcing the pin into a position in which it has virtually no leeway whatever. However, it can also "guide" in the sense of providing a much wider pathway, provided that pathway has limits, and the fact that those limits cannot in practice he exceeded, or even encountered by the pin due to the constraints on the movement of another part of the same piece does not prevent the wide groove providing a "guide". It is true that the guide is in that context is of little or no real practical benefit, but the "guide" nonetheless exists. Similarly, the fact that the groove is substantially wider than the pin does not prevent the pin passing or resting the groove. As a matter of ordinary language, the much greater width of groove does not appear to me to prevent the pin being "mateable with" the groove.

  47. So far as integer (n) is concerned, the requirement that the lower pin "detachably engag[es] with the lower guide groove" appears to me to raise a rather different point. In the end, however, I do not think it takes the argument further. It means what it says: the lower pin must in some way engage with or be "stuck into" the lower guide groove (or lower vertical groove). As Mr Birss points out, this may have limited value or point in relation to a single piece valve, but it is a described integer of the claim which is not stated to have any specific function.

  48. Finally, there is integer (r). On the face of it, as Mr Birss says, it seems a little difficult to understand the inclusion of this integer at all, irrespective of whether the claim extends to a two piece valve only, or to a two piece valve and single piece valve. Having postulated the problem, Mr Birss, I believe, satisfactorily answers it by explaining that one naturally concentrates on the functioning of the valve while being removed from the neck ring, whereas integer (r) is really concerned with what happens when one is inserting the valve.

  49. Secondly, much, albeit by no means all, of Mr Burkill's argument, and in particular in relation to the earlier integers he relies on, is based on the purpose of providing for a lower guide groove which is mateable with the lower pin, namely to prevent rotating of the lower cylinder in a two piece valve. As Mr Birss points out, there is nothing in Claim 1 which expressly indicates that that is the purpose, let alone the sole purpose, of the lower guide groove or of the lower pin being mateable therewith. While I accept that, in the case of a two piece valve, it is necessary to have the lower guide groove and the lower pin, and the guide groove has to be relatively tight as against the pin, it seems to me that much of Mr Burkill's argument involves reading requirements or functional purposes into the claim which have not actually been expressed in that claim.

  50. Having considered the arguments relating to Claim 1 in its unamended form, and without casting my eyes any more widely, I would summarise my views as follows. The more natural meaning of Claim 1, particularly in light of the integers upon which the defendants' argument concentrates, is that it is directed towards a two piece valve. The notional reader of the Patent would, I think, consider that some of the features of Claim 1, partly because of their inclusion in the claim in the first place, and partly because of the way in which they were expressed, indicated that the draftsman of the claim was more likely to have in mind a two piece valve, rather than a single piece (or indeed a multi-piece) valve. However, I do not consider that it can fairly be said that Claim 1 in its unamended form would clearly and unambiguously have indicated to the reader that it was directed solely towards a two piece valve. In other words, the language of Claim 1 in its unamended form is capable, without any real violence to the language or artificiality, of applying to a single piece valve, although I take the view that the more natural reading of Claim 1 in its unamended form, would have indicated a two piece valve.

  51. Both common sense and authority indicate that Claim 1 cannot be read in isolation. It must be read in the context of the very document in which it is found, namely the Patent. I have already considered the various provisions of the specification, including the figures, which bear on this issue. In my judgment, the natural reading of the specification would have pretty clearly indicated to the reader that the claimed invention is intended to apply to a single piece (indeed to a multi piece) valve which has the various functions and attributes described in the specification, as well as to a two piece valve.

  52. In these circumstances, I reach the conclusion that, on its true construction, particularly when one takes into account the specification and commercial common sense, Claim 1 in its unamended for extends to single piece valves as well as two piece valves.

    CONSTRUCTION OF THE AMENDED PATENT

  53. Turning to the amended patent, it appears to me that there are two areas which have to be canvassed. The first area concerns points raised by the defendants to the effect that the claimant's proposed amendments to Claim 1 are unsatisfactory in various ways. The second area is the overall meaning of the amended Claim 1. I propose to consider those two areas in turn.

  54. I turn first to integer (c), which the claimants propose to add. First, the defendants contend that it is unclear which of the component parts are to be "integral" or for what purpose. I do not agree. To my mind, once one appreciates that the claim relates to a valve of a specific design which is to fit into a neck ring which has two pins, but is otherwise of a standard design, it seems clear that what one is being told is that the valve is a single piece. That view is supported both by consideration of figure 1 of the Patent, which identifies a two piece valve (as well as illustrating to other pieces, namely the neck ring and circlip). It is also supported by the passage I have already quoted on page 8 lines 16 to 23. Although Mr Burkill contends that this passage describes only one design of valve, I take a different view, as already indicated. The opening words of the passage suggest that more than one design is contemplated, and that is reinforced by the important word "however" at line 20. As I have also mentioned, it appears to me that the idea that a single piece valve is being described in the first sentence is supported by the natural meaning of the words "integrally formed", a conclusion reinforced by the obvious meaning of the same expression a few lines later on the same page.

  55. The defendants also make a number of criticisms of integer (s). First, it is said that the claim is a product claim, whereas the integer appears to be drafted as if it were a method or process claim. As Mr Burkill accepts, this is not, on its own, a valid criticism of the integer. The criticism is that the integer does not put forward any standard by which one judges whether it is satisfied. In other words, the defendants are effectively contending that the amended Claim 1 is insufficient in this connection.

  56. Mr Burkill seeks to make good his insufficiency argument in relation to integer (s) by reference to three examples.

    • The first is that it is unclear whether the valve would be within the claim if "a sufficiently large opening" was formed, even if it thereafter re-sealed.

    • Secondly, he points out that, on the evidence, whether resealing occurs depends on a number of factors, including the level of pressure within the keg, and in this connection he points to the evidence which shows that, while the pressure within a beer keg is normally around 2.5 bar, the tests which are in some way or another in evidence in this case show that testing may occur up to 4 bar, 5 or 6 bar, or even 9.5 bar, on the basis that such a high pressure is sometimes, albeit very rarely, found in kegs.

    • Thirdly, he says that, if the claim is that resealing is prevented, it is unclear from the integer whether it is to be prevented every occasion, on most occasions, or on some occasions.

  57. While these points are not without force, I have come to the conclusion that they do not render integer (s) insufficient or unsatisfactory. Although the word "suitable" is not to be found before the word "for" in integer (s), it seems to me that one starts with a passage at paragraph 6.53 of Terrell on the Law of Patents, 15th Edition:

    A claim to an article "suitable for" a particular purpose is a claim to such an article, whatever its intended purpose, so that the actual use which is intended need not be shown. In Buehler v Satake [1997] RPC 232, the claim was to a "roller mill for the milling of cereals or the like". It was common ground that this was a limitation so that a device was only covered if it was capable of milling cereals or the like. A dispute as to how capable it had to be was resolved on the basis that the claim covered a mill if it could be shown to be suitable for milling cereal without alteration to make it so.

  58. While that passage does not in terms deal with the point that no quantitative guidance is given in the amended Claim 1, either as to pressure at which resealing should be prevented or the reliability with which resealing should be prevented, it appears to me that the general thrust of that passage emphasises that, as with virtually any product claim, one is concerned with an alleged invention in the practical, commercial world. Almost inevitably, unless the claim gives quantitative guidance, there will be "fuzzy edges" or "puzzles" which are "at the edge of the claim". In such a case, the patentee, as I see it, has two alternatives. The first is to incorporate quantitative requirements, which in a case such as this would be possible but somewhat arbitrary. Thus, it would be unrealistic, and, indeed, unfair to the patentee, to expect him to claim that the device would work on every occasion, but once he departs from 100%, there is no particular reason for adopting a particular lower percentage. Equally, it would be a matter almost of taste as to what minimum and maximum pressure to insert into the claim. The alternative course, which was taken by the claimant, is to phrase the claim in quantitative terms. At least in a case such as this, as I see it, there can be no real doubt but that the notional addressee would appreciate that the aspect described in integer (s) was intended to function in a keg where the pressure was between 1 and 2.5 bar, and little doubt that it was also intended to function if the pressure was between 2.5 and 4 bar. There is plainly room for more argument once the pressure gets above 4 bar, but I do not need to decide, and therefore do not propose to decide at what point above 4 bar there could be sufficient uncertainty to be within the "puzzle" or "fuzzy" area of the claim. In such a case, if an alleged infringement is only within the fuzzy area, it will not infringe.

  59. It appears to me that the degree of reliability required for integer (s) to be satisfied is also somewhat flexible. As I see it, it is plainly right that the claim does not require the lower pin to prevent resealing on every occasion. Provided that it is capable of preventing resealing, and, provided that it achieves this end on more than the odd "one-off" occasion, it will be within integer (s).

  60. The final criticism advanced on behalf of the defendants in relation to integer (s) is that the reference to "the first dismantling step" is unclear. That appears to me to be a bad point. While I accept that, if one confines oneself to Claim 1 on its own, there could be argument as to what precisely constitutes "the first dismantling step", it seems to me that the notional addressee, reading Claim 1, would work out what it meant. Even if that is wrong, it is clear, to my mind, that Claim 1 cannot be read on its own: like any provision in a document, it can, indeed must, be read in its context. That context includes the specification, and, on page 4 in the description of figure 5, in the description as to how the valve works on page 7, the addressee would, to my mind, be in no doubt as to what constitutes "the first dismantling step".

  61. In these circumstances, I consider that the two integers which are proposed to be added to Claim 1 pursuant to the amendment are clear and not, of themselves, open to criticism.

  62. I now turn to the second area of construction issues relating to the amended Claim 1, namely the consequences of the amendments on the meaning of Claim 1. There are two apparent consequences. The first apparent consequence arises from the new integer (c) which makes it clear that the valve claimed is a single piece valve, and that two piece or multi-piece valves are excluded. The second effect arises from integer (s), which explains the purpose of the lower pin, namely to prevent resealing, and therefore to enable venting or depressurising to occur, at the end of the first dismantling step. I do not consider that the meaning or effect of these integers is called into question by any of the integers of Claim 1 in its original form (all of which are included in the amended claim). Indeed, I do not understand Mr Burkill to suggest otherwise, at any rate with any force. The concern or puzzlement which the notional addressee might have about some of the integers, and in particular the reference to the lower guide groove in integer (h), the lower pin being mateable in integer (k), the lower pin detachably engaging in integer (n) and the lower pin sliding along the underside of the lower cylinder in integer (r) would all apply to the amended Claim 1 as well.

  63. In one way, the concern of the notional addressee in relation to these integers would be less strong in relation to the amended Claim 1. That is because, in light of the two new integers, he would know that, whatever concerns he had about the meaning or consequences of the other integers, he was plainly being told that the valve was a single piece and the purpose of the lower pin could not have been in doubt. On the other hand, his concern might be greater, because, at least in relation to the claim in its present form, the addressee could see that the integers which would be somewhat hard to explain in relation to a single piece valve would have some point in relation to a two piece valve. Given my conclusion that Claim 1 in its present form extends to a single piece valve, I do not see any problems as to the meaning of the original integers in the amended Claim 1. Indeed, it is fair to record that Mr Burkill did not suggest that the meaning of the original integers in Claim 1 in its original form changed in the amended Claim 1, if I concluded that the original Claim 1 extended to a single piece valve.

    CONTRIBUTION TO ART

  64. Particularly in light of the proposed amendments, it is, I think, appropriate to turn to the contribution to the art (or, to put it another way, the inventive concept) which is claimed for the Patent. The claimant contends that the contribution to the art disclosed by the Patent is to be found in integer (s) which it seeks to add by amendment at the end of Claim 1. In other words, the claimant's case is that the disclosure of the Patent is that a valve constructed in accordance with its teaching will ensure that at the end of the "first dismantling step", the pressure in the keg will have been released so that thereafter the second and third dismantling steps can be taken with the reasonable certainty of the pressure in the keg having been released at the end of the first dismantling step.

  65. At the end of the first dismantling step, the upper pin will be kept in place so long as the pressure in the keg is pushing the valve out of the neck ring, and therefore, as it were, discouraging the rotation of the valve in the neck ring due to the upwards pressure on the upper pin from the underside of the longitudinal groove in the valve neck. The function of the lower pin at that stage is to prevent self sealing (and consequential removal of this pressure on the upper pin) by virtue of the rubber gasket accidentally resealing during this process. The lower pin effectively keeps the gasket from resealing by ensuring that there is a hole through which the pressure can escape, particularly at the end of the first dismantling step.

  66. The claimant contends that, while a valve constructed in accordance with the teaching of the Patent cannot be guaranteed to achieve this end, it is very likely to do so, and that, in effect, it will be only if the person seeking to remove the valve is determined to ensure that the pressure in the keg is not all substantially released before the second dismantling step is started that a valve designed in accordance with the teaching of the Patent would not achieve the desired result. The claimant contends that, although RS-SOS has a very similar design, and that the lower pin may well sometimes do the job claimed for the lower pin in integer (s) of Claim 1, it is far less reliable. To support its contention on this issue, the claimant relies not only on common sense, but also on experiments. So far as common sense is concerned, the claimant's case is that the existence of a slanted, as opposed to a horizontal, latitudinal groove renders the positioning and/or effectiveness of the lower pin less consistent and reliable than in the case of a strictly horizontal latitudinal groove.

    INFRINGEMENT

  67. So far as infringement is concerned, there are really two questions. The first is the issue of whether the DSI valve infringes the Patent. The second question is whether each of the two defendants is liable for infringement. As to the first question, I have already considered the various arguments which Mr Burkill advances to support his contention that the DSI valve does not infringe the Patent when determining the issues of construction. As I hope is clear from my discussion on the principal issue of construction which divides the parties, he has forceful case for contending that, as a matter of construction, various integers of Claim 1 do not apply to the DSI valve. However, essentially for the reasons which persuaded me that the Patent does extend to a single piece valve, I am of the view that the DSI valve does infringe the Patent. To set out my reasons for this conclusion would involve repeating the various contentions on the issue of construction, and therefore I do not propose to say more on this aspect. I turn, therefore, to the question of the liability of each of the two defendants.

  68. The action for infringement arises out of a contract under which a brewer, Carlsberg-Tetley, were supplied with kegs fitted with the DSI valves. DSUK admits in its defence that it made the offer to supply those valves, and that the offer was made on behalf of the defendant. I see no reason for doubting that that is indeed the case. On that basis, if there is no other defence to the claim for infringement, it appears to me that each of the two defendants infringe.

  69. On behalf of the two defendants, Mr Burkill contends that, while the two defendants "admit that they acted in concert with each other", he says that "neither is alleged to be the primary tortfeasor". It does not seem to me that there is anything in that point. An offer to supply imported infringing material in this country by A on behalf of B appears to me to involve the inevitable conclusion that A and B are liable for infringement as against the patentee.

  70. Additionally, Mr Burkill contends that, by discontinuing against each of the other six defendants, the claimant has released the two defendants, because of the principle that a release of one jointly liable person must carry with it the release of those who are jointly liable with him. This was not a point which was developed at any length, but I do not think there is anything in it. While it is true that the claimant alleged that all of the original eight defendants were jointly liable, the claimant does not contend in these proceedings that any of the other six defendants were in fact liable, and the two defendants have not put forward any evidence to suggest that any of the other six defendants were liable. Accordingly, there is simply no evidence upon which I can conclude that any of the defendants other than the two defendants were in fact guilty of infringement. Quite apart from this, discontinuance is not the same as release. It appears to me that the fact that the claimant has issued proceedings against one defendant but has then discontinued against that defendant is no more significant in the present context than if the claimant had not begun proceedings against that defendant in the first place. If the claimant had not begun proceedings against the six defendants at all, I do not see how it could be argued on behalf of the two defendants that they were thereby released because the claimant had released any of the other six defendants. The claimant would merely have elected to proceed against two of several jointly liable parties, which is undoubtedly a course open to a claimant in such circumstances.

  71. Accordingly, subject to of validity and amendment, it appears to me that each of the two defendants has infringed the Patent.

    EXPERIMENTS

  72. The claimant applied to put in evidence of an experiment very late, but I decided that, provided that the defendants were given a proper opportunity to attend repeat experiments, and a right to apply for an adjournment if it was reasonably required, it would be wrong to shut out that evidence. The experiment was effectively designed to test the reliability of the RS-SOS valve. In the presence of representatives of the claimant and of the defendants, ten runs were carried out. They all involved inserting one of twenty RS-SOS valves into a neck ring in a pressurised keg, and fixing in the valve with a circlip. Thereafter, the circlip was removed, whereafter the valve was rotated. In each experiment, the parties tested whether the pressure reduced by escape of air ("venting"), by listening. They tested at four stages, namely when the circlip was removed, and at three stages of rotation, namely after about 45, 190 and 275 degree respectively. Out of the ten experiments, the records show that, at the first stage (removal of the circlip), there was no venting on five runs, slight venting on two runs, and significant venting on three runs. On all ten runs, there was significant venting at 45 degree; the results at 190 degree and 275 degree were fairly inconsistent.

  73. The claimant contends that this experiment should effectively be taken at face value, and that it indicates that the RS-SOS device cannot be relied upon to ensure that venting takes place at the end of the first dismantling step. In so far as there were a significant number of runs upon which venting did occur during the first dismantling step, that is said either to be because resealing was not occurring or because the lower pin was doing the job. However, says the claimant, the number of runs where venting did not take place during the end of the first dismantling step indicates that the lower pin cannot be relied upon to achieve the desired result in the case of the RS-SOS valve.

  74. The defendants attack the reliability of the experiment. 

    • First, they contend that, with the exception of one of the runs, namely run 5, the circlip was not removed in the way in which it would have been removed in the real world, and on that occasion there was immediate venting, i.e. venting before the second dismantling step. 

    • Secondly, as I have mentioned, the defendants suggest that there was or would have been venting in all runs, which the person conducting the experiment, Mr Winther, was able to mask or prevent by turning the extractor to a position where resealing could occur. 

    • Thirdly, on the runs where resealing did not occur, it would appear that the lower pin was doing its job, and consequently the RS-SOS valve is within the claim.

    I shall pass over that third point, which is not in itself a criticism of the experiments.

  75. Although it is right to mention that I have doubts as to the evidential value of the experiment (as to which more below), I do not consider that these particular criticisms of the way in which the experiment was carried out are well-founded. As he acknowledges, Mr Burkill's second criticism involves impugning Mr Winther's honesty. At an early stage of the trial, it appeared that there was considerable reason to doubt the good faith of Mr Winther, and possibly of the claimant. The defendants' solicitors, and indeed the Court, had been informed that Mr Winther was very ill, and would therefore be unable to attend to give evidence. However, a telephone call from the defendants' solicitors established that Mr Winther was in fact at work and apparently in reasonable health. I am satisfied that there was a genuine misunderstanding, in that Mr Winther did fall unexpectedly ill, and, after making a speedy recovery, did not inform the claimant's solicitors that he might be available. He was cross examined on video link, and it was not established to my satisfaction that he was dishonest. I consider that the experiment was performed by Mr Winther in good faith, and, indeed, I do not accept that the two criticisms I have mentioned have any force.

  76. There is nothing to suggest that the slight difference between run 5 and the other nine runs was significant in terms of what was being investigated in the experiment. In particular, it is to be noted that the immediate venting which was observed on run 5 also occurred on a number of the other runs. Further, there is nothing to suggest that the precise way in which the valve was removed was of any significance. Indeed, cross examination of the defendants' own expert witness, Mr Doelman, established that, in 1984, he had carried out a similar experiment himself on the DS-SOS valve, with apparently fairly similar results. He saw nothing significant or unrealistic about the way in which the nine runs were carried out and the way in which a valve might be removed from a keg in the real world. In particular, the suggestion that Mr Winther somehow "pushed down" the valve to mask any apparent venting, as was suggested in the cross-examinations of Mr Partington and Mr Winther himself, faces difficulties in light of the evidence of Mr Doelman. Although he did express the view that he thought that Mr Winther may have been exerting a downward pressure on the valve when appearing to turn it, and that it may have made a difference, Mr Doelman said "I do not think it [sc. the difference] was significant".

  77. Two other smaller criticisms of the experiment are also advanced. First, there is the point that the experiment was effected in an unrealistic way because in each run Mr Winther ensured that the rotation continued past the safety detent. The safety detent is not a feature of the contribution to the art; anyway, pushing past the detent is something which, according to the evidence, an unauthorised and uniformed person might seek to do. Secondly, it is said that, because the experiment involved having an artificially induced dry air pressure in the keg, rather than pressure due to the presence of beer, the conditions were not typical, and in particular the gasket would act differently with regard to self-sealing. As Mr Birss says, the evidence suggested that the relative dryness of the gasket could go either way, but, given that there was clear evidence that resealing occurred in practice (i.e. with beer in the keg) and the experiment showed that self-sealing occurred during the runs, any difference is irrelevant: the only question is whether the particular valve, and in particular its lower pin, prevents resealing. I note that Mr Doelman carried out his tests in 1984 on a dry pressurised keg as well as a wet keg.

    LACK OF NOVELTY

  78. The defendants' contention that the Patent should be revoked on the grounds of anticipation appears to me to involve two issues. The first is an issue of construction, namely whether the Patent (in its present form or as proposed to be amended), requires a strictly horizontal latitudinal groove, or whether a deviation, and in particular a small deviation, such as 3 degree from the horizontal, would nonetheless be comprehended within the claim. Unless Claim 1 requires a strictly horizontal latitudinal groove, then the defendants argue that it would suffer from lack of novelty, because it would extend to the RS-SOS valve, which was clearly a prior use. Secondly, even if the Patent is limited to a valve with a strictly horizontal latitudinal groove, it is still contended that the disclosure involved in the RS-SOS valve anticipated it. I shall deal with these two alleged allegations of anticipation in turn.

    Does the Patent as currently drafted extend to the RS-SOS valve?

  79. As Mr Birss accepts, there is no express specific reference in the Patent, and in particular there is no express specific reference in Claim 1, limiting the claim or disclosure to a valve whose latitudinal groove is strictly horizontal, as opposed to helical. In support of his contention that the latitudinal groove must be horizontal, Mr Birss relies on two main factors.

    • First, he says that it was common ground between the expert witnesses, Mr Partington for the claimant and Mr Doelman for the defendants, that the Patent was "only concerned with devices with a horizontal channel".

    • Secondly, he says that a device, such as the RS-SOS valve, with a helical groove would not have an "upper and lower cylinder part spaced by a recess", i.e. integer (g): the argument is that only a device with a strictly horizontal groove would have that feature.

  80. I am not persuaded by either of these arguments. Expert evidence does not appear to me to be of central significance on a question of construction. It is for the Court to decide what a patent means, and, save where it is claimed or agreed that one or more of the words in the Patent has a special technical meaning (not the case here) evidence as to what the Patent means is strictly inadmissible. However, because the Court is construing a patent through the eyes of the notionally skilled person, it can sometimes be helpful for the Court to hear what the expert witnesses have to say about construction, but it is fair to say that this can often be more dangerous than helpful. In the present case, I accept that it is possible that, while it would be obvious to a person with engineering experience that the "recess" described in integer (g) would be strictly horizontal, that might well not be apparent to a person with no engineering experience (such as most lawyers or judges).

  81. I turn, then, to the evidence upon which Mr Birss relies. In the passage relied on in Mr Doelman's evidence he was comparing the disclosure of the Patent with an earlier disclosure. In that passage, he said that he understood that the Patent "describes a recess formed at the joining of two cylinder parts" i.e. a two piece valve. He went on to say "I understood this recess would stretch all the way around". It is in that context that he then stated that "the recess in the Patent appears to be horizontal". I do not regard this as supporting what Mr Birss contends. The whole passage proceeds on the basis that the Patent only extends to a two piece valve, in which case, for practical reasons, a strictly horizontal channel would be inevitable or at any rate much more likely. It would be expected to go all the way around the valve, i.e. it would span 360 degree, and would therefore have to be horizontal. As I see it, the claimant cannot have it both ways; if, as the claimant contends and I, albeit with some hesitation, accept, the Patent extends to a single piece valve as well as a two piece valve, then the claimant cannot contend that, because the latitudinal groove would have to be strictly horizontal in a two piece valve, it would also have to be strictly horizontal in one piece valve.

  82. Quite apart from this, Mr Doelman expressed the view that "there is no difference from an engineering point of view because .... the slope of [a slight helix] makes no difference to the principle of operation which relies on the need for rotation but not on any axial movement during rotation".

  83. As to the evidence of Mr Partington relied on by Mr Birss, he referred to "a "vertical-horizontal-vertical" groove" which is "confirmed by the drawings which describe a horizontal groove". If anything, this seems to me to highlight the difficulty faced by the argument that the latitudinal groove must be strictly horizontal. It is inconsistent for the claimant to rely on the detail of the figures in the Patent, given that it is such an essential part of the claimant's case, in connection with its contention that the Patent extends to a single piece valve, that the figures are merely illustrative. Indeed, it goes further than that. Given that the figures (with the exception of figures 2 and 3) all show a two piece valve, and not a single piece valve, it is inevitable that that latitudinal groove is horizontal because, as I have explained, it goes all the way around.

  84. As to the claimant's reliance on integer (g) I do not consider that it is of the essence of a "cylinder part" that it has an absolutely flat top. I agree with Mr Burkill's submission that "the essence of a cylinder is straight sides and a constant circular cross section" and that "two cylinders with slightly inclined tops are still two cylinders and a recess between them which is not horizontal is still a recess", at least where the variation from strictly horizontal is a matter of a few degrees. Further, it is to be noted that where the draftsman of the Patent wishes to be relatively precise, he is: contrast the use of the adverb "longitudinally" in relation to the two vertical grooves in integer (h) with the absence of any description of the "recess" in integer (g). It is also unattractive, indeed inconsistent, for the claimant to insist on such precision when considering the word "cylinder" as compared with its relaxed approach to the meaning of words such as "mateable" or "guide groove".

  85. Indeed, the inconsistency in the claimant's case may go wider than that. When challenging the defendants' reliance on the reasons for including certain integers, such as integers (k), (n) and (r), Mr Birss justifiably relies on the contention that the defendants were invoking a purpose which was not anywhere expressed in Claim 1: it could only be implied as a reason for explaining the inclusion of integers which might otherwise be relatively hard to justify. Yet, subject to his argument based on integer (g), Mr Birss has to accept that there is nothing expressed in Claim 1 which limits or requires the latitudinal groove or "recess" to be strictly horizontal in the case of a one piece valve. Again, he needs to imply that requirement. In this connection, it seems to me that his argument faces greater difficulties than those faced by Mr Burkill's argument as to the purpose of the integers he relied on for justifying the contention that the Patent only extends to two piece valves. At least Mr Burkill was able to justify his argument by reference to functional features which are almost self-evident, once one appreciates how the one piece valve and two piece valve function. However, subject to carrying out experiments which did not even occur to the claimant or its advisers until nearly the last minute in these proceedings, there does not appear to me to be anything in the Patent which provides ammunition of a similar sort for Mr Birss's case for implied requirement that the latitudinal groove or recess be strictly horizontal. As I have said, it appears, somewhat ironically, that the only basis for contending that the latitudinal groove or recess is to be strictly horizontal is if the Patent is limited to two piece valves, which, in accordance with the claimant's contention, I have held that it does not.

  86. Quite apart from considering the language used in the Patent, it appears to me, in agreement with Mr Burkill, that one also has to consider the essence of the invention as disclosed by the Patent. Whether one looks at the specification or Claim 1, it appears to be a requirement that the valve is removed in two (or possibly three) stages, the second of which involves a rotation. Again to quote from Mr Burkill's skeleton argument "whether or not there is some axial travel during the rotation is immaterial to the way the invention works".

  87. Accordingly, I conclude that the Patent, at least as originally drafted, comprehends the RS-SOS valve.

    Does the amended Patent extend to the RS-SOS valve?

  88. The claimant contends that, even if the Patent as currently drafted extends to the RS-SOS valve, it will not do so if it is amended, and in particular if Claim 1 is permitted to include integer (s). Partly relying on the experiments to which I have made reference, the claimant contends that, once it is made clear in Claim 1 that the purpose of the lower pin is to prevent resealing during or at the end of the first dismantling step, then RS-SOS does not anticipate the disclosure of the Patent, because, unlike the type of valve described in the Patent, the lower pin in the RS-SOS valve does not achieve the result described in the Patent, and in particular in integer (s) of Claim 1 as proposed to be amended.

  89. In my judgment, the purpose of including the second pin in the neck ring in connection with the RS-SOS device was to do precisely what is described in integer (s). Indeed, I believe that this was made clear when the RS-SOS valve was being marketed to potential consumers before the priority date claimed by the Patent. Before turning to the positive evidence in this connection, I find it hard to conceive what purpose the second pin in the neck ring can have had in connection with the RS-SOS device other than with a view to preventing the gasket self-sealing. However, I accept that that does not necessarily get the defendants home on this issue, given that the latitudinal groove was not horizontal. Because the latitudinal groove is helical, the position of the lower pin relative to the gasket would change as the RS-SOS valve was rotated in the neck ring, and consequently the lower pin might conceivably not prevent self-sealing at the end of the first dismantling step: the gasket might be too high relative to the lower pin at the end of the first dismantling step, and it might only be sometime during the rotation (possibly even when the upper pin was in the detent) that the lower pin would engage with the gasket.

  90. More importantly, although the documentation provided by the claimant has been sparse, some documents dating from 1986 and 1987 relating to the RS-SOS valve, which appear to me to be significant, have been produced. Thus, in a document relating to the RS-SOS device dated May 1987 and headed "Product Operating Data" one finds this:

    The keg neck has two lugs, one of which acts as a retaining lug and the other which allows full venting of the keg to occur.

    Indeed, the lower lug is referred to as a "venting lug".

  91. The document goes on to explain that it is hard to take out the valve, largely owing to the difficulty in removing the circlip. It then continues:

    Assuming persistent determination does remove the [circlip] then the extractor tube will "jump" into its venting position (figure 3). In this position the extractor tube cannot be rotated for removal until the keg is fully safe.

  92. Although figure 3 referred to in the document is not a particularly good drawing, it shows the top lug or pin in the position which it would be at the end of the first dismantling step (i.e. at the beginning of the latitudinal groove and at the bottom of the second or upper vertical groove) and the lower lug or pin partly above and partly level with the gasket. Although it was suggested on behalf of the claimant that this pin was on the wrong side of the gasket to prevent resealing, I consider that, looking at the drawing together with the extracts from the document which I have quoted, it is and would have been obvious that the purpose of the lower pin at that point is to assist in venting, and the only way it could sensibly have achieved that end would have been to interfere with the gasket so as to prevent resealing.

  93. As I have mentioned, the claimant's documents relating to the RS-SOS valve that have been made available appear to be dated after the priority date. The amount of documentation disclosed by the claimant has been surprisingly sparse. Indeed, I have considerable sympathy with the suspicion advanced on behalf of the defendants to the effect that the claimant has not given full disclosure, and therefore may well have damaging documents, possibly of a similar nature to those that I have just referred, and which originate from before the priority date. The defendant have particular grounds for their suspicion bearing in mind the fact that even the document which I have described emanated in part from the defendants rather than the claimant, and in part was prompted as a result of the defendants producing its part of the document. Nonetheless, although there are obvious grounds for suspicion, it seems to me that there is insufficient to justify concluding that the claimant has been anything other than lax in producing relevant documents in its possession.

  94. Although this document relating to the RS-SOS valve is dated after the priority date, I am quite satisfied on the evidence that similar documents existed before the priority date and were made available to consumers and potential consumers by the claimant and its salesmen. While Mr Bernt Ipsen of the claimant was slightly equivocal on the topic, it seems to me clear from the evidence that, for almost a year before the priority date the claimant was marketing the RS-SOS valve on the basis of literature identical or similar to the document which I have described, and through salesmen who understood that the purpose of the lower pin was to prevent the gasket resealing at the end of the first dismantling step, and who would have informed potential customers of this feature and its purpose. After all, any customer would have been interested in this issue partly because it would have involved a change to their keg neck rings and partly because it would have been put forward as a positive advantage of the RS-SOS device, and therefore of obvious potential benefit to the claimant in selling its product. It is clear from the evidence that the claimant employed marketing people who were technically competent.

  95. It is also clear on the evidence, to my mind, that Mr Doelman knew of the existence and function of the lower pin, and was free to discuss it, and indeed that Mr Partington was evaluating the second pin in about 1984 and was told of its function around the end of 1984. At one time, it was suggested that what Mr Doelman knew was confidential. I do not think that is maintainable. Although Mr Doelman was not prepared to commit himself on the point, Mr Ipsen accepted that Mr Doelman had "always .... been" free to discuss the RS-SOS device and how it worked.

  96. In my judgment, therefore, an important - probably the important - new feature of the RS-SOS device, and in particular the use of the lower pin in the neck ring, was to achieve precisely that which Claim 1 of the Patent, as amended claims to achieve. Of course, the fact that an earlier device, namely the RS-SOS valve was designed to achieve precisely that which the alleged invention claims to achieve, does not of itself mean that the earlier device anticipates the invention. However, in the present case, as I have already explained, the similarities between the allegedly anticipating prior use, the RS-SOS valve, and the claimed invention are very striking. The only difference is the slight departure from the strictly horizontal so far as the latitudinal groove in the RS-SOS device is concerned.

  97. In principle, I do not see why this should make a difference. What is crucial to the functioning of the lower pin, whether for the purpose of integer (s) or for the purpose described in the RS-SOS literature, is that it is located in such a position as to be able to interfere with the gasket so as to prevent resealing when the upper pin is at the bottom of the upper vertical groove, i.e. at the end of the first dismantling step. I cannot see any conceptual or practical difference between the exercise involved in locating the lower pin for this purpose in the RS-SOS device and the similar exercise in a device which is identical in all respects save that the lateral groove is strictly horizontal. On the face of it, the RS-SOS device has the advantage that, if the lower pin is just a little too high to achieve its function at the end of the first dismantling step, or even when the upper pin is in the detent it may nonetheless be able to perform its function later on in the rotation as the valve moves slightly higher in the neck ring due to the helical nature of the groove. Because the groove is horizontal in the patented device, this "second chance" for the lower pin to function is not available. I suppose it can also be said that there is a greater risk of "slippage" at the end of the first dismantling step if the lateral groove is slightly angled, as opposed to it being strictly horizontal.

    What if the Patent is limited to a strictly horizontal groove?

  98. Even if the Patent is to be read as being limited to a valve with a strictly horizontal latitudinal groove, it seems to me that the distinction between the RS-SOS device and the claimed valve, even if Claim 1 is amended as sought, is so slight as to render the RS-SOS valve novelty destroying. The only thing which appears to stand in the way of that conclusion is the experiments, to which I have referred. Although I acquit the claimant of "fixing" the experiments, I am not persuaded that they establish that RS-SOS is not novelty destroying. It seems to me quite possible that the lower pin managed to achieve precisely what it was designed to achieve in five of the ten runs carried out in the experiment: apart from run 5, four of the remaining nine runs demonstrated some venting the moment the circlip was removed. It is conceivable that there was no resealing at that point, but it appears to me more likely than not that on at least some of those occasions, the lower pin of the RS-SOS device achieved its intended aim of preventing resealing.

  99. Further, although I acquit the claimant of "fixing" the experiments, I am not persuaded that the experiments carried out were entirely reliable. As Mr Winther candidly admitted, the experiments were carried out with the aim of seeking to show resealing, and the only keg used was a keg whose neck ring was one where the claimant had already established that resealing occurred. Thus, the experiment now relied on was not merely carried out by the claimant with a view to establishing its case, but was carried out with equipment which had already been tested as being satisfactory from the claimant's point of view. In particular, the obvious reason (or at any rate, an obvious reason), as I will mention later, for the RS-SOS device not working is that the lower pin is not in the right position to prevent resealing. That could well have been a problem with the design of the neck ring in the keg used in the experiment. As I have mentioned, while a number of different RS-SOS valves were used, it was always the same keg, and therefore the same neck ring.

  100. This is certainly not to say that the claimant has acted dishonestly, but I am not persuaded that the experiments are typical of the performance of the RS-SOS valve. In this connection, there is no suggestion on the part of the claimant that any customer who has been supplied with the RS-SOS valve has complained about its efficacy. Indeed, now, more than 15 years after it was first introduced, the RS-SOS device is still on the market. The apparent absence of any complaint is particularly significant given that at least one of the main selling features, and quite possibly the only selling feature, of the RS-SOS valve over its predecessors was that the lower pin prevented resealing, and therefore substantially reduced the risk of the pressure in the keg causing the valve to be expressed therefrom violently.

  101. I also bear in mind the point which Mr Birss emphasised when considering the interpretation of the Patent, namely that it is no part of the claimant's case that any valve designed in accordance with the teaching of the Patent can always be guaranteed to prevent resealing at the end of the first dismantling step. I am prepared to accept (although I am dubious) that the use of a strictly horizontal latitudinal groove, as opposed to an almost horizontal groove as in RS-SOS, may somewhat improve the reliability of the intended functioning of the lower pin to prevent resealing, but, as I see it, the furthest the claimant can go on this issue is to say that the idea of a strictly horizontal groove is a small workshop improvement on the helical, albeit very nearly horizontal, groove in the RS-SOS device.

    Conclusion

  102. In my judgment, therefore, the defendants' case on anticipation succeeds. It seems to me that there are severe difficulties in the way of the claimant's contention that Claim 1 is not anticipated by the RS-SOS prior use. Even if Claim 1 is amended so as to incorporate integer (s), I consider that it would be anticipated by the RS-SOS prior use, albeit that it is fair to say that the experiment carried out by the claimant gives it some grounds for arguing the contrary. The same conclusion applies to Claim 6. If I am wrong on anticipation, the defendants rely on obviousness, to which I now turn.

    OBVIOUSNESS

  103. When one turns to obviousness, as Mr Burkill rightly observes, one is immediately faced with a problem, because, in relation to the unamended Claim 1, the solving of the resealing problem is not part of the claim, and is therefore irrelevant. Accordingly, when seeking to identify the inventive concept for the purpose of answering the four well known questions in Windsurfing International Inc. v Tabur Marine (Great Britain) Ltd [1985] RPC 59 at 73 to 74, there are obvious difficulties. In his evidence, Mr Ipsen of the claimant, who was responsible for the claimed invention, said that the inventive concept was the new function of the lower pin, namely to retain the lower cylinder separately from the top cylinder. If that were admissible on the issue of interpretation, it would be highly damaging to the claimant's case that the Patent relates to a single piece valve. However, that evidence is inadmissible on construction.

  104. I must identify the difference between the RS-SOS valve and the valve disclosed by the Patent (if construed as the claimant contends, namely so as to include a one piece valve and so as to limit it to a valve whose latitudinal groove is strictly horizontal). On that basis, the only difference between the unamended Claim 1 and the RS-SOS device is that the former is vertical-horizontal-vertical, to use Mr Partington's helpful shorthand, whereas the latter is vertical-helical-vertical, albeit that the helical aspect is precious close to (namely 3 degree from) the horizontal. In my view, for the reasons already given, the difference is an immaterial one, and, possibly to put the same point a slightly different way, there is no inventive step in moving from one to the other.

  105. In my view, the choice between a strictly horizontal groove or a helical groove, particularly when the helical groove varies from the horizontal by only a few degrees, represents an ordinary engineering alternative. There is no significant difference in terms of engineering function. In the case of the RS-SOS valve, it is true that there is some axial movement if the valve is rotated. However, despite a rather faint suggestion to the contrary in the evidence, I am wholly unconvinced that the axial movement would help, or would have been believed by an engineer to help, when the valve was being assembled into the neck ring. As Mr Burkill argues, the essential point is that the spear containing the valve still has to be pushed downwards so that the underside of the valve neck presses against the gasket as the final stage before the circlip is fitted. As to dismantling, rotation is involved in the second dismantling step, but I am wholly unpersuaded that a small departure from the horizontal so far as the groove is concerned would make, or would be perceived by the notional reader to make, any difference to the functioning of the valve.

  106. Indeed, as I see it, a person considering reproducing the RS-SOS valve might well decide to change it slightly with a view to the latitudinal groove being strictly horizontal, simply on the basis that it would make no difference whatever to the functioning of the valve and because it would be, according to the evidence I heard, slightly cheaper to manufacture. In other words, it seems to me quite obvious, on general engineering or manufacturing grounds, to vary the RS-SOS device so as to render the latitudinal groove which is 3 degree from horizontal to being strictly horizontal.

  107. I do not consider that the claimant's case on obviousness is significantly improved when one turns to Claim 1 as proposed to be amended, and in particular with integer (s). The purpose of the lower pin, even as to be described in integer (s) was itself part of the state of the art - see the RS-SOS device and the literature relating to it. As Mr Burkill points out, the RS-SOS valve has worked well enough to have been extensively marketed, for over 15 years; indeed it remains on the market. There apparently have been no complaints as to the way in which the RS-SOS valve worked, and one would have expected the claimant to know if there were any such complaints, because it was the claimant who marketed the valve, and there would have been no reason for the claimant not to have put forward any such complaint in evidence. Accordingly, to put it at its lowest, either the lower pin in the RS-SOS valve system performs its function as intended (i.e. to ensure venting at the end of the first dismantling step) or it performs the function at a somewhat later stage (e.g. when the upper pin was in the detent). In any event, there does not appear to have been any obvious demand for a device which superseded or improved on the RS-SOS valve.

  108. The only reason which could logically be put forward for the RS-SOS device not working as intended was that the lower pin was in the wrong position to prevent the gasket resealing when the circlip was first removed. If that happened, it would be technically obvious how to sort this out: one relocates the lower pin in the right position to achieve the desired end. Indeed, this point was put in terms to Mr Partington in cross examination, and he accepted that "a mechanical engineer would have been capable of putting the [lower] pin .... in a position where he knew how to prevent resealing". He also accepted that "if he was not satisfied that the pin achieved the objective claimed by Micromatic .... he would raise it".

  109. As I think I have already indicated, it appears to me that Mr Burkill is right to say that it is equally obvious to "flatten the helix" particularly in light of the fact that, as he puts it, the "slope is substantially horizontal anyway".

  110. Mr Birss understandably relies on the evidence of Mr Doelman to repudiate the contention that the Patent was invalid on the ground of obviousness. It will be recalled that Mr Doelman carried out experiments at the end of 1984 on the RS-SOS device, and found that it sometimes resealed at the first dismantling step and sometimes did not. His evidence was that "it seemed an unreliable method especially at higher pressures" but that he was happy to accept the RS-SOS device on the basis that, at worse, it "would be no worse than with the existing neck ring which had a single pin". On the basis that Mr Doelman was a person with more than ordinary skill in the art, Mr Birss submits that he was seeking to deal with the problem with resealing, and it did not occur to him to vary the RS-SOS device to the extent of changing the helical groove to a horizontal groove, it must be an inventive step.

  111. It is fair to say that that argument gives one food for thought. However, in the end, it does not cause me to change my view that, whether in its present or proposed amended form, Claim 1 fails on grounds of obviousness. The impression I got of Mr Doelman's state of mind in 1984 was that he was sceptical as to whether RS-SOS would work, but he was prepared to give it a go. As I have said, the evidence of the real world (as opposed the claimant's experiments) suggests that consumers were happy with the RS-SOS valve: they had no complaints. On that basis, Mr Doelman's evidence takes matters little further. While he was somewhat pessimistic about its potential effectiveness, he was prepared to try it, and the implication, particularly given that the RS-SOS device is still on the market, is that his scepticism did not appear to him to be justified by events. In those circumstances, it is not clear to me why he should have been giving his mind to the possibility of improving the RS-SOS device. Nonetheless, I cannot pretend that Mr Birss's argument based on what did not occur to Mr Doelman has no weight, but in the end it does not persuade me that the obviousness argument is wrong.

  112. Quite apart from that, Mr Birss's reliance on Mr Doelman's thought processes in 1984 does not get round the fact that I consider that modification of the RS-SOS valve so that the latitudinal groove is strictly horizontal was and would have been technically, and indeed practically, obvious to the notional reader of the Patent. In other words, it does not answer the point that it was obvious, particularly on grounds of being relatively cheaper to manufacture with no resultant disadvantage, to alter the RS-SOS valve so that the helical valve was strictly horizontal. Thus, it was an obvious modification in the sense that it "would be obvious to the unimaginative skilled addressee in the art" - see Hallen v Brabantia (UK) Ltd [1989] RPC 307 at 327 per Aldous J (upheld at [1991] RPC 195) .

  113. In these circumstances, I conclude that, whether in its present form or as proposed to be amended, if Claim 1 was not anticipated, it fails for obviousness over RS-SOS. The same applies to Claim 6.

    AMENDMENT

  114. It is now necessary to deal with the question of whether I should permit the proposed amendment to the Patent. Given my conclusion that, even in its amended form, the alleged invention of the Patent is not novel or in the alternative obvious, it seems to me plain that I ought not to allow the amendment. However, in case this matter goes further, and my conclusions on anticipation and obviousness are rejected, it is right to deal with the question of whether I would have allowed the amendment. First, have I jurisdiction to allow the amendment; secondly, should I exercise that jurisdiction; thirdly, should I grant relief under Section 63(2) of the 1977 Act.

    Jurisdiction

  115. It is common ground that the Court cannot allow an amendment if it would add matter, if it is not clear, and/or if it is not concise. On behalf of the defendants, Mr Burkill contends that the proposed amendments in the present case suffer from all three defects.

  116. So far as added matter is concerned, I believe that the issue is resolved by the conclusion that Claim 1 in its current form extends to a single piece valve. On that basis, it seems to me that the amendment does not add matter. That is self evidently correct in relation to integer (c): it serves to limit the amended claim to a single piece valve, whereas previously, at least on my view, it also extended to a two piece valve. So far as integer (s) is concerned, while it plainly does add something to Claim 1, the function of the lower pin as described in integer (s) is already included in the current specification, namely at page 7 lines 24 to 32. It should be added that, if I am wrong on the construction of the Patent, and Claim 1 does not extend to a single piece valve, then, as I understand it, Mr Birss accepts that the amendment would add matter, because it would not merely limit Claim 1 to a single piece valve: it would extend the claim to a single piece valve.

  117. As I have mentioned, Mr Burkill also contends that the proposed amendments are neither clear nor concise. When dealing with the construction and effect of integers (c) and (s) of the amended Claim 1, I have already dealt with his points. In particular, I do not have any difficulty with the meaning of "the first dismantling step". In these circumstances, but for the point that the Patent would still fail for want of novelty or obviousness, I would have considered that I have jurisdiction to permit the proposed amendment.

    Discretion

  118. I turn to the question of whether, as a matter of discretion, I should allow the amendment. In this connection Mr Burkill relies on the proposition that the purpose of the amendment is, as I would accept, to "validate" the Patent. In those circumstances (Terrell, op.cit. paragraph 9.32):

    The onus to establish that amendment should be allowed is on the Patentee and full disclosure must be made of all relevant matters. If there is a failure to disclose all the relevant matters, amendment will be refused.

  119. Mr Burkill contends that the claimant has failed in its obligation to give such disclosure, and that in those circumstances, if I had jurisdiction to permit the amendment, I should nonetheless refuse it. Essentially, the defendants' case is that the claimant knew, or must have known, about the RS-SOS prior use and the difficulties to which it gave rise in relation to the validity of the Patent, in that, as made clear during the currency of this case, albeit that it was not clear from the contents of the Patent, the contribution to the art alleged by the claimant to have been made by the Patent is the use of the lower pin to prevent resealing when dismantling. Yet, argue the defendants, the claimant has at every stage failed to do what it ought to have done in relation to the RS-SOS device. To quote from Mr Burkill's skeleton argument:

    Since the purpose of the amendment was to distinguish that prior use, the Patentee was under a duty to explain with proper candour the circumstances of that prior use and the reasons why the claim was not originally framed so as to distinguish it.

  120. Mr Burkill points out that, in its Statement or Reasons to support the application to amend (dated 28th January 2000) and in its Statement in Reply of 30th June 2000, the claimant did not explain the full effect of the RS-SOS valve, and in particular did not explain the purpose of the lower pin in that device. Further, at least in his original witness statement, Mr Ipsen of the claimant failed to explain the purpose of the second pin in the RS-SOS device.

  121. There is considerable force in these submissions. It appears to me that the claimant was either not very competent or realistic in relation to the RS-SOS device, or that the claimant appreciated the difficulty which its case might face in light of the prior use, and has therefore been guilty, if not of dishonesty, then of lack of candour in these proceedings in relation to the RS-SOS device.

  122. In favour of the conclusion that there was lack of candour are the following factors. First, the fact that it was the claimant who invented and marketed the RS-SOS valve. Secondly, having listened to and considered the evidence in this case, the relevance of the RS-SOS valve to the issues raised in the proceedings seems almost self-evident. Thirdly, it is not as if the claimant has missed only one occasion on which one would have expected it to refer to and explained the functioning of the RS-SOS device. Fourthly, one can also take into account the possible failures of the claimant on questions of disclosure (to which I will turn in a little more detail below).

  123. On the other hand, the following factors would tend to support incompetence or not properly considering the matter, as opposed to lack of candour. First, the unsatisfactory drafting of the Patent itself: this is relevant not so much because it fails to refer to the prior art and prior use, but because the claim does not include the very feature which is said to be inventive. Secondly, wisdom of hindsight can be dangerous; while it may seem obvious after a four day action that the functioning of the RS-SOS device, and in particular its lower pin, is of central significance in this case, that does not mean that it must have been obvious (indeed it does not even mean that it should have been obvious) to the claimant before the trial began. Thirdly, I must refer to Mr Ipsen, who gave evidence for the claimant in these proceedings. He has been an employee of the claimant for at least 20 years. He was closely involved at all times with the claimant, and indeed was the inventor under the Patent, and, indeed, I think of the RS-SOS device. Mr Ipsen was also responsible for the conduct of these proceedings for the claimant. He struck me as an honest person. That is not only based on his appearance in the witness box (not by any means always a reliable touchstone). It is based more upon the fact that, considering both the individual answers he gave and his testimony as a whole, his evidence was generally reasonably convincing, and was certainly not demonstrated to be dishonest. Fourthly, while the Court should certainly not shy away from finding dishonesty if so satisfied, it should nonetheless be careful before concluding that there has been dishonesty: it finding to make. I suppose that there could be circumstances in which lack of candour does not actually amount to dishonesty, and therefore the Court need not be cautious before finding lack of candour. Nonetheless, lack of candour involves some degree of moral opprobrium.

  124. If the issue was whether the claimant should have appreciated the significance of the RS-SOS device, and in particular of the function of the lower pin, then I would have no hesitation in saying that the claimant failed to satisfy what was required of it. However, it seems to me that lack of candour must go further than that. In all the circumstances, I have concluded that the claimant is not guilty of lack of candour.

  125. I mentioned that, in support of his contention that the claimant was guilty of lack of candour, Mr Burkill relied on the claimant's failure to disclose relevant documentation. In my judgment, the only point which can be established against the claimant is that it failed to give disclosure of a relevant drawing within the two-year window referred to in paragraph 9.2(2) of the Patents Court Practice Direction. However, there was a perfectly reasonable explanation for this. The drawing in question was one which had been produced outside the two year window, but which had been amended within the two year window. In my view, this amended drawing was a document which was produced within the two year window, and should therefore have been disclosed. However, there is no authority to that effect, and the fact that the claimant and its advisers reached a different conclusion does not mean that they can be fairly criticised.

  126. Furthermore, in so far as it is relevant on this issue (and I rather doubt that it is) it is right to mention that Mr Birss points out that the rather late evidence relating to the RS-SOS device did not take the two defendants by surprise. First, unknown to the claimant, the defendants had an RS-SOS device; secondly, they had the benefit of the advice of Mr Doelman, who had actually carried out experiments on the RS-SOS device in 1984; thirdly, it was the defendants who found the claimant's literature relating to the RS-SOS device. As I say, given that one is concerned with the candour of the claimant as the person seeking the amendment, I am not sure that any alleged secretiveness on the part of other parties or the fact that they were not taken by surprise by late information is really relevant. If it is relevant, then it seems to me that Mr Birss is entitled to rely on these points.

    Section 62(3) of the 1977 Act

  127. Accordingly, were it not for the fact that I consider that, even as amended, the Patent is invalid on the ground of lack of novelty or alternatively on the ground of obviousness, I would have given the claimant leave to amend as it seeks. However, I would not have permitted the claimant to seek damages from the defendants for any past infringement, in light of Section 62(3) of the 1977 Act. This section, of course, entitles a patentee to claim damages for infringement of an amended patent on the basis of the amendment and retrospective to the amendment only if the Court is satisfied that the patent has been drafted in good faith and with reasonable skill and knowledge. Essentially for the reasons I have already considered, it appears to me that the Patent was drafted in good faith. However, I do not consider that it was drafted with reasonable skill and knowledge. The claimant has taken what I regard as the proper course in this connection, in calling the person who gave instructions to the patent agent, Mr Ipsen, and the patent agent, Mr Edvard Holme, who drafted the patent in question.

  128. In my view, assuming in the claimant's favour that the Patent was indeed intended to concentrate on the function of the lower pin in preventing resealing, then it cannot be said that it was drafted with "reasonable skill and knowledge". The only function of the lower pin is now said to be to prevent resealing: there is nothing whatever about that in any of the claims in the Patent. The more complicated and novel a particular invention, the easier it is to understand and forgive an error. Although it is fair to say that I did not fully understand the inventive concept of the Patent, until I heard an oral explanation from Mr Birss in opening, that does not cause me to conclude that this was not a simple case. First, the case was made considerably more difficult by the time it came to me, because of the unsatisfactory drafting of the Patent. Secondly, as I have said, it is one thing to read about a mechanical invention: it is much more satisfactory to have an oral explanation with a sample of the product in one's hand. Thirdly, the patent agent was being instructed by Mr Ipsen and he would have had considerably more experience of mechanical inventions than I have had. Fourthly, the essential point is a simple one. The patent agent, Mr Holme, in the present case said that he believed that the Patent did not require amendment to explain the functioning of the lower pin. I find that view surprising, but I acquit him of dishonesty: having drafted the Patent, he was scarcely likely to be objective.

  129. It is particularly striking that there is no mention of the RS-SOS device, given that there is reference in the Patent to prior use, namely the old threaded joint system, without any pins or detent. Either Mr Holme was told about the RS-SOS device when he was instructed to draft the Patent, or he was not. If he was not, then he ought to have been, and if he was, then it is even harder to understand why the Patent is drafted as it is. Mr Birss makes the point that Mr Burkill did not expressly put to Mr Ipsen that he should have told the patent agent about the RS-SOS valve, but I do not think that alters the thrust of the point. Mr Birss suggests that the Court is precluded from concluding that the Patent in this case was drafted without "reasonable skill and knowledge" because the defendants who so allege have not called an independent patent agent to support this conclusion. If this were a borderline case, there might be force in that but, particularly bearing in mind that the onus under Section 63(2) is firmly on the patentee, I do not think that there is anything in that point.

  130. Accordingly, if I had permitted amendment of the Patent, I would not have allowed the claimant to seek any damages against either of the two defendants.

    CONCLUSIONS

  131. In these circumstances my conclusions are as follows:

    1. The Patent in its present form does extend to a single piece valve, as the claimant contends;

    2. The defendants have both infringed the Patent by marketing the DSI valve;

    3. However, the Patent as drafted is invalid on the ground of anticipation through prior use of the RS-SOS valve;

    4. If the Patent is not invalid on the ground of anticipation, then it is invalid on the ground of obviousness over the RS-SOS valve;

    5. The proposed amendment to the Patent does not save it from invalidity;

    6. However, if the proposed amendments would have saved the Patent, I would have allowed the claimant to amend the Patent as sought;

    7. However, even if the Patent was amended as sought and was valid, I would not have permitted the claimant to seek damages from the defendants;

  132. In the event, I believe that it follows that the various claims raised by the claimant should be dismissed, and the Patent should be revoked.


Cases

Windsurfing International Inc. v Tabur Marine (Great Britain) Ltd [1985] RPC 59; Hallen v Brabantia (UK) Ltd [1989] RPC 307

Legislations

Patents Act 1977, s.62(3)

Patents Court Practice Direction, para 9.2(2)

Authors and other references

Terrell on the Law of Patents, 15th Edition

Representations

Colin Birss for the claimant (instructed by Messrs. Bird & Bird)
Guy Burkill for the defendants (instructed by Messrs. Freshfields Bruckhaus Deringer)


all rights reserved