In re Crash of Aircraft N93PC on July 7, 2013, at Soldotna, Alaska

WO IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF ALASKA
In re Crash of Aircraft N93PC ) No. 3:15-cv-0112-HRH
) [Consolidated with
on July 7, 2013, at Soldotna, Alaska ) No. 3:15-cv-0113-HRH and
_______________________________________) No. 3:15-cv-0115-HRH]
O R D E R
Honeywell’s Motion to Exclude the Testimony
of Colin Sommer and Arthur Lee Coffman
Honeywell International Inc. moves to exclude the testimony of plaintiffs’ experts,
Colin Sommer and Arthur Lee Coffman.1 This motion is opposed.2 Honeywell has
withdrawn its request for oral argument,3 and the court deems oral argument unnecessary.
Background
On July 7, 2013, a deHavilland DHC-3 “Otter” airplane operated by Rediske Air, Inc.
and piloted by Walter Rediske crashed shortly after take-off from the Soldotna Airport.
1Docket No. 232.
2Docket Nos. 278 and 280.
3Docket No. 358.
-1-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 1 of 22

Rediske and all of the passengers on board were killed in the crash. Plaintiffs, which are the
estates of the passengers and Rediske, assert wrongful death, negligence, strict liability and
breach of warranty claims against Honeywell.
The accident aircraft was powered by a Honeywell TPE 331 turbine engine, which “is
a lightweight fixed-shaft engine designed to provide primary power for fixed wing aircraft.”4
“Output drive for the aircraft propeller is provided by gear trains which are driven by the
rotating group of the engine. The engine main rotating group is made up of the compressor
impellers, turbine rotors and shouldered shaft (main shaft).”5 “A torsion shaft, positioned
concentrically inside the main shaft, extends through the length of the main shaft.”6 “The
torsion shaft is driven by a spline at the rear of the main shaft and drives the matched bearing
and shaft set (high speed pinion) through a splined coupling at the front.”7 “The torsion shaft
is designed to twist slightly with the application of power by the pilot; a torque sensor
measures this twist to provide the pilot with an indication of engine power.”8 “The torsion
4Overhaul Manual, Exhibit A at 1, Honeywell International Inc.’s Motion to Exclude
the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
5Id.
6Id.
7Id.
8Honeywell’s Memorandum in Support of its Motion to Exclude the Testimony of
Colin Sommer and Arthur Lee Coffman at 4, Docket No. 233 (citing Honeywell TPE 331
Line Maintenance Manual at p. 7-10, Exhibit B, Honeywell International Inc.’s Motion to
Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232).
-2-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 2 of 22

shaft has two bushings that sit in ‘lands’ on the shaft. The bushings are used to position the
torsion shaft concentrically within the main shaft.”9
Post-accident examination of the engine showed that the torsion shaft was sheared.
Honeywell’s position is “that the torsion shaft broke as a result of the load forces created
when the propellor struck the ground during the crash sequence.”10 Plaintiffs’ position is that
the torsion shaft broke in flight. Plaintiffs offer the testimony of two experts to support their
position, Colin Sommer and Arthur Lee Coffman.
Sommer is the “Vice President and Aircraft Mishap and Failure Investigator with
Aeroscope, Inc.”11 Sommer has “a Bachelor of Science [degree] in engineering and civil and
environmental engineering with an emphasis in structural design. . . .”12 Sommer testified
that he took basic course work in aerospace engineering but did not take advance
aerodynamic course work.13 He testified that his post-graduate formal training in accident
investigation consisted of a two-week course at the NTSB Academy, a two-week course at
9Id.
10Id. at 5.
11Curriculum Vitae of Colin A. Sommer, P.E., Appendix 2, Report of Findings,
Exhibit C, Plaintiff’s [sic] Responses to Honeywell’s Motion in Limine to Exclude the
Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 278.
12Deposition of Colin Sommer at 8:21-24, Exhibit C, Honeywell International Inc.’s
Motion to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No.
232.
13Id. at 11:13-12:2.
-3-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 3 of 22

the Southern California Safety Institute, and a week-long course at Teledyne.14 Sommer
testified that he did not think he had ever flown a DHC-3 aircraft with a TPE 331 engine.15
Sommer has extensive experience investigating aircraft accidents, having “personally
investigated over 350 different” accidents.16
Arthur Lee Coffman has 52 years of experience as a mechanic, inspector,
maintenance administrator, and investigator.17 But, Coffman is not an engineer.18 Coffman
testified that he does a little hands-on work on an aircraft “occasionally. I assist somebody
with a rigging issue or something else it seems like every few months. But as far as . . .
making a living working on an aircraft or engine on a daily basis, I haven’t done that since
about ‘91.”19
In his expert report, Coffman opined that “[t]o a reasonable degree of probability the
subject engine failed in-flight and was not producing power at the time of impact, due [to]
14Id. at 12:6-18.
15Id. at 22:4-10.
16Appendix 2 -- Sommer CV at 2, Report of Findings, Exhibit C, Plaintiff’s [sic]
Responses to Honeywell’s Motion in Limine to Exclude the Testimony of Colin Sommer and
Arthur Lee Coffman, Docket No. 278.
17Coffman Report at 1, Exhibit 1, Memorandum in Support of RAC’s Motion in
Limine No. 3 (Coffman), Docket No. 341.
18Video Deposition of Arthur Lee Coffman at 8:19-21, Exhibit D, Honeywell
International Inc.’s Motion to Exclude the Testimony of Colin Sommer and Arthur Lee
Coffman, Docket No. 232.
19Id. at16:5-9.
-4-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 4 of 22

the failure of the torsion shaft bushing and the torsion, causing the accident.”20 Sommer
opined in his expert report that
4. The torsion shaft aft bushing failed and caused heavy
damage and scoring to the torsion shaft after bushing
land.
5. The torsion shaft runout was far beyond normal specifi-
cations for type of application.
6. The torsion shaft in the subject Honeywell International,
Inc. TPE331-10R-511C failed in-flight, shortly after
departure from the Soldotna Airport in Soldotna, AK.
This is evidenced by the damage to the torsion shaft
bushing land and condition of the inner diameter of the
main shaft.[21]
And it is Sommer’s opinion that the failure of the torsion shaft, which caused the engine to
lose power, combined with the pilot’s improper loading of the accident aircraft, caused the
crash.22
Honeywell now moves to exclude Sommer’s and Coffman’s opinions and testimony
as they relate to the torsion shaft.
Discussion
Rule 702 of the Federal Rules of Evidence provides that expert
opinion evidence is admissible if: (1) the witness is sufficiently
20Coffman Report at 5, Exhibit 1, Memorandum in Support of RAC’s Motion in
Limine No. 3 (Coffman), Docket No. 341.
21Report of Findings at 21, Exhibit C, Plaintiff’s [sic] Response to Honeywell’s
Motion in Limine to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman,
Docket No. 278.
22Sommer Deposition at 128:7-10, Exhibit C, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
-5-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 5 of 22

qualified as an expert by knowledge, skill, experience, training,
or education; (2) the scientific, technical, or other specialized
knowledge will help the trier of fact to understand the evidence
or to determine a fact in issue; (3) the testimony is based on
sufficient facts or data; (4) the testimony is the product of
reliable principles and methods; and (5) the expert has reliably
applied the relevant principles and methods to the facts of the
case.
City of Pomona v. SQM North America Corp., 750 F.3d 1036, 1043 (9th Cir. 2014). “Before
admitting expert testimony into evidence, the district court must perform a ‘gatekeeping role’
of ensuring that the testimony is both ‘relevant’ and ‘reliable’ under Rule 702.” United
States v. Ruvalcaba-Garcia, 923 F.3d 1183, 1188 (9th Cir. 2019) (quoting Daubert v. Merrell
Dow Pharmaceuticals, Inc., 509 U.S. 579, 597 (1993)).
“Relevancy simply requires that ‘the evidence logically advance a material aspect of
the party’s case.’” Id. (quoting Estate of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 463
(9th Cir. 2014)). “Expert testimony which does not relate to any issue in the case is not
relevant and, ergo, non-helpful.” Daubert, 509 U.S. at 591 (citation omitted).
“‘[R]eliability’ . . . requires that the expert’s testimony have ‘a reliable basis in the
knowledge and experience of the relevant discipline.’” Ruvalcaba-Garcia, 923 F.3d at 1188-
89 (quoting Barabin, 740 F.3d at 463). “The district court must assess whether ‘the
reasoning or methodology underlying the testimony is scientifically valid’ and ‘properly can
be applied to the facts in issue[.]’” Id. at 1189 (quoting Daubert, 509 U.S. at 592–93). “‘The
district court is not tasked with deciding whether the expert is right or wrong, just whether
his testimony has substance such that it would be helpful to a jury.’” City of Pomona, 750
-6-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 6 of 22

F.3d at 1044 (quoting Alaska Rent-A-Car, Inc. v. Avis Budget Group, Inc., 738 F.3d 960,
969-70 (9th Cir. 2013)). “The court must assess the expert’s reasoning or methodology,
using as appropriate[,] criteria such as testability, publication in peer-reviewed literature,
known or potential error rate, and general acceptance.” Id. “‘But these factors are meant to
be helpful, not definitive, and the trial court has discretion to decide how to test an expert’s
reliability as well as whether the testimony is reliable, based on the particular circumstances
of the particular case.’” Id. (quoting Primiano v. Cook, 598 F.3d 558, 564 (9th Cir. 2010)).
Honeywell first argues that Sommer’s testimony that analysis of the iPhone video
reveals the moment that the torsion shaft broke is not reliable. Sommer testified that at
approximately 50.5 seconds into this 3 minute and 16 second video taken by one of the
passengers with an iPhone, there is a metallic sound and that this is when the torsion shaft
failed.23 Honeywell argues that Sommer is not qualified to offer this opinion because he is
not a metallurgist and did not perform any tests to determine that the sound at 50.5 seconds
was in fact “metallic.” Honeywell also argues that Sommer is not qualified to offer this
opinion because he has no specialized training or experience in performing audio analysis.
Honeywell argues that these deficiencies are highlighted by the fact that plaintiffs retained
a metallurgist, Mark Hood, and an expert in audio analysis, Frank Graham, neither of whom
were asked to identify the so-called metallic sound.
23Sommer Deposition at 55:24-56:7, Exhibit C, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
-7-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 7 of 22

This argument fails. Sommer’s training as an engineer and his training and experience
as an aircraft accident investigator provide the necessary background for him to opine as to
what a sound on the iPhone video represents.
But even if Sommer is qualified to opine as to what the metallic sound on the iPhone
video represented, which he is, Honeywell argues that his opinion that the sound at 50.5
seconds was the torsion shaft failing should still be excluded because it is not supported by
any reliable methodology. At his deposition, Sommer provided the following testimony:
Q. Okay. And what is it about the torsion shaft failing that
creates that particular sound?
A. The only significant damage that we see in the engine is
on the aft land of the torsion shaft other than the failure
itself. The only explanation that is available for that
metallic, screeching type noise would have to be the
destruction of that land due to the interference and
relative motion between the torsion shaft and the main
shaft because we couldn’t find anything else in the
engine that would make that kind of a noise.
Q. Did you give any consideration to the possibility that that
metallic sound might be something from the rest of the
aircraft and not from the engine?
A. We explored the different options, but you don’t really
have any other components that I could conclude that
would do that. The flaps wouldn’t make that noise, and
we don’t see them change position anyway. The trim
wheel wouldn’t make that noise. The boost pump would-
n’t make that noise. Control surfaces wouldn’t make that
noise. There isn’t retractable gear so that couldn’t make
that noise. There isn’t anything else that I know of in
that aircraft that would be operating that would make that
type of noise.
Q. Okay. Did you actually take a DHC-3 Otter and try to
create some of those noises and then compare the sounds
you hear on the tape?
-8-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 8 of 22

A. You mean to break a torsion shaft?
Q. Or the other noises that you were talking to, the boost
pump, the trim wheel, the flaps, the gear.
A. The boost pump, actually, you hear later in the video . . .
after the aircraft crashes, and it is a significantly different
noise. The flaps didn’t change position because we can
see that in the video. The trim wheel -- I’ve used dozens,
hundreds of trim wheels, and I can’t imagine how it
would make that [noise]. But no, to answer your ques-
tion, I did not do it in a DHC-3.[24]
Honeywell argues that this testimony indicates that Sommer summarily concluded that the
noise at 50.5 seconds on the iPhone video must be the torsion shaft failing but that he
performed no analysis or sound comparisons to reach this conclusion.
Honeywell acknowledges that Sommer contends that Frank Graham’s opinion that
there were “engine variations” heard right after the metallic sound supports his theory that
the torsion shaft broke in flight. At his deposition, Sommer testified that Graham’s
[s]ound spectrum analysis showed that there were fluctuations
in the prop/engine signatures. My belief is that that is in
accordance with the engine spooling down and the governor
trying to control the propeller speed. So as the power’s lost to
the drivetrain, the governor’s going to flatten out the blades in
an attempt to maintain RPM. And it’s going to do that some-
what cyclically because it’s a hysteresis based system that is
going to chase the appropriate RPM. And that would be
consistent with the loss of the torsion shaft.[25]
But when asked further about the significance of the engine fluctuations to his theory that the
24Sommer Deposition at 56:8-57:25, Exhibit C, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
25Id. at 59:21-60:6.
-9-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 9 of 22

torsion shaft sheared in flight, Sommer testified that he would defer to Graham for further
analysis “as [Graham] did the sound spectrum analysis.”26 However, Honeywell points out
that Graham testified that he did not reach any conclusions about the cause of the RPM
changes that he detected.27 This, Honeywell contends, is yet another indication that
Sommer’s opinion about the metallic sound being the torsion shaft failing is not supported
by any reliable methodology.
Sommer’s opinion about the metallic sound being the torsion shaft breaking is
reliable. This is a case of two experts disagreeing. Honeywell’s expert determined that the
metallic sound was caused by movement of the trim wheel and Sommer testified that
movement of the trim wheel would not make such a sound. Contrary to Honeywell’s
contention, Sommer did eliminate other possible origins of the sounds. As set out above,
Sommer considered whether the boost pumps or flaps could have caused this sound and
explained why they did not.28 That Sommer did not do the testing that Honeywell is
suggesting he should have done is not fatal. Sommer’s opinion as to what the metallic sound
represented is based on his experience and knowledge. The lack of testing may be a basis
26Id. at 60:19-24.
27Videotaped Deposition of Frank Richard Graham, Jr. at 120:18-20, Exhibit E,
Honeywell International Inc.’s Motion to Exclude the Testimony of Colin Sommer and
Arthur Lee Coffman, Docket No. 232.
28Sommer Deposition at 56:10-57:8, Exhibit B, Plaintiff’s [sic] Response to
Honeywell’s Motion in Limine to Exclude the Testimony of Colin Sommer and Arthur Lee
Coffman, Docket No. 278.
-10-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 10 of 22

for Honeywell to attack Sommer’s opinion on cross-examination, but it is not a basis for
excluding Sommer’s opinion about what the metallic sound on the video represented. See
Hemmings v. Tidyman’s Inc., 285 F.3d 1174, 1188 (9th Cir. 2002) (“objections to the
inadequacies of” an expert’s analysis “are more appropriately considered an objection going
to the weight of the evidence rather than its admissibility”). Sommer’s opinion about the
metallic sound being the torsion shaft failing is not unreliable and thus will not be excluded.
Honeywell next argues that Sommer’s and Coffman’s opinions that the torsion shaft
broke during flight are unreliable because they are not based on reliable principles and
methods. Honeywell argues that Sommer’s and Coffman’s opinions are based on nothing
more than speculation about what might have caused the torsion shaft to break and that they
did not do any metallurgical or materials analysis, shear strength testing, or functional testing
on exemplar torsion shafts to either confirm or rule out causes of the torsion shaft failure.
In fact, Honeywell points out that at his deposition, Sommer testified that in some cases, “it’s
possible that the torsion shaft fails because it fails. That’s not a very engineering, scientific
approach, but they don’t indicate any reason. So we at least have some history that that can
happen.”29 Coffman testified in response to the question, “[d]o you have an opinion about
what caused th[e] fracture of the torsion shaft?”, as follows:
I have no way to prove this or anything. I think it’s possible
29Sommer Deposition at 106:23-107:2, Exhibit C, Honeywell International Inc.’s
Motion to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No.
232.
-11-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 11 of 22

since this bushing causes more damage than I’ve ever seen
before, did it freeze up prior to spinning? I can’t -- I have no
way of proving that except it is a possibility. In other words, we
know that bushings can stick to one surface or the other. Or I’ve
seen that occur, that they kind of bound up against the torsion
shaft or against the main shaft. In this one after it spun, I can’t
say what it did, what it looked like prior to the spinning. I know
that the bushing -- I don’t know whether this bushing totally
deteriorated, whether there’s pieces in there that fell out that I
can’t -- I haven’t seen.[30]
Honeywell argues that Sommer’s and Coffman’s failure to use objective testing or any other
analytical methods to confirm a theory of liability is especially suspect given that plaintiffs
retained Hood, a registered professional engineer in materials and metallurgy, who testified
that he was unable to find any material defects in the torsion shaft through metallurgical
materials analysis.31
Plaintiffs, however, argue that Honeywell has failed to look at the overall methodol-
ogy employed by Sommer and Coffman. In his original expert report, Sommer stated that
he followed “the guidance and investigative techniques found in numerous widely accepted
investigation manuals[.]”32 This, along with Sommer’s review of the physical evidence,
30Coffman Deposition at 22:1-16, Exhibit D, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
31Deposition of Mark B. Hood, P.E., at 35:22-36:1, Exhibit H, Honeywell Interna-
tional Inc.’s Motion to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman,
Docket No. 232.
32Report of Findings at 21, Exhibit C, Plaintiff’s [sic] Response to Honeywell’s
Motion in Limine to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman,
Docket No. 278.
-12-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 12 of 22

including the propellers, the compressor section, the turbine section, and the torsion shaft,
as well as his analysis of the iPhone video, shows, according to plaintiffs, that Sommer’s
opinion was based on sufficient facts and data and on reliable principles and methods.
Similarly, plaintiffs argue that Coffman based his opinion on the physical evidence, such as
the light rubbing within the engine and very light blade splatter, as well as his 52 year
experience as a mechanic, inspector, maintenance administrator, and investigator.33 Plaintiffs
argue that Coffman’s and Sommer’s opinions are further supported by the testimony and
opinions of Hood, who opined that the engine failed in flight, an opinion that is based on
Hood’s examination of the physical evidence.34
To the extent that plaintiffs are arguing that Sommer’s experience and expertise alone
provide a sufficient basis for his opinions, Honeywell disagrees. Honeywell contends that
Sommer does not have any training in aircraft propulsion, has never designed a component
for a turbine engine, has never piloted an Otter aircraft, and has never piloted an aircraft with
a TPE331 engine. In other words, according to Honeywell, Sommer does not have the
requisite experience and expertise to offer opinions about what happened to the engine in the
accident aircraft. As for Sommer’s contention that he is relying on Graham’s conclusion that
the engine speed was fluctuating between 97-101% RPMs before impact, Honeywell again
33Coffman Report at 1, 4-5, Exhibit 1, Memorandum in Support of RAC’s Motion in
Limine No. 3 (Coffman), Docket No. 341.
34Hood Deposition at 12:24-15:8, 19:19-20:2, 21:18-22:7, Exhibit E, Plaintiff’s [sic]
Response to Honeywell’s Motion in Limine to Exclude the Testimony of Colin Sommer and
Arthur Lee Coffman, Docket No. 278.
-13-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 13 of 22

contends that Graham did not opine as to why these fluctuations were happening or try to
correlate them to any actions by the pilot. Honeywell argues that this means that Sommer
has no support for his testimony that a torsion shaft failure would have caused fluctuations
within that range and that he has dismissed, without any basis for doing so, the possibility
that the fluctuations could have been caused by pilot input.
Honeywell also argues that the physical evidence that plaintiffs are contending that
Sommer and Coffman relied on had nothing to do with the existence of a defective torsion
shaft or bushing but rather go to the question of whether the engine was producing full power
at the time of impact. Although Honeywell acknowledges that Sommer and Coffman are
opining that the engine was not producing full power because of the torsion shaft failure,
Honeywell contends that in doing so they are ignoring all of the other potential causes for a
reduction in engine power. In short, Honeywell argues that Sommer’s and Coffman’s review
of the physical evidence does not provide sufficient support for their opinions about when
the torsion shaft broke but rather illustrate that Sommer’s and Coffman’s theories are not the
product of reliable principles and methods and sufficient facts and data.
Sommer’s and Coffman’s opinions that the torsion shaft failed during flight are
sufficiently supported by their experience and knowledge and their review of the physical
evidence. Honeywell’s arguments on this issue go more to the weight that Sommer’s and
Coffman’s opinions should be given than to the admissibility of their opinions.
-14-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 14 of 22

Honeywell next challenges Sommer’s reliance on the bend in the torsion shaft, which
was found on post-accident examination. Honeywell disputes that this is evidence that the
torsion shaft broke in flight, as Sommer opines. At his deposition, Sommer testified that
if you lost the support of the shaft and the shaft bent to where
instead of having the shaft supported on both ends and the shaft
supported with both bushings, if you now have this long,
unsupported section of a bent shaft, then perhaps you could have
enough side load due to the very high speed of the shaft to
where you could break the shaft.[35]
Sommer testified that there must have been a manufacturing defect in the torsion shaft back
in 1998 when it was installed in the engine because he did not “know of any other way that
it would get bent like that, that is the only conclusion that one could make.”36 But Honeywell
argues that Sommer had not tested his theory, even though he admitted that it would be
possible to do so.37 Honeywell insists that any opinion Sommer has about the bend in the
torsion shaft being a manufacturing defect is based on nothing more than speculation and that
because Sommer is not a metallurgical expert, he is not qualified to offer opinions about
whether there was a manufacturing defect in the torsion shaft. Honeywell emphasizes that
Sommer offered this opinion without even having seen the manufacturing/engineering
35Sommer Deposition at 109:25-110:7, Exhibit C, Honeywell International Inc.’s
Motion to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No.
232.
36Id. at 106:2-8.
37Id. at 109:18-23.
-15-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 15 of 22

drawings.38 At his deposition, the following exchange took place:
Q. Is it your opinion that the bend in the torsion shaft was
present when the engine was operating in the air?
A. I’m not sure that it’s 100 percent possible to know
whether or not the bend in the torsion shaft could have
occurred during the fracture of the shaft. Because the
shaft should be supported by the bushings in two differ-
ent locations which are supposed to keep the centerline
of the shaft uniform, I wouldn’t expect a shaft of this
hardness to bend like that just from a torsion break. The
torsion break would be somewhat uniform. The shaft is
supported. I wouldn’t think that it would be likely, and
would be more likely than not that it was bent prior. But
post accident I’m not sure that there is a 100 percent
guaranteed way to know that.
Q. So your opinion is it’s more likely than not that that bend
in the torsion shaft was present in operation?
A. Yes.
Q. And how did it get bent?
A. I can’t answer that. I don’t know what the specs were
because I’ve never seen any manufacturing documenta-
tion on the shaft so I’ve never seen where they did the
runout check and then compared that runout check to
what the drawing says. And actually, in this case I’ve
never seen the drawing for the shaft so I don’t know
exactly what the runout should be, but it should be on the
order of a few thousandths. But it could have been bent
during manufacturing. And maybe it was material
reviewed and approved, or maybe it was bent during
operation because of the loads that are on it. Of course,
that’s going to be dependent upon the bushing maintain-
ing its concentricity. And as the bushing fails, that
concentricity is going to be compromised.[39]
38Sommer had not reviewed the drawings because Honeywell had not yet produced
them.
39Sommer Deposition at 42:20-44:4, Exhibit C, Honeywell International Inc.’s Motion
(continued...)
-16-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 16 of 22

Honeywell also emphasizes that Hood did not attempt to determine the cause of the bend in
the torsion shaft. In response to a question as to whether he did any analysis to determine
why the torsion shaft bent, Hood testified that this “wasn’t part of [his] analysis,” although
he testified that he “didn’t see anything that indicated that there was post-accident damage
that would be responsible for that bend.”40
To the extent that Sommer is opining that the bend in the torsion shaft was a
manufacturing defect, that opinion is excluded as unreliable. Sommer’s opinion on this issue
is nothing more than speculation. Sommer has offered no support for this opinion. He
testified that the bend in the torsion shaft was present during operation but offers no
explanation as to how the accident aircraft operated for years without incident despite this
alleged defect. As Sommer acknowledged, he could have tested his theory that the bend in
the torsion shaft was a manufacturing defect. His failure to do so in this instance results in
an unsupported, unreliable opinion.
Honeywell next argues that the fact that the aft bushing was destroyed does not
support Sommer’s and Coffman’s opinions that the torsion shaft broke during flight.
Honeywell argues that Sommer and Coffman are speculating about the possible effect of one
of the torsion shaft bushings binding in flight. Specifically, Sommer testified that
39(...continued)
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
40Hood Deposition at 51:1-23, Exhibit H, Honeywell International Inc.’s Motion to
Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
-17-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 17 of 22

[i]f the bushing is to bind because of its location in the rear of
the engine where it’s much hotter due to the turbine section -- if
the bushing is to bind either because the bushing is improperly
sized or because of the excess heat and coking, which
Honeywell indicates can happen to both the shaft, the torsion
shaft and the main shaft, you are now going to get an improper
torque indication in the cockpit. And a pilot is going to look at
his torque gauge and believe that he has less torque on his
engine than he really has. And if he is to believe that torque
gauge and then to apply greater torque to the engine because he
believes he has a margin there -- and more likely than not, this
engine would have been torque-limited on this flight as opposed
to temp-limited -- then he could possibly overtorque the engine,
breaking the torsion shaft because of the bound bushing between
the two components.[41]
Honeywell argues that Sommer offers many “ifs” and “possibilities” but has no evidentiary
basis for his speculation that the aft bushing on the accident aircraft was improperly sized or
subjected to excessive heat and coking or that there was an erroneous torque gauge reading
on the day of the accident. Honeywell acknowledges that Hood testified that if the bushing
“adhered to both” the main shaft and the torsion shaft “momentarily and then released,
[there] would be a sudden . . . change in the torque loading at the aft end of . . . the shaft. I
can see that as one way to . . . shear the shaft.”42 But, Honeywell contends that Hood did not
test his theory either. Rather, Hood testified that he had not tried to calculate what the
momentary torque load would be if the bushing were momentarily to adhere to both the main
41Sommer Deposition at 48:5-22, Exhibit C, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
42Hood Deposition at 25:4-8, Exhibit H, Honeywell International Inc.’s Motion to
Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
-18-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 18 of 22

shaft and torsion shaft and then release.43 Honeywell contends that Coffman also did nothing
to test this theory. Coffman testified at his deposition that
[i]f we have binding at the rear bushing and it’s not free to
measure the twist from one end to the other, it could be causing
a low torque reading in the cockpit, which the pilot may be
applying more torque. And it’s just one of those things. I say
it’s possible. Have I proved it? No. It’s just I try to look at all
things that I’ve seen in the ag world and various things and say
I believe it’s probably possible.[44]
Thus, Honeywell argues that Sommer’s and Coffman’s opinions as to the effect that the
binding of the bushing might have had on the torsion shaft are completely unsupported and
as such, are not reliable.
Sommer’s and Coffman’s opinions as to the effect of the binding of the aft bushing
are reliable. Sommer explained at his deposition that
it is obvious that there was significant binding between the rear
bushing and the main shaft. We don’t know what happened to
the rear bushing. It should have been there, or pieces of it
should have been there during the Honeywell teardown. I can’t
imagine that the bushing was completely washed away and
gone, or else you would have found pieces of it in the oil filter.
But you should have found pieces of it during the teardown. I
don’t know what Honeywell did with the components. I would
expect them to be there.
But the binding of that bushing is so significant because
of the heavy scoring and damage to the rear land of the torsion
shaft that it is indicative of both relative rotation of the torsion
43Id. at 30:3-21.
44Coffman Deposition at 25:8-16, Exhibit D, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
-19-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 19 of 22

shaft to the main shaft and some sort of a defect in the bushing
or the land of the torsion shaft prior to installation. [45]
In other words, Sommer has explained the significance of the physical scoring of the bushing
landing, that this would account for an over torque situation. Sommer’s opinion about the
effect a bound bushing might have is supported by sufficient facts and data.
Similarly, Coffman’s bushing opinion is reliable because it is supported by sufficient
facts and data. Coffman testified that “[i]f we have binding at the rear bushing and it’s not
free to measure the twist from one end to the other, it could be causing a low torque reading
in the cockpit, which the pilot may be applying more torque” and that this overtorque could
cause a shock to the torsion shaft, causing it to shear.46
The fact that Sommer and Coffman may have used the words “probably” and “if”
when expressing and explaining their opinions does not mean that their opinions are
unreliable. An expert does not have to testify that he is 100% certain that something would
happen. Rather, an expert may opine as to probable causes, after eliminating unlikely or
improbable causes.
Finally, Honeywell argues that Sommer’s and Coffman’s opinions that the torsion
shaft failed in flight are not relevant. “The jury is not permitted to speculate in choosing one
45Sommer Deposition at 41:20-42:12, Exhibit B, Plaintiff’s [sic] Response to
Honeywell’s Motion in Limine to Exclude the Testimony of Colin Sommer and Arthur Lee
Coffman, Docket No. 278.
46Coffman Deposition at 25:8-26:10, Exhibit D, Plaintiff’s [sic] Response to
Honeywell’s Motion in Limine to Exclude the Testimony of Colin Sommer and Arthur Lee
Coffman, Docket No. 278.
-20-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 20 of 22

of alternative possibilities, but is restricted to reasonable inferences based upon facts.”
Victorian Condominium Owners Association v. Fireman’s Fund Ins. Co., Case No.
C11-1002-JCC, 2012 WL 13028242, at *3 (W.D. Wash. Oct. 2, 2012) (citation omitted).
Honeywell argues that Sommer and Coffman are offering opinions that suggest a possible
cause of the accident and thus if the jury were allowed to hear their opinions, the jury would
be left to speculate about the cause of the torsion shaft failure. Because their opinions would
only allow the jury to speculate, which is not permitted, Honeywell argues that Sommer’s and
Coffman’s opinions as to the cause of the torsion shaft failure are not relevant.
Plaintiffs argue that if the jury were to accept Sommer’s and Coffman’s opinions, the
jury would not be forced to speculate. Plaintiffs insist that Coffman and Sommer have
opined that the torsion shaft failed in flight and explained why they believe it failed at that
time, rather than on impact. These opinions, according to plaintiffs, are based on reliable
investigative principles and reasonable inferences and the jury should be allowed to hear
them.
In reply, Honeywell argues that in their response, plaintiffs are attempting to merge
two hypothetical manufacturing defects into a singular, untested theory of causation.
Honeywell points out that Sommer testified that there were two possible manufacturing
defects that could have caused the torsion shaft to break. At Sommer’s deposition the
following exchange took place:
Q. Okay. So that’s -- so we now have -- we have two
mechanisms by which the shaft could be broken, am I
-21-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 21 of 22

correct? One is if you have erroneous torque indications
and the pilot overtorques the engine? The other is side
loading because the shaft is bent?
A. It’s possible.
Q. Okay. Any others?
A. Other than bad metal, which Mr. Hood didn’t find, I
would say that’s about it.[47]
Honeywell contends that plaintiffs are now trying to pass these two theories off as one in
order to avoid the argument that Sommer’s and Coffman’s opinions would require the jury
to speculate in choosing one possible alternative.
Honeywell’s relevancy argument is not persuasive. Sommer’s and Coffman’s
opinions that the torsion shaft broke during flight are relevant.
Conclusion
Honeywell’s motion to exclude the testimony of Sommer and Coffman as it relates
to the torsion shaft48 is granted in part and denied in part. The motion is granted as to
Sommer’s opinion that the bend in the torsion shaft was a manufacturing defect. The motion
is otherwise denied. Sommer’s and Coffman’s opinions that the torsion shaft broke in flight
are both relevant and reliable.
DATED at Anchorage, Alaska, this 22nd day of April, 2020.
/s/ H. Russel Holland
United States District Judge
47Sommer Deposition at 110:8-17, Exhibit C, Honeywell International Inc.’s Motion
to Exclude the Testimony of Colin Sommer and Arthur Lee Coffman, Docket No. 232.
48Docket No. 232.
-22-
Case 3:15-cv-00112-JMK Document 363 Filed 04/22/20 Page 22 of 22