[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Taylor v Ishida (Europe) Ltd & Anor [2001] EWCA Civ 1092 (12 July 2001) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2001/1092.html Cite as: [2001] EWCA Civ 1092, (2001) 24(11) IPD 24070

[New search] [Printable RTF version] [Help]

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM CHANCERY DIVISION
MR JUSTICE PUMFREY

		Royal Courts of Justice Strand, London, WC2A 2LL Thursday 12th July 2001

ALDOUS LJ:

1. In the judgment handed down on 30th July 1999, Pumfrey J held European Patent (UK) No. 0165819 to be valid and infringed by two packaging machines marketed by the defendants Ishida (Europe) Ltd and Ishida Co. Ltd. There is no need to differentiate between them and I will refer to them as Ishida. Against the order made by the judge, Ishida appeal. They contend that the patent has not been infringed and in any case it is invalid as it is obvious and insufficient. The patentee, Mr Taylor, supported the judgment and reasoning of the judge on infringement and validity, but cross-appealed against the ambit of the injunction granted.
2. The Background - The patent claims priority from an Australian application filed on 20th June 1984. It is entitled "Packaging apparatus". As stated in the specification, it relates to a packaging apparatus, particularly adapted to package food products such as potato crisps. It is directed to persons who in 1984 were engaged in producing packaging machines. Such persons would know of the general background to the patent which was set out by the judge in his judgment:

"Background to the invention.

7. The invention is concerned with a machine for continuously packaging food products. A familiar example of the kind of package with which the patent is particularly concerned is the ubiquitous potato crisp packet. The package consists of a generally tubular bag which is heat sealed at each end. The tube is itself formed from a flat web of material, and the edges of the flat web are joined together with a continuous longitudinal seal. The machine that forms the bag, fills it and seals it is called a form, feed and seal or FFS machine. There are two classes of machines, horizontal (HFFS) and vertical (VFFS). The former are used for regularly shaped solid products, such as chocolate bars and the like. For loose confectionery and snack products, on the other hand, VFFS machines are used. In a VFFS machine the weighed product passes down the centre of a tubular former round which the web which will make the bag is wrapped. At this stage, the longitudinal seal is made in the web to form a cylinder. The product drops into the bottom of the web, which is sealed, and the upper seal is formed above the product, so forming the bag.

8. In the case of dense products (sweets, dried fruit, nuts and so on) there is no problem in getting the product into the tube and into the bag. With light products, such as potato crisps and other snackfoods, there is a problem getting the product into the bag in a reliable manner. Where the product is not dense, and has a propensity to float, the manufacturer must allow sufficient time, or take deliberate steps, to ensure that all the weighed product is in the bag before the top is sealed. It is most important to avoid what is called the "crisp in seal" problem. The main classes of product which exhibit this difficulty are potato crisps, tortilla-like products and various extruded products such as Hula Hoops and cheese puffs. I shall call these snackfood products.

9. By the priority date, it was known to use a technique called "stripping". Stripping is essential in VFFS machines used for snackfood products. Stripping in its simplest form consists of mechanically pinching the walls of the cylinder above the point at which the upper seal will be formed, and so to speak running the pinch down the cylinder to pick up any laggard items and push them into the region below the point at which the seal will be formed. Other methods (including devices like rammers in the filling and forming tube) have been proposed.

10. In a continuous VFFS machine, it will be understood that stripping and sealing are two essential operations. Sealing must be carried out by heated jaws of some description (nearly all bags are made without adhesive) which must remain in contact with the moving web for long enough for a reliable seal to be made. At the same time the bag must be stripped before it is sealed, and thus some method of moving a pinch faster than the bag is moving must be found, which will be capable of moving through the region where the jaws will close to form the seal, all this being done as the bag is in continuous motion. The evidence shows that a great deal of ingenuity has been expended by many manufacturers in seeking to come up with a solution to this far from straightforward mechanical problem, and it is the problem with which the patent in suit is concerned.

11. The evidence was that there were a large number of manufacturers of VFFS machines in the world in the early 1980's. The industry is competitive, and Mr Poley (Ishida's expert) and Mr James (Mr Taylor's expert) both gave evidence that so far as possible the competing manufacturers not only kept an eye on the state of the market but tried to find out as much as they could about the way in which their rivals' machines operated.

….

14. There were three basic classes of VFFS machines in use in the snack industry: reciprocating jaw types, reciprocating frame (or former) types and belt advance machines. In all three types, the web which will form the bag is run to a former round which it is shaped to form a tube, and the longitudinal seal is formed. The product is supplied through the former to come to a rest above the last seal formed. In the first type (reciprocating jaws) the former is fixed, and the seal is formed by a pair of reciprocating jaws which move in an essentially rectangular path. The jaws move towards each other, clamp the web and then move with the web for the period of time needed to form the seal. They unclamp, move away from each other and back towards their initial position. In the second class of machines, the web on the former is drawn along the former by two belts which grip the web. The belts move a suitable length of web through the jaws, and stop the movement. The jaws move in and seal, and move out. The belts move again. In the third class of machines the jaws close: the whole former with its tube moves upwards, pulling a further bag length, and permitting it to be filled; the jaws open and the whole descends to feed the new bag length through the jaws.

15. In each of these types of machine, a stripping mechanism could be combined with the reciprocating motion of the jaws or belts. Of these types of machines, the evidence was from Mr James that the most important were the reciprocating jaw types (he calls them draw-bar machines) and the intermittent belt drive with horizontal jaws (he calls them pull-belt machines).

16. Outside the snack food industry, with its requirement for verticality and stripping, there were a number of other machines. HFFS machines were very common. These commonly used a rotary movement of the jaws, as did VFFS machines for use without stripping for products other than snackfoods. These machines, of which the pleaded Rose Forgrove RF98 appears to have been an example, used jaws which rotated continuously. In a rotary jaw machine, the jaws rotate about parallel axes so that the face of each of the jaws describes a cylindrical locus. The web passes through the tangent line or lines of intersection of the cylinders and are thus sealed at regular intervals."


3. The Patent - The specification identifies in column 1 at line 6 the problem which the invention is said to solve.

"In the packaging of many products, bag material is delivered to a sealing head, in tubular form. The product to be packaged is delivered to the sealing head and is located within the bag material. The bag material is sealed transverse of its longitudinal direction of extension, and then the material is located within the bag material. Thereafter, the tube is closed, a stripping apparatus condenses the volume occupied by the product to be packaged, and the bag material again sealed, so that the packaged material is located within a discrete bag. The bag is then severed from the tubular bag material. This packaging operation is generally intermittent, that is not continuous.

It is a disadvantage of known apparatus that the sealing, closing, stripping and severing of the bag is achieved by relatively complex apparatus which increases the cost of the machine as well as rendering the machine unreliable. Additionally these known apparatus operate intermittently, and accordingly are slow."

The specification then goes on to acknowledge and distinguish between 5 prior art patents. I do not need to deal with the first as it is irrelevant to the issues in this appeal. The second, called Crawford, discloses a rotary HFFS machine with no stripping means. It therefore provides no guidance as to how to produce a VFFS machine with stripping.

4. The third, Wilson, discloses a VFFS machine without stripping. The fourth, Bartlo, also discloses a VFFS machine without stripping. It is not rotary and is of the type which the specification says has disadvantages. The last patent, referred to as Zwight, discloses a VFFS machine which is not rotary. It operates intermittently in the sense that the feed stops to allow stripping and the jaws move upwardly and downwardly and in and out.
5. After setting out the consistory clause, the specification describes two preferred forms of the invention with the aid of 5 figures. I cannot do better than adopt the judge's description of those figures.

"27. ….

Figure A: Figure 1 of the patent



This figure shows the simplest schematic embodiment of the invention. The strippers are mounted in an assembly which slides over the sealing heads and are urged outwardly by the springs. The specification says (column 4 line 29) that the heads are moved towards each other and are moved in the direction of movement of the material 12 at a velocity greater than that of the material 12. Accordingly the heads will cause stripping of the bag material 12 to thereby reduce the volume occupied by the product 11. Upon the heads 14 reaching a predetermined position, the bag material 12 is sealed.

28. It should be noted that in order to achieve the movement described, the rotational speed (angular velocity) of the shafts will have to vary: the heads must travel faster than the web before sealing, and slower than the web after sealing. This problem forms the basis for one of Ishida's allegations of insufficiency.

29. The second embodiment (Figure 2 of the patent) looks like this:

Figure B: Figure 2 of the patent



30. The basic scheme of things is still the same, but now the stripper heads are constrained to move by cam tracks. The stripper bars are at 22 and it may be seen that again they are telescopically mounted with respect to the sealing heads. Again the specification makes it clear that the strippers must move faster than the web until the sealing heads engage.

31. The nature of the sealing heads themselves gave rise to a great deal of controversy at the trial. It will be observed that nothing is said about them at all in the specification. The reason they are so important is this. The only real question on construction of the claim is what the words "spaced parallel fixed axes" mean in the claim. The problem arises since it is not in dispute that in Ishida's machines the axes of rotation of both the heads move during a cycle in equal and opposite amounts towards and away from the web. They move under the control of the software, to provide that the heads moves in generally D-shaped paths in a way which I shall discuss in detail below. This enables the strippers, which are attached to the heads, to contact the web at a suitable point for stripping and enables the heads, which have a fixed radius of rotation, thereafter both to contact the web at a suitable point after stripping and remain in contact with the web for a suitable period of time to effect the seal. Ishida's first argument is that the words of the claim mean what they say, and that the axes are spaced apart from each other and fixed in position with reference to each other. To this contention, the reply is that such a contention is manifestly wrong. Sealing is not possible otherwise than by exerting a pressure between the faces of the sealing jaws. Such a pressure cannot be exerted otherwise than permitting a degree of movement and a conventional mechanical device for this purpose is shown (although not provided with an index number and not described) in Fig 3.

Figure C: Fig 3 of the patent—the linear guidance mechanisms



Moreover, it is pointed out that any designer of sealing machines would know that a rolling contact between two cylindrical jaws, as shown in the specification, can only apply pressure and heat to any one region of the web momentarily, certainly not long enough to form a seal. This is so even if the face of the jaws are as they always are, that is, serrated and complementary, so as to provide a wavy path for the web between them and stretch and compress it so as to assist the seal. Finally, it is said that there is no suggestion that only cylindrically faced jaws should be used: if flat faces are used, then the skilled man would realise that he must provide some means of providing a period of time during which the web is sandwiched between the flat jaw faces and the jaw faces are applying a pressure and moving at the same speed as the web. This, it is said, can only conveniently be done by permitting the axis to move laterally so as to accommodate the movements of the jaws necessary to conform with the movement of the web while providing a seal."

6. Claim 1, split up into integers for convenience, is in this form:

"A stripping and sealing assembly, for packaging apparatus,

(1) said apparatus including a product delivery head and a drive assembly

to pass tubular bag material (12) past said delivery head so that product delivered from said head is located within said tubular bag material (12),

(2) said stripping and sealing assembly including a pair of opposing sealing

and stripping means (14, 40, 41, 42) located on opposite sides of said bag material (12) at a position downstream from said delivery head relative to the direction of movement of said bag material (12) through said apparatus,

(3) said sealing and stripping means (14, 40, 41, 42) being adapted to

cooperate to sealingly close portions of said bag material (12) and strip same,

(4) a first arm means (15, 45) supporting one of said sealing and stripping

means (14, 40.41, 42) and a second arm means (15,45) supporting the other sealing and stripping means (14, 40, 41, 42),

(5) a pair of generally parallel rotatably driven shafts (16.43.44) from each of

which there extends radially outwardly therefrom one of the arm means (15, 45),

(6) cutting means (27, 46) mounted so as to be adjacent the extremities of the

arm means (15, 45) and adapted to cooperate to sever said sealed portions from said bag material (12) to thereby form discrete bags of said product and wherein

(7) each sealing and stripping means (14, 40,41, 42) includes co-operating

stripper bars (22, 49, 50) which strip the bag material (12); wherein the arm means (15, 45) via said shafts are rotatably driven continuously through complete revolutions in synchronism in opposite directions about spaced parallel fixed axes generally transverse of the direction of movement of the bag material (12)

(8) so that prior to sealing said bag material (12) the sealing and stripping

means (14, 40, 41, 42), are moved along said bag material (12) to cause the stripper bars (22,49, 50), to strip same,

and wherein

(9) each arm means (45) is provided with one of the stripper bars (22, 49, 50),

and the assembly further includes pivotable supports (19, 20, 53. 55, 47, 48, 57) for each stripper bar (49, 50) pivotally mounting the stripper bars (49 50) relative to said arm means (15,45) so as to provide for relative movement therebetween."

7. The Issues - The main issue on infringement turns on the construction of integer (7): in particular whether in the Ishida machine the sealing and stripping means, which are supported on arm means, are rotatably driven about fixed axes. In the Ishida machine the axes move in equal and opposite amounts towards and away from the web, and Ishida argue that integer (7) is missing and there can be no infringement. The patentee submitted before the judge that the word "fixed" in integer (7), when properly construed, did not mean that there was no movement of the axes. Just as the House of Lords had in Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 refused to construe the word "vertical" as excluding a part which was 6^o off the true vertical, the word "fixed" should be given a practical width. The skilled reader would realise that in the embodiments described in the specification, there had to be some movement of the axes and therefore the word "fixed" when construed in context did not exclude movement of the axes as in the Ishida machine. The judge accepted the reasoning of that argument. He went on to make findings of fact as to the extent and type of movement in the Ishida machine and concluded that there was infringement.
8. This is a case where construction needs to be approached with knowledge of the alleged infringement as it is only with that in mind that the parties' submissions can be considered in context.
9. The Ishida Machines - Two Ishida machines were alleged to infringe. There is no need to differentiate between them as the issues on infringement are the same for both machines. It is agreed that they have all the integers of claim 1, except integer (7). It is therefore only necessary to describe in detail the means for and method of moving the stripping and sealing assemblies. As the judge's description was not criticised, I adopt it.

"43. In general terms the Ishida machines are VFFS machines in which the seal is effected by rotating sealing jaws. The jaws have flat faces and the faces are constrained to remain vertical and facing each other through an entire rotation by a planetary gear system. Strippers are fixed below the heads so that they engage with the film before the heads do. The film passes between the jaws at (for any particular mode of operation of the machine) a predetermined speed which is not necessarily constant. Each jaw rotates on a shaft and each shaft is mounted in a frame. The shafts may be moved together and apart under the control of a servo motor, the so-called turnbuckle motor. The shafts are geared together (their mutual movement being accommodated by couplings called Schmidt couplings clearly visible in the model) and are driven by a servo motor. The shafts rotate at a speed which varies through a rotation: they must both rotate so that the head follows the film during stripping but and so that the strippers move faster than the film during stripping. The turnbuckle motor moves the shafts so that the heads and their associate strippers move during the cycle to appropriate positions with respect to the film, and also drive the heads together during sealing (the so-called torque mode of the motor). The machine may be considered to have three servos (film drive, arm drive and turnbuckle drive) which are controlled by software."

10. It is convenient at this stage to emphasise the main differences between that arrangement and the specific embodiments described in the patent. First, the jaws have flat faces, whereas the specific embodiments have curved jaws. Second, rotary shafts are moved together and apart under the control of a servo motor by a ratchet system called a turnbuckle, whereas the shafts in the specific embodiment move under the control of the spring shown in Fig. 3 of the specification and depicted in Figure C of this judgment as "linear guidance mechanisms".
11. Ishidas machine works in 3 modes: with the strippers working (stripping on), with the strippers not working (stripping off) and at high speed (high speed mode). The judge described the movements with stripping on in the following way:

"44. The arms rotate in a generally D-like locus. ….. When stripping is switched on, the movement is very D-like. This is the most general case (Figure D).

Figure D – Stripping on



In this Figure, the movement of one jaw only is shown by the heavy line (including the dotted portions CA and BD.) The jaw shown rotates anticlockwise. If one starts from point D: the arm, driven by the arm servo, rotates from D to C, and the axis remains stationary where X_CD intersects the horizontal axis. From C to A the axis moves from X_CD to X_A [and accelerates]. The jaws move from A-F as the axis moves away from the film, from X_A to X_F. Then the jaws move together to seal, and between G and B the axes again move to keep the jaws in clamping relationship with the film. Then from B to D the jaws move towards the web [and the movement of axes decelerates]. During the whole of the cycle, the rotational speed of the jaws varies. In the parts of the cycle denoted by a thick line they move essentially at constant angular velocity, but CA and BD are regions of the cycle where there is [normally] angular acceleration [of the arm]."

12. The words in square brackets are clarifying suggestions by Mr Thorley QC counsel for Ishida. The movement with stripping off is simpler and is illustrated in Figure E.

13. From D to C the head moves anti-clockwise at a constant speed determined by the bag length, speed of production and the distance. The axis is stationary at X_CD. From C to G the speed changes (if necessary) to the speed of the web and the axis moves from X_CD to X_GB with the result that the sealing means come into contact with the web. From G to B the vertical velocity of the head is constant and equal to the web speed. The axis moves from X_GB to X_H and then back to X_GB. Sealing takes place during that period. Movement from B to D is the reverse of that from C to G.
14. As the judge held in paragraph 46 of his judgment, passive stripping occurs with stripping off. He held that with stripping on, the amount of axial movement was between 2mm and 5mm during the cycle. That was challenged by Mr Thorley upon two bases. First, the results relied on by the judge did not take account of the movement from X_CD to X_GB and X_GB to X_CD and second, they could not be derived from X8, which contained Mr Taylor's jaw path calculations as stated by the judge. The second challenge must be rejected as the figures are those calculated for representative runs. The first is correct, but not pertinent for reasons to which I will come.
15. There is no need to describe the High Speed mode as it is similar for the purposes of this case to that for stripping off.
16. Construction – There was no dispute between the parties as to the approach to construction. As section 125 of the Patent Act 1977 makes clear, the claim has to be construed through the eyes of the notional skilled person as part of the whole specification in accordance with the Protocol on Interpretation of Article 69 of the European Patent Convention. Further, both parties agreed that the approach to construction adopted by the judge was correct. But Ishida submitted that the judge had come to the wrong conclusion because he did not have adequately in mind the common general knowledge of the skilled person, had misunderstood the way the Ishida machines worked and had failed to appreciate the differences between the type of movement of the axes in the Ishida machines and that contemplated by the patent when it used the word "fixed".
17. Mr Thorley accepted that the judge had in paragraphs 19 to 20 of his judgment properly instructed himself as to the difference between common general knowledge which the notional skilled person would have in mind when reading the specification, and what was just public knowledge. He also did not criticise the judge's description of the basic types of machine nor the statement in paragraph 21 of the judgment that the notional skilled person would be aware "of the general type of machine available which I have already described, and will certainly be aware of the fact that a snackfood machine must be a VFFS machine and must incorporate stripping". According to Mr Thorley, the judge's mistake was to believe that rotary machines without stripping formed part of the common general knowledge as stated in paragraph 16 of the judgment. Also, he failed to point out that VFFS rotary machines with flat jaws did not form part of the common general knowledge.
18. I reject both criticisms. There was ample in the evidence in chief of Mr James to support the finding in paragraph 16 of the judgment and the judge had no need to say what did not form part of the common general knowledge. He sensibly made findings as to what did form part and by inference excluded matter which he did not refer to.
19. The parties agree that the acontextual meaning of the words "about … fixed axes" is that the axes will not move at all. They also agree that those words, when construed in context, cannot have that meaning. It is only necessary to look at Fig. 3 of the patent to see why that is so. There springs are shown as linear guidance mechanisms for the axes. A mechanism which allowed movement was necessary in all rotary machines. As the judge held:

"35. All rotary machines, on the evidence, permit one or both shafts a degree of movement to accommodate the fact that the surfaces of the jaws must touch through the web. The web is thin and high precision is not attainable, and so a degree of "give" must anyway be incorporated. But to apply pressure, the pitch circles of the jaws must be deliberately arranged to intersect, so that on contact one or both jaws recoils slightly. It is important not to exaggerate this effect, as the radii of curvature are comparatively large. A great deal of confusion was caused in the evidence by the suggestion that if the pitch circles marked out by the jaws intersect there will be a so-called D-shaped movement, so that the web is continuously pressed by the sealing jaws. Mr Taylor makes this suggestion in Appendix 5 to his report, where a contrast is made between a "point contact" arrangement and a spring loaded shaft. But there is no distinction as X3, put to Mr Taylor, makes clear. In fact what happens geometrically is that if the pitch circles of the jaws overlap, the jaws touch and remain just touching at the leading edge until horizontal dead centre is reached, when they roll past each other. The contact remains a rolling contact, but under higher pressure. Practically, the position is not so simple because at the leading edge the jaws are nipping the web, and beginning to form a seal, and the same happens at the trailing edge. In fact, the contact time of the web with the jaws is largely influenced by the profile of the surface of the jaws. But this is obvious, and all jaws for rotary machines appear on the evidence to have a generally wavy profile cut in them. To this extent paragraph 10.2 of Mr Taylor's report is not correct as a matter of geometrical fact. I should observe that Mr Taylor has conducted or is conducting litigation against Ishida in a number of jurisdictions. From time to time, particularly when giving evidence in relation to the disclosure of the patent, he stepped over the line and became an advocate. On the other hand, particularly when he was giving evidence in relation to the history of his development, and as to how, based on his observations, the Ishida machines operated, he was a satisfactory and reliable witness."

20. The judge went on to deal with flat jaws. He said:

"36. It is convenient to deal with the question of flat-faced jaws at this stage. Flat-faced jaws (of the kind for example used in Zwight and Crawford) cannot be used in a purely rotary machine unless they are arranged to pivot, so that the faces of the jaws are parallel and move into positive engagement with the film as their support rotates. Such a movement is difficult to achieve mechanically (Crawford gets over it with his cam track for the heads and his barrel cam for varying the speed of rotation). With flat faced jaws, sealing time is directly measured. The position is different with jaws with a cylindrical surface. In the latter case, the relevant time is the time during which the softened, or molten, web material is in the nip between the jaws. Although the times are not directly comparable, Mr Taylor gave evidence that in fact the times were comparable, and he supported this contention by reference to his sketch X4.

37. Mr Taylor said that a machine with a "kissing" contact between the jaws would not work. He was adamant, under repeated cross-examination, that as a practical matter some give was absolutely necessary, and that it would be obvious to the skilled man that he could (within limits) increase the time during which the film was subjected to a sealing pressure by increasing the overlap of the pitch circles of cylindrical-faced jaws. So far as flat jaws are concerned, obviously they required substantial overlap of the pitch circles and (equally obviously) flat jaws were used to increase sealing time. On this question I accept Mr Taylor's evidence. The arguments to the contrary seem to me to turn excessively on a purely geometrical analysis of the problem (see for example X3) in which insufficient attention has been directed to the real dimensions of the components concerned and the realities of machinery rotating rapidly."

21. Mr Thorley did not challenge the findings of fact made by the judge in those paragraphs. However, he submitted that the judge was wrong to construe the claim as covering flat jaws. Rotary machines with flat jaws did not form part of the common general knowledge and therefore would not be in the mind of the skilled person when reading the specification. That submission is unreal. Whatever may have been within the knowledge of the notional skilled person in 1984, the patent directs his attention to the Crawford and Zwight patents which had flat jaw machines. Further he would realise that the embodiments described in the specification were as Mr Taylor, the patentee, said in his witness statement – "based on another well-known form of rotary machine, namely, flat jaw machines. These machines are described in prior art references including Crawford, Klopfenstein and Gausman."
22. Having come to the conclusion that some movement of the axes was inevitable and was by implication disclosed in Fig. 3 of the specification, the judge concluded:

"38. In my view, if the axes move in a manner which is irrelevant to the basic requirements of the invention, such a movement is within the contemplation of the claim. I consider that a movement of the jaws normal to the web would be seen by the skilled man as inevitable if a seal was to be formed using cylindrical sealing faces and he would further appreciate that such a movement would be provided by the conventional spring guide arrangement shown in Figure 3 of the patent. I consider, therefore, that a movement normal to the web of the axis of rotation of one or both of the jaws the function of which is to ensure that the sealing jaws (a) follow the web as they rotate and (b) move so as to control the pressure or duration of the clamping action is within the claim. I think that this construction is really to put a sensible meaning on the word "fixed" in its context. It is certainly a variant on the acontextual meaning of the word, and it seems to me that the skilled man would both appreciate that it was a meaning which could have no effect on the way the invention worked—indeed, it is necessary for the invention to work practically—and it is certainly not excluded by the specification itself."

23. It is at this stage of the judgment that the submissions of the parties violently diverge. Mr Thorley submitted that the judge had failed to keep in mind the problem stated in the patent and the solution proposed. The object of the invention was to produce a VFFS machine with continuous rotary movement of the arms with strippers and sealers which had no active intermittent movement. The extent of any movement was not the material issue. To support his submission he referred us to the opening parts of the specification where, he submitted, the disadvantages of what had been proposed before were stated to be intermittent movement and complex apparatus. That was explained by the references to Zwight which was said to have a reciprocal motion so that the movement of the jaws was "inherently intermittent and slow". The solution was the rotary movement of the arms with strippers and sealers mounted on them and passive means (the springs) to allow movement normal to the web to the extent required. The claim should be construed to meet those objects of the invention. Thus the word "fixed" allowed passive or compliant movement, but excluded active intermittent movement as such movement introduced complex apparatus which had caused difficulties in the prior art machines. Such a construction excluded that which the invention sought to avoid.
24. Those submissions of Mr Thorley introduce difficulties which do not arise from a correct understanding of the specification. The specification draws attention to the complex and intermittent operation of the prior art machines. The invention is having a continuous rotary machine with a stripper and sealer on the same arm. In that way the whole operation required for a VFFS machine is carried out by rotary movement of arms about axes. Those axes are fixed in the sense that the movement is not that of the type described in Zwight and the other prior art machines referred to. However they do have movement to enable the jaws, which rotate, to strip and seal. Such a construction gives effect to what I perceive to have been the obvious intention of the patentee. No skilled person reading the specification would give the word "fixed" its acontextual meaning. Such a person would realise that the advantage described was the machine with continuous rotary stripping and sealing and for that to be achieved some movement was required. The extent would depend upon the design of the actual machine. The actual extent of movement that falls within the ambit of the claim is that which the notional skilled person would realise would obviously have no material effect upon the way the invention worked taken at the level of generality of claim 1. Such a construction gives fair protection to the patentee and reasonable certainty to the public.
25. I have expressed my view as to the ambit of the word "fixed" in different words to that used by the judge to meet the submissions made in this Court. But I find no fault in the conclusion reached in paragraph 38 of the judgment.
26. Infringement – I have already described the working of the Ishida machines. In view of the submissions made it is necessary to consider infringement by the machines when working in the different modes and I do so starting with stripping off.
27. The basic submission of Ishida was that their machines did not infringe as the movement of the axes was activated and controlled by a servo motor. The positive movement obtained from the servo motor was needed to obtain the desired acceleration of movement of each axis and the position of the jaws. That positive movement was necessarily an intermittent and reciprocating movement which introduced complexity which was undesirable. The word "fixed" should not be construed as encompassing such positive control as was used in the Ishida machines. That was emphasised by an analysis of the movement. One component, that between C and G in Figure E was attributable to the use of flat jaws and there was no equivalent in the patented embodiments of the movements between C to G and B to D.
28. Mr Thorley also submitted that there was no infringement because the Ishida machines had flat jaws. He accepted that there was nothing in the claims which excluded flat jaws, but submitted that the claim should be so read. That submission was hopeless. The claim covers all types of jaws. The fact that the preferred embodiment used curved jaws cannot be used to input into the claim a restriction as to shape. To do so would not give effect to the inventive concept disclosed, namely a continuous rotary stripping and sealing machine.
29. I also reject the distinction sought to be drawn between what can be considered to be the passive or reactionary action of the springs shown in Fig. 3 of the specification and the active positioning of the axes in the Ishida machines. There is no limitation in the claim requiring that any movement should be passive or reactionary as opposed to active or control by a servo motor. The difference between the spring means of the patent and the servo mechanism of Ishida is an obvious variant. Professor Limebeer, who was called by Ishida to give expert evidence, made that clear in his cross-examination. He said that pressure was necessary to permit sealing to occur and to permit38.

movement. Obvious ways of doing that were springs and servo motors. He believed that their use was undergraduate mechanics.

30. By 1984 servo motors to control movement were well-known. The notional skilled person would realise that substitution of a servo mechanism for the springs in Fig. 3 of the patent was an obvious variant. If so, I can see no reason why the word "fixed" in claim 1 should not be construed as covering such a machine. With rotary machines, the jaws must move, as must the axes as well. That movement is in one sense intermittent, but it is different to that which the specification describes as intermittent. All that the servo motor does is to anticipate the movements that would happen using springs and to provide equivalent movement.
31. The movement provided by a machine made like Fig. 3 of the patent, but with a servo motor, would not be identical to that provided by the servo motor for the Ishida machines, because of the flat jaws. That difference cannot be material. The way the invention works, considered at the level of generality of the claim, is not affected by the shape of the jaws. No doubt the means for allowing movement of the jaws and exerting pressure would have to be adapted. Thus the anticipatory movement between C to G and B to D. A servo motor to move the axes was an obvious variant of the springs of Fig. C which could not affect the way the invention worked. That would be evident to the skilled person.
32. I conclude that the judge was correct to conclude that the Ishida machines in the stripping off mode infringed the patent. As there is no marked difference between that mode and the High Speed Mode, infringement of that mode was also established.
33. Comparison between Figures D and E show the differences between stripping on and stripping off. Professor Limebeer stated in his supplemental report "the stripping on case is only a trivial variant of the stripping off situation." True, as Mr Thorley pointed out, the added movement that can be seen in Figure D of the judgment can only be provided by a servo motor, but that added movement cannot affect the way the invention worked when considered at the level of generality of the claim. When properly construed the claim enables movement of the axes to enable stripping and sealing in a rotary machine. The patentee could not have intended to exclude movement that did just that, but in a way that was possible using a servo motor. The Ishida machines have a sophistication that was not excluded from claim 1.
34. Obviousness – Although a number of patents were pleaded as rendering the invention of claim 1 obvious, at the trial Ishida only alleged that it was obvious having regard to the common general knowledge available at the priority date.
35. As the judge pointed out, the question of obviousness is a jury-type question. He said that he was swayed by the often quoted words of Lord Moulton in British Westinghouse Electric and Manufacturing Company Limited v Braulik [1910] 27 RPC 209 at 230 and those of Lord Diplock in Technograph Printed Circuits Ltd v Mills and Rockley (Electronics) Ltd [1972] RPC 346 at 362 which contain warnings against the use of hindsight. He concluded that obviousness had not been established.
36. In the light of this Court's decision in David J. Instance Ltd v Denny Bros. Printing Ltd 20th June 2001, which applied the guidance given by the House of Lords in Designers Guild Ltd v Russell Williams (Textiles) Ltd [2001] FSR 113 at 122, Mr Thorley accepted that this Court should be slow to overturn the conclusion of the judge on obviousness unless he had erred in principle. He submitted the judge had done just that. To understand that submission, it is necessary to set out in outline the case advanced by Ishida.
37. Ishida relied upon the evidence of Mr Poley. He had worked from 1977 to 1979 on VFFS machines when he was employed by Pedigree Pet Foods. He subsequently joined P.A. Consulting Group. In 1986 he was the leader of a development project commissioned by Wright Machinery from the Technical division of PA Consulting to provide a design of a VFFS machine which would be a quantum leap forward. Wright Machinery were well-known manufacturers of VFFS machines for packing such things as potato crisps. The contract was placed by Wright Machinery as they were interested in producing faster VFFS machines quickly to meet the needs of the market.
38. Over 4 months Mr Poley and his team studied what was on the market and worked to achieve the needs of Wright Machinery. Their proposal was included in a report. Mr Poley gave evidence that in the report, they arrived at a configuration for a machine for stripping and sealing which he believed had all the key features of the patent. His view, expressed in his evidence, was that the machine was obvious. That view was stated to be based on his experience at Wright Machinery.
39. Mr Thorley submitted that the judge had erred in principle in that he had failed to identify properly the inventive concept of the patent and had not taken proper account of the evidence, instead he had relied on the steps that Mr Kitchin had submitted had been taken by Mr Poley. Mr Thorley identified paragraphs 64 and 67 as containing the error in principle.

"64. Mr James, who worked for Hayssen at the material time and was apart from Mr Taylor the only witness who had experience of this packaging business from the inside did not consider that the solution of attaching the strippers to the jaws was obvious. He had himself experience of one arrangement (the Gausman 1 patent, US 2,950,588) in which flat-faced sealing jaws rotated. Gausman had a slightly later patent for a very elaborate stripper arrangement which passed between jaws of a reciprocating-jaw machine but the strippers were independently driven of the jaws which appear not to have rotated (US 2956383). Gausman 1 incidentally shows another method of accommodating the necessary movement of flat-faced sealing jaws: see Figure 7. So Mr James was conscious of the problem, and was aware of flat-faced rotary jaw machines and was aware of the need for stripping, but an arrangement within the patent in suit escaped him. The same goes for Bartlo, also employed by Hayssen and Mr Gausman's successor. He had invented the stripper mechanism disclosed in his patent which was pleaded, but he did not make the connection either."

40. Those steps were relied on by the judge when coming to his conclusion. He said:

"67. Of course, Mr Poley's report is an actual rather than a conjectural series of steps, and it is right also that Mr Poley did not start with any particular end in view. In the present case I think that the sum of the steps which I have outlined above in considering Mr Poley's report is, cumulatively, indicative of invention, albeit not of the flash of inspiration variety. But to go from the start to the end was in my judgment inventive, and my conclusion is that the claim is not invalid for obviousness over the common general knowledge."

41. I reject Mr Thorley's submission as it isolates paragraph 67 of the judgment from the rest of the judge's reasoning. He stated:

"62. The approach of Mr Poley's report is rigorously systematic, but in no sense merely mechanical. He is himself an inventor with a number of patents to his name. The development was investigated in advance of the availability of weigh feeders capable of supplying 120 bags per minute from a single head. Phase 1 of the investigation, which was a design study, was the only work done by Mr Poley. He is a good engineer introduced to a new art, and he approaches the problem which he was given with none of the preconceptions of prejudices of the workers in the art themselves. He considered that it was obvious to rotate the strippers on the jaws, and Mr James disagreed.

63. Mr Poley's report proceeds by successive refinement. He sets out with a set of logical options for phase 1. He investigates how potato crisps drop in tubes. He identifies the need to avoid high accelerations and heavy components. He looks at the physics of film sealing. The report refers to brainstorming sessions for generating innovative concepts without consideration of the prior art. The results of these sessions were presented to Wright's for discussion. The report in fact contains an excellent step by step analysis of the design of a machine. Mr Poley did not come to the idea of attaching the strippers to the arms carrying the sealing heads immediately: his figure 6.15 in the first part of the report does not contain this feature, and it was after he had investigated the locus of the jaws and the locus of the stripper blade that he concluded that selection of a rotary stripping mechanism would allow elimination of a separate strip drive motor. I should record that Mr Poley's evidence taken as a whole both in his report or under cross-examination left me with the clear impression that the invention was not obvious. Mr Kitchin identified I think seven successive distinct steps in Mr Poley's design: (1) the brainstorming sessions (2) the investigation of the drop characteristics of potato crisps (3) the production of a number of different design concepts, including a design with independently driven strippers and sealing jaws (4) selecting between the different approaches (5) investigation of the different approaches to stripping through the sealing zone, and identification of reciprocating, rotary and track strippers and sealers, but without any particularly driving means (6) analysis of kinematics of the stripping and sealing mechanisms (7) the final design. Mr Kitchin points to the fact that PA considered that at the step 6 stage they had made a significant development."

42. In my view paragraph 67 of the judgment must be read in the context of the case being advanced by Ishida, namely that Mr Poley got to the invention without knowledge of the patent and without invention; therefore the invention of claim 1 was obvious. The judge rejected and was entitled to reject the second element of that argument with the consequence that the third element failed. There was no error of principle and for that reason the appeal on obviousness fails. In any case I believe that the judge was bound to conclude that the evidence did not establish that claim 1 was obvious.
43. The parties agreed that the inventive concept was encapsulated in integer (7) of the claim. I have referred to it as continuous rotary stripping and sealing. I am prepared to accept that Mr Poley arrived at a construction which included that concept, albeit the evidence suggested that the machine proposed would not have had all the integers of the claim. However, his evidence did not establish that claim 1 was obvious for a number of reasons which I will summarise.
44. First Mr Poley was not the notional skilled addressee. His skills far exceeded those of such a person. It was for that reason that he was the person in charge of the team of consultants. Thus his views do not necessarily equate to those of the notional skilled person. Second, Wright Machinery failed to think of the Poley concept. They were well-known machinery manufacturers who would be expected to think of obvious concepts. Instead they had hired P.A. Consultancy. I accept they did this to speed up development, but if they had had the idea they would have fed it to Mr Poley for his team to consider. Third, Mr Poley's brief was to develop a new concept of machine to pack crisps as there was an opportunity to make significant advances on competitors' machines. What was wanted was a machine which was a "Quantum Leap" above current art. That Mr Poley believed he achieved. He said in cross-examination that his proposal was a new concept and was a significant advance. That being so, it is surprising that it was also obvious. Further, the project proceeded in the seven steps identified by Mr Kitchin. If it had been obvious, it was a surprisingly laborious way of moving to the obvious. Fifth, Mr Poley recommended that the machine concept should be "patented comprehensively as soon as possible". A surprising recommendation if it was obvious. Sixth, there was a demand for a new concept as the report makes clear. If it had been obvious, why had it not been proposed before by one of the packaging machine manufacturers, particularly as the patented machine has been successful? The answer may be, as Mr Thorley suggested, that other inventive steps were needed before major advances in speed could be made. But the demand and the success add weight to the conclusion that the concept that Mr Poley proposed was not obvious. Seventh, there was no evidence to support the attack of obviousness other than that of Mr Poley. He relied solely on the report. It follows that if the report does not establish an obvious route to the invention, which it does not, the attack must fail. Eighth, there is evidence from Mr James that neither he nor his employers had come up with the invention despite their interest.
45. The matters to which I have referred above point to invention rather than obviousness.
46. Insufficiency – No separate argument was advanced against the conclusions of the judge that the attack of insufficiency failed. In those circumstances the appeal on insufficiency fails.
47. The Cross-appeal – The judge granted a limited form of injunction. The parties accept that he should have granted it in the usual form and therefore the cross-appeal will be allowed.
48. Conclusion – The appeal should be dismissed and the cross-appeal allowed so as to substitute an injunction in the usual form.

HALE LJ:

49. I agree.

JONATHAN PARKER LJ:

50. I also agree.
ORDER: Appeal dismissed and cross-appeal allowed as indicated in the Judgments; permission to appeal to the House of Lords refused; counsel to lodge a draft minute of order. (Order does not form part of approved Judgment)