UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OF NEW YORK
December 7, 1971
Anna M. O'KEEFE, as Administratrix of the Goods, Chattels and Credits of Michael F. O'Keefe, deceased, et al., Plaintiffs,
The BOEING COMPANY, Defendant
Cannella, District Judge.
The opinion of the court was delivered by: CANNELLA
CANNELLA, District Judge.
In January, 1963, Lieutenant Colonel Joe R. Simpson, Jr. and Major William W. Gabriel of the United States Air Force arrived from Walker Air Force Base, New Mexico to instruct instructor crews for B-52 bombers at Westover Air Force Base, Massachusetts in "terrain avoidance procedures" through the use of "advance capability radar" or ACR, which was a relatively recent development at the time and which was being installed in B-52's. Two "standardization" crews were to make a checkout flight together, starting at 8 a.m. on January 24, 1963, but maintenance problems with regard to the plane to be used, a model B-52C, number 53-406,
necessitated a delay of about four hours. Once the plane finally had become airborne shortly after noon with the two instructors
and both crews
aboard, it remained in the Westover vicinity for the next one and a half hours or so while an attempt was made to calibrate and correct the ACR equipment, whereupon the following radio conversation took place between the plane and the Westover weather station:
1830 Frosh 10 -- Westover Metro, this is Frosh 10
Metro -- Aircraft calling Westover Metro, this is Westover
Metro go ahead.
Frosh 10 -- Roger, this is Frosh 10 and we are still in
the local area here, will you give me a reading on the
northern Pokerdeck Area at the present time, is that
improving any up there.
Metro -- Roger 10 this is Westover Metro, stand by while
I check the very latest on that. Roger stand by.
1831 Metro -- Westover, this is Westover Metro, (Pause),
Pokerdeck generally scattered in the flatlands still
looks like however, it will be scattered to broken with
bases from around 2,000 to 2,500 in the south, in the
with tops around 6,500 MSI. Also received a report,
1832 expect anyway, moderate turbulence, with occasionally
severe turbulence in some of the hills. We do not have
Mt. Washington to the south, it is still zero, zero, fog,
and blowing snow. And we also have a report on Greenville,
showing that it is scattered, the restricted visibility is
down and blowing snow. But the general condition at
Greenville is 2500 scattered and ten. It looks like
the flat country will be scattered, occasionally broken
in the hills and some of the hills obscured over.
Ten this is Westover Metro did you read, Over.
Frosh 10 -- Roger, 10 here, be advised you are coming in,
your transmissions are very, very garbled and chewed up,
and hard to read.
Metro -- 10, Westover Metro, 10 Westover Metro, how do you
read me now.
1833 Frosh 10 -- Your coming in better now, Metro.
Metro -- This is Westover Metro, I repeat, the flatlands
in the northern pokerdecks route is generally high thin
scattered clouds, bases around 2500 feet, tops about
5,000. In the higher terrain you can expect 2500 scattered
to broken with a few isolated snow showers, that is MSI
on the 2500 feet. Tops about 6,000 still expect moderate
turbulence change to occasionally severe turbulence in the
hills. We have a report from Mt. Washington to the south
of your route it shows that they are still zero, zero and
fog and blowing snow. Over.
1834 10 this is Westover Metro, did you read me that time, Over.
Frosh 10 -- Thank you very much Metro, this is Frosh 10 out.
The background of this conversation was that two alternate routes had been planned for the low altitude training mission, one over central Maine, classified "Poker Deck 8-3" or northern route, and the other over the Carolinas, classified "Poker Deck 8-5" or southern route. The primary determining factor as to which route would be flown was apparently to be forecast(s) of weather conditions in the given area, and they resulted in an initial determination to fly the southern route. However, the delay in taking off combined with the airborne delay, the fact that the particular route was to be flown three or four times and under visual flight rules, and the seeming inaccuracy of the forecast with respect to the weather in the immediate Westover area led Colonel Bulli to opt to fly Poker Deck 8-3. The plane then proceeded to enter this route by way of the Princeton, Maine VOR at about 2:30 p.m., assuming an altitude of 500 feet above the terrain lying below the route flight path of 45 DEGREES 07' N 67 DEGREES 57' W to 45 DEGREES 35' N 69 DEGREES 26' W. Indicated airspeed was 280 knots. The aircraft encountered light to moderate turbulence right from the start, and Colonel Bulli elected to terminate the run as the plane approached the vicinity of Elephant Mountain. He caused the plane to assume a nose-up angle, and just as it began to climb, there occurred what sounded like an explosion to him; the aircraft went into a bank to the right with the nose down. The crew's immediate efforts to get the right wing and the nose up proved to no avail, and the plane crashed. Colonel Bulli and Captain Adler
The team sent by the Air Force to investigate the mishap arrived at the crash site near Upper Wilson Pond
on January 26, 1963 and discovered that the wreckage was at the one place except for the vertical tail which was subsequently found essentially intact approxmately a mile and a half back along the flight path.
The plaintiffs filed their complaint on January 23, 1964, alleging that B-406 crashed as a result of negligence and/or breach of warranty on the part of the defendant.
This Court has jurisdiction pursuant to 28 U.S.C. § 1332.
The defendant argues that the various claims asserted by the plaintiffs are barred by the applicable statute of limitations.
Since jurisdiction herein is based solely upon diversity of citizenship of the parties, this court must look to the pertinent state statute of limitations. See Guaranty Trust Co. v. York, 326 U.S. 99, 65 S. Ct. 1464, 89 L. Ed. 2079 (1945); Hodge v. Service Machine Co., 438 F.2d 347 (6th Cir. 1971).
B-406 was designed, manufactured and sold by the defendant in Washington. It was also turned over to the Air Force in Washington. The plane was subsequently stationed in Massachusetts; it crashed in Maine while being manned by military personnel whose assignments with the Strategic Air Command (SAC) caused them to continually traverse, if not live in, numerous states of the Union. This action was brought in New York, and the court is therefore required to refer to New York's conflict-of-laws rules
in order to determine which period of limitation is controlling. The New York "borrowing statute," N.Y.C.P.L.R. § 202, which applies to this case,
reads as follows:
An action based upon a cause of action accruing without the state cannot be commenced after the expiration of the time limited by the laws of either the state or the place without the state where the cause of action accrued, except that where the cause of action accrued in favor of a resident of the state the time limited by the laws of the state shall apply.
The threshold issue then is what cause or causes of action accrued as a result of the crash on January 24, 1963. The plaintiffs allege causes of action for wrongful death and for personal injury. The rule once firmly settled with respect to such actions was that, being unknown to the common law, they derive solely from statutes and that the statute which governs is that of the place of the wrong. See, e.g., Turner v. Capitol Motors Transportation Co., 214 F. Supp. 545, 547 (D. Me. 1963); Kilberg v. Northeast Airlines, Inc., 9 N.Y. 2d 34, 38, 211 N.Y.S. 2d 133, 135, 172 N.E. 2d 526, 527 (1961). In this case, or course, all of the injuries and deaths complained of occurred in Maine, which, like most states, had statutory provisions for the maintenance of the causes of action alleged herein. See Me. Rev. Stat. Ann. title 18, §§ 2501, 2551-53. The present case is typical of most, if not all, cases involving complex choice of law considerations in that the positions of the parties plaintiff and defendant vary with regard to the traditional rule from one locus delicti to the next. It is the defendant here which argues that this court is compelled to apply Maine law.
However, after the New York Court of Appeals (per Desmond, C.J.) had first unsettled in Kilberg the "law long settled," it rendered an even more famous decision in Babcock v. Jackson, 12 N.Y. 2d 473, 240 N.Y.S. 2d 743, 191 N.E. 2d 279 (1963), holding that "[justice], fairness and 'the best practical result' . . . may best be achieved by giving controlling effect to the law of the jurisdiction which, because of its relationship or contact with the occurrence or the parties, has the greatest concern with the specific issue raised in the litigation." 12 N.Y. 2d at 481, 240 N.Y.S. 2d at 749, 191 N.E. 2d at 283. Thereafter, the Court held in Long v. Pan American World Airways, Inc., 16 N.Y. 2d 337, 266 N.Y.S. 2d 513, 213 N.E. 2d 796 (1965), that a New York forum, essentially neutral in a given death action, need not necessarily look at all to the lex loci delicti. The Court's opinion concludes as follows:
There was no suggestion in Babcock that its approach and principle were inapplicable to actions for wrongful death, and indeed, in a number of cases in which the question has been considered, the courts - including the Supreme Court of Pennsylvania (Griffith v. United Air Lines, Inc., 416 Pa. 1, 7, 203 A. 2d 796 . . .) - have indicated that the law to be applied is the law not necessarily of the place where the fatal accident occurred but rather of the place having the most significant relationship with, and the greatest interest in, the issue presented. (See, e.g., Fornaro v. Jill Bros., 22 A.D. 2d 695, 253 N.Y.S. 2d 771, affd. 15 N.Y. 2d 819, 257 N.Y.S. 2d 938, 205 N.E. 2d 862; Tramontana v. S.A. Empresa De Viacao Aerea Rio Grandense, [121 U.S. App. D.C. 338] 350 F.2d 468, 471; Gianni v. Fort Wayne Air Serv., 7 Cir., 342 F.2d 621; Watts v. Pioneer Corn Co., 7 Cir., 342 F.2d 617; Mertens v. Flying Tiger Line, Inc., 2 Cir., 341 F.2d 851, 858, cert. den. 382 U.S. 816, 86 S. Ct. 38, 15 L. Ed. 2d 64; Fabricius v. Horgen, [257 Iowa 268] 132 N.W. 2d 410). It would be highly incongruous and unreal to have the flexible principle of Babcock apply in a case where the victim of the tort is injured but not where he is killed.
Maine's relationship to the case at bar is certainly no less accidental nor more significant than Maryland's relationship to Long. For example, a B-52 is hardly an instrument of commercial carriage flying regularly scheduled commercial routes. And while Poker Deck 8-3 may have been a regularly used training route, it was flown on January 24, 1963 only after an airborne change of plan by the flight commander. Then again, there is no contention that any of the members of the crew(s) were citizens of Maine. In short, the inescapable conclusion to be drawn is that this court is not compelled to refer to any part of the law of Maine in this case.
But this does not mean that the deaths and injuries in Maine did not give rise to the causes of action for wrongful death and personal injury asserted by the plaintiffs since all of the states with more significant relationships to B-52 bombers in general and B-406 in particular allow for such actions, to wit, California, Kansas, Massachusetts, Nebraska, Ohio, Oklahoma and Washington.
See Cal. Code Civ. Proc. § 377; Kan. Stat. Ann. §§ 60-1801 to 60-1905; Mass. Ann. Laws chs. 228, 229; Neb. Rev. Stat. §§ 30-809, 30-810; Ohio Rev. Code Ann. §§ 2125.01-2125.04; Okla. Stat. tit. 12, §§ 1051-54; Wash. Rev. Code §§ 4.20.010, 4.20.020, 4.20.046. And New York, the forum state, also provides for such actions. See N.Y. Decedent Estate Law §§ 119, 130 (superseded Sept. 1, 1967 by N.Y.E.P.T.L. §§ 5-4.1, 11-3.2(b)).
Within the Air Force, responsibility for the design conformance, "depot level" maintenance, use and safety of B-406 was divided, respectively, among the B-52 Systems Project Office (SPO) at Dayton, Ohio, the Oklahoma City Air Materiel Area (OCAMA) at Tinker Air Force Base, Oklahoma, SAC headquarters in Omaha, Nebraska and the Office of the Inspector General at Norton Air Force Base in California. Routine maintenance of the plane while stationed in Massachusetts was the responsibility of SAC. "Engineering responsibility" within the defendant corporation for all B-52's, including the ill-fated B-52C herein, was apparently transferred in 1958 from Seattle, Washington to the defendant's installation at Wichita, Kansas. Quite unsurprisingly in view of the complexity of the B-52's, their apparent importance to the United States, and the large number produced,
the defendant maintains "direct relationships" on a "first name basis" with all of the above Air Force commands, with the Air Force maintaining, in turn, a "plant representatives office" (AFPRO) in Wichita.
The plaintiffs set forth in their brief 18 ways in which they contend the defendant was negligent. Two of these presumably relate ultimately to activities in California.
Eleven of the alleged acts of commission or omission appear to relate to the defendant's activities at the Air Force bases in California, Nebraska, Ohio and Oklahoma and at the defendant's facility at Wichita,
as well as possibly to Westover in Massachusetts. The plaintiffs propose as their initial conclusion of law, however, that the "law of Washington applies to issues of liability." And not only do their remaining five allegations of negligence appear to relate essentially to the defendant's activities in Washington, but with the possible exception of the two relating to California, the other eleven contentions appear to relate as much to activities in Washington as to acts in the other states. The specific question here then is not whether the defendant was negligent in any of the ways alleged or whether or not the standard of care varies among any of the above-mentioned states,
but rather whether the plaintiffs' first proposed conclusion of law is correct. The court concludes that it is. If in a case such as this the "niceties of . . . legal legerdemain do not concern us,"
but rather justice, fairness and the best practical result, then the law of Washington must apply here. B-406 was designed, manufactured, sold and delivered in Washington. If the defendant did not live up to a duty to manufacture it free from latent (defects), that duty was breached in Washington. Almost all of the alleged acts of negligence on the part of the defendant are referable, directly or indirectly, to Washington which is the defendant's principal place of business. Then again, the defendant is one of Washington's principal businesses. In short, the court concludes that Washington is the state with the greatest concern with the specific issues
raised in the litigation at this time. Of the other states, only Kansas appears to have a very significant concern with the issues raised in view of the shift of engineering responsibility for the B-52's from Seattle to Wichita in 1958 - or prior to the accident complained of. Only one of the two theories of liability raised by the plaintiffs relates to Kansas, however, and, in any event, the court is not persuaded that the interests of Kansas outweigh those of Washington. The liability questions are therefore controlled by Washington law.
The court concludes further that for purposes of applying the New York borrowing statute the plaintiffs' causes of action accrued in Washington as a result of the crash in Maine. The Court of Appeals for the Second Circuit predicted in George v. Douglas Aircraft Co.
that for borrowing statute purposes New York would hold that the cause of action "arose"
in the state of manufacture, sale and delivery and compare that state's statute of limitations. So far as this court is aware, the New York Court of Appeals has not yet specifically affirmed the prediction, but the decision in Long clearly underscored its logic. Cf. Braniff Airways, Inc. v. Curtiss-Wright Corp., 424 F.2d 427, 430 (2 Cir.), cert. denied sub nom. Addabbo v. Curtiss-Wright Corp., 400 U.S. 829, 91 S. Ct. 59, 27 L. Ed. 2d 59 (1970). Thus, even if New York courts would consider the issue of the statute of limitations separate and apart from the issue of which wrongful death statute applies,
this court is not persuaded that the "decision of reasonable intelligent lawyers, sitting as judges of the highest New York court, and fully conversant with New York 'jurisprudence'"
would be to refer to the statute of limitations of any state other than Washington. Referring then to Washington, the statutory period applicable to the causes of action asserted by the plaintiffs is three years which began to run at the time of the crash, January 24, 1963. See Wash. Rev. Code § 4.16.080; Robinson v. Baltimore & S. Mining & Reduction Co., 26 Wash. 484, 67 P. 274 (1901).
Returning to Section 202 of the N.Y.C.P.L.R., the New York limitation periods apply to plaintiff O'Keefe, a New York resident. These periods are two years for the wrongful death action and three years for the personal injury or survival action, and they normally begin to run as of the date of death or injury. See N.Y. Decedent Estate Law § 130; N.Y.C.P.L.R. §§ 214, 203. Plaintiff O'Keefe and all of the other plaintiffs base their causes of action, however, on two theories of liability, one predicated on negligence and the other sounding in products liability. Insofar as their causes of action are based on negligence, they are not time-barred by the statutes of limitations of either New York or Washington, the plaintiffs' complaint having been filed within one year of the accident. On the other hand, the New York Court of Appeals held in Blessington v. McCrory Stores Corp. that a cause of action for breach of implied warranty of fitness for use "gets the benefit of the six-year limit of subdivision 1 of section 48 of the Civil Practice Act,
as being an implied contract obligation or liability." 305 N.Y. 140, 147, 111 N.E. 2d 421, 423 (1953). The Court of Appeals, relying in part on Blessington, subsequently held that insofar as a cause of action for personal injury was based on a claim of breach of implied warranty of a product, the period of limitation runs from the date of the sale.
See Schwartz v. Heyden Newport Chemical Corp., 12 N.Y. 2d 212, 237 N.Y.S. 2d 714, 188 N.E. 2d 142, motion to amend remittitur granted, 12 N.Y. 2d 1073, 239 N.Y.S. 2d 896, 190 N.E. 2d 253, cert. denied, 374 U.S. 808, 83 S. Ct. 1697, 10 L. Ed. 2d 1032 (1963). In 1969, the Court refused to overrule Blessington, albeit by the narrowest of margins, 4-3. See Mendel v. Pittsburgh Plate Glass Co., 25 N.Y. 2d 340, 305 N.Y.S. 2d 490, 253 N.E. 2d 207 (1969). The majority's opinion reads, in part, as follows:
The appellants argue that Blessington does not apply to the instant case because our decision in Goldberg v. Kollsman Instrument Corp., 12 N.Y. 2d 432, 240 N.Y.S. 2d 592, 191 N.E. 2d 81, created in favor of the third-party strangers to the contract, a cause of action in tort and not in warranty and, therefore, the three-year-from-the-time-of-the-injury, rather than the six-year-from-the-time-of-the-sale, limitations period should apply. We do not agree. When Goldberg was before us, we were confronted with the issue of whether or not a cause of action other than in negligence should exist in favor of those persons not in privity, with the contract of sale. After determining that the cause of action should exist, two avenues were open to us - either to establish, as other jurisdictions already had, a new action in tort, or to extend our concept of implied warranty by doing away with the requirement of privity. While there is language in the majority opinion in Goldberg approving the phrase "strict tort liability," it is clear that Goldberg stands for the proposition that notwithstanding the absence of privity, the cause of action which exists in favor of third-party strangers to the contract is an action for breach of implied warranty. The instant action being one for personal injuries arising from a breach of warranty, it is our opinion that Blessington controls and, therefore, the applicable Statute of Limitations is six years from the time the sale was consummated (CPLR 213, subd. 2).
We would merely add that both parties appear to agree, and we believe correctly, that strict liability in tort and implied warranty in the absence of privity are merely different ways of describing the very same cause of action.
The Court also relied on subdivisions (1) and (2) of Section 2-725 of the N.Y.U.C.C. to support its decision although that section and its 4-yr. period of limitation did not specifically apply to the case, which is also true here. See N.Y.U.C.C. §§ 10-101, 10-105.
The causes of action the Court of Appeals dealt with in both Blessington and Mendel were for personal injuries, although the former case did involve a death. However, whatever the degree of persuasiveness and logic in the dissenting opinion in Mendel,30 this court, to paraphrase Long, supra, concludes that it would be highly incongruous and unreal to have the rule for which the two cases stand apply where a person is injured but not where he is killed or to apply only to plaintiff Adler but not to the other plaintiffs. Hence, the plaintiffs in this case were required to assert their products liability claims within six years of the date of delivery
of B-406 to the Air Force. That date was August 17, 1956.
The plaintiffs argue that "even if New York's contract statute of limitations applies, the statute was tolled because of defendant's modification of the airplane throughout its existence."
While there may be some authority generally supporting this type of proposition,
the court is unaware of any precedent controlling the type of case at bar. More importantly, however, liability in this case, as will be seen hereinafter, relates specifically to one section of an empennage bulkhead, and there is no showing that this particular section of the bulkhead was modified in any way subsequent to the date of delivery and prior to the date of the accident. Indeed, the plaintiffs claim that the defendant was negligent in not modifying the bulkhead. Thus, whether or not the defendant modified or improved upon other parts of B-406 or even other sections of the bulkhead in question, there is no showing (or even contention) that any such changes affected or changed the section at issue from the date of delivery until the crash.
In view of the foregoing, the court is constrained to conclude that plaintiff O'Keefe's causes of action, insofar as predicated upon products liability, are time-barred under the law of New York, and the court reaches the same conclusion through application of Section 202, N.Y.C.P.L.R. with respect to the products liability claims of the other plaintiffs.
See Braniff Airways, Inc. v. Curtiss-Wright Corp., 411 F.2d 451 (2d Cir.), cert. denied, 396 U.S. 959, 90 S. Ct. 431, 24 L. Ed. 2d 423 (1969), modified on petition for rehearing, supra.
OCAMA B-52 structures engineer Kenneth Denney, who was sent to Maine as a member of the team investigating the crash, testified that he went to the site where the vertical tail of B-406 had been found and determined that the "fin had failed from the left side and folded toward the right side"
of the aircraft. The fin apparently bore markings which indicated that it had struck the vortex generators on the right horizontal stabilizer in separating from the plane. In any event, this case is unlike many tragedies involving men and their complex machines in that there is no dispute about how B-406 crashed; the court finds that it crashed due to a loss of directional stability as a result of the vertical tail's separation from the fuselage during flight. There is a very definite dispute, however, as to what caused the tail to break away.
The vertical fin (and rudder) on a B-52C, representing a total area of some 555 square feet and a total exposed area of 460 square feet, is attached to the fuselage at three primary points - two forward at "body station" 1655 and one aft at station 1719
- with all of the loads on the exposed area being transmitted into the fuselage through these three points. See, e.g., TM, p. 2364. There are bulkheads at both stations, with the function of the station 1655 bulkhead characterized as follows by its designer, the Airframe Components Division of the A.O. Smith Corporation:
Station 1655 Bulkhead is the empennage support bulkhead. Its function is to distribute fin loads, stabilizer jack screw loads, and ground handling loads into the body structure. The bulkhead must also sustain the longitudinal components of above mentioned loads, which are reacted by main and auxiliary longerons, stabilizer jack screw support tube, and lower beam stability reinforcement.
The 1655 bulkhead in B-406 was manufactured by the Rohr (Aircraft) Corporation from AOS 4405, 4335 - AMS 6428, 6434 steel alloy, with the crown of the bulkhead an open web type, pin-ended truss welded together with SW-151 welding rod. Attached to and part of the crown assembly were two fin attachment lugs.
* * *
On January 19, 1961, a model B-52B bomber,
number 53-390, was engaged in a celestial navigation run at an altitude of about 36,000 feet when it suddenly went out of control and crashed north of Monticello, Utah. The team which investigated the accident was able to determine that the "vertical fin failed first. It failed sharply to the right and contacted the horizontal stabilizer resulting in [the] vertical fin leading edge being smashed to the left. [The] fin failed in [the] steel bulkhead adjacent to fin terminals." Exhibit BH; TM, p. 2257. Parts of the crown of the 1655 bulkhead in plane 53-390, which was also manufactured by Rohr, were examined and tested thereafter in laboratories of both the defendant and the Air Force, as well as the National Bureau of Standards (NBS) in Washington, D.C. The examination and tests conducted at NBS under the auspices of its Chief of the Mechanical Metallurgy Section, John A. Bennett, led to a conclusion that the origin of the plane's downfall was a "fatigue crack".008 of an inch deep in a fillet weld located near the bottom of a vertical weld relief hole below and immediately outboard of the left fin attachment lug. The summary of the NBS report, dated April 16, 1961, reads as follows:
1. The primary fracture in the parts submitted had originated at a small fatigue crack in a weld bead.
2. The design of the part resulted in significant stress concentration, and this was an important factor in the initiation of the fatigue cracks.
3. A large section of the fractured member had been fabricated by building up with weld metal.
4. The relatively small growth of the fatigue crack prior to the initiation of fast fracture is attributed to the presence of the stress concentration and to the very low fracture toughness of the weld metal at the operating temperature.
5. It is believed that the presence of the fatigue crack caused a significant decrease in the static strength of the part.
6. The results of the examination provide no basis for believing that this failure is unique.
* * *
Mr. Denney testified and the court so finds that the 1655 bulkhead of B-406 was found generally intact with the exception of the left hand fin attachment lug which remained attached to the fin when it separated from the plane.
A subsequent examination at NBS led to a conclusion on its part that the fracture of the butt-welded heavy members just below the heretofore mentioned weld relief hole was "nearly identical"
to that which had occurred earlier in bomber 53-390 over Utah. For a rough sketch of the approximate planes of the two fractures of the butt welds and heavy members, see line A-A on Exhibit 21.
* * *
The plaintiffs argue that there was a defect in the fillet weld of the B-406 1655 bulkhead. They further argue that the defendant was negligent in "specifying and using a welded, rather than a forged, bulkhead; designing a bulkhead with welds in close proximity to each other; burning a weld relief hole through the bulkhead and thereby creating local stresses and areas of severe stress concentration; designing a bulkhead which, because of its geometry, the proximity of the welds, and existence of the weld relief hole, contained an area of severe stress concentration; improperly welding, so as to create a defect; failing to use accepted and available inspection techniques, such as zyglo, magnaflux, dye penetrant, and x-ray to discover the existence of flaws or defects; failing to discover the existence of the defect during the manufacturing process and before the bulkhead boxed section was completed; removing the web, or part of the web, from the bulkhead, in all production bulkheads, after the 1954 static tests showed that a crack developed at 75% of ultimate load; failing to install an access hole or otherwise make it possible to inspect the weld relief hole area on the bulkhead, in the design of the aircraft, even after the 1954 static tests showed that a crack developed there at 75% of ultimate load; failing to order or recommend the installation of an access hole, or other adequate means of inspection, after cyclic tests also demonstrated that a crack developed at the weld relief hole; failing to order or recommend the installation of an access hole, or other adequate means of inspection, after the Monticello, Utah accident in January, 1961 when the bulkhead failed at the weld relief hole area in the same place as the crack had developed in the 1954 static test and in the cyclic tests; failing to take other appropriate measures, such as recommending modifications to or replacement of the welded bulkhead, and relating such modifications or replacement to the Monticello accident and the occurence [ sic ] of cracks in the same area as the Monticello failure, following receipt of information and opinions indicationg [ sic ] a weakness of the bulkhead at the weld relief hole; failing to alert the Air Force to the existence of a safety in flight problem relating to the bulkhead, and failing to relate such a problem to the need for a retrofit on B-52s in service; affirmatively advancing theories on the cause of the Monticello accident to the Air-Force-industry investigating board on that accident, which theories were known to be tenuous, and contradicted by the known facts and circumstances of that accident; advancing 'overload' as a cause of the Monticello, Utah accident when the known weather data would not support such a theory and when it was known that the aircraft had not yawed to the left as it would have done in an overload situation; affirmatively excluding, from discussions of the 1655 bulkhead, reference to the Monticello, Utah, failure . . .; ignoring the warnings of Berman, Bennett, and others that there was an area of severe stress concentration in the bulkhead, that the Monticello fracture was not unique, and that other aircraft would fail in the same way, despite knowledge that there was no evidence of overload failure in the form of ductility or permanent deformation on the Monticello fracture face; [and] failing to adequately review and reconsider the manufacturing processes of the bulkhead, including the burning of the weld relief hole, following the Monticello accident and failing to discover those processes which were creating severe stress concentrations." Plaintiffs' Brief, pp. II-3 to II-5.
At the outset, the court takes note of the fact that, even if the plaintiffs' causes of action based on breach of warranty and/or strict liability are time-barred, as indeed the court has concluded, failure to consider these alternate theories on the merits would be to presume that the court's conflict-of-laws analysis is infallible and to do a disservice to both sides who so thoroughly prepared the case. Moreover, products liability is almost invariably related to negligence, or stated another way, "an honest estimate might very well be that there is not one case in a hundred in which strict liability would result in recovery where negligence does not." Prosser, The Assault Upon the Citadel (Strict Liability to the Consumer), 69 Yale L.J. 1099, 1114 (1960).
* * *
In referring initially to the law of Washington as the state with the most significant relationship to this case, the court has not found a recent aviation case tried and decided in a Washington forum, applying the law of that state, nor have the parties cited any such precedent.
Looking elsewhere then, perhaps the most succinct statement of the rules applicable to a case such as this is to be found in the opinion of the U.S. Court of Appeals for the Eighth Circuit in Nicklaus v. Hughes Tool Co., 417 F.2d 983 (8th Cir. 1969); it reads as follows:
A manufacturer has a duty to exercise reasonable skill and care in the design and manufacture of its product, commensurate with the risk of harm flowing from normal use of that product. . . . This duty also binds the manufacturer with respect to component parts incorporated into its final product, but manufactured by another. . . . As a necessary corollary, a manufacturer has an affirmative duty to make such tests and inspections, during and after the process of manufacture, which are commensurate with the dangers involved in the intended use of the product. . . .
Where, as here, the plaintiff's claim against the manufacturer sounds in both negligence and breach of implied warranty, he must establish by a preponderance of the evidence these basic elements: (1) that the product was defectively designed or manufactured . . . (2) that the defect was the proximate cause of the damage complained of . . . and (3) that the defect existed at the time the manufacturer parted with possession of the product . . . To permit recovery for negligent manufacture, ordinarily plaintiff must show, in addition to lack of reasonable skill and care in the process of manufacture, that the manufacturer failed to make a reasonable inspection or test to discover defects.
The law of Washington is in general accord. See, e.g., Di Pangrazio v. Salamonsen, 64 Wash. 2d 720, 393 P. 2d 936 (Dep't 2 1964); Sutton v. Diimmel, 55 Wash. 2d 592, 349 P. 2d 226 (Dep't 2 1960). In addition, on March 29, 1969, the Supreme Court of Washington rendered its en banc "decision to discard the terminology of 'implied warranty' and adopt the language of strict liability contained in the Restatement (Second) of Torts § 402A,"
. . . (1) One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property, if
(a) the seller is engaged in the business of selling such a product, and
(b) it is expected to and does reach the user or consumer without substantial change in the condition in which it is sold.
(2) The rule stated in Subsection (1) applies although
(a) the seller has exercised all possible care in the preparation and sale of his product, and
(b) the user or consumer has not bought the product from or entered into any contractual relation with the seller.
Since this decision, parties plaintiff have been free to assert strict liability as well as negligence and/or breach of warranty,
and the complaint herein has been amended by way of the Pre-Trial Order to incorporate this change of nomenclature.
The Washington Supreme Court had earlier adopted the language of Section 395 of the Restatement of Torts in Di Pangrazio. The section reads as follows:
A manufacturer who fails to exercise reasonable care in the manufacture of a chattel which, unless carefully made, he should recognize as involving an unreasonable risk of causing substantial bodily harm to those who lawfully use it for a purpose for which it is manufactured and to those whom the supplier should expect to be in the vicinity of its probable use, is subject to liability for bodily harm caused to them by its lawful use in a manner and for a purpose for which it is manufactured.
In addition to the specific acts of negligence alleged above, the plaintiffs also seek to rely on the doctrine of res ipsa loquitur. New York law determines whether this approach is permissible or not here,
and the New York Court of Appeals has concluded that "the introduction of some evidence does not render an instruction [to a jury] on res ipsa impermissible or improper." Abbott v. Page Airways, Inc., 23 N.Y. 2d 502, 514, 297 N.Y.S. 2d 713, 722, 245 N.E. 2d 388, 395 (1969) (emphasis added). The dilemma posed by the case at bar then is whether the significant amount of evidence adduced at trial on the issue of the defendant's negligence or lack thereof precludes the plaintiffs from attempting to rely on the res ipsa inference. The court concludes that it does not, if only because of the perhaps unfounded assumption that this juror is capable of distinguishing the logical nuances of such an approach. See generally Trihey v. Transocean Air Lines, Inc., 255 F.2d 824 (9th Cir.), cert. denied, 358 U.S. 838, 79 S. Ct. 62, 3 L. Ed. 2d 74 (1958), and the cases cited therein.
The within-the-exclusive-control-of-the-defendant requirement of the res ipsa doctrine is satisfied in a products liability context "if there is evidence of control by the defendant at the time of the negligent act complained of, i.e., creation of the defect . . . although the defendant's control is not exclusive at the time of the accident, provided plaintiff proves that the condition of the product had not been changed after it left defendant's control."
* * *
The defendant "takes the position that the decedents and the injured plaintiff assumed the risk of the happening of the accident" and "contends that decedents Morrison, Hanson, Simpson, Gabriel and plaintiff Adler were contributorily negligent in that they were in positions of responsibility and could have prevented the aircraft from going into an area of severe turbulence." Pre-Trial Conference Memorandum Submitted on Behalf of the Boeing Company, pp. 4-5. Unfortunately, the defendant has never seen fit subsequent to trial to provide the court with the benefit of its briefed reasoning on these points, which remain issues to be determined pursuant to the Pre-Trial Order.
Assumption of risk is a defense in a products liability case under Washington law. See, e.g., Stark v. Allis-Chalmers & Northwest Roads, Inc., 2 Wash. App. 399, 402, 467 P. 2d 854, 856 (Div. 1 1970). See also Northwest Airlines, Inc. v. Glenn L. Martin Co., 224 F.2d 120 (6th Cir. 1955), cert. denied, 350 U.S. 937, 76 S. Ct. 308, 100 L. Ed. 818 (1956). On the other hand, the Washington Supreme Court has pointed out that:
the fact that a danger is patent does not automatically free the manufacturer from liability, but does so only if the plaintiff voluntarily and unreasonably encounters it. Restatement (Second) of Torts § 402A, comment n at 356 (1965). It could never be said as a matter of law that [one] whose job requires him to expose himself to a danger, voluntarily and unreasonably encounters the same.
The court finds that plaintiff Adler and his fellow servicemen were doing nothing more than their duty on January 24, 1963, which was, of course, to conduct inflight training at low level. That this assignment involved peril is history; that plaintiff Adler and the plaintiffs' decedents did not assume the risk as a matter of law in this case is patent, and assumption of risk is therefore not a viable defense.
To be sure, it is a matter of record that the B-52 Flight Manual warns pilots to avoid clear air turbulence "whenever practical by all means available,"
that experienced pilot Bulli was well aware of this,
that turbulence is encountered more frequently at low altitude than at high altitude
and that disaster followed Colonel Bulli's decision to fly Poker Deck 8-3. It is also history that Colonel Bulli's guess as to the nature of the weather over Maine based on the discovered discrepancy between the Westover forecast and fact was wrong. However, before the conclusion could be drawn that the crew(s) of B-406 unreasonably exposed themselves to danger, there would at least have to be a showing that there were no reasonable grounds for the flight commander's guess,
that training flight(s) over Poker Deck 8-5 were still possible that afternoon (or that none at all were really necessary on January 24, 1963) and that the military minds aboard could have functioned in an abnormal fashion. The evidence adduced at trial does not even support inferences in this respect.
As for the defense of contributory negligence, even if Colonel Bulli's decision amounted to negligence (although the court makes no such finding
), such negligence would not bar the crew's present suit.
On the other hand, even if it is contributory negligence to fail to look out for danger when there is reason to apprehend it,
there is again no showing that those individuals among the crew(s) who should have had reason to apprehend danger could have countermanded the flight commander's decision. The court therefore concludes that contributory negligence is not a viable defense in this case either.
The court further concludes that the concept of "obviousness of the peril" is inapposite here. See generally Pike v. Frank C. Hough Co., 2 Cal. 3d 465, 473-474, 85 Cal. Rptr. 629, 634-635, 467 P. 2d 229, 234-235 (1970), cited with approval in Palmer v. Massey-Ferguson, Inc., 3 Wash. App. 508, 517, 476 P. 2d 713, 719 (Div. 2 1970). The record is utterly devoid of anything showing that the crew(s) had or should have had reason to suspect that B-406 would never return to Westover. Colonel Bulli, by way of example, had had more than 3,000 hours of flying time in B-52's prior to January 24, 1963.
* * *
American military planners apparently first conceived the design of the B-52 in the late 1940's. Intended by the government to be a high altitude, long range bomber, the first prototype built pursuant to government contract specifications by the defendant flew in 1952, and many of the later models are still in constant use today. At trial, there was no refutation of the boast on the part of the defendant's military programs manager that of the "major airplanes in the Air Force inventory, the B-52's . . . have had the best safety record." TM, p. 1699. But this obviously provides no solace to the plaintiffs, nor can there be any consolation in the fact that the crew(s) in B-406 on January 24, 1963 were learning to fly in "a completely new environment"
as a result of improved Sino and Soviet high altitude electronic detection capabilities.
There is no question, and the court so finds, that ultimate responsibility for the design and use of the B-52 bomber rests and always has rested with the United States government. The court concludes, however, that this fact, in itself, neither exonerates the defendant, nor has it in any way altered the defendant's duty as a manufacturer in this case
where there has been no showing that the defendant was totally oblivious of and/or aloof from the genesis of the design specifications in the first place or that the specifications represented either something less than the uppermost level of the art or a compromise of safety.
Cf. Littlehale v. E.I. du Pont de Nemours & Co., 268 F. Supp. 791, 804 n. 17 (S.D.N.Y. 1966), aff'd, 380 F.2d 274 (2d Cir. 1967). On the other hand, this does not mean that the court has overlooked the relevant axioms herein that the lighter the plane, the longer its range or the lower the altitude, the rougher the ride. See, e.g., TM, pp. 921, 934, 2794. Cf. id. at 1727. Indeed, the court finds that B-406 was not specifically designed and built for sustained flying at low altitude.
The court does not find that the evidence adduced in this case supports the plaintiffs' allegations that the defendant was negligent in specifying and using the welded 1655 bulkhead in B-406 containing welds in close proximity to each other and the weld relief hole(s)
alluded to heretofore. This is so even though the evidence clearly shows that a forged 1655 bulkhead is superior to the welded assembly
because the evidence shows, in addition, and the court finds, that heavy-duty, precision presses of the size and number capable of forging the quantity of 1655 bulkheads needed in the early to mid-1950's were unavailable in the United States. See, e.g., TM, p. 1704. To be sure, the record indicates that this situation had changed prior to January 24, 1963, that is, that there was no reason from a technological standpoint why B-406 could not have been retrofitted with a forged bulkhead before it crashed. See TM, pp. 1048, 2218. But, it is also clear, on the one hand, that the defendant was not called upon to effect a retrofit while it is anything but clear, on the other hand, that a retrofit would have made a difference in this case. To deal first with the defendant's inability to effect a retrofit, the defendant sent the following communication to the B-52 SPO on July 28, 1961 (or after the crash at Monticello, Utah):
. . . The contractor proposes a review of the fuselage station 1655 bulkhead to determine the practicability of a redesign to a forging as discussed herewith.
2. The subject bulkhead for the XB-52 airplane was designed of heat-treated welded steel in the upper portion for attaching the fin spar and the horizontal stabilizer jack screw. However, as tooling and the development of a heat-treating and welding technique made steel bulkheads unavailable in time for the first production airplanes, all B-52A airplanes and the first six B-52B airplanes were equipped at considerable cost with an aluminum alloy assembly in the upper portion of the bulkhead instead of the welded assembly. The aluminum alloy assembly was machined from a very large hand-forged billet. The bolted on fin lugs were of steel. Due to insufficient large forging press capacity at that time, the design was not changed from the welded assembly to a better and less costly aluminum alloy forging on the later B-52 airplanes. The welded steel assembly was continued, beginning with B-52B, 52-010. Several of these early airplanes with the machined aluminum alloy assembly in station 1655 bulkhead are still in service with no reported difficulties.
3. After approximately sixty B-52 airplanes were assembled, sufficient forging press capacity became available and many of the large welded steel assemblies were changed to aluminum alloy forgings, namely, the fuselage landing gear bulkheads. As very large precision forgings were relatively new to the aircraft industry and in order to avoid exceeding the press capacity, station 1655 bulkhead was not redesigned at that time.
4. To date, difficulties, such as small cracks, have been experienced in this welded assembly in approximately forty-five B-52C through B-52F airplanes. These cracks have been repaired by grinding and stop drilling. Production ECP 951 and 951M (KIT), and ECP 951-12-M (KIT) have been issued to improve this bulkhead.
5. At the time improvements were engineered for the B-52G, the cracks in the B-52C through B-52F bulkheads noted in paragraph 4 above had not been found. Accordingly, the station 1655 welded bulkhead for the B-52G was only revised to facilitate manufacturing. To date only one minor crack has been reported on the B-52G aircraft. However, with more flight time on the B-52G and B-52H airplanes, especially low-level time, additional structural maintenance is expected.
6. When this bulkhead was designed for the XB-52 and early production airplanes, the use of a heat-treated welded steel assembly was the only practical approach from a size, load, and cost standpoint. It has now been determined that the type of service presently being experienced on the aircraft and the planned long term usage will result in a significant increase in structural maintenance in this area. Further, the internal portions of the welded box section cannot be visually inspected, and the entire assembly is difficult and/or nearly impossible to repair in certain critical areas. We are of the opinion that the design of the bulkhead should be reviewed to determine if an increase in reliability of follow-on, follow-on B-52H airplanes is desirable considering their anticipated more severe service use. The objective in a redesign to less costly aluminum alloy forging of 7079-T6 with bolted on fin lugs of heat-treated steel would be trouble-free structure from a maintenance standpoint. The properties and consistency of forgings can be more uniformly controlled during processing than is possible on welded assemblies. The use of a forging will allow complete surface inspection of the part in the field and in many cases can be repaired with straps or angles by bolting and/or riveting if required. This new forging would also be designed to repair any station 1655 bulkhead on existing B-52 airplanes by replacing the welded assembly if required.
7. If action on this matter is not started until a follow-on follow-on B-52H contract is received, insufficient lead time will exclude the production incorporation of an improved station 1655 bulkhead. The results of a preliminary engineering investigation of the cost difference between the welded bulkhead and a forged bulkhead, including new tooling, appear favorable. Accordingly, the contractor requests authorization to submit an ECP for the subject change so that the cost difference, design practicability, testing requirements, and the feasibility of production incorporation may be determined.
Appended to the message was the following comment from the AFPRO:
To the best of our knowledge, the only signs of fatigue damage to the 1655 bulkhead are the area already covered by the 951 and the 951-12 fixes. We believe the cyclic testing done on the fin revealed satisfactory proof of the 951 change up to the limit of the testing done (10,000 cyclic test hrs.). We have never been completely convinced that 951-12 was even required, as these cracks on the cyclic test were stop drilled about midway thru the test and showed no progression from the stop drilling at the end of the test. We further believe that the contractor is optomistic [ sic ] in saying that the cost for the new forging would compare favorable [ sic ] with existing part because of the limited number of parts over which the die casts would have to be ammortized [ sic ]. Therefore, we cannot recommend the expenditure of effort on the proposed ECP.
On August 30, 1961, the Air Force denied the requested authorization.
See Exhibit Y; TM, p. 1741. Cf. id. at 1742-43.
The weld relief holes were clearly points of stress concentration, but the court does not attribute any specific significance to this fact, in itself, since all holes are points of stress concentration, namely, "[regions] of discontinuity in a structure where the stresses are higher than the average [load] P . . . divided by the average area [A],"
nor does the court find that it was negligence to have incorporated them into the design of the crown assembly.
Indeed, it appears that the weld relief holes were necessary to avoid creation of other stress concentrations. Cf. TM, p. 2119.
As originally designed, the crown assembly of the 1655 bulkhead centered on a solid plate of steel.179 of an inch thick. However, at the request of the defendant, the outboard panels on this plate delineated by the truss members were cut out to within approximately one inch of these members "with a view to eventual weight reduction on the upper crown section"
prior to static testing of a 1655 bulkhead in 1954 by the A.O. Smith Corporation. The plaintiffs, as set forth above, allege that it was negligence to have subsequently used bulkheads with the reduced webs when crack(s) developed during the static testing at approximately 75 percent of design ultimate load -- the load which will break a structure. But the crack(s) which developed were in an external fillet weld at the top of the box assembly, and the plaintiffs have not proven that the reduction of the area of the web plate caused the crack(s). Compare photo 1731 contained in Exhibit 25 with page 7 thereof and TM, p. 2795. Then again, there is no contention that the crash of B-406 stemmed from that particular fillet weld.
* * *
This is yet another case of extraordinary technological subtilization. B-406, a machine made up of untold numbers of component parts, was found in Maine in a shattered (and burned) condition by the investigators. Numerous parts fractured, resulting in myriad so-called fracture faces. Even the vertical fin, essentially anchored at only the three points heretofore described, left nevertheless numerous such faces when it tore away from the fuselage. Still, the scientists (and the lawyers) have melted down all of the fractures in this case to one through the butt weld and heavy members on the left-hand side of the crown of the 1655 bulkhead and, at least insofar as the plaintiffs are concerned, to what amounts to a speck in the fillet weld alluded to above.
The defendant does not dispute the plaintiffs' contention that there was a flaw in the fillet weld at or near the bottom of the forward weld relief hole below and immediately outboard of the left-hand fin attachment lug, and the court finds that the defendant could not dispute the contention since the record is replete with evidence supporting it. For example, Professor Charles Nash of the Department of Mechanical Engineering at the University of Maine examined the pertinent fracture face under magnification on January 29, 1963 and discovered what he characterized as a "definite flaw . . . a very dark flaw."
Mr. Bennett testified at trial as follows with respect to his subsequent inspection of the fracture:
Q . . . tell us what you observed on the part that was submitted to you. . . .
A . . . I observed the marked similarity between the fracture in this assembly and that which I had examined two years previously which was reported to be from the Monticello, Utah accident.
In each case there were chevron markings on the fracture surface which I felt gave me the opportunity to determine accurately the area where the fracture had started.
On looking more closely at that area I realized that the detailed features at the origin weren't similar to the Monticello accident exactly in that . . . I had observed a small fatigue crack at the origin of the Utah accident, and I didn't observe any such feature at the origin of this fracture.
When I say a fatigue crack I refer to what is known as a high cycle fatigue crack which has a characteristic appearance.
However, in the origin area on the fracture from the Maine accident I did observe a discolored area which I believed to be a flaw that had existed prior to the time that the fracture had occurred.
Q Would you describe the flaw to us?
A Well, it was a dark area which had a surface texture quite different from the rest of the fracture face. It was irregular in outline. . . . But the outstanding feature was the dark surface layer which I interpreted to be an oxide layer, and the definitely different texture of the surface as compared with the remainder of the fracture face.
Q Can you estimate the size of the flaw that you observed?
A . . . The depth of the flaw was about 500ths of an inch, somewhat longer than that on the surface . . . about a tenth of an inch. . . .
In other words, he found what he considered to be a "welding defect" in the 1655 bulkhead crown of B-406 as opposed to what he considered to be a "fatigue crack" in plane 53-390.
See TM, p. 590. See also id. at 726, 728, 729. Mr. Bennett further testified that he believed that the oxide formed when the bulkhead was originally heat treated and that it couldn't have formed subsequent to the crash on January 24th and prior to his examination. Compare page 330 of the trial minutes with page 329. The court finds in view of this and other testimony that the flaw or discontinuity originated during the process of manufacturing by the Rohr Corporation.
Rohr's manufacture of the particular bulkhead in question here was carried out under the auspices of the defendant's quality control personnel. But even if the fabrication of the part had been carried out completely independently, the defendant would still be responsible for any deficiency in it. See, e.g., Nicklaus v. Hughes Tool Co., supra; Boeing Airplane Co. v. Brown, 291 F.2d 310, 313 (9th Cir. 1961). The plaintiffs' position is that the court "can infer negligence, for which the defendant is responsible, from the mere existence of [the] defect." Plaintiffs' Brief, p. 10. Indeed, they produced no evidence at trial tending to detail the methods and procedures utilized by Rohr.
As has been alluded to above, the plaintiffs are entitled to rely on such a position "if the facts and circumstances of the case raise a reasonable inference that the defendant must have been negligent,"
but the mere permissibility of drawing such an inference does not necessarily make that inference decisive where inferences of a lack of negligence are equally permissible and of equal or greater reasonableness or where there is specific evidence tending to refute the plaintiffs' inference. Cf. Montgomery v. Goodyear Aircraft Corp., 392 F.2d 777, 779 (2d Cir.), cert. denied, 393 U.S. 841, 89 S. Ct. 121, 21 L. Ed. 2d 112 (1968); Trihey v. Transocean Air Lines, Inc., 255 F.2d at 827. The court finds that the existence of the flaw does give rise to a reasonable inference of negligence on the part of the defendant, but the court also concludes that it is not the most probable (and reasonable) inference and therefore it is not decisive in this case in view of the applicable standard of care.
One point the plaintiffs sought to emphasize via Mr. Bennett as a witness in rebuttal to the defendant's case is the "inherent variability"
of cast metal alloys, both in composition and mechanical attributes. The court finds that the homogeneity of weld metals is generally even less perfect and that this fact, combined with the inherent difficulty (if not impossibility) of drawing perfect weld beads under all circumstances, often results in welds with certain discontinuities.
Thus, it is not only conceivable, but probable, that discontinuities will occur from time to time in weldments which have been made as carefully as is humanly possible. In short, the inference asserted by the plaintiffs is by no means more reasonable than an inference that Rohr (and the defendant) fabricated the bulkhead with reasonable care.
The court, in fact, finds the latter to be more probable.
Even if the standard of care for aircraft manufacturers is or ought to be as near to perfection as possible and failure to live up to this standard is negligence, the court is not aware that the standard is perfection.
If this be true, then the court is unable to conclude that the defendant was negligent, that is, that the defendant failed to live up to a lesser standard of care, be it more-than-reasonable-care or otherwise, in the fabrication of B-406. However, this fact does not exonerate the defendant under the concept of strict liability, which will be dealt with hereinafter.
* * *
In Noel v. United Aircraft Corp., 342 F.2d 232 (3d Cir. 1964), the Court of Appeals held, in effect, that the defendant propeller system manufacturer was under "a continuing duty" to improve the system after sale and delivery. See especially the various opinions with regard to the petition for rehearing, 342 F.2d at 240-244 (1965). The Court of Appeals for the Second Circuit has not found it "necessary to adopt the rule" of Noel. See Braniff Airways, Inc. v. Curtiss-Wright Corp., 411 F.2d at 453. However, the court finds that the record in this case reflects an apparent assumption of such a continuing duty by the defendant, and the defendant's conduct will therefore be measured against this duty with respect to the plaintiffs' allegations of post-manufacture negligence.
The NBS's analysis of the crash of plane 53-390 in Utah was that it was caused by the fatigue crack in the fillet weld. The Technical Director of the Directorate of Flight Safety Research at Norton Air Force Base, one Sydney D. Berman, agreed with this analysis. The engineers and metallurgists of both the defendant and the Air Force did not agree. The defendant's initial analysis of the cause was that the vertical fin "failed catastrophically" in clear air turbulence and that "there was no evidence of fatigue." Exhibit BH; TM, p. 2257. The position of the Air Force (and, to a certain extent, of the defendant) was modified at a meeting with the NBS on March 15, 1961 with respect to the existence of a fatigue crack. See Exhibit 39. However, neither the defendant nor the Air Force ever changed their position that the crash was caused by overload and that the fatigue crack was of no real consequence,
although, at the same time, the defendant did recognize "that the sequence of failure suggested by the NBS [was] certainly possible." Exhibit 32, p. 1. See also TM, pp. 1993-94.
Mr. Berman apparently considered the position of the Air Force (and the defendant) to be "ridiculous,"
a "comic version;"
his friend, Mr. Bennett, "did not use any part of it as a basis for [his] judgment;"
the plaintiffs allege that the defendant was, in effect, negligent in taking such a stand. They refer, for example, to transcripts of telephone conversations between members of the defendant's staff during January and February, 1961
and to a recommendation in a June 16, 1961 inter-office memorandum of the defendant that "[all] future discussion of the 1655 bulkhead should exclude all reference to the Monticello failure"
in seeking to discredit the defendant's contention. However, a review of these exhibits in their entirety as well as the complete record in this case reveals not only that the defendant's evaluation of the Utah crash was reasonable (and therefore not negligent
), but that any dogmatism with respect to that accident was clearly on the side of Messrs. Berman and Bennett. Certainly, courts are generally not reservoirs of scientific wisdom, but in attempting to correlate the metallurgical phenomenon that is a fatigue fracture to a lawsuit coincidentally involving another A.O. Smith design, the Supreme Court of Washington has pointed out that such a fracture, "standing alone, does not establish a defect in manufacture." Arrow Transp. Co. v. A.O. Smith Co., 75 Wash. 2d 843, 850, 454 P. 2d 387, 391 (Dep't 1 1969). Then again, to infer that the NBS knew more about the design and use of the B-52 than the defendant and the Air Force or that the defendant should have readily concurred with the NBS's evaluation of the Utah crash even though a failure of a similar nature had not been reported earlier in hundreds of other B-52's
would be wholly unwarranted.
Prior to the Utah crash, the defendant had reevaluated its design of the 1655 bulkhead. The initial result of this reevaluation following the cyclic testing was ECP 951-0, promulgated as Air Force Technical Order (TO) 1B-52-1287 on January 13, 1961. See Exhibit 35. The projected modification was reinforcement "of the bar connector on the upper chord of bulkhead station 1655, to prevent possible fatigue failure." Shortly thereafter, on February 3, 1961, TO 1B-52-1350 was issued in order "to strength the station 1655 bulkhead to carry the fin load if cracks occur in the weld relief at water line 240 in the bulkhead." TM, p. 1036. See Exhibit 36. The modifications were carried out on the 1655 bulkhead of B-406 during March and November, 1961. See TM, pp. 1035, 1039. Neither, however, was directly connected with the box section of the crown truss. See, e.g., Exhibit 35, p. 2; Exhibit 36, p. 3.
The plaintiffs point to the defendant's recognition that "fatigue cracks of any kind, anywhere in the airplane, cannot be tolerated,"
and they argue that there was "sufficient notice" to the defendant "from the 1954 static test, from the cyclic tests, and from Monticello to warrant a careful inspection of the soundness of the bulkhead at the particular place where the Monticello bulkhead had failed and where the test bulkheads had developed cracks. This area could not readily be observed,
because it was in a boxedup section and no inspections were done in that area between the Monticello and Maine accidents . . . But [the defendant], as the expert in B-52 technology, could have and should have recommended installation of an access hole in the particular place that it then knew to be dangerous, or at least suspect, so as to make inspection feasible, as it did after the Maine accident . . ." Plaintiffs' Brief, pp. 64-65.
It is true that the defendant did not recommend that the B-52's be retrofitted with forged 1655 bulkheads after the Utah crash and prior to the one complained of herein
and that no fleet inspection was either recommended
or carried out during that period, but that subsequent to the loss of B-406 (and the one in New Mexico), the defendant not only proposed a procedure for inspecting the welded bulkheads,
it carried out a retrofit of the entire fleet with new forged bulkheads. But the court does not find that the defendant either ignored or negligently misinterpreted the Monticello crash and the other factors known to it prior to January, 1963 or that it was negligent in not taking the above steps prior thereto. Indeed, the very hindsight the plaintiffs seek to rely on belies their position that the defendant did not act as a more than reasonably prudent aircraft manufacturer would have under similar circumstances.
In summary then, the court finds that the plaintiffs have failed to sustain their burden of proof with respect to each of their specific allegations of negligence on the part of the defendant, either prior to delivery of B-406 in 1956 or to its downfall in 1963, and the court concludes that the evidence adduced at trial does not permit the drawing of a conclusive inference of negligence with respect to any of the allegations relating to the manufacturing process.
This case is thus a rare one in that the plaintiffs have proven the existence of a flaw, albeit hardly visible to the naked eye, but not that its existence is attributable to negligence on the part of the defendant manufacturer. Recovery therefore can only be predicated upon the concept of strict liability in tort,
which, of course, has not done away with the requirement of proving proximate cause.
See, e.g., Swain v. Boeing Airplane Co., 337 F.2d 940, 942 (2d Cir. 1964), cert. denied, 380 U.S. 951, 85 S. Ct. 1083, 13 L. Ed. 2d 969 (1965); Vandermark v. Ford Motor Co., 61 Cal. 2d 256, 260-261, 37 Cal. Rptr. 896, 898, 391 P. 2d 168, 170 (1964); Greenman v. Yuba Power Products, Inc., 59 Cal. 2d 57, 62, 27 Cal. Rptr. 697, 700, 377 P. 2d 897, 900 (1962). The plaintiffs clearly recognized this; to quote their astute counsel, the "critical question here . . . is whether the accident was caused by overload or whether it was caused by material failure." Post-Trial Hearing Minutes, p. 9. See also plaintiffs' Brief, p. 9. And the burden of proof with respect to this issue lies, of course, with the plaintiffs in a case such as this. See, e.g., Krause v. Sud-Aviation, Societe Nationale de Constructions Aeronautiques, 413 F.2d 428, 431 (2d Cir. 1969). Cf. Montgomery v. Goodyear Aircraft Corp., 392 F.2d at 779.
The defendant contends, inter alia, that it built B-406 pursuant to government specifications which contemplated use of the aircraft as a long range, high altitude bomber; that the government thereafter changed this mission requirement; that the Air Force subsequently encountered greater turbulence, hence loads and resultant stresses, at low altitude; that the pilot of B-406 chose to fly it into an area where severe turbulence had been forecast and which was, in fact, encountered in the form of a mountain wave; and that this turbulence resulted in a load on the vertical tail which placed greater stress on the 1655 bulkhead than it was designed to withstand. See generally TM, pp. 35-49. The plaintiffs argue, among other things, that the flaw in the weld substantially weakened the bulkhead by causing a crack to propagate therefrom through the butt weld and eventually through the base metal; that B-406 was designed to withstand severe turbulence; that the topographical and weather conditions in Maine were not conducive to the creation of gusts strong enough to knock the tail off; that the smallest margin of safety with respect to overload was at another point in the fuselage; that the aircraft did not yaw to the left as it should have if the fin had separated as a result of overload; and that if the 1655 bulkhead had experienced an overload, the fracture face in question would have shown evidence of ductility in the form of permanent deformation, which it did not. See generally plaintiffs' Brief pp. 1-2, 4-5.
* * *
The vertical tail for the X and Y prototype B-52's was designed in accordance with Army Air Forces Specification No. R-1803-6B,
promulgated April 2, 1946,
which read, in pertinent part:
E-3a. Gust Load on the Vertical Tail Surfaces. The limit load shall be computed from a sharp edged gust of . . . 50 feet per second, acting normal to the plane of the tail at the level-flight high-speed of the airplane. . . .
Based on this requirement, the defendant had calculated the design ultimate gust load
for the exposed area of the vertical tail on the prototypes to be 80,000 pounds. A stress analysis using this figure resulted in a calculated gross area stress of 108 thousand pounds per square inch (ksi) for that part of the box section of the 1655 bulkhead in question in this case. See Exhibit B, pp. A-71, A-73.
On December 14, 1954, the 50 fps sharp edge gust specification of R-1803 was superseded by Air Force specification MIL-S-5702
requiring computation of limit load on the vertical tail through use of a conceptual "discrete gust" in the configuration of a one-minus-cosine curve 25 chord lengths long
and with a derived equivalent velocity (Ude) of 55 fps. Initial reevaluation of the design ultimate side gust load on the exposed vertical tail pursuant to MIL-S-5702 resulted in the figure of 64,570 pounds for the B-52A. See, e.g., Exhibit 24, p. F-2. Continuing analysis eventually led to a calculated design ultimate side gust load for the entire vertical tail for the B-52C, including B-406, of 87,500 pounds. See Exhibit BD, pp. 3E-6, 3E-6a. The exposed side area of the tail, to wit, 460 square feet, represents approximately 83 percent of its total area, yet analysis apparently showed that the exposed area was subject to 86.5 percent of the total side load. See TM, pp. 2360-62. The court therefore finds that the design limit side gust load for the exposed area of the vertical tail of B-406 was 50,500 pounds
and that the design ultimate side gust load for the same area was 75,700 pounds,
both figures rounded off to the nearest hundred pounds. The defendant's dynamic loads expert, John Dempster, testified as follows with respect to this ultimate load:
Q . . . can you tell . . . what would happen to a B-52C aircraft if that aircraft sustained the horizontal load on the exposed tail in excess of the 75,700 pounds . . . design ultimate load . . .?
A If the B-52 vertical tail load exceeded 75,700 pounds on [the] exposed area, that would constitute an overload condition, and we would expect structural failure in some part of the B-52 that was originally designed from a strength standpoint for the fin gust condition . . . that's the vertical tail, . . . the portions of the aft body, . . . the portion of the wing center section and portions of the 1655 bulkhead. . . .
The court finds that, based on the design ultimate load of 75,700 pounds, the box section of the 1655 bulkhead at issue herein was designed to withstand gross area stress in the amount of approximately 102 ksi. Cf. TM, pp. 2365, 2370, 2568-69.
Between April 16, 1965 and February 15, 1966, the defendant conducted a test program, one of two primary objectives of which was to "[determine] B-52 transfer functions for loads response to atmospheric turbulence [for comparison] with theoretical transfer functions." Exhibit 62A, p. 4. The plane used for the testing was a B-52E, number 56-632, "standard in all important structural aspects"
and equipped, among other things, with a gust boom and boommounted probe on the nose containing pressure, acceleration and attitude transducers. "Gust velocities and airplane responses were processed to obtain power spectra and various statistical parameters. Response transfer functions were computed by both direct and cross-spectral techniques." Id. See generally Exhibits 62B, 62C and 62D.
Load transfer functions derived from experimental flight test data were required for comparison with theoretical load transfer functions used in the dynamic analysis of the B-52 airplane. A transfer function was defined as the square root of the ratio of the power spectral density of the output function to the power spectral density of the input function. The input function for this testing was considered to be the vertical or lateral gust velocity. Output functions were the structural loads, stresses, and accelerations recorded throughout the B-52 test airplane. Transfer functions of angles of pitch, roll, and yaw were also determined.
Data for use in computing load transfer functions were selected from 36 test conditions, each approximating 10 minutes in duration, conducted during a total of six flights. The best 5-minute part of each condition was ultimately chosen for processing. From three 10-minute conditions, a total of six conditions were edited for comparative and cross spectral purposes. Gross weights, altitudes, airspeeds, fuel distribution, and associated items . . . [precede] the data plots for each test condition. . . .
Also . . . preceding the data plots for each condition . . . [are meteorological] data, recorded at the nearest weather station at the closest time to the time interval during which each transfer function test condition was conducted . . . The information, including temperature, humidity, barometric pressure, and winds aloft data were provided in the form of Adiabatic Charts . . . Winds aloft were not obtained from the B-52 test instrumentation because of unsatisfactory Doppler radar operation. Data pertinent to and correlated with the load transfer function test conditions were extracted from the Adiabatic Charts . . .
At trial, Mr. Dempster calculated the transfer function for correlating random lateral gusts of a magnitude of 70-130 fps in an area of severe turbulence with loads on the vertical tail of a B-52 flying at 280 knots, an altitude of 500 feet above the terrain and a gross weight of approximately 350,000 pounds
to be 1150 pounds per foot per second, rounded off to the nearest ten pounds. See TM, pp. 2434-37; Exhibit BP9. He arrived at this result by using the root mean square of the side bending moment at vertical fin station 135, with data therefor having come from a test flight of plane 56-632 over western Kansas on November 17, 1965 at a pressure altitude of 3,700 feet, gross weight of 349,000 pounds and an airspeed of 270 knots. See TM, pp. 2636, 2744-47; Exhibit 62B, Fig. 952. Then by simply dividing the design ultimate gust load for the vertical tail of B-406, 75,700 pounds, by the calculated transfer function, 1150 pounds per fps, Mr. Dempster reached a conclusion that a peak, true gust
velocity of 66 fps
[or more] on January 24, 1963 at the specific time and point in question here would have created an overload
on the vertical tail of B-406. See TM, pp. 2440-41, 2509-11, 2711.
The plaintiffs' turbulence consultant, U. Oscar Lappe, testified in rebuttal that in his opinion Mr. Dempster's directspectral approach was not the proper procedure for computing the pertinent transfer function, but that he should have used a cross spectrum, primarily "[because] it discriminates against incoherent noise," that is, "[any] signal source which is not directly relatable to the input, any unrelatable signal appearing in the output function which is not directly relatable to the input function or caused by the input function." TM, p. 3295. Cf. id. at 3340, 3353-54, 3356, 3433. He testified further that a cross spectral determination of the transfer function leads to a conclusion that a peak gust velocity of 80 fps would have been required to create an overload on the vertical tail of B-406 and that, furthermore, in his "best opinion," 100 fps plus or minus 10 fps would have actually been necessary to produce ultimate load.
See id. at 3302, 3303, 3307.
* * *
All but 17 percent of the exposed area of the smaller vertical tail of a B-52H, number 61-023, which had been "highly instrumented to measure the response of the airframe to turbulence at low-level",
was broken off
during a test flight as the airplane passed adjacent to "East Spanish Peak" (located near Walsenburg, Colorado) in the Sangre de Cristo Mountains. See TM, pp. 2070, 2386. This occurrence led to a total of 61 test flights flown under the joint auspices of the defendant and the Air Force between March 9 and April 28, 1964 "in search of atmospheric turbulence in the Sangre de Cristo mountains, generally between Albuquerque, New Mexico, and Pueblo, Colorado." Exhibit W19, p. 10. See id. at 45. The test plane was an Air Force F-106A fighter equipped with a gust boom. Among other things, the recorded lateral gust velocities, which ranged from a low of 25 fps to a high of 175 fps, were subsequently plotted graphically as against the wind velocities recorded by the U.S. Weather Bureau station at Alamosa, Colorado. See Exhibit W19, p. 78.
In 1967, the Department of Atmospheric Science of Colorado State University published what apparently was the doctoral dissertation of one Harry P. Foltz
in which he sets forth a methodology for predicting clear air turbulence
as it relates to mountain lee waves. See, e.g., TM, pp. 1378-79. The tool devised by Dr. Foltz for determining the degree of turbulence is a graph showing the wavelength of a lee wave as a function of the wave's maximum vertical velocity. See Exhibit W11, p. 94. In order to determine whether a mountain wave exists (or existed) in the first place, it is necessary to determine, among other things, the profiles of the given terrain and of the wind(s) and temperature(s) in the atmosphere.
At trial, the defendant's chief meteorologist, William Moreland, concluded from the weather data gathered on January 24, 1963 that the wind(s) were blowing essentially from the northwest at approximately 312 degrees at the time of the crash of B-406.
He also concluded from two geological survey maps, Exhibits W7 and 50, that the plane crashed in an area of mountain ridges, and he constructed a profile of the terrain (Prong Pond Mountain -- Elephant Mountain -- Horseshoe Pond) in the vicinity of the crash site based on an orientation of 312 degrees. See Exhibit W8; TM, p. 1351. In addition, Mr. Moreland plotted a psuedo-adiabatic diagram
and wind profiles
for January 24, 1963 based on rawinsonde temperature and wind speed readings obtained by weather stations at Portland and Caribou, Maine and Maniwaki, Canada. He then interpolated in order to obtain temperature and wind profiles upwind from Elephant Mountain at approximately 2 p.m. on January 24th.
Relying on these graphs, combined with application to Elephant Mountain of a theory apparently first propounded by British scientist R.S. Scorer that an important factor in mountain wave formation is the relationship of a mountain's height h to its half-width b, led to a conclusion on Mr. Moreland's part that a "strong mountain wave" existed in the area where B-406 crashed. See, e.g., TM, pp. 1278, 1629.
Having concluded that a mountain wave existed at the time and place of the crash, Mr. Moreland followed the Foltz methodology
in computing the wavelength and vertical velocity to be seven kilometers and 10.9 meters per second, respectively. See TM, p. 1327. See also Exhibit W17. These two calculations plot a point on the Foltz graph within the severe turbulence curve,
thereby leading to a conclusion by Mr. Moreland that that degree of random clear air turbulence existed at and around Elephant Mountain. See TM, pp. 1278, 1629. Mr. Moreland also sought to correlate the F-106 data to the conditions in Maine on the basis that the "absolute height of . . . mountains doesn't make any difference in . . . mountain wave problems"
and that the Elephant Mountain h over b ratio of.6 was comparable to such ratios for the various Sangre de Cristo peaks. See TM, pp. 1342, 1384. Cf. Exhibit W11, p. 135. Since two lateral gust velocities of 70 fps, one of 80 fps and one of 135 fps had been recorded when wind velocities over the Sangre de Cristo range approximated 40 knots,
Mr. Moreland stated his ultimate conclusion as follows:
. . . we can't specify exactly . . at the exact time of the aircraft accident . . . the specific speed that we had, but I think we can state, in my opinion, that the gusts based upon my analysis, first that there was a mountain wave occurring, secondly that qualitatively we know there was what is defined as . . . severe turbulence, and thirdly we know that the wind speed was 40 knots, so that we can estimate that the range of values of the maximum lateral gusts would be between 70 and 130 feet per second.
While repeatedly disclaiming any expertise with respect to mountain waves,
Mr. Lappe nevertheless also disclaimed both Mr. Moreland's thesis and his attempt to prove it. Whereas Mr. Moreland was "absolutely sure"
that a mountain wave with associated severe random turbulence existed in the vicinity of Elephant Mountain at or about 3 p.m. on January 24, 1963, Mr. Lappe opined that it was "extremely unlikely"
that this was so. He did agree with Mr. Moreland's evaluation that the wind direction was generally conducive to the formation of a mountain wave,
but he was skeptical about the validity of the Scorer premise with respect to h and b;127 he did not find the "topographical conditions in particular" in Maine in conformity with the topography of areas where mountain waves have been reported on "numerous occasions";
he estimated the velocity of the winds aloft to be 30 knots which he did not consider strong enough to create a mountain wave;
and he did not think that a wind speed gradient of two knots per one thousand feet of altitude existed at Elephant Mountain.
See TM, pp. 3232-33. Mr. Lappe concluded, on the other hand, that even if a mountain wave existed, he "would expect" peak lateral gust velocities of about 50 fps, with a margin of error of perhaps plus or minus ten percent or a little more. Id. at 3277. See id. at 3289.
Mr. Lappe, among others, sought to point out the significant degree of subjectivity still involved in his particular field of endeavor, yet he ignored Colonel Bulli's testimony in making his judgment(s)
even though pilots are apparently still a primary source of information with regard to turbulence. The court, for its part, could not justify not referring to Colonel Bulli's testimony or, for that matter, the testimony of the two other people who were fortunate enough to live through January 24, 1963 in Maine and who appeared at trial. Colonel Bulli testified as follows:
Q This turbulence that you . . . had been experiencing all along . . gave you a bumpy ride up to this point, did it not, sir?
A Yes, that's correct.
Q When you experienced this increase in turbulence, when you were at the point three miles from Elephant Mountain, this was a sudden increase in the turbulence, was it not, sir?
A Yes, it was a jolt rather than a steady increase, if I remember correctly.
Q Somewhat unexpected, was it, sir?
A Yes, like a bump in the road, you know.
Q As a matter of fact, it was a fairly severe bump, wasn't it?
A Yes, it was. To the best of my knowledge -- recollection, that is.
. . . Q I think that you referred to these in your statement . . as "real jolts," didn't you?
A That's right, a jolt. Like you hit a chuck hole in the road when you are driving a car.
Q Subsequently, after receiving those two real jolts, you subsequently hit a few more real jolts, didn't you, some time about five seconds and a half a mile later?
A That's what I said before, and my recollection seems to verify that. . .
Q There were continuing real jolts that you were receiving at that time?
A Yes, sir. They were intermittent, yes.
Q These were fairly severe jolts, Colonel?
A I would say they were, yes. It's hard to categorize them. But again, I can only state they were like hitting a chuck hole in the road, you know. You feel the airplane give a little bit.
Q At this point in time you were still at the altitude of 500 feet; is that correct, sir?
A We were 500 feet above the terrain. . . .
Q Shortly after you established your climb attitude you then encountered very heavy turbulence, didn't you?
A I believe we continued to get these heavy jolts, yes.
Q As a matter of fact, you felt very heavy turbulence, did you not, Colonel?
A Yes. That's what it says here and I am sure that is correct.
Q And it was just after you felt this very heavy turbulence that you heard this loud, sharp explosion that you described . . .?
Q Immediately after that, the aircraft went into this 40 degree bank; is that right?
A That's correct.
Q At that point, sir, I believe you looked out to see whether or not you were going to clear this mountain that we have been referring to?
Q You had cleared it, in fact, had you not, sir?
Q At the time you were three miles away from . . . Elephant Mountain, was it your intention to continue on a straight path over the peak of that mountain?
A Yes, it was.
. . . Q Eventually, your aircraft did in fact proceed over the peak of this 2300 foot
A Oh, yes, very definitely.
Plaintiff Adler, who had been an airman for over nine years at the time of the crash of B-406, testified on direct examination as follows with regard to the plane's last seconds aloft:
Q . . . [had] you been through severe turbulence in the past?
A Yes, sir.
Q How many times?
A I had been flying the B-52 since 1959 - I suspect it must have been a dozen times. . . .
Q . . . getting back to your flight, did you at this time experience anything like what you have just described to us as severe turbulence?
A Yes, sir.
Q Tell us what.
A We had been climbing for, I don't know, half a minute, maybe, and the aircraft started being shaken around quite a bit . . . . I was being tossed around. I was getting scared. . . . over a period of a very few seconds, I was being shaken up pretty badly. I would describe that as severe.
Q And you described the length of time that that existed a very few seconds?
A . . . it could have been maybe two seconds, maybe three seconds, I don't know, maybe five seconds, it was just a very short period of time.
. . . Q . . . you felt this very brief period of severe turbulence, and tell us what happened.
A After this turbulence started, the aircraft suddenly snapped into a nose low -- well, I know a right wing low bank. I know that because I was tossed over in my seat on the right side, my shoulder harness had snapped taut. A shoulder harness will do these things with a sudden burst . . .
Mrs. Howe, who was present in Greenville on January 24, 1963, testified that she remembered that day "very clearly"
and that it was "blowing gustily,"
it was "extremely gusty . . . throughout the day."
She also testified that her personal flying experience in the vicinity of Elephant Mountain, albeit in a very light aircraft, had been that the air was "[very] turbulent even on a clear day." TM, p. 3328.
This court readily leaves the esoterica of mountain waves, in general, to another niveau. The court also concedes that the terrain immediately to the east of Moosehead Lake in Maine, over which B-406 flew, should not remind many people of the Himalayas or even the Rockies, but the court finds that the topography in question here is clearly mountainous and rugged within the general meaning of those terms. See, e.g., TM, 3328; Exhibit W7. The court further finds what the evidence in this case clearly shows, namely, that an area of severe random clear air turbulence existed to the lee of Elephant Mountain between the hours of 2 and 3 p.m. on January 24, 1963 just as could have been expected for that time of year
and as the Westover Metro had predicted, the historical weather data tended to show and the airmen discovered; and that B-406 flew into this area of severe turbulence
wherein the vertical tail separated from the aircraft. Compare, for example, the testimony of Colonel Bulli with Exhibit H31.
* * *
In Mr. Bennett's opinion, the flaw in the internal fillet weld of the 1655 bulkhead "would significantly decrease the strength of the part." TM, p. 542. Indeed, his opinion prior to trial had been that "there had been some slow propagation of [a] crack from the bottom of the flaw . . . until the crack became large enough so that fast fracture occurred . . . over the entire face." Id. at 543. This position was modified, more or less on the eve of trial, when a viewing of electron micrographs of the pertinent fracture face,
apparently for the first time, led Mr. Bennett to conclude that the fracture occurred not in two, but rather three stages, the first being the "slow propagation" from the flaw, supra, the second being rapid propagation of a crack along the entire length of the butt weld down to the interface of that weld with the base metal, which crack then stopped, and the third being rapid propagation of a crack through the base metal. See id. at 563. See also id. at 537, 540-41. This change of opinion did not alter Mr. Bennett's conclusion, however, that the fracture "definitely was not the result of a one-time overload"
despite the following facts: (1) He did not know of his own knowledge the specific part of the bulkhead from which the fracture face came;
(2) he only saw one half of the fractured crown;
(3) he did not know the loads and stress criteria for the B-52;
(4) he relied on the "professional guess" of a physicist with the Naval Research Laboratory who "lacked certain important pieces of information to make an accurate estimate" that approximately 125,000 psi were necessary to start a fast fracture through the weld metal;
and (5) he never determined the amount of stress necessary to cause the base metal to fracture once the crack had propagated through the weld metal.
The defendant's expert metallurgist, Professor Alan Tetelman, stated that "fracture mechanics is an engineering discipline that has been developed over the last ten years that attempts to predict the strength of a cracked part in terms of three very simple parameters: The stress acting on the part, the size of any flaw that is in the part, and the toughness of the material." TM, p. 1167. See also id. at 2881. Under Professor Tetelman's supervision, the same steel alloy -- 4335 -- that had been used for the heavy members of the crown of the B-406 1655 bulkhead was heat treated to a strength of 200 ksi.
Four tests to determine the metal's toughness at minus 65 degrees Fahrenheit, the "Air Force standard for the lowest operating temperature that a structure would have to be expected to see in service,"
resulted in values ranging from 82,400 to 91,000 psi square root inches
with an average value of 86,800 psi square root inches. See TM, p. 1179.
Professor Tetelman testified that he completely concurred with Mr. Bennett's amended opinion that the fracture at issue herein occurred in three stages, with the crack having stopped after having gone through the weld metal. See TM, pp. 1188, 1233-34. However, he contended that the "real question" of whether or not the bulkhead would fail was not whether the crack went through the weld metal, but rather whether there was enough stress to fracture the base metal of the heavy members;
or stated another way: "There are different stages in a fracture. In this case there were three, and the ultimate strength of the part is going to be the highest of those three strengths that you have to overcome to make the part break in two." TM, p. 1249. The depth of the base metal was calculated to be.91 of an inch,
the depth of the crack in the butt weld.13 of an inch. See, e.g., id. at 1189. Knowing the parameters of fracture toughness and the size of the crack, Professor Tetelman's "best judgment" was that the average gross area stress
acting on the B-406 1655 bulkhead at the time it failed was 108,800 psi. See TM, pp. 1190, 1222, 1241, 1242, 1243, 1255, 1259. The court finds that this average figure is clearly in excess of the design ultimate load of 102 ksi for the pertinent section of the bulkhead and would be an overload, as would the lowest of the four specific gross area stresses calculated on the basis of a temperature of -65 DEGREES F.
In support of their contention that the crash of B-406 was not caused by overload, the plaintiffs sought vigorously throughout the trial and afterwards to have their Exhibit 1, which was characterized by their counsel as "the most important thing we have ever seen on this case"
and which apparently was turned over to them by the defendant on the eve of trial, admitted into evidence on the grounds that it represents an admission on the part of the defendant and that it also is a business record of the defendant within the meaning of 28 U.S.C. § 1732. However, the plaintiffs did not and have not persuaded the court that the document (D6-6170), entitled "Material Investigation -- Fin Terminal Attachment -- Body Station 1655 Bulkhead -- Model B-52 Aircraft,"
is admissible in toto on either ground. See generally the minutes of the post-trial hearing. However, be that as it may, portions of the document were duly admitted into evidence during the course of the trial, to wit, Exhibits 1A, 3, 19 and 21, and the court has since reviewed generally the remainder of the exhibit, which might be termed a melange relating not only to the crash herein, but also to the heretofore mentioned accidents at Monticello, Utah and Mora, New Mexico as well as to a discovery made during inspection of a 1655 bulkhead of a B-52 at Amarillo, Texas on February 7, 1963. In particular, the court reviewed those parts of the document considered most important by the plaintiffs,
and the court does not find that "it confirms, in specific detail, the plaintiffs' theory"
or that it "resolves the big question in the case: causation."
Page 7, for example, to which the plaintiffs specifically refer, contains the "Metallurgical Findings of all Investigators" with respect to the Monticello crash. The only one of these brief summaries which conceivably could otherwise readily qualify as an admission in this case, namely, the one attributed to the defendant, reads, however, as follows: "The final failure occurred due to an overload. No evidence of fatigue was found." Moreover, the authenticity, i.e. reliability, of the document is by no means inescapably clear.
The orientation of the diagram on page 4 (Exhibit 19) for example, to wit, "Looking Aft," is wrong. Compare Exhibit 15; TM p. 110. As drawn, the sketch shows the right hand fin attachment lug areas of the 1655 bulkheads of plane 53-390 and B-406 as having failed first, whereas the evidence clearly shows that failure occurred on the left hand side first.
* * *
The plaintiffs also seek to rely on the rebuttal testimony of Mr. Stieglitz to the effect that the crash of B-406 was not caused by overload. See, e.g., TM, p. 2941. Mr. Stieglitz based his opinion essentially on Mr. Bennett's analysis in general and, more specifically, the claim that the fracture face in question did not show much evidence of ductility;
on his understanding of the aft fuselage stress analysis for the B-52;
and on the absence of any reported yaw of B-406 at the moment the vertical tail blew off.
See TM, pp. 2941-42.
Mr. Bennett had defined what he was talking about in the following manner:
In cases of service failures particularly, if a part is subjected to excessive load it will generally deform. Most structural metals are reasonably ductile. They have a reasonable ability to deform under load, and if a part is simply loaded beyond its yield strength it will deform, and this deformation is generally observable after, say, an accident or catastrophe of some kind, so that I would say the typical evidence of overload is excessive deformation, permanent deformation of the metal.
However, Mr. Bennett had also testified that, in general, the ductility of a low strength metal will be "much greater" than that of a high strength material such as the 4335 steel alloy used in the 1655 bulkhead. See TM, p. 755.
Mr. Dempster had testified that, from a theoretical or analytical standpoint, the stress analysis of monocoque section 47 running from body station 1237 aft to station 1655 indicated to him that a gust overload on the vertical tail would result in a failure at the area with the lowest margin of safety, to wit, the skin between the upper and the lower longeron, before the 1655 bulkhead would fail.
The court concurs with Mr. Stieglitz's evaluation of the record in this case that buckling or wrinkling of section 47 skin of B-406 was neither reported nor testified to,
but the court by no means draws the conclusion implicit in Mr. Stieglitz's testimony - namely, that there was no such buckling or wrinkling - in view of the shattered and charred condition in which the fuselage was found in Maine.
Plane 61-023 apparently yawed a little when struck by the gust which broke off its vertical tail. See TM, pp. 2070, 2388. Mr. Dempster testified that yaw is not beyond the sensitivity of a pilot, although a pilot has "less reference" in yaw than he does in roll and pitch. See id. at 2646-47. See also id. at 2627. Cf. id. at 2411. Mr. Dempster characterized an angle of from 3 to 5 degrees as a "big yaw". Id. at 2649. This court is not persuaded that the absence of any reported yaw of B-406 as opposed to the reported 40 degree bank to the right with a nose down attitude
is significant in view of the aircraft's size, weight, speed and altitude in an area of both severe turbulence and mountains and thus the crew members' understandable immediate anxiety and even fright,
with at least one of them apparently ejecting for the first time.
Then again, it is not clear that a yaw should be expected under all circumstances when a vertical fin fails as a result of a lateral gust. Cf. TM, pp. 2066, 3046.
* * *
Neither Mr. Bennett nor Professor Tetelman was able to say how long the crack in the butt weld had stopped before the base metal fractured. See id. at 564, 764, 1240. If the pause at the interface had been more than ephemeral, Professor Tetelman testified it would have taken "one blow" of 108,800 psi to break the B-406 1655 bulkhead. See id. at 1243-44. Indeed, the right-hand crown assembly of the bulkhead, in which "[some] cracks were found in the vicinity of the weld relief holes,"
was static tested after the accident at a temperature of -65 DEGREES F with the following result:
Failure of the upper right hand portion of the B.S. 1655 bulkhead assembly of B-52C, 53-406, occurred at 105 percent of design ultimate load. Failure did not occur at any existing crack.170
At the same time, a vertical fin from a B-52A was static tested in combination with a fleet bulkhead from a B-52F. During the destruction phase, the assembly was cooled to -65 DEGREES F. The result:
The load applied to the vertical fin at the time of the failure of the Station 1655 bulkhead was 81,100 pounds. . . .
The failure of the Station 1655 bulkhead occurred at the left-hand fin attach lug. The lug pulled out of the crown truss in a similar manner to the failures experienced on fleet airplanes. . . .
* * *
In view of the foregoing, the court is constrained to conclude that the plaintiffs have not met their burden of proving by a fair preponderance of the credible evidence that the crash of B-406 was proximately caused by the flaw at the weld relief hole. To be sure, "[fact] finding does not require mathematical certainty,"
and this rule implies that a party plaintiff's burden of proof does not involve any such requirement. However, the court finds that whatever mathematical credibility exists in this case overwhelmingly supports the defendant's theory that overload proximately caused the crash of B-406.
On the other hand, however distinguished the careers of the plaintiffs' "experts" may have been and whatever their general reputations might be, the court finds the way in which they reached their conclusions in this case to have been incredibly unscientific, for the most part -- not because they were generally unfamiliar with the B-52, which is understandable, but because they made little, if any, effort to find out before stating their opinions at trial.
In summarizing this opinion, which represents the court's findings of fact and conclusions of law pursuant to Rule 52 of the Federal Rules of Civil Procedure, with regard to the specific issues in the Pre-Trial Order which were tried, the court concludes that the law of Washington applies to all questions of liability, that the plaintiffs stated certain claims upon which relief could have been granted, but that the claims for personal injury on the part of the decedents are barred by Washington law and the claims based upon strict liability in tort (and breach of warranty) are barred by the applicable statute of limitations.
Furthermore, the court concludes that neither plaintiff Adler nor the other plaintiffs' decedents either assumed the risk of the accident or were contributorily negligent, but also that the defendant was not negligent in any respect.
The court therefore concludes that the defendant is entitled to judgment dismissing the plaintiffs' complaint, and the parties are hereby directed to settle an order to this effect within ten days in conformity with this opinion.