T 1069/00 (Controller/PRESSTEC) 31-10-2002

Reasons for the Decision

The patent proprietor's main request

1. The prior art

D4 (cf fig.1 and associated text) describes a press having multiple imaging stations according to the preamble of claim 1 of the opposed patent. In such a press an array of image spots is produced in a series of axially sequential, circumferential imaging swaths under the control of a controlling means 14. Each swath comprises a series of circumferentially spaced-apart image spots formed during one revolution of a cylinder 102, successively produced spots being circumferentially spaced from one another by a distance determined by rotation of the cylinder between successive discharges. Furthermore, angular position data specifying the location of the discharge means relative to the cylinder are provided by a magnetic detector 122 coupled to the controlling means 14 (cf column 10, lines 45 to 64). A look-up table storing the x and y coordinates (ie in the circumferential and axial directions) of all dots is used also to store x and y offsets for dots of individual colours in order to eliminate possible deviations of the colour lines from their theoretical true position (cf paragraph bridging columns 12 and 13). Such offsets are entered during a calibration step performed at the factory during the final check-out phase of press manufacture. Not explicitly mentioned is the problem that varying dimensions of the plate cylinders result in poor registration of the different colour plates.

2. The patent proprietor has argued that the invention according to claim 1 involved an inventive step when compared with D4 because, according to the invention, the storage means contained position correction data on the basis of which the intervals between discharges were varied. One aim of the invention was to provide an imaging apparatus capable of varying the circumferential size of the image applied to a plate while avoiding dimensional errors. In this way misregistration between colours could be avoided. D4 was concerned with cyclical mechanical errors (cf. D4, column 12, lines 62 to 64), which would be the same for all colours, whereas the invention provided a more general kind of correction.

3. In the Board's view, however, the kind of correction described in D4 (column 12, line 62 to column 13, line 13) corresponds closely to the one defined in claim 1. A dot position look-up table stores x and y coordinates of "all dot positions". Dot positions are corrected "for each colour". The corrections are based on "end-to-end tests" by printing test patterns and determining deviations of colour lines from their theoretical position. It appears that this correction can be applied to any type of positional error, including the "slanted swath" condition mentioned in the patent-in-suit which is "characterised by lines in the image which run in the axial direction as opposed to the circumferential direction, and which appear 'sawtoothed' or jagged instead of straight" (cf column 1, lines 44 to 48 of the patent-in-suit). In an on-press imaging system a positional correction necessarily corresponds to a variation of the intervals between discharges and will have an impact on the size of the image in the circumferential direction. Furthermore, a look-up table for all dot positions and for each colour permits mutually independent, dot-by-dot corrections.

Summing up, all features in claim 1 are covered by the disclosure of document D4. Claim 1 is therefore not allowable (Articles 52 and 54 EPC).

4. In claim 19, also directed to a press, the independence of the positional corrections is further emphasised. Position-correction data specifying circumferential offsets are stored "for different image spots within the array" and "the circumferential spacing between successive discharges can be varied independently of one another". However, for the reasons already indicated, D4 is regarded as disclosing means providing independent corrections, namely a look-up table for every dot position. Thus, this claim is also not allowable (Articles 52 and 54 EPC).

5. Claim 23 is directed to an imaging method. The method comprises in particular the steps of printing a test pattern at each station (ie for each colour), determining how much their sizes differ, calculating how much to shrink or increase the image, storing suitable offsets, and adjusting the time interval between discharges to move the discharge locations forward or backward in the circumferential direction in order to achieve images of the same size.

D4 discloses printing a test pattern and measuring the differences of deviating colour lines from their theoretical true position (see the passages already indicated). The known method and the claimed method can be expected to yield the same result. The adjustment of the time interval between discharges is believed to be known from D4 since dot positions on a rotating drum can only be modified by varying the timing of the discharges. Furthermore, Claim 23 specifies that it is the size of the test pattern which is measured (rather than some other parameter) but this does not appear inventive for the following reasons. It was a known problem in the art that the size of the images will vary due to unavoidable differences in the radius of the respective cylinders. As is stated under the heading "Discussion of the prior art" in the patent-in-suit (column 2, lines 17 to 23), "/m/anufacturing tolerances also produce variations in the dimensions (i.e. circumferences) of the printing plate cylinders. Thus, there is a likelihood that in a four-color imaging system which incorporates four separate cylinders (each which is paired with its own set of imaging devices) the four circumferences will not be the same. Accordingly, adjustments must be made... in order to produce four printing plates whose images are the same size in the circumferential direction". Also D2, a document relating to a scanning/plotting device of the rotatable drum type, mentions this problem (cf pages 2 to 9): "Thus, if the effective radius is different from ideal by length dr, the distance between pixels will vary... For an oversized drum... the result is an image which is too large; for an undersized drum the image is too small". The skilled person would expect such size errors to occur also in the press described in D4 (cf. the abstract of D4: "...the diameters of all of said plate and blanket cylinders being substantially the same...", italics added). Appropriate corrections would have to be determined and the obvious starting point would be first to determine the error in size.

Thus, the method of claim 23 does not involve an inventive step (Article 56 EPC).

The patent proprietor's auxiliary request

6. Claim 1 of the "first" auxiliary request differs essentially from the main request firstly in that the storage means (30) for storing correction data is explicitly referred to as second storage means, ie the means are separate (in some meaning) from the means (4) "for storing image information representing the image". Secondly, a result to be achieved is specified at the end of the claim: "so compensating for angular position error between said sensing means (16) and said discharge means (8)".

7. As to the first addition, the Board cannot see that the provision of separate storage means for the correction data would in this case involve inventive considerations. Nor does the description suggest that this feature would solve a particular technical problem. Furthermore, as Opponent II has pointed out, the fact that in D4 calibration is performed only once may suggest that the look-up table containing the corrections is not identical with the memory used to store images, which will have to be rewritten each time a new image is to be printed.

The second addition, based on claim 2 as filed, refers to angular position errors between the sensing means (16 in Figure 1) and the discharge means. To elucidate the meaning of this feature the patent proprietor has referred to column 13, lines 13 to 15 of the description where it is stated that "/it/ is this skewing which effectively compensates for any error between the angular encoder 16 and the imaging devices". Similarly, at column 6, lines 5 to 24 it is mentioned that the electrodes (ie imaging devices) of the writing head should ideally be "matched" to the encoder (which is however impossible to achieve in practice). At column 2, lines 5 to 16 it is explained that if "multiple imaging devices are used for imaging, the circumferential distances between such devices must be precisely fixed to represent an integral number of units of circumferential distance" (as determined by the sensing means), otherwise a slanted swath condition will result. The Board understands all these passages to refer to one and the same technical problem. Thus, in the light of the description, the "position error" mentioned in claim 1 necessarily involves the distance between two discharge devices of a writing head having multiple discharge devices (there are for example 16 discharge devices in a head according to the described embodiment; cf. Figure 2). Claim 1 however also covers presses comprising heads having a single discharge device (D4 might be of this kind). Thus, and leaving aside any objections against the claim under Article 84 EPC, claim 1 is formulated in such a way that it encompasses subject-matter which does not solve the above problem. It has not been argued and does not seem probable that the feature in question contributes to the solution of some other problem. But if a technical problem is not solved there can be no inventive step (cf. T 26/81 OJ EPO 1982, 211, point 9 of the reasons: "the inventive step may be considered as a step from the technical problem to its solution").

8. It may be added that even if claim 1 were limited to a press comprising a writing head having multiple discharge devices is it unlikely that its subject-matter would be inventive. In D4 the corrections applied will compensate for any errors which may be detected by inspection of test images (cf. paragraph 3 above). A slanted swath condition would be compensated for even if the reason for it was not known. Therefore the result specified in claim 1, viz to compensate for angular position errors between the sensing means and the discharge means, would be achieved as a matter of course when applying the teaching of D4 to a press with writing heads having multiple discharge devices. It is noted that such heads are well known in the art (cf. patent-in-suit, column 1, lines 32,33: "Multiple imaging devices may be used to produce several lines of image spots simultaneously").

9. Thus claim 1 cannot be allowed (Articles 52 and 56 EPC) and the patent proprietor's auxiliary request must also be refused.