England and Wales High Court (Patents Court) Decisions

BETWEEN

Hearing date: 25-28 November 1997

I direct pursuant to RSC Order 68 rule 1 that no official shorthand note shall be take of this judgment and that copies of this version as handed down may be treated as authentic.

1. Judgment Date: 18th December 1997

1. This is an action for infringement of UK patent 2 110 157 in respect of an invention which is entitled 'Manufacturing a series of curved moulded objects from a thermally mouldable material'. In fact, the curved moulded objects concerned are principally mudguards for vehicles varying in size from bicycles to lorries. The lorry mudguards have been referred to in the evidence as 'fenders'. The defendant ('Jonesco') is alleged to have infringed the patent by making a machine and using that machine to make fenders.

2. Jonesco denies infringement and alleges that the patent is invalid for various reasons. At the hearing, Mr Fysh QC accepted that claim 1 of the patent was invalid, but said that claim 5 was independently valid. The plaintiff ('Parlok') applies to amend claim 1 in a manner which I shall discuss in more detail below. Parlok also applies to add a new independent claim (new claim 6) which had no counterpart in the specification before amendment, and which it alleges is infringed. The amendments are not suggested by Jonesco to offend against the provisions of section 76 of the Act, but it was said that the discretion of the court should not be exercised to permit them. By the time of the plaintiff's reply, the only objection taken was in relation to the addition of new claim 6.

3. Accordingly, the issues at trial were infringement of claim 5 (or claim 1 as proposed to be amended); validity of claim 5; validity of new claim 1; validity of new claim 6; and whether the amendments should be allowed in the exercise of the court's discretion.

4. I was given an opportunity on the first day of the trial to inspect a machine substantially the same as the allegedly infringing machine in operation at Lille. I found this inspection very helpful.

5. Before the priority date, it seems that fenders for lorries were made by a variety of discontinuous processes. Mudguards for bicycles, however, had been made by continuous extrusion and forming (to use a neutral term) of plastic. The patent makes no distinction between different sizes of mudguards, and the only mudguards actually referred to are bicycle mudguards (page 1 line 5), so the prior art methods of making bicycle mudguards are relevant. The addressee of the specification (the 'person who is likely to have a practical interest in the subject matter of the invention') is assumed to have a reasonable practical knowledge of comparatively simple plastics extrusion techniques and plant.

6. It is necessary to give some explanation of plastics extrusion. All continuous extruders work in the same basic way. Plastic granules or pellets are fed into the machine in which they are passed by an Archimedean screw or screws along a heated barrel. The screw is often provided with a decreasing pitch so that the plastic is compressed as it passes along the barrel. The plastic is melted and pressurised. At the end of the barrel is an aperture, called the die, through which the molten plastic is forced under pressure. This die may have a particular shape ('profile') which may be either the intended profile of the finished product or a shape intended to provide the extrudate with a suitable profile for further manipulation.

7. If molten plastic is merely allowed to emerge from the die, it will twist and curl. It is invariably taken up by some form of driven pulleys or rollers so that it emerges from the mouth of the die in a well controlled fashion. This involves applying some tension to the extrudate. This tension stretches the extrudate in the direction of flow, and tends to align the long polymer molecules in that direction. There is a corresponding reduction, generally called 'draw down', in the transverse dimensions of the extrudate.

8. Unlike other materials, plastics soften and melt over a temperature range which may be considerable. Their properties change substantially as temperature changes. Generally, as the plastic emerges from the mouth of the die, it is still molten or very soft and mouldable, but will have some degree of elasticity. As it cools, it becomes more resilient until it ceases to be mouldable at all. Extruded plastics material which has been stretched while soft and mouldable and allowed to cool under tension has 'frozen-in' stress, and, if it is warmed to a temperature close to the original processing temperature, will tend to 'relax' and shrink back.

9. Both the expert witnesses in the case, Mr Webb for Parlok and Mr Kliene for Jonesco, had practical knowledge of plastics extrusion and the construction of extrusion machinery. Mr Webb has extensive experience from 1955 up to the present day in injection moulding and extrusion. Mr Kliene's experience seems in recent years to have been largely in the rotational moulding field, but he is well acquainted with extrusion techniques and apparatus. It is indisputable, I think, that Mr Webb has more day-to-day hands-on experience of extrusion than Mr Kliene, but both were valuable experts and were doing their best to assist the court.

10. Prior to the amendments which are sought to the claims of the patent in suit, the invention, in its widest sense, can be ascertained from page 1 of the specification lines 7 - 13:

11. The preferred embodiment of the invention describes a machine which is constructed so that the extruded web of material is wider than the drum. The tension applied by the drum is greater at the centre of the web than at the edges, with the result described as follows (page 5 line 19):

12. Obviously, what is happening is that the centre of the web is being made longer than its edges. This disparity can be compensated for if the web takes up an arc of a circle, and the radius of the edges is less than the radius of the centre. This is shown clearly in Figure 1 of the specification.

13. The die at the end of the extruder 2 of Figure 1 may be profiled (i.e. be other than rectangular): see page 5 lines 14-22, page 8 lines 15-24. In such a case the calendar rollers (rather like a mangle and used to give the web a smooth surface) are omitted, and the web passes onto the drum via guide rollers. In the described embodiment, the drum is free-running and the drive is imparted by the driven roller 7.

14. Curvature can only be imparted to the web if it passes over the drum in what the claim calls a 'mouldable condition'. By the time a point on the web has passed to the roller 7 in Figure 1, it will have cooled and set, and the resulting mudguard will have a radius 0-2% less than the radius of the drum (page 6 line 20), that is, it will have substantially the same radius as the drum.

15. It is convenient to set out claim 1 as proposed to be amended.

16. New claim 6 is dependent on new claim 1, and adds a feature not present in the claims of the specification before amendment:

17. This claim is merely the counterpart to old claim 7 (in which the web was extruded with a profile) but of course excludes profiles any more complicated than a rectangle from the scope of the monopoly.

19. Mr Hicks does not press his objections to the amendments to claim 1 and the corresponding amendments to existing claims 5 and 6, which are essentially 'deletion' amendments. The onus placed on an opponent resisting amendment is high. I conclude that Mr Hicks was right, and these amendments should be allowed.

20. On a narrow reading of the statement of objections, the defendant raised no discretionary objections to the amendment to introduce new claim 6. However, I allowed Mr Hicks to advance any reasons which he thought the facts supported for not exercising my discretion to permit this amendment.

21. Because the claim sought to be introduced is dependent on new claim 1, it is a 'validating' amendment only if new claim 1 is invalid. It is advanced for the purpose of distinguishing the invention from three prior uses, which I shall consider below, referred to in the case as 'Silvatrim', 'Cagnion' and 'Dover/Bluemels'. Although the Silvatrim prior use had been in the case since 1994, the plaintiff was unable to form any concluded view on the material available to it of the way in which it worked, the surviving apparatus being very incomplete. The Cagnion apparatus is another example of the Silvatrim, still in use at Fontainebleau. It was inspected in October 1996. The Dover/Bluemels apparatus is no longer available. The significance of these prior uses is that they seem to have used profiled extruder dies. The plaintiff was unaware of the Cagnion and Dover/Bluemels prior uses until the receipt of the draft re-amended Particulars of Objections on 25 April 1996.

22. The application to amend was decided on by the plaintiff in November 1996, and the amendment formulated by February 1997.

23. It is not suggested that the plaintiff should have been aware of the Cagnion and Dover/Bluemels uses at any earlier date with the exercise of any reasonable degree of diligence. Mr Hicks's objection is based purely on the plaintiff's delay. The period of delay is short, and he does not suggest that the defendant (or any other member of the public) has been prejudiced by the delay which has taken place.

24. I cannot discern any reasons for refusing to exercise my discretion to permit this amendment, notwithstanding its introduction into the claims of features not present in the claims prior to amendment.

26. It is convenient at this point to describe the alleged infringement. There is no drawing which is entirely satisfactory for this purpose, but I shall use the drawing called "Annexe 1"to the defendant's response to the plaintiff's Notice to Admit. (Figure 2). Figure 2 is not quite accurate, since the arc between the first forming wheel 11 and the outer tension roller is considerably shorter than the right angle shown, being only about 55º (Day 3/12C-D).

27. The extruder has a simple rectangular die. All the tension is applied to the web between the brake rollers 6 and the tension rollers 15 and 16 by the latter. The electric heater is permanently on and serves to soften the web before it passes over the first forming wheel 11. The defendant suggests that the web passes over 11 in a tight curve (rather tighter than Figure 2 suggests) which then flattens out in the approach to the adjustable rollers. This agrees with what I saw in Lille and was confirmed by Mr Webb Day 3/13A-B. This angle can just be seen in Photograph 8 in Annex 1 to the report of Mr Webb. Mr Webb stated that the radius of curvature of the web between roller 11 and the adjustable rollers was greater than the radius of curvature of the resulting mudguard (3/14E-F). The web, therefore, takes up three distinct radii of curvature after the heater 9: a radius tighter than the radius of curvature of the finished article at the roller 11: a radius greater than that of the finished article between 11 and the first roller 13, and finally a third radius corresponding to that of the finished article. For reasons which I give below, this third radius is not necessarily that of the curve tangent to the outsides of the rollers 13.

28. At the same inspection, it was clear that the valences, which had begun to form shortly after the start of the heater, were substantially completely formed as the web left the roller 11 (see the cross-examination of Mr Webb at Day 2/45E-F, 3/13C-D).

29. The adjustable forming rollers 13 are shown more clearly in Mr Webb's photograph 10. Without objection from the plaintiff, at the view I was shown what was, in effect, an experiment, as follows. At the beginning of the view, the mudguards being produced had a radius of 640mm and a span of 1280mm. Their overall length was 1930mm. The apparatus was then modified. The pull at the rollers 15 and 16 was increased from 113 to 140 rpm. The roller 17 was moved inwards by about 50mm, and the roller 8 was moved so as to increase the angle which the web made with the horizontal at the first forming wheel 11 from 9º to 30º. No other changes were made to the machine. In the result, a mudguard with a radius of 490mm and an overall length of 1555mm was produced. It had no visible imperfections in the valances.

30. The manner in which the alleged infringement produced its results was not, in the end, a matter of substantial dispute between the experts. It is quite clear that the fundamental mechanism for producing the valence on the mudguard is that described in the patent in suit, viz stretching the centre of the web by an amount greater than the edges while imparting a curve to the stretched web so that the valance was not puckered. It is equally clear that the final radius of the product is determined by the temperature of the web at different points along its travel and the tension applied to it (3/14A-C). It appears from Mr Webb's evidence that these are the crucial parameters (3/15F):

31. As it leaves the roller 11, the web is still mouldable, because although the roller 11 is cooled by its water bath, the cooling of the web is on the surface, and the cooling fan 14 is probably necessary for the web to be in a condition to be cut at 18: Webb 3/17F - 18A/B, Kliene 3/44F - 45B.

32. There was no dispute that all the features of amended claim 1 were present in the alleged infringement, save that it was asserted that there was no rotating drum in the alleged infringement which could be used for the following steps of the method:

33. This may be summarised by asking whether the allegedly infringing apparatus (see claim 9) has a rotatable drum with the characteristics necessary for the foregoing steps of the method to be carried out.

36. The specification describes the invention (page 1 lines 12) as follows.

37. The specification does not suggest that anything other than the drum is responsible for the shape taken up by the web. Where the making of the valences is described, they are made by stretching the central part of the web on the drum, so the drum is still responsible for moulding the resulting article. In claim 1, the drum has the mouldable material passed round it. The effect of the tension is inter alia to ensure close contact between the web and the drum. The material is separated at a position on the drum at which the web has cooled sufficient for it not to be mouldable.

38. I think that it is clear that the mouldable material is intended to take up a shape which broadly matches the shape of the drum - see page 6 line 10. This is consistent with the suggestions in the specification that the material is moulded by the drum. I also think that it is clear that the reason why the web is to be kept in close contact with the drum is to enable it to cool by its contact with the drum and provide an effective pull for the web - see page 6 lines 11-13, which is general description of the invention. It follows that I reject the submission that the roller 11 should be considered as the drum for the purpose of the method of the claim. As the evidence which I have summarised above makes clear, there is no point on the roller 11 at which 'the web has cooled sufficiently for it not to be mouldable'.

40. At trial, the objections of obviousness based on US 4,161,383, US 2,980,161 and FR 958,822 were not persisted in. Anticipation was alleged by the Silvatrim patent (FR 2,281,264) and its corresponding application, the 'Silvatrim', 'Cagnion' and 'Dover Bluemels' prior uses, and the publication of a brochure entitled 'Activities of the Bluemel Dover Group of Companies'.

41. There is no material difference between the disclosure of these two documents, and I shall consider only the patent.

42. Silvatrim is concerned with the manufacture of plastic mudguards for bicycles. The mudguards in question may have a metal sheet or strip embedded in them. The process is continuous. The general description of the invention (page 2 of the translation) contains the following passage:

Silvatrim Fig 1:

43. Figure 1 of the patent is set out above. A number of points should be noted. The extruder 1 uses a so-called cross-head die. In such a die, a metal strip passes directly through the die to the take-up rollers, and the plastic is extruded effectively from the side, so as to surround the strip. The item 13 is itself described (page 3 of the translation) as a conformator. Mr Webb considered that the item 13 and the rollers 3 and 4 acted to shape the strip, and that there was no suggestion that the mudguard was formed by the effect of over-stretching the centre of the web by overrunning the wheel any more than was necessary to prevent the extrudate from sagging excessively on its way to the wheel. Mr Webb considered that 13 looked like what would normally be called a calibrator, that is a shaping tool subsequent to the extrusion die. Mr Kliene did not know 'whether to trust the draftsman or not': 3/54E. It is clear from the evidence of Mr Kliene subsequent to this answer that the disclosure of this document is far from certain. I believe that there is a clear description in this document of shaping by the rollers 3 and 4, and the device 13, and no description of stretching the extrudate so as to shape it. It follows that there is no clear description of any apparatus in this document the use of which would fall within amended claim 1 of the patent in suit.

44. This is described in two witness statements from M Rovis and a statement from M Damaz. Neither of these witnesses was produced for cross-examination and their statements were admitted under hearsay notices. Remains only of the apparatus have been available for inspection. Both plaintiff and defendant have put photographs of this apparatus in evidence by notices of models and photographs. It may be observed that the apparatus appears to differ from that shown in the Silvatrim patent in that the device 13 has become four tools mounted around the top of the wheel.

45. The defendant has obtained from M Damaz a sample of bicycle mudguard which, it is alleged, was made on the Monaco apparatus. This sample was made of ABS and had no metal strip embedded in it. It was subjected to an experiment which shows, as Mr Webb accepts, that it 'was made by a process in which the centre of the web was stretched more than the edges'. It follows that there was a degree of differential stretching in the Silvatrim apparatus, and unamended claim 1 of the patent in suit is anticipated, or at the least, obvious.

46. M Damaz says that the apparatus in Monaco was used to produce mudguards both with and without an embedded metal strip. The mudguards without the embedded strip were made as follows. A generally planar web is extruded from the extruder. It passes through the shaper 3 on his diagram which produces a generally channel shaped profile corresponding to that of the wheel. He says that 'the channel shape was already produced when the plastic material strip arrived at the 4 tools'.

47. I approach evidence of a prior use which occurred more than 15 years ago with considerable reserve unless it has been tested by cross-examination. But in any event, this evidence does not suggest that any part of the profile produced by the shaper was 'stretched less than the central region of the web thus causing the or each side portion to be inclined radially inwardly with respect to the curvature of the drum' as the claim requires. Mr Webb's conclusion in his report was not challenged in cross-examination:

48. I accept this summary. The use of the Silvatrim apparatus does not fall within amended claim 1.

49. The Cagnion apparatus was inspected by both parties. The only evidence of the manner of its use before the priority date is given by M Cagnion's statement, again adduced by hearsay notice, and not cross-examined. The use of the machine relied on is use to make cellulose acetobutyrate mudguards with embedded metal strips. As in the Silvatrim apparatus, the extruded web is shaped by two shaping units before passing to the shaping wheel. The description of the machine given by M Cagnion is not entirely satisfactory. Photographs taken on the inspection show some form of tool at the top of the wheel (defendants' Notice of Models Set E). This tool is also shown in Mr Geoffrey Shaw's sketch of the apparatus (Annex 3 to Mr Webb's statement) but is not shown in M Cagnion's own sketch of the apparatus. M Cagnion says that there is differential stretching, but again there is no suggestion that the side portions of the web were inclined radially inwardly by the action of the differential stretch. There is no dispute that there is some differential stretch in the Cagnion apparatus. Accordingly I conclude that there is no anticipation of amended claim 1.

50. Again this apparatus was used for making mudguards for bicycles. The use pleaded ceased in 1970-71. Bluemels made two different types of mudguard, one without and one with an embedded metal strip. The apparatus for these mudguards was different.

51. Evidence was given by Mr Warmington under a hearsay notice and by Mr Huckvale, who gave evidence before me. Mr Warmington exhibits drawings of the different apparatus used for the different types of mudguard, as SDW4 and SDW5. He says (paragraphs 13 and 14 of his witness statement) that in the case of the plain mudguard, a 'profile was extruded generally to the required shape, to be finished and curved on a final forming wheel'. In the case of the mudguard with a metal strip, 'a flat strip was extruded which completely enveloped an aluminium foil ribbon. This flat strip was then profiled generally to the shape required in a water-cooled and lubricated calibration unit. This was then finished and curved on a final forming wheel.' At the trial, most attention was concentrated on the first of these methods.

52. Mr Warmington's evidence was that in this method the extrusion dies were made with an aperture thickness about 25% more than the final thickness of the mudguard, to allow for drawdown at the exit of the die under tension applied by the forming wheel. The forming wheel kept the web under tension and stretched it, and the plastic was still sufficiently hot to be mouldable until about 90º from the top of the wheel. Mr Warmington's evidence (paragraph 29 of his statement) is that the rollers 'pressed the web onto the take-off wheel. Usually we would use full width rollers. Sometimes we used partial width rollers which pressed only on the central portion of the web. This would encourage the sides of the web to curve round to the shape of the forming wheel.'

54. It seems to me that the effect of this evidence, if I accept it, is that the flattened extrusion was differentially stretched, and the effect of the differential stretching was that the sides of the soft, mouldable material were pulled down onto the wheel, so that they acquired their profile from the wheel.

55. Faced with this evidence given under cross-examination, the plaintiff applied to recall Mr Webb to give evidence in rebuttal. I allowed Mr Webb to be recalled, but I do not think that there was anything in the further evidence which he gave which persuaded me that Mr Huckvale's evidence should not be accepted because it was either impossible or highly unlikely that the machine operated as he remembered. I have carefully considered Mr Huckvale's evidence. He seemed to me to be a man who knew his job, and he had no axe to grind. I am not entirely surprised that he had a clear memory of the machine, particularly as it had been stimulated by the brochure, which showed him as a young man operating the machine in question, and by cross-examination. I believe that I should accept this evidence. It follows that all of the features of claim 1 as proposed to amended were present in this prior use (Mr Fysh did not suggest the place of cutting the mudguard was relevant to inventive step) and claim 1 as proposed to be amended would be obvious and invalid.

56. Claim 6 is not anticipated. The Dover/Bluemels prior use would render claim 6 obvious if it were obvious to use a simple horizontal die orifice rather than one which is profiled, and to add calendar rollers. Mr Kliene's evidence is that the use of calendar rollers is conventional where the web is to be precisely dimensioned and provided with a smooth or patterned surface (paragraph 18 of his report, which is unchallenged). I believe that if use of the profiled die is not new, it would be a workshop variation without inventive significance to use a simple rectangular die, to produce simple mudguards in the Bluemels apparatus unless there is some aspect of the Bluemels process which depends upon the presence of the profile and whose presence would induce the skilled man to consider that the apparatus was not suitable for use with a simple die and wheel. Mr Fysh pointed out that Mr Huckvale's evidence is that the profiled die was used for two reasons. First, it enabled the top surface of the mudguard to be detailed. Second, it helped the forming on the wheel, by providing the web with some memory of its extruded shape (see Day 3/32E). I have considered this carefully, but I think that use of the Dover/Bluemels apparatus, but with a simple die and wheel with a simple profile and calendar rollers cannot amount to an inventive step over the process which I have found to be prior used. Amended claim 6 is obvious and invalid.

57. Claim 8 is alleged to be independently valid. This is the substantially as described claim. I do not believe that it adds anything to claim 6, and is probably of similar scope. It is invalid.

58. Claims 1 and 6 as proposed to be amended and claim 8 are invalid. Had they been valid, they would not have been infringed. I would have permitted amendment had the claims as proposed to be amended been valid.