Monday, December 14, 2020 T 483/17 - 'more up to the hilt' is not needed

The opponent in this case alleged that one of its own products is novelty destroying prior art. As this prior use lies within the sphere of the opponent, proof 'up to the hilt' was required. The board's analysis of this proof is very case-specific of course, but makes for interesting reading (if you're into that kind of thing). The proprietor makes a good effort to find inconsistencies in the  provided proof, which leads the Board to state at one point that if more evidence was provided, it 'would merely have been proven even more "up to the hilt"'. 

In the end the opponent manages to prove to the Board's satisfaction that it really did sell a product that falls under the scope of the claim. 

(...)

Reasons for the Decision

1. Admission of the documents filed in appeal

The documents PU1-B*, PU1-C*, PU1-D*, PU1-G and PU1-Ad were filed with the grounds of appeal in response to the impugned decision, in which the opposition division, contrary to their preliminary opinion, found that the documents on file in the opposition proceedings did not prove a prior use of a retainer segment according to claim 1 of the patent in suit. In particular, in points 2.4.2 to 2.4.3 of the decision the opposition division expresses doubt about the "chain of evidence" derivable from documents PU1-B to PU1-D due to the partial deletions and removal of content in these documents.

The filing of documents PU1-B*, PU1-C*, PU1-D*, which correspond to documents PU1-B, PU1-C and PU1-D with less redaction, and of PU1-G and PU1-Ad, which are directed to providing the information considered by the opposition division to be missing from the already provided evidence of the very same sale, is thus seen as a legitimate response by the appellant. These documents are therefore admitted into the proceedings (Article 12(4) RPBA 2007 in conjunction with Article 25(2) RPBA 2020).

2. Main request - Novelty in view of public prior use

2.1 Since the sale and delivery of five roller bearings from the appellant/opponent (SKF) to a third party (Hansen Transmissions) is completely out of the sphere of the respondent/patent proprietor (NTN Corp.), according to the case law of the Boards of Appeal, proof has to be provided "up to the hilt". The determination of the prior use thus requires that the facts giving rise to the prior use must be established without gaps (see T 472/92, OJ 1998, 161, point 3.1 of the Reasons); however, this does not mean that the opponent has to provide each and every theoretically possible piece of evidence. Instead, it is sufficient that the evidence put forward convinces the Board.

On the other hand, a prior use has not been sufficiently proven if the proprietor successfully pointed out any inconsistencies or drew attention to any gaps in the opponents chain of proof (see T 472/92, op. cit.).

2.2 Applying these principles, the Board considers it established that the alleged prior use took place.

The provided documents sufficiently prove that, on 23 May 2006, five "Waelzlager" with the designation BT-28125 A/HA1 and a total weight of 13.900 kg were transported by Hans Geis GmbH + Co from the appellant (opponent) SKF in Schweinfurt (DE) via Aachen to Hansen Transmissions in Lommel (BE) and were charged at €43 760.58 per unit. Furthermore, the Board is convinced that the engineering drawings according to appendix B, C and D - provided with corresponding product designation numbers - show the items actually sold.

2.3 The doubts about the prior use expressed by the opposition division in points 2.2.3.1 and 2.2.3.2, page 5 of the contested decision, and the arguments provided by the respondent (patent proprietor) do not point out a gap in the opponent's chain of proof for the following reasons:

2.3.1 Even if the opponent had submitted a bank statement concerning the payment for the delivered bearings, the transaction would merely have been proven even more "up to the hilt"; however, this does not change the facts already established (see point 2.2 above). A further possibly existing piece of evidence not having been provided does not reduce the significance of the evidence actually submitted. The same applies to the respondent's further allegation that, due to the kind of product, it was not plausible that an offer with technical specification was irretrievable.

2.3.2 The fact that the date "07-09-2004" is printed on the "Versandauftrag", appendix PU1-Ab, has been sufficiently explained by the appellant (opponent). The explanation given, namely that 7 September 2004 was the day when the form was created and 23 May 2006 was the day when it was actually used in the current case, is convincing and cannot be invalidated by mere reference to the unusual nature of the two dates.

2.3.3 The different designations "0001/0004B" and "0002/0004C" (although the quality of the filed invoices is not particularly high, it can be seen that the last one actually is "4G") on the invoice PU1-A or A* appear under "regel", which is indeed unclear. However, all items have exactly the same "Omschrijving", i.e. designation and price. The Board is therefore convinced that the invoice relates to only one type of bearing and that five bearings of the same type were delivered. The respondent's assumption that the separate items of three and then two bearings pointed to a sale for testing purposes and the further assumption that it could therefore be assumed that there was an at least tacit secrecy agreement therefore has no basis. Furthermore, the appellant has given a convincing explanation of the additional designations U60-BT28125E501L and 40 in the second and third lines of the "Omschrijving", namely that they are product numbers which were added on the customer Hansen Transmissions' request.

2.4 The respondent also doubted that Mr Clement, in his declaration on oath dated 13 April 2017 (PU1-G), was able to reproduce all the details about the delivery of bearings more than 10 years ago and pointed out that the appellant, who obviously pre-formulated this declaration, must have had more information/documents than it was willing to submit in these proceedings. However, Mr Clement's affidavit is not needed to convince the Board that five bearings designated "BT-28125 A/HA1" have been delivered from SKF to Hansen Transmissions. Therefore any "inconsistency" within this declaration is of no relevance.

2.5 The respondent also disputed that the technical drawings according to appendixes PU1-B or B* to PU1-D or D* corresponded to the allegedly sold items. It argued that due to the redactions in appendix PU-1B and B* it was not clear that the cage in appendixes PU1-D and D* in fact corresponded to or was identical to the cage shown in appendix PU1-B and B*.

2.5.1 In this respect the Board is convinced by the respondent's argumentation that documents PU1-B or B* to PU1-D or D* were the construction drawings of the specific bearing sold and delivered to Hansen Transmissions and that the bearing cage of this bearing was made of the material "PEEK nach 1054077", as indicated in drawing PU1-D or D*.

The Board notes in this context that the drawings PU1-B and B* of the cross section of the complete bearing carry the same product designation "BT2-8125 A/HA1" as the bearings on the invoice PU1-A. The table of parts of this drawing list a "Cone" designated "AJ-BT-8125 A-1/HA3". The respective drawings PU1-C and C* showing the "inner ring, rollers and cage" have exactly this product designation. They in turn list a plastic cage, corresponding to the retainer segment of the patent in suit, having the designation "CJ-BT2-8125" as part of this. The drawings PU1-D or D* showing a "Kunststoff-Segmentkäfig" have exactly this designation. They also indicate the material (of the cage) as "PEEK nach 1054077". Against this background, the time difference of nine months between the dates indicated on PU1-C and C* and PU1-D or D* is of no particular significance.

The above numbers are very specific indicators which consistently identify the product in consecutive engineering drawings. They thus cannot be compared to general model names for products like cars, for which the customer subsequently chooses the exact specification, e.g. a particular engine or gear box. Indeed, the trade name of the prior use bearing is "Nautilus", and not the long product designation on the engineering drawings. There is thus no indication that the product identified by this number could be provided with different retainers made of a material different from PEEK according to standard 1054077.

2.5.2 Furthermore, due to a "standard" (albeit an internal one) having the very purpose of standardizing, the appellant's argument that standard 1054077 could have changed within the short time between its definition and the sale of the prior use bearing is not convincing. The standard PU1-E clearly mentions PEEK-GF as the material on page 1, under point 2 "Werkstoff". Thus the information "PEEK nach 1054077" on the drawing PU1-D or D* implies the material to be PEEK-GF, which is then further specified on page 12(13), "Tabelle 3", to comprise 14+-1 wt% glass fibres.

2.6 Finally, the Board is not convinced by the respondent's argumentation that, due to the redaction in appendixes PU1-B and B*, it was not proven "up to the hilt" that any alternative cages, i.e. cages made of other materials, could have been used in the bearing of the prior use. The respondent pointed to the "Abstandshülsen" identified with their designation in the top left part of the drawing and submitted that the long redacted part following "Käfig" in the drawing PU1-B* could describe alternative cages ("Käfigsegmente"). However, the "Abstandshülsen" and "Scheiben", which have their respective designations indicated directly in the drawings, are listed with exactly the same designations in the table of parts of the bearing. Thus, if the redacted part following "Käfig" were hiding a designation number, this number would, for consistency reasons, also be found in the table of parts. This is not the case.

Instead, the indications in the lower left corner under the heading "Kontrolle der Käfigelemente" all relate to sizes and dimensions. This applies to the information given for "Axialluft" and "Radialluft", for example, as well as to the information given for "Käfigluft in Unmfangsrichtung". In this context, it would not be consistent if, contrary to all other information given in this part of the drawing, the redacted information about the "Käfig" did not contain information on sizes and dimensions but on materials.

2.7 The Board thus comes to the conclusion that the prior use is sufficiently proven and that retainer segments according to drawings PU1-D or D* formed of PEEK according to standard 1054077 (PU1-E) were made available to the public. As also accepted by the appellant, such retainer segments exhibit all the features of claim 1, which is thus not novel.

This decision T 0483/17 (pdf) has European Case Law Identifier:  ECLI:EP:BA:2020:T048317.20200911. The file wrapper can be found here. Illustration by  Alexander Lesnitsky (AlLes) obtained via PixaBay  under the PixaBay license (no changes made).