Appeal 2006-1768 Application 10/389,327 Because burner port area is a result effective variable, it is reasonable to expect success in developing workable or optimized burner port areas. Appellants cite to a non-precedential Board decision, Ex parte Sullivan 2003 WL 23014513 (Bd. Pat. App. & Interf.), to support their argument that Aller is not applicable to the facts of the present appeal. The Sullivan decision does not support the Appellants’ position in the present appeal for two reasons. First, it is a non-precedential decision that is not binding on the Board.1 Second, the facts in Sullivan are inapposite to the facts of the present appeal. As we determined above, burner port size is an art recognized, result-effective variable. As such, it would have been prima facie obvious to develop workable values for this variable. Woodruff, 919 F.2d at 1578, 16 USPQ2d at 1936-1937; Boesch, 617 F.2d at 276, 205 USPQ at 219; Aller, 220 F.2d at 456, 105 USPQ at 235. Appellants allege the following statement shows unexpected results: “Intuitively, it would be expected that raising the tip flow area would proportionally reduce tip velocity, but instead, it is found that the drop in velocity can be mitigated by the fact that raising tip flow area raises FGR” (Specification, ¶ [0033]). Regarding this allegation, we agree with the Examiner’s analysis in the Answer. Under the circumstances of this appeal, Appellants’ single sentence statement in the Specification, without more, is not sufficient to prove unexpected results. Appellants’ quoted statement above provides no express statement of unexpected results and fails to explain why the increase in FGR in response to an increase in burner port area is unexpected. Additionally, 1 See, Board of Patent Appeals and Interferences Standard Operating Procedure 2: Publication of Opinions and Binding Precedent § VII(F) (August 10, 2005, Rev. 6). 11Page: Previous 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 NextLast modified: November 3, 2007