T 0537/03 10-11-2005

Reasons for the Decision

1. The appeal is admissible.

2. Claim of priority (Article 87 EPC)

The English translation of the Danish priority document was considered in these proceedings, the accuracy of the translation was not contested.

The thermally insulating unit, as defined in the patent, is not covered in the priority document.

Claim 1 according to each of the submitted requests contains the following two features, which are neither explicitly disclosed nor implicitly derivable by the person skilled from the priority document.

2.1 The first feature concerns the upper limit of the vacuum pressure range: "a pressure of the magnitude 0.1-25 mbar". The sole reference in the priority document, on page 6, discloses values in the order of 1/10 mbar or less for the vacuum pressure. The claimed range thus clearly lies outside the range indicated in the priority document.

2.2 The priority document also provides no support for the second feature of the patent relating to the upper limit of 0,5 mm for the cell size. In the priority document the cell size is defined as being smaller than 0,3 mm and preferably smaller than 0,1 mm (see page 5 of the English translation).

2.3 In conclusion, the invention as claimed according to all the requests for maintenance of the patent, is not the same as that disclosed in the priority document (G 2/98, OJ 10/2001, 413). The claim of priority does not meet the requirements of Article 87(1) EPC and is thus not valid.

Hence the relevant date of the patent, either as granted or as amended according to the auxiliary requests, is the effective filing date, 19 January 1995 of PCT/DK95/00028, and not the priority date (19 January 1994).

2.4 Document D1, which was published on 16 March 1994, thus after the claimed priority date but before the effective filing date of the patent, is state of the art as defined by Article 54(2) EPC.

2.5 During the oral proceedings before the Board, the appellant (patentee) agreed that the full scope of the claimed subject-matter according to each of the requests is not disclosed in the priority document.

3. Main request

3.1 The aim of the invention disclosed in D1 is to create high-quality vacuum insulation (column 1, line 56). This is achieved by using a small, cheap and energy saving vacuum pump 18 in communication with hermetically sealed spaces 14 filled with insulating material (column 2, lines 2 to 18). The pump is activated at least during the initial operation of the machine (which is for example a refrigerator), and this can extend over a relatively long period (one week to several months) until the desired vacuum pressure is achieved (less than 0,1 mbar, column 2, line 50).

After the initial period, the pump may be either stopped or further activated. This is carried out either continuously or periodically in order to maintain the desired vacuum pressure (column 3, lines 7 to 20). The insulating material is a foam having closed cells (column 2, lines 35 to 40).

3.2 The following features of claim 1 are thus explicitly disclosed in D1:

a thermally insulating unit or, respectively, a device or apparatus with such a unit, e.g. a refrigerator cabinet (see e.g. column 1, lines 1 to 11 and also claim 1), said unit being built with a core plate of a cellular material 14 located between tight cover plates 12,13 (see Figure 1) and encapsulated for use in evacuated condition (see column 2, lines 29 to 35 and lines 40 to 44, Figure 1), in which

- the unit is permanently connected with an associated vacuum pump 18 (see column 3, lines 7 to 9);

- the vacuum pump being adapted to maintain, after an initial phase, a pressure of the magnitude 0.1-25 mbar in the unit (actually the purpose of the arrangement of D1 is the same, i.e. upholding a vacuum pressure of 0,1 mbar after a week, see column 2, lines 48 to 53; this implies that the pump is able to maintain a pressure of between 0,1 and 25 mbar at least during the initial phase).

3.3 The remaining features of claim 1, not explicitly known from D1, are the following:

(a) the encapsulation is not absolutely hermetically tight;

(b) the cell sizes are predominantly in the range of 0.5-0.1 mm or less.

The feature relating to the "open celled type" of the foam is not taken into account for the assessment of novelty, since this feature is defined as optional.

3.4 In dispute is whether features (a) and (b) mentioned above are implicitly disclosed by the general teaching of D1 when read by the person skilled in the art.

3.4.1 The appellant submits that the passage of column 3, lines 7 to 20 of D1, which indicates that the pump may be kept continuously or sequentially activated, only refers to the initial phase of operation. The function of the pump is defined as enabling the end user to obtain the desired vacuum pressure, and not, as is usually the case, to set the vacuum during the manufacturing process. However, the skilled person would also understand the aforementioned passage to include activation and deactivation during its life time after the initial setting of the machine, since the pump remains in the machine and can be operated in conjunction with the compressor.

The purpose of the controlled activation of the pump as disclosed in D1 is to maintain the vacuum pressure during the life time of the machine, taking into account that pressure may increase due to ingress of air at relatively untight areas of the unit.

The appellant further argues that the expression "hermetically sealed" used in D1 cannot be construed as meaning "not absolutely hermetically sealed", because a hermetically sealed product is usually made absolutely hermetically sealed in the sense that it is sealed at least for the life time of the product.

The qualification "not absolutely hermetically tight" (underlining added for emphasis) of claim 1 has no generally recognised or accepted specific meaning in the field of vacuum isolating panels. This expression leaves completely open, when and to which extent the encapsulation is not tight when compared with the "normal" hermetical encapsulation described in D1. Neither can a distinction between absolutely hermetical and not absolutely hermetical units be derived from the other cited documents. Contrary to the appellant's opinion, the mere fact that the insulation unit of D1 uses foams with closed cells does not provide evidence that the unit is absolutely hermetically tight, even though it is generally accepted that it would be more difficult to pump air out of such a foam material in the event of a leak.

It may be additionally mentioned that no details of the meaning of the expression are given in the description of the patent. According to the embodiment defined in dependent claim 5, a controllable leak is deliberately provided in the encapsulation; the expression can therefore be construed to mean any leaking area occurring during the life time of the product.

The Board also agrees with the respondent that an absolute sealed space can in practice rarely be achieved, and then only with high production costs and long production and testing methods. This is certainly not the case in D1, which seeks to find a balance between thermal insulation efficiency and limited production costs (see for instance column 1, line 38 to column 2, line 1).

The Board therefore reaches the conclusion that the qualification "not absolutely hermetically tight" in the context of claim 1 cannot provide a clear and unambiguous distinction over prior art D1.

3.4.2 According to established case law of the Boards of Appeal, in assessing novelty there must be a high degree of certainty that a feature not contained explicitly in a document is inevitably disclosed therein. In this case it is not disputed that values for the average cell size are not explicitly mentioned in D1. However, the skilled person would inevitably arrive at the values defined in the claim for the following reasons.

D1 indicates that vacuum pressures in the foamed material should be about 0,1 mbar (see column 2, lines 48 to 53). The skilled person, when performing the invention disclosed in D1, would have thus to design the foam material by using general knowledge in the field.

The parties agree that the person skilled in the art knows the general principle that establishes a close relationship between pressure and cell size, in the sense that the higher the pressure of the gas, the smaller the mean free path length is before a gas molecule strikes an object e.g. the wall of the cell. In addition, it is general knowledge that thermal insulation is improved if the cell size is smaller than the mean free path length; this interrelationship is illustrated for instance in E1 (line 24 of page 5 to line 9 of page 6), D2 (column 3, lines 13 to 28), D3 (page 1, lines 20 to 29) and is acknowledged as basic knowledge in the priority document itself (English translation, page 5, lines 8 to 23) and in the patent (page 2, paragraph [0003]).

By applying this common knowledge, the skilled person, starting from a preselected vacuum pressure, is thus able to determine the average cell size.

As explained by the appellant himself, the person skilled in the art generally knows that vacuum pressures about 0.001 mmHg (0.0013 mbar) require cell sizes in a foam material usually comprised between 0.3 and 1.0 mm (as illustrated for instance in D1), and that, by means of a straightforward extrapolation, cell sizes in the order of 0.003 to 0.010 mm must be chosen for pressures about 0.1 mbar (pressure described in D1).

These results for the cell sizes are thus inevitably disclosed in D1 in the light of the general knowledge attributed to that technical field. This implicit disclosure of cell sizes of 0.010 or less is fully in line with the characterising feature of claim 1, which defines cell sizes in the range of 0.5-0.1 or less,

i.e. there is no lower limit of the range defined in claim 1.

The appellant argued that the invention reveals surprising effects, namely meeting satisfactory insulation properties, while departing from the aforementioned general theory for dimensioning cells with respect to the vacuum pressure.

According to the invention, the cells are bigger than they should be when determined in line with said theory, and it is accepted that the unit is not sealed but connected to a pump. Even though this general technical concept may be understandable as such, there is, as discussed above, no clear limitation in claim 1, which could define such a distinction over the prior art.

3.5 Since the remaining features a) and b) are implicitly disclosed by D1, the unit defined by claim 1 as granted lacks novelty when compared to D1.

4. First auxiliary request

The amendment to claim 1, namely the deletion of the expression "or less" in the last feature of the claim as granted, infringes the requirements of Article 123(2) EPC.

Since the value of 0.1 mm for an average cell size is disclosed in the originally filed documents of the patent application (see for instance claim 1 and page 2, line 6), the question arises whether such an exemplified feature can form the basis for a new range as claimed. The value of 0.1 mm for the cell size is cited in page 1, line 51, but only in connection with pressures of about 1 mbar. However, the claimed device refers to pressures between 0.1 to 25 mbar, thus significantly lower or greater than 1 mbar. No indication of any preferred cell size range corresponding to the full range for the vacuum pressure is given in the originally filed documents of the patent. On the contrary, from the patent application as a whole and from the aforementioned general theory, the skilled man would have readily recognized that values for cell sizes lower than 0.1 mm, which are now excluded, were presented as having a close functional dependency on the values of the upper part of the defined pressure range, i.e. for pressures between 1 and 25 mbar. Consequently, in this case the cell size of 0.1 mm cannot be detached from an example, and used to form a basis for a generalised lower limit for the cell size range, without taking into consideration the corresponding pressure range. It is clear that the ranges indicated in claim 1 as granted, i.e. between 0.1 and 25 mbar for the pressure and open ended range (0.5-0.1 mm or less) for the average cell size are to be understood as closely linked or interactive with respect to the results to be achieved, i.e. good thermal insulation properties.

Hence, claim 1 of the first auxiliary request is not based on the application as filed.

The appellant argues that the basic merit of the invention is the finding that there is not always a strict accordance between the theory discussed above and the practice relating to the dependency of cell size and pressure; this is set out in the disclosure of the patent, see for instance page 2, lines 21 to 26.

However, in the absence of any additional information in the application documents concerning the extent to which the invention departs from the rigid scheme suggested by the theory, it cannot be concluded that the values lower than 0.1 mm for the cell size, though admittedly greater than the values implied by pure application of the theory, would not be expected for higher pressures within the claimed range.

5. Second auxiliary request

The amendments do not meet the requirements of Article 84 EPC.

5.1 Firstly, the combination of features of amended claim 1 does not clearly define the device. Some of the added features were comprised in dependent claim 4 as granted, and define a potentially untight area in the encapsulation. The qualification "potentially untight" means that there may or may not exist an untight area, and as such does not constitute a limiting feature for the claimed device. Since the added features are linked to the original feature that the encapsulation is "not absolutely" hermetically tight, it is not unambiguous if, how and where the unit should actually be untight.

In addition, this now leads to a lack of clarity concerning the feature relating to the connection of the pump with the unit. According to claim 1, the pump should be connected to the unit at a relatively large spacing from the potentially untight area. However, as it is questionable whether such an untight area exists, the meaning of a feature defining the connection of the pump at a large spacing from the untight area cannot be clearly determined.

5.2 Secondly, claim 1 of the second auxiliary request is not supported by the description. The patent as a whole describes one embodiment having a deliberately provided untight area in form of a controllable leak (see paragraph [0013] of page 3), and wherein the pump is connected to the unit opposite to said leak.

The unit according to the definition of claim 1 now comprises, on one hand, a potentially untight area distant from the pump and, on the other hand, a controllable leak which may be provided anywhere, even in close vicinity to the pump. Such duplication of untight areas is not supported by the description, and even appears to be in contradiction with the detailed embodiment.

6. Third auxiliary request

This request corresponds to the auxiliary request considered in the opposition procedure, the subject-matter of which was deemed to lack inventive step in the impugned decision. It was not maintained as a request with the appeal as filed, but was only introduced at the very end of the oral proceedings.

The sole amendment made in comparison with claim 1 of the main request is that the open celled form of the foam material is no longer an optional feature.

The patentability of the subject-matter of such a revised claim, in particular when compared to D1, would still be questionable since the meaning of an open celled foam is broad, as indicated in the patent itself, e.g. foams including only 15% of open cells are also considered to be of the claimed type (see page 3, lines 44 to 46 of the patent).

Because of the late filing and since its subject-matter does not prima facie appear to meet the requirements of Article 52(1) EPC it was not admitted by the Board.