Friday, February 3, 2012

Eurocopter v Bell Overview

Eurocopter v Bell Helicopter Textron Canada Limitée 2012 FC 113 Martineau J

Eurocopter v Bell is a familiar case name from numerous visits to the Federal Court on discovery matters, but now the infringement decision itself is available. It has not yet now been posted to the Federal Court website, but Alan Macek has made it available through his IPPractice site. Eurocopter’s patent in question, 2207787, relates to sled-type helicopter landing gear which was known as “Moustache” gear. Matineau J held that one claim was valid and infringed by Bell’s prototype “Legacy” landing gear, but no valid claims were infringed by Bell’s current “Production” gear. The most interesting legal aspect of the decision is that Martineau J held that Eurocopter as the successful patentee was entitled to punitive damages, which is extremely unusual in Canadian patent law. I will deal with that issue in a separate post. This post will provide a synopsis of the remainder of the decision.

Moustache landing Gear
Everything on a helicopter must be light, including the landing gear, in order to maximize the useful load. At the same time, the landing gear must absorb the shock of landing, and it must also resist the phenomenon known as “ground resonance” which arises with four-rotor helicopters for which the patented landing gear was designed. Eurocopter’s patented landing gear was a substantial improvement over the prior art in all these respects [341], [347]. In developing its new Bell 429 helicopter, Bell sought to optimize the landing gear. Bell leased a Eurocopter EC120 helicopter that was equipped with the patented Moustache landing gear for testing and “benchmarking” [172]. They liked what they saw, and copied the Moustache landing gear [426] to make their prototype Legacy landing gear. Only twenty-one sets of the Legacy gear were ever manufactured, as Eurocopter commenced this infringement action [177], and in response, Bell redesigned the landing gear to the new Production version. The Production version was not as good as the Legacy gear [402], but Bell hoped it would not infringe. Eurocopter alleged infringement of its patent by both the Legacy and Production gear.

Production Gear
That the Legacy gear infringed was ultimately conceded by Bell [252]. The Production model was held not to infringe because the patent required an “integrated front cross piece” connected to the skids through a transition zone with a “double curvature,” while the Production model used a saddle joint. Eurocopter bravely argued that a sharp joint, like the saddle joint, could be considered to have a double “curvature”. Martineau J rejected this [228]. The main question was whether this double curvature was an essential element [229]. Martineau J concluded it was [246], and consequently the Production model did not infringe. I did not find this part of the decision entirely satisfactory. For example, Martineau J stated at [246] that “The entire concept of the Moustache landing gear is clearly oriented around the necessity of having an ‘integrated front cross piece’ in order to have a more economical, lighter and a more flexible landing gear having the other related benefits disclosed in the patent.” That may be so, but it does not tell us whether, to paraphrase Catnic [1982] RPC 183, 236 (HL), a skilled person “would know that for all practical purposes a [saddle joint] would be as effective as [a double curved joint] and that all the essential features of [the landing gear] would still be present.” The evidence showed that “a saddle joint is stiffer and works differently than the double curvature of the inclined transition zone” [245], but it is not clear to me that it works differently in an essential respect. For example, in McPhar Engineering [1956 - 60] Ex CR 467 the defendant’s means of securing the transmitter in a vertical orientation worked differently than did the means specified in the patent, but Thorson P’s conclusion was that this difference was not essential to the invention. I am not suggesting that Martineau J’s conclusion on this point was wrong, but from the discussion, I am not sure it was right. This is not a criticism of Martineau J’s decision; rather, I suggest that it illustrates that the essential elements doctrine remains obscure, for all that the SCC has called it the “key” to purposive construction (see Whirlpool v Camco 2000 SCC 67 at [45]).
Saddle Joint

Bell argued it was entitled to the regulatory or experimentation exception based on subsection 55.2(1) in respect of the Legacy gear, because the Legacy gear had been used for various tests related to certification of the Bell 429. However, Martineau J noted that a non-test helicopter displayed at trade shows was equipped with the infringing gear, and perhaps more importantly, Bell had solicited advanced orders and actually signed agreements with clients for helicopters equipped with the infringing gear. Accordingly, Martineau J held that the use of the Legacy gear was not “solely” for regulatory purposes, as the section requires.

Bell’s argument on anticipation was weak, based on a single artist’s representation of a concept helicopter, which did not show the landing gear clearly, and alternately on documents which Martineau J found were not publically available, and did not anticipate in any event. The obviousness argument was also weak. While non-obviousness is a notoriously slippery concept, both the expert evidence and the objective (“secondary”) evidence that the patented invention was a real inventive advance were compelling [308]-[309].

Bell did mount a successful attack based on utility against a number of claims. In the preferred embodiment, used by both Eurocopter and Bell, the front cross piece is offset forward. Many of the claims encompassed gear with the front cross piece offset backwards, and one claim specifically claimed it. Eurocopter had never tested gear with a backwards offset, and Martineau J held that it was not established that the gear would work with that design. A number of claims were therefore invalid [371]. This was a hollow victory, as Bell did not use a backwards offset in either version of its landing gear, presumably because it was not useful. One claim did cover a forward offset, as used by Bell, and Martineau J had no difficulty holding that the utility of this claim was established [354] and the claim was therefore valid [371].


  1. It was pointed out to me that this is one of the first cases - possibly the first - in which an assessment of sound prediction (in particular the requirement for an explicitly stated "sound line of reasoning") had been applied to subject matter outside the chem/pharma space.

  2. Good point. It is true that sound prediction was considered, but the decision didn’t really turn on that doctrine. For claim 16, which was held to be valid, utility was based on demonstrated utility [354], not sound prediction. The claims that were invalid encompassed an embodiment with a backwards offset front cross piece. This variant was completely untested [363], nor had any theoretical modeling been done, and even the expert witness for the patentee was unwilling to testify that the landing gear would work in that configuration [364]. In short, there was no evidence of any kind, whether in the specification or produced at trial, that that embodiment would work [368]. There was also some evidence that it would not work, as a backwards offset front cross piece would be more likely to buckle on impact [366]. Martineau J therefore concluded that the claims were invalid because neither demonstrated utility nor sound prediction had been established [371]. It seems to me that on that evidence the same conclusion would have been reached even on a traditional analysis that did not mention sound prediction at all.

    With that said, this case does suggest how sound prediction might be raised in a mechanical case; for example, if the backwards offset cross piece had not been actually tested, but there had been computer modeling done which showed it would be expected to work in the same way as the front offset, which had been tested, then validity might have been established based on sound prediction.