IN THE SUPREME COURT OF BRITISH COLUMBIA
Citation: | Advantage Helicopters Inc. v. Heliponents, Inc., |
| 2014 BCSC 674 |
Date:
20140417
Docket: S097339
Registry:
Vancouver
Between:
Advantage
Helicopters Inc. and Rilpa Enterprises Ltd.
Plaintiffs
And
Heliponents,
Inc. and John Doe #1
Defendants
Before:
The Honourable Mr. Justice Willcock
Reasons for Judgment
Counsel for the Plaintiffs: | Christopher Harvey, |
Counsel for the Defendants: | R.J. Randall Hordo, Daniel Fetterly |
Place and Date of Trial: | Vancouver, B.C. September 4-6 2013
|
Place and Date of Judgment: | Vancouver, B.C. April 17, 2014 |
TABLE OF CONTENTS
A Brief Description of Component Parts. 3
Evidence With Respect to the Accident 5
Evidence Collected at the Scene. 7
Later Post-Accident Examination of Parts. 10
Events Following the Crash. 10
The Overhaul Completed by Heliponents. 11
The Plaintiffs Evidence Re: Maintenance Procedures. 16
Plaintiffs Expert Evidence. 17
Damage to the Duplex Bearing and Pinion. 18
Defendants Expert Evidence. 24
Damage to the Duplex Bearing and Pinion. 27
The Responsive Opinion of Dr. Elvis
Cepu. 41
Damage to the Duplex Bearing and Pinion. 41
The Quality of the Overhaul. 48
Pre-Accident Location of the Cork in the Pinion. 55
Damage to the Duplex Bearing and Pinion. 64
Criticism of the Plaintiffs Theory. 68
Introduction
[1]
On October 6, 2007, a 1964 Bell 204B helicopter owned by Rilpa
Enterprises (Rilpa) and leased by Advantage Helicopters Inc. (Advantage),
experienced a sudden and total loss of tail rotor drive and crashed while being
used to conduct heli-logging operations near Kaslo, British Columbia. The main
transmission assembly of the helicopter had been overhauled by the defendant,
Heliponents, Inc. (Heliponents) at its Mesa, Arizona facility in May 2007. The
plaintiffs allege the accident occurred as a result of the negligent overhaul
of the transmission, which resulted in starvation of oil supply to bearings in
the quill assembly of the transmission, failure of the duplex bearing set,
seizure of that portion of the transmission, and the fracture of the tail rotor
driveshaft.
[2]
The defendant says the evidence is inconsistent with the plaintiffs
theory of causation and, as a result of inadequate investigation of the
accident, it is impossible to establish the cause of the loss of tail rotor
control and the crash.
[3]
The pilot of the helicopter was injured, but fortunately survived the
crash. A claim arising out of his personal injury has been settled elsewhere.
The parties have agreed to the quantum of damages, and only liability is in
issue.
The Evidence
A Brief Description of Component Parts
[4]
Appended to these reasons for judgment as Appendix A are figures 1 to
4 from the December 6, 2011, report of Dr. J.W. Morris, an expert witness
called to testify at the trial. Figure 1 illustrates the location of the tail
rotor drive quill in a Bell 204B helicopter. The drive quill is attached to the
helicopters main transmission. The transmission is at the base of the mast and
at the forward end of the tail rotor drive shaft. The function of the tail
rotor drive quill is to transmit drive power from the base of the transmission,
via the drive shaft, to the tail rotor. The function of the tail rotor is to
limit and prevent rotation of the helicopter, which is a consequence of the
force generated by the rotation of the main rotor. Figure 2 depicts the parts
of the tail rotor assembly from the transmission to the tail rotor. The tail
rotor drive quill assembly is at the forward end of the tail rotor assembly.
Figure 3 is a schematic drawing of the tail rotor drive quill. At the forward
end of the drive quill there is a beveled gear which rotates a hollow steel
pinion, running through the centre of the assembly, as it is driven by the
gears in the transmission. The rotation of the pinion is transmitted through a
splined coupling at the aft or rear end of the tail rotor drive quill and
drives the rotation of the sections of the drive shaft aft of the quill
assembly. From the compartment in which the quill assembly is located (referred
to by one of the expert witnesses, Dr. Morris as the fan bay, which I will
refer to as the transmission bay) the drive shaft passes through an aperture in
a fire wall, then, contained in a titanium tunnel, through the engine bay, and
then to the rear of the helicopter.
[5]
The pinion is secured in the quill assembly by three bearings. At the
forward end of the pinion there is a roller bearing. Aft of the roller bearing,
at the middle of the pinion, there is a conjoined set of ball bearings, one aft
and one forward, referred to as the duplex bearing set. The balls in each of
the bearings in the duplex set should spin freely between the concave surface
track or race of the outer half of the bearing (the outer race) and the
concave surface of the inner half of the bearing (the inner race) while contained
in a cage. The inner races of the duplex bearing should spin with the pinion,
onto which they are affixed by pressure.
[6]
When the assembly is functioning normally the duplex bearing is
lubricated by oil from two sources. Some oil flows to the duplex bearing
through the roller bearing. Some oil flows through the hollow center of the
pinion and into the duplex bearing via a small hole in the pinion, through
which oil is moved by centrifugal force. Oil is sprayed into the centre of the
pinion by a jet in the transmission. At the aft end of the hollow core of the
pinion there should be a cork which seals the system and prevents leakage of
oil through the end of the drive quill assembly. Figure 4 is a detailed
depiction of the parts of the tail rotor drive quill. For reasons set out below,
figure 4 does not depict the cork or its appropriate location in the pinion.
[7]
The plaintiffs say the cork was misplaced at the front end of the pinion
at the time of the Heliponents overhaul, preventing oil from travelling down
the centre of the pinion, obstructing the principal source of oil to the duplex
bearing, causing oil starvation, the overheating and failure of the duplex
bearing and, eventually, the failure of the pinion itself.
[8]
Also appended to these reasons for judgment as Appendix B are figures
20 and 21 from the February 24, 2010 investigation report of David Rupert, an
expert retained by the plaintiffs. Those figures illustrate the effect of
incorrect cork placement on the flow of oil through the tail rotor drive quill.
Evidence With Respect to the Accident
[9]
At the time of the accident, the helicopter was being flown by Ron
Batty, an experienced pilot who had flown Bell medium line helicopters similar
to that involved in the accident for approximately 3,000 hours. He says the aircraft
was in very good condition and apparently well-maintained. He had started
flying the helicopter at the Kaslo job site in September 2007 and had flown it
without incident for approximately 200 hours since then. On October 6, 2007, he
started working at around 7:30 a.m., flying the helicopter from the base of the
work site up to a cut location on the hill, where he would pick up a log with
the helicopter, fly back down to the log landing area, drop the log and then
repeat the cycle or turn. He was using a 200-foot line equipped with a
grapple hook and was lifting logs approximately 40 feet in length. The average
lift per cycle was about 3,000 pounds.
[10]
At the end of the fourth turn of the morning, as the helicopter was
using its maximum power, hovering at a point where it was about to unload a log
above the landing zone, Mr. Batty smelled an unfamiliar burning smell,
like a burning pot on a stove. He had not noted any warning lights illuminated
in the cockpit. The smell grew stronger quickly. He released the log and began
to steer toward the service landing site when he began to lose tail rotor control
or authority. The airframe started to rotate aggressively. Mr. Batty
entered auto-rotation by putting the collective, or pitch, of the blades down
and rolling the throttle off so that no engine power was being applied to the
main rotor of the helicopter. This was intended to stop the spin of the
helicopter and to permit the helicopter to fall, causing upcoming air to drive
the rotor. The fall was initiated from a height of about 200 feet, too little
distance to successfully bring a helicopter down under auto-rotation. From the
point at which the log was dropped, the helicopter was capable of travelling
only a very short distance over land. Mr. Batty had little choice but to
find the closest acceptable landing location, a flat spot near the loading site
at the side of a logging road.
[11]
Helicopter pilots are trained to increase the collective shortly before
impact so as to increase air resistance and create some lift, or cushion. Mr. Batty
cannot recall doing so; he has no specific recollection of the last seconds of
his fall or the landing. With what must have been great skill, he brought the
helicopter down in a flat landing zone and managed to keep the rotor blades
clear of obstacles until at or about the point of impact, at which time the
blades appear to have struck an object on the right side of the aircraft,
fortunately exerting backward force on the engine and main rotor, pressing it
downward behind the cabin, crushing the cowling above the engine bay. The
landing was hard and the helicopter was substantially damaged, but Mr. Batty
survived and was able to walk into court to testify and provide us with the
only account of the malfunction and the crash.
[12]
Mr. Batty acknowledged that he did not have a clear memory of the
whole sequence of events after the airframe began rotating aggressively, but
his evidence with respect to events prior to that point is clear and apparently
unaffected by subsequent events, including his injury.
[13]
Some of Mr. Battys evidence was recorded immediately. The first
record of the accident refers to a burning smell that became quickly more
noticeable followed, within seconds, by the helicopter beginning to spin. It was
not suggested to Mr. Batty in cross-examination that his recollection of
this sequence was mistaken. He was questioned with respect to whether there was
any loss of power before he lost tail rotor authority. He acknowledged that he
had no recollection of looking at his gauges at the moment he lost tail rotor
authority, but given that he was hovering, he assumed that there was no loss of
power to the main rotor before the helicopter began to spin.
[14]
Mr. Batty accepted the
proposition that the maximum gross weight limit for the Bell 204 is 9500
pounds. The helicopter itself weighed 5000 pounds. He usually flew with 1000
pounds of fuel and refueled when the fuel load dropped below 200 pounds. The
longline and the grapple hook used in this case weighed 180 pounds. A pilot
weight of 180 pounds had to be considered. The helicopter, longline, hook and
pilot had a total weight of 5300 pounds. After making allowance for fuel, the
maximum external load on the first flight or cycle was 3200 pounds. When the weight
of fuel onboard fell to 300 pounds, the maximum external load was 3900 pounds. Mr. Batty
says he began the cycle assuming he could lift 2500 pounds and that on his last
cycle his limit was about 4000 pounds. In order to stay within those limits he
would reject a 4000 pound log on the first cycle.
[15]
Mr. Batty was shown his
flight logs. These indicate that he occasionally lifted logs weighing 4200
pounds. Mr. Partridge, the president of Advantage, a qualified aircraft
maintenance engineer, acknowledged that a load of 4200 pounds would cause the
helicopter to exceed its maximum gross lift capacity.
Evidence Collected at
the Scene
[16]
Mr. Partridge testified for
Advantage on examination for discovery and at trial. He was at the scene at the
time the helicopter crashed. He took Mr. Batty to the hospital, about 40
minutes away. He contacted Ken Mizera, Rilpas Chief Executive Officer, and
returned to the accident scene with him on the following day, October 7, 2007. Mr. Mizera
has been a qualified aircraft maintenance engineer for 35 years. He testified
as a representative of Rilpa on examination for discovery and at trial. He is
familiar with the characteristics of Bell 204 helicopters and with procedures
for their maintenance, having bought them and refurbished them in the past.
[17]
Mr. Mizera performed what he
described as a cursory examination of the helicopter at the scene. When he
opened the engine cowls to examine the transmission bay, a segment of the tail
rotor drive shaft and the coupling were lying on the deck. The tail rotor drive
quill portion of the transmission was burnt and discolored. The remaining tail
rotor drive shaft was in pieces in the adjacent engine bay. He concluded the
quill had been extensively overheated, had become malleable and twisted off. He
thought that damage to the quill was pre-accident damage, because the quill had
clearly overheated while functioning; the damage to the drive shaft and tunnel to
the rear of the quill assembly, on the other hand, was considered to be post-accident
damage (in the sense that it was a consequence of the failure of the quill). Mr. Mizera
probably expressed that opinion to Mr. Partridge at the scene and, later,
to Mr. Ward, an aviation insurance adjuster in Salmon Arm.
[18]
Mr. Mizera and Mr. Partridge both thought the accident had
occurred because of a problem with the tail rotor drive quill. For this reason,
they took relatively few parts of the helicopter for further examination and,
from an early date, their inquiry was focussed on what was thought to be the
obvious cause of the accident.
[19]
Mr. Partridge took numerous photographs at the accident scene, some
of which have been examined by experts and have become exhibits.
Inspection at Salmon Arm
[20]
The helicopter was loaded on a
flatbed truck and arrangements were made to take it to the Advantage hanger in
Salmon Arm on the following day, October 8, 2007. It was Mr. Partridges
evidence that all of the pieces of the broken tail rotor drive shaft, with the
possible exception of a small piece of the drive shaft that can be seen in the
photographs of the transmission bay of the helicopter taken at the scene of the
accident, were taken to the Advantage hanger and are shown in photographs that
depict the equipment on the hangar floor.
[21]
The quill assembly was disassembled and examined in Salmon Arm by Mr. Mizera,
Mr. Partridge, and Mr. Ward, who met them there. Mr. Mizera
thought it was glaringly obvious what had caused the accident. When the quill
assembly is manufactured or overhauled, a cork plug must be inserted in the hollow
shaft of the pinion. The location of that plug is important; it must be seated
at the aft end of the quill assembly so as to permit oil sprayed from the jet
in the transmission to travel through the hollow centre of the pinion, and as
far as the small hole at the aft end of the pinion through which oil should
flow into the duplex bearing. After the accident, the cork plug was found to be
at the forward end of the pinion gear, about three inches forward of the
position that it should have occupied; this would have prevented the
circulation of oil through the pinion to the duplex bearing. Noting the plug at
that location, Messrs. Mizera and Partridge, both of whom are familiar with
helicopter engines and transmissions, concluded that the duplex bearing had
been starved of oil, leading great heat to accumulate at the aft end of the
quill, causing the pinion to deform and fail.
[22]
All parts taken from the accident scene, other than a few parts retained
for examination by experts, including most of those examined at the Advantage
hanger, were later shipped to the Rilpa facility in Calgary. The main rotor
blades were disposed of. The main rotor hub and all of the drivetrain went to
Rilpa. Some of these parts cannot now be located or accounted for. Mr. Mizera
was examined at trial with respect to the late and piecemeal location and
production of parts for examination by the defendant. He confirmed that at his
May 2011 examination for discovery he had been asked to locate additional parts
from the helicopter, in particular, pieces of the tail rotor drive shaft and
the drive shaft tunnel depicted in photographs taken at the accident scene. In
November 2011, Mr. Mizera found a missing part depicted in photographs
taken at the Advantage hanger – a yellow main rotor drive shaft adapter – but
he could not find the parts of the main input drive shaft and the tail rotor
drive shaft depicted in the photographs marked as Exhibit 6 at trial. There was
poor control of the parts. Following the October 2007 inspection Mr. Mizera
was uncertain whether parts would be taken from Salmon Arm by the insurance
adjuster or kept by Advantage and there is an incomplete record of the receipt
and disposal of those parts that were ultimately sent to Rilpa.
[23]
Mr. Partridge testified that on May 26, 2011, following several
demands and the production of two lists of documents, Advantage produced, for
the first time, many of the photographs taken at the accident scene. In late
2011 additional photographs were produced. These appear to have been
photographs recorded on a USB stick provided to counsel shortly before the
November 1, 2011, examination for discovery.
Later Post-Accident Examination of Parts
[24]
A few critical parts, the failure of which were thought to have caused
the accident, were forwarded to R.J. Waldron & Co., where they were
examined by David Rupert, an expert who was called to testify at trial, and
George Heath. Those parts included the tail rotor drive quill assembly and a
section of the forward tail rotor drive shaft. Numerous other parts of the tail
rotor drive mechanism were photographed but not retained.
[25]
Experts for all the parties have subsequently examined the retained
parts in detail, have conducted destructive testing, and have expressed
opinions on the damage depicted in photographs without having the opportunity
to examine all parts of the helicopter and, in particular, without having the
opportunity to examine all portions of the drive train of the tail rotor.
Events Following the
Crash
[26]
After the accident, Mr. Mizera
telephoned Martin Doss, a principal of Heliponents, to advise him of the crash.
He advised Mr. Doss it was his opinion that the transmission had failed due
to incorrect placement of the cork in the tail rotor drive quill assembly. He
sent an e-mail to Heliponents on October 26, 2007, with pictures of the
accident scene, depicting how the transmission looked when it was removed from
the helicopter. At about the same time, the parts of the quill assembly were
being sent to Vancouver. Mr. Mizera is uncertain whether he thought they
were being sent to RJ Waldron & Co. or for examination by the
Transportation Safety Board of Canada. In any event, they were being removed
from Rilpas possession and Mr. Mizera did not invite Heliponents to
examine the parts before they were sent to Vancouver.
[27]
On January 9, 2008, Heliponents
sent an e-mail to Bell Helicopter, notifying Bell that Heliponents copy of the
Bell Maintenance and Overhaul Manual (the Manual) did not provide
directions for the appropriate installation of a cork in the quill assembly. On
January 10, 2008, Bell Helicopter advised Heliponents that the assembly should
have a cork installed, and provided its part number with instructions for its
installation.
[28]
Rilpa eventually rebuilt the 204B
helicopter. There is an agreement between counsel that the value of the claim
for damages in relation to the helicopter is $1.175 million. Counsel advised the
court that they would come to an agreement with respect to the calculation of
pre-judgment interest on the claim. The plaintiffs asked for, and have received,
leave to address that matter or any other issue in relation to the quantum of
damages they believe ought to be included in a judgment, in the event the
plaintiffs succeed.
The Overhaul
Completed by Heliponents
[29]
Heliponents records establish
that the overhaul of the Advantage transmission was performed by its employees Russell
Mayhugh and Patrick Felice.
[30]
Mr. Mayhugh was examined for
discovery in Mesa, Arizona, on April 13, 2011, and testified at trial. Martin
Doss, Heliponents president, was examined for discovery on April 14, 2011.
Excerpts from testimony at examinations for discovery were read into the
record. Mr. Felice is no longer employed by Heliponents and did not
testify.
[31]
The overhaul of the transmission
is recorded on a number of documents produced by Heliponents, including:
a)
a magnetic particle and
fluorescent penetrant inspection sheet dated April 24 and April 25, 2007;
b)
a labor instructions/inspections
record dated May 16, 2007
c)
a Bell 212 transmission assembly
overhaul kit record;
d)
a Bell 212 main transmission
assembly overhaul worksheet;
e)
a traveler form; and
f)
an authorized release certificate.
[32]
The magnetic particle and
fluorescent penetrant inspection sheet was completed by Mr. Mayhugh. It is
a record of the non-destructive testing of components before the reassembly of
the transmission. That work appears to have been performed on April 24 and 25,
2007.
[33]
The labor instructions/inspections
sheet is a document that lists 14 steps to be performed in the course of a
transmission overhaul. It bears the initials of Mr. Mayhugh beside each
step in the process and appears to have been signed by Mr. Felice on May
16, 2007. Although this document appears to describe a step-by-step process and
requires initials and the stamp of an inspector at the conclusion of each step,
it was Mr. Mayhughs evidence that this document was initialed at the
conclusion of the overhaul of the transmission, and was not completed after
each step in sequence.
[34]
The transmission assembly overhaul
kit document is a list of items assembled by Heliponents for the purpose of
performing a transmission assembly overhaul. The kit used in this case was
prepared for the overhaul of a Bell 212 transmission. Mr. Mayhugh was
unable to say why a 212 transmission assembly overhaul kit was used for the
overhaul of the transmission of a Bell 204 helicopter. He could not say whether
the kit typically includes instructions on how to assemble or install
components, including corks. The list of the contents of the overhaul kit used
in this case includes four corks, bearing parts numbers 7A, 18A, 22A and 15A.
[35]
The form of worksheet completed in
this case, similarly, was for the overhaul of the transmission of a Bell 212
helicopter, rather than a 204 helicopter. It bears the initials of Mr. Mayhugh
and Mr. Felice beside each step in the process. Like the labor
instructions/inspections record, this is a document that appears to be intended
to be completed as the overhaul of the transmission progresses. Pages 6 and 7,
which include a record of the first 14 steps to be taken during the buildup of
the transmission, are missing from the document produced. The document bears a
certification by Mr. Felice, as inspector, dated May 17 2007 that The
component described on this form has been inspected and assembled in accordance
with Bell helicopter 212 component repair and overhaul manual.
[36]
Mr. Mayhughs evidence with
respect to the timing of the completion of this document was inconsistent. At
trial he testified that it was his practice to initial the worksheet as work
progressed. He had testified at his examination for discovery that this sheet
was probably completed when the transmission overhaul was finished. Ultimately,
at trial, he agreed that he could not reliably say when the sections of this
worksheet were completed and initialed.
[37]
The traveler form is a document
that should stay with the transmission during the course of the overhaul. The
form appears to initially have been opened on November 27, 2006. It bears the
initials of Mr. Mayhugh, and Mr. Felice. Mr. Mayhugh appears to
have initialed the form on April 25, 2007 when the non-destructive inspection
was completed, as documented by the magnetic particle and fluorescent penetrant
inspection sheet. All other initials on the form, including initials meant to
confirm the incoming inspection, preliminary maintenance inspection,
disassembly inspection and final inspection appear to have been written on the
form by Mr. Felice on May 17, 2007.
[38]
The authorized release certificate
was completed by Patrick Felice on May 16, 2007. It confirms the overhaul had
been completed in accordance with the directions in the April 7, 2006 revision
of the Manual.
[39]
Mr. Mayhugh was first
employed at Heliponents in the component overhaul shop from March 1995 to March
1998. During that period he did not participate in the overhaul of the transmission
of a Bell helicopter. After leaving Heliponents he worked with other employers as
an airplane mechanic for a number of years until he was laid off. He returned
to Heliponents as a mechanic on February 12, 2007. In the interim, from March
1998 to February 2007, he did not work on helicopters. Mr. Mayhughs only
training in the performance of transmission overhauls was on-the-job training.
The work he did on the Advantage transmission may have been the first
transmission overhaul he did at Heliponents. When he was examined for discovery,
he was unable to say whether he had overhauled the transmission of any Bell 204
helicopter other than the Advantage transmission.
[40]
An assessment of Mr. Mayhughs work
and training prepared by his supervisor, Chad Barta on December 3, 2007 includes
a note: Employee needs to have manufacturer training in overhaul of components
[and] Part 145 training. That is a reference to the American Federal Aviation
Regulation containing the rules a certified repair station must follow.
[41]
Mr. Mayhugh confirmed, by
looking at the documents, that he was the employee primarily responsible for
the overhaul of the Advantage transmission, but he does not remember working on
the job. Although Heliponents certified the work was done in accordance with the
Manual he acknowledged that the maintenance and overhaul instructions
require work to be completed in accordance with the Manual as
supplemented or modified by alert service bulletins and other directions issued
by Bell helicopter. He is not sure he knew of that requirement at the time he
performed maintenance in this case.
[42]
Mr. Mayhugh is not able to
say whether he followed the Manual and supplements when he did this
work. He agreed that it is possible he did not follow it item by item. There
must have been some departure from the Manual because the parts list at
figure 441 of the April 7, 2006, does not include a cork. He agreed that at
least in that respect it was inaccurate to certify that all the work in this
case was done in accordance with the Manual.
[43]
The worksheet that he completed
for the overhaul of this transmission was a worksheet for the overhaul the Bell
212 transmission, on which that number was crossed out and 204 was inserted in
its place. Mr. Mayhugh does not know if Heliponents had an overhaul
worksheet for a Bell 204 helicopter.
[44]
Mr. Mayhugh has no specific
recollection of installing a cork in the pinion of the Advantage transmission. When
he was asked if it was possible, given his experience in 2007 that he could
have installed the cork in the wrong end of the pinion, he said, I do not know.
When he was asked if he could say one way or the other, he answered, I cant
say one way or the other.
[45]
He cannot remember whether he used
shellac or epoxy to cover and seal the cork. At trial he said that it continued
to be his practice to insert a plug from the forward, beveled end of the
pinion, despite the April 30, 2008, revision of the Manual that
expressly directs that the cork plug should be inserted in the rear, splined end
of the pinion gear shaft. He testified that it was his practice, after
inserting a cork, to clean the inside surface of the bore of the pinion using
his finger and acetone, or another cleaning solution, but he agreed that he
could not get his finger into the bore so far as the small hole through which
oil is designed to travel to the duplex bearing.
[46]
He is now aware that the April 30,
2008 revision of the Manual expressly describes the installation of an
epoxy or cork plug in the pinion. It contains a parts diagram at figure 441.
The part bearing number 25 on that diagram is an epoxy or cork plug. The
revision directs the mechanics attention to the hole permitting oil passage
into the duplex bearing. Mr. Mayhugh says he was aware of the requirement
that the bearings be adequately lubricated to ensure proper functioning at the
time he is recorded to have performed the overhaul, but, at his examination for
discovery, he was not sure of the direction of the flow of oil through the
transmission into the tail rotor drive quill assembly.
[47]
The revised Manual directs
that the forward surface of the epoxy or cork plug should be located 3.68
inches aft of the shoulder or lip on the inside of the forward end of the
pinion. Mr. Mayhugh acknowledges that the cork in the pinion gear at the
location where it was found after the Advantage accident is less than 3.68
inches from the shoulder. He acknowledges that the cork at the position in
which was found after the accident was in the wrong place.
[48]
Mr. Doss acknowledged that on January 9, 2008, Chad Barta wrote to
Bell Helicopter and indicated that the Manual for the 204 helicopter did
not refer to a cork and asked how the maintenance and repair facilities were to
install that cork without guidance in the Manual. He acknowledged receipt
of directions from Bell Helicopter in January 2008. He also acknowledged
receipt of notice of the crash by way of a telephone call from Mr. Mizera and
admits he did not speak with the employees responsible for overhauling the
transmission, Mr. Mayhugh or Mr. Felice, when he received that notice.
He did not ask either of them whether they had installed the cork in the pinion
gear or, if so, how they had done the job. Heliponents did not conduct an
internal investigation.
[49]
Mr. Mayhugh confirmed he was not advised when Heliponents received
notice of the crash in this case. He was not part of a formal or informal
investigation or meeting to discuss the crash.
[50]
Mr. Doss was not certain
whether this was the first Bell 204 main rotor transmission that had been
overhauled by Heliponents. The company did not do a great deal of work on 204
helicopters. He cannot say why a 212 worksheet was used on this job, rather
than a 204 worksheet. He does not know if Heliponents had a 204 worksheet. He acknowledged
that a 212 transmission assembly overhaul kit was used for this overhaul,
rather than a 204B overhaul kit.
The Plaintiffs Evidence Re: Maintenance Procedures
[51]
Mr. Mizera was asked about
inadequacies in the Manual used by Heliponents. He confirmed that the Manual
does not describe the cork in the quill assembly. He knew from experience
where to install that cork. He has received new gears with a cork installed at
a location just aft of the hole in the pinion that permits oil to flow into the
duplex bearing. He says the quill assembly requires a cork to work properly. He
has always inserted the cork from the spline end, at the aft end of the pinion,
rather than the beveled (front) end. He was aware that when the cork is
inserted it should be soaked in shellac and pushed forward in the bore until it
is appropriately placed. After installation, the installer should check to see
the oil hole is clear. He uses a lock wire to do so.
[52]
When the overhaul of the
transmission was completed, it was returned to Rilpa and to Advantage for
installation in the helicopter. Mr. Partridge was asked about the
transmission-to-engine alignment check that is normally done by the helicopter
owner on receipt and installation of an overhauled transmission. He could not
find the Advantage record documenting the completion of that alignment check. Mr. Partridge
considered that to be unusual. He has no clear memory of the alignment check
being done. It is his evidence, however, that if the transmission was out of alignment
the input shaft would overheat. That could cause a failure of the input drive
shaft.
Plaintiffs Expert Evidence
[53]
George Heath of R.J. Waldron & Co., a company primarily engaged in
aircraft accident investigation and failure analysis, prepared a preliminary
investigation report on December 5, 2007, at the request of the plaintiffs. Mr. Heath
did not testify; his opinion was only in evidence to the extent that it was
adopted or commented upon by other experts. Mr. Heath documented a number
of observations, all of which are borne out in the photographs that were
admitted into evidence and the testimony of other witnesses (with the minor
exceptions noted):
a)
the pinion assembly was bent and fractured adjacent to the aft duplex
bearing;
b)
the shaft was heavily rubbed at that point;
c)
the roller bearing was undamaged and ample clean lubricant was present;
d)
the duplex ball bearing assembly was seized and heat-discolored, and the
inner race was worn and deformed;*
e)
the large gland nut that loads the duplex bearing set was severely
damaged by the flailing pinion shaft;*
f)
the tail rotor drive shaft coupling had some spline damage from
overtravel;*
g)
the temp-plates did not record any abnormal temperature;
h)
the plug was found installed in the interior of the pinion assembly
about 1.25 inches aft of the bevel gear forward face; and
i)
the tail rotor drive shaft was fractured about 12 inches aft of the
coupling. There was heavy contact wear at the fracture for about half the
circumference, indicating eccentric rotation as the duplex bearing set failed
to support and position the pinion assembly.*
* (The accuracy of these
observations is unquestioned, but the conclusions expressed by Mr. Heath
with respect to the cause of the damage observed are not properly in evidence.)
[54]
David Rupert was called as an
expert witness on behalf of the plaintiffs and testified to the opinions
contained in his report of February 24, 2010. Mr. Rupert obtained a
three-year mechanical engineering technology diploma before he was hired by the
Transport Safety Board of Canada in 1971. He worked with that organization
until 1984 as an analyst/technologist and technical investigator, among other jobs.
In 1984 he moved to R.J. Waldron & Co.
[55]
Mr. Rupert examined documents,
including photographs of the helicopter at the accident scene and photographs
later taken at the Advantage hanger, and also examined the component parts of
the quill assembly brought to R.J. Waldron & Co. for examination. He was
not retained to investigate the accident generally but, rather, to examine the
parts provided to him and report on conclusions that could be drawn from
examination of those parts.
Damage to the Duplex Bearing and Pinion
[56]
Mr. Rupert says the inner
races of both elements of the duplex bearing were blackened by exposure to high
temperatures and had been rubbed circumferentially by the pinion gear shaft.
The inner races are not designed to rotate relative to the shaft. He says the
inner race of the aft duplex bearing was severely deformed. The balls were
discolored and only a few pieces of the ball cage were visible.
[57]
Mr. Rupert says asymmetric
rub damage was found on the outside diameter of the pinion shaft adjacent to
the fracture. That rub damage extended around approximately 260° of the drive
shaft circumference with the heaviest rub damage being opposite the areas with
no rub damage. Mr. Rupert thought this was a one-time overload fracture.
There was, in his view, no evidence of progressive fatigue prior to fracture.
Damage to the Coupling
[58]
Mr. Rupert thought there were
three anomalies in the coupling assembly. First, there was rub damage on the
spacer adjacent to the outer coupling, thought to be due to the failure of the
pinion gear shaft. The second anomaly was asymmetric damage, deformed or chipped
spline teeth, to the mating splines of both the inner and outer couplings. Mr. Rupert
believes the spline damage was caused by over-angling of the couplings relative
to each other as result of the pinion gear shaft failure. The third anomaly
noted by Mr. Rupert was blackening of only one of the two temperature
indicators mounted on the exterior of the coupling. Mr. Rupert attributes this
anomaly to a defective temperature indicator, the one that has been blackened. It
is his view that where one indicator is black and the other is not, that is
suggestive of defective installation, rather than the presence of elevated
temperature. He says that is more likely than a scenario where more heat has
been generated on one side of the coupling than the other side. In
cross-examination, Mr. Rupert suggested that hot grease could result in
discoloration of one of the temperature indicators. That was not canvassed as a
possibility in his report and appeared to be an afterthought on his part, but
this opinion was also expressed by a defence expert, Dr. Morris.
Installation of the Cork
[59]
It is Mr. Ruperts opinion, based on his examination of the pinion,
that the cork plug appeared to have been installed flush with the forward end
of the taper on the inside diameter of the pinion gear shaft, approximately
0.75 inches aft of the front end of the taper. It should have been installed
3.68 inches aft of the front of the taper. Mr. Rupert thought the cork may
have moved approximately a quarter of an inch from its original position, but
no more, as a result of the bonding material having partially melted during the
failure sequence of the pinion shaft. The evidence that the cork moved slightly
after installation appears to be found in the edge of the epoxy that extends
past the ridge but still conforms to its shape.
Failure Sequence
[60]
Mr. Rupert believes incorrect
installation of the cork blocked the passage of oil to the duplex bearing.
There was some oil provided to the duplex bearing, through the roller bearing,
but not enough to lubricate the part for its designed lifetime. Over time, at
least one element of the two duplex bearings failed. The bearing seized,
disrupting the interference fit (or pressure fitting) that ordinarily prevents the
inner race from rotating relative to the pinion shaft. The inner race then
began to rotate relative to the pinion, resulting in metal-to-metal contact and
the rapid accumulation of heat on the unlubricated interface between the pinion
and the inner race. That heat weakened the shaft material and it deformed under
torsional and centrifugal stresses.
[61]
Deformation caused misalignment of
the tail rotor drive shaft. The forward tail rotor drive shaft segment came
into contact with the helicopter structure where the shaft passes through the
firewall between the transmission bay and the engine bay, and failed. At that
point, there was a complete loss of tail rotor drive. The pinion gear continued
to rotate in the quill assembly and rub against the duplex bearing and sleeve
nut and it failed completely within a matter of seconds.
Overhaul Procedure
[62]
Mr. Rupert was critical of the
overhaul procedure. In particular, he pointed to the following concerns: reliance
on the Manual without reference to the installation of a cork; the
failure to properly implement the worksheet to ensure that the overhaul was
done correctly; the failure to use the proper worksheet; and the failure to use
the proper overhaul kit containing necessary parts and, apparently, some
instruction.
[63]
Mr. Rupert says the traveler
form used by Heliponents was initialed at items 14 through 18, 20 and 22 on May
17, the date of the airworthiness approval tag and authorized release
certificate. This suggests that all but one of the entries on the record were
initialed after all the work was completed on the transmission. It suggests
that inspection was not conducted, as required, on a periodic basis as the work
progressed. The records of the service are not helpful in determining
whether, and if so, where the cork was inserted in the drive shaft in the
course of the overhaul.
[64]
In this case, a Bell 212 overhaul
kit was used. The 212 overhaul kit calls for the use of what is described as a 15-A
cork in the 212 tail rotor drive quill. The latest Bell 204 parts breakdown
manual calls for a 17-A cork. It is unclear what cork was in fact installed
in this case because of the lack of records and the damage suffered by the
cork. A 15-A cork is smaller in diameter than a 17-A cork. Dr. Morris, the
expert witness who testified for the defendant, explained that when he examined
the cork he found its diameter was not as large as it should have been. There
is a possibility the wrong cork was used. In my view for reasons set out below,
resolution of this question is unnecessary to address the plaintiffs claim.
Time in Service
[65]
When Mr. Rupert prepared his
expert report he was well aware of the circumstances surrounding the crash of a
Bell ADH1G Cobra helicopter owned and operated by Garlick Helicopters at
Lake Pend Oreille, Idaho, on October 18, 1991 (the Cobra accident). He had
investigated that accident and prepared a report with respect to its cause on
May 1, 1992. Mr. Ruperts report of the Cobra accident was referred to
extensively because the Cobra accident was acknowledged to have been caused by
the misplacement of a cork in the quill assembly of a Bell helicopter with very
similar design features. Defence counsel emphasized that the Cobra accident
occurred very soon after the helicopter was returned to service (after about
six hours of flight time) and argued strenuously that no helicopter could
function for long if the cork was misplaced in the pinion at overhaul.
[66]
Mr. Rupert knew the Advantage
transmission had accumulated 559.9 hours, on a 1500 hour time-between-overhaul
schedule, after the completion of the overhaul and before the accident. It was
suggested to him that it was unusual that the Cobra accident was not mentioned
in his initial report in this case and suggested that he did not mention the Cobra
accident to his colleague Mr. Heath. Defence counsel put it to Mr. Rupert
that because he was unable to account for the time lag between the overhaul and
the accident in this case, he did not wish to draw attention to the quick
failure of the Cobra helicopter. Mr. Rupert denied this suggestion, noting
that the report on the Cobra accident had been released by R.J. Waldron &
Co. to the Transportation Safety Board and he did not expect it to be
concealed.
[67]
In fact, Mr. Rupert had some difficulty
reconciling the differences between the present case and the Cobra accident. In
cross-examination he identified what he suggested were a number of distinguishing
features of the Cobra accident: the longer flight time in this case might be
accounted for by different use of the equipment; perhaps the Cobra helicopter
had been put to heavier use. However both accidents occurred while the
helicopters were carrying significant external loads. The Cobra helicopter was
being used to carry water on firefighting missions. There was no reason to
suppose that placed greater demands on the equipment than heli-logging. He
suggested that the Cobra helicopter might have experienced loss of tail rotor
control while it was a higher altitude but later learned that the Cobra
helicopter had been about 50 or 60 feet above the surface of a lake when it lost
control, whereas the Advantage helicopter was about 200 feet off the ground.
[68]
Some differences in the failure
sequence and the physical evidence were said by Mr. Rupert to be
insignificant. In the Cobra accident the pilot heard a loud high-pitched squealing
sound for two or three seconds prior to the failure but did not report the
presence of a burning smell. In the case at bar the pilot smelled something
burning but did not report any significant sound. Mr. Rupert explained there
are always variances in the sequence of events when a part fails. It was not
suggested to Mr. Rupert that Mr. Battys account of events in the
case at bar was less suggestive of a bearing failure than the pilots account
in the Cobra accident.
[69]
Comparing the Advantage pinion to the
Cobra pinion, Mr. Rupert expressed the view that the bend in the shaft in
both pinions is at approximately the same location. In the Cobra accident the
pinion became deformed and the drive shaft aft of the quill assembly then
fractured. The Cobra pinion did not fail completely. Mr. Rupert believes
the failure sequence continued further in the Advantage accident; the fracture
of the drive shaft was followed by further stress upon, and ultimately the
fracture of the pinion itself.
[70]
Mr. Rupert acknowledged that
when the Cobra accident was investigated all of the major components of the
helicopter were saved and inspected. The investigation conducted by Mr. Rupert
was more extensive and complete than that conducted in this case. The bearing
balls were heat discolored and demonstrated varying degrees of skid damage and
deformation. The bearing outer rings were also heat discolored and demonstrated
transfer of material onto the ball raceways. The metal of the raceways was
severely smeared and distorted. All this damage was consistent with the duplex
bearing having failed as result of inadequate lubrication.
[71]
In his report in relation to the Cobra
accident Mr. Rupert wrote that when the plug is installed at the forward
end of the gear shaft no oil is able to flow down the shaft and through
the oil hole to lubricate the duplex bearing. Mr. Rupert now modifies
that opinion, noting that the primary source (but not the only source) of
lubrication for the duplex bearing is oil from the number four jet that flows
through the centre of the pinion.
[72]
Mr. Rupert says a factor that
might lead to the quill lasting longer in the Advantage case is oil seepage
past the plug to the duplex bearing. Each bearing is somewhat different,
bearings have different dimensional tolerances. Different lubricants might have
been used. The plug or cork in the Cobra pinion appeared to completely seal the
hollow center of the pinion.
Adequacy of Investigation
[73]
Mr. Rupert was criticized for
not having formally analyzed how long the duplex bearing would last without its
designed lubrication, under external load conditions. He acknowledged that he
did not have all the pieces of the tail rotor drive shaft in his possession
when he wrote his report. He looked at photographs of the drive shaft but had
not analyzed the photographs to see what failure mechanism could be identified
in them. When he prepared his report in February 2010 he did not know there
were multiple fractures in the tail rotor drive shaft.
[74]
He was not aware that Advantages
maintenance records make no reference to the performance of an
engine-to-transmission main drive shaft alignment check having been performed
when the transmission was installed in June 2007. He acknowledged that check is
designed to ensure that the engine and transmission operate properly within
stipulated tolerances following an overhaul. He agreed misalignment can lead to
damage to the input drive shaft but he was not asked about evidence of such
damage in this case.
Defendants Expert Evidence
[75]
The defendant relies upon the reports of two experts: Joseph Sislo, a
very experienced helicopter pilot and accident investigator, and Dr. J.W.
Morris, a very well-qualified metallurgist and material scientist.
Joseph Sislo
[76]
Mr. Sislo says the accident was not thoroughly investigated and
that it is not possible to rule out causes other than failure of the tail rotor
drive quill. It is his opinion, based on the reported hours of operation of the
main transmission after its installation, and the appearance of the duplex
bearing set, specifically the aft segment of that set, that the accident was
not the result of failure of the quill assembly,
[77]
The fact the helicopter flew for about 500 hours, approximately
one-third of the ordinary interval between overhauls of the quill assembly,
strongly suggests to Mr. Sislo that there was adequate lubrication. The
aircraft was engaged in repeated heavy lift operations, apparently including
lifts in excess of the maximum lift capacity of the helicopter.
[78]
On the basis of his limited examination of the bearing. He says it does
not show the appearance of lubricant-starved bearings he has seen in the past.
The balls were retained within the bearing, and their appearance, in his view,
does not evidence the sliding, flattening, skidding, or metal transfer that he
would have expected to see in the case of a catastrophic failure.
[79]
Given the cursory nature of his examination of the bearing and the
significantly greater expertise of other witnesses, I give relatively little
weight to Mr. Sislos testimony with respect to the evidence of oil
starvation on the bearing.
Dr. J.W. Morris
[80]
Dr. Morris taught at the University of California at Berkeley from 1971
until his retirement in 2011, and is an expert in the field of materials
science and failure analysis. His first report related to the accident was
dated August 10, 2010. When he prepared that report, Dr. Morris had
visited Heliponents and examined an exemplar of the tail rotor drive quill and
discussed its assembly. He had visited the offices of R.J. Waldron & Co. and
examined the available parts of the tail rotor drive quill then in their possession.
He performed some tests on an epoxy compound in order to compare the tested
material with that affixing the cork to the inside walls of the pinion. He
examined documents, including the Heath Report of December 5, 2007, and a
Transportation Safety Board of Canada Safety Advisory, apparently generated as
a result of the Heath Report. While his investigation was limited, and
described as ongoing when he wrote his first report, he expressed the opinion
that he had sufficient metallurgical evidence to state that the Heath
hypothesis was almost certainly wrong.
[81]
Dr. Morris reached the following conclusions:
(1)
Although he had not yet conducted any destructive testing or opened the
duplex roller bearing set, he concluded that it had not seized and was of the
view that it did not cause the failure. He believed the duplex bearing set was
operational until sometime after the pinion shaft failed.
(2)
He could see no significant evidence of oil starvation on the visible
parts of the roller bearing set.
(3)
He believed the metallurgical evidence to be consistent with the
conclusion that the pinion shaft was deformed by some event that happened aft
of the roller bearing set (by which he meant the duplex ball bearing set). There
was evidence of a one-sided mechanical interference with the duplex bearing
inner diameter, leading to its deformation and failure. The inner race of the
bearing had been deformed by a mechanical cause and the tail rotor shaft was
fractured as a result of torsional load. The mechanical load on the bearing and
the quill assembly was a result of a force applied from the rear that could not
be determined because of the loss of the relevant parts. He concluded that the
cork seal did not play a causal role in the failure.
[82]
From the outset, it was Dr. Morris view that Mr. Heaths
theory was not consistent with the following evidence:
(1)
the bearing was not seized when he first examined it;
(2)
the visible bearing balls were not significantly distorted;
(3)
the inside surfaces of the bearings were not scored;
(4)
the thermal damage was local; and
(5)
there is persuasive, affirmative metallurgical evidence that the ball
bearing set was functional at the time the pinion shaft broke and continued to
rotate for some time thereafter.
[83]
Dr. Morris second report, dated September 4, 2011, was prepared
after the delivery to him of the 1992 report on the 1991 Cobra accident.
Comparison of the evidence collected in relation to each of the two accidents
confirmed Dr. Morris view that they were the result of different causes.
[84]
Dr. Morriss third report was dated December 6, 2011. This further
and more comprehensive report was prepared after review of additional documents,
the examination of parts (including the main parts of the tail rotor drive
quill and remaining parts from the gearbox, shaft, and fuselage of the
helicopter); and the performance of tests on the properties and behaviour of
the epoxy compound used to fix the cork in the drive quill assembly. Among the
documents Dr. Morris had reviewed when he wrote his December 2011 report
were extensive sets of photographs of the Advantage helicopter.
[85]
Dr. Morris concluded that the evidence made available to him was
not consistent with a lubrication failure. The metallurgical evidence in
support of his conclusion was said to establish:
(1) the duplex bearing set was
lubricated at the time of failure;
(2) the drive shaft did not
rotate within the bearing inner races;
(3) the duplex bearing did not
disintegrate;
(4) the aft duplex bearing
rotated with the shaft at the time of the fracture;
(5) the time-to-failure is not
consistent with a lubrication failure;
(6) the failure was rapid
and traumatic rather than progressive; and
(7) other
failures apparently preceded the fracture of the pinion gear.
[86]
It should be noted that when he testified at trial, Dr. Morris was careful
to qualify every reference to his view that the evidence was not consistent
with a lubrication failure by saying what he meant was the evidence was not
consistent with a failure initiated by a lack of lubrication.
[87]
Finally, Dr. Morris prepared a responsive report on January 13,
2013, entitled Metallurgical Analysis of Damage in the Failure And Hard
Landing Of Bell Helicopter 204B, CGREL.
Damage to the Duplex Bearing and Pinion
[88]
Dr. Morris says he could easily rotate some of the balls in the aft
duplex bearing with finger pressure when he first examined the bearing set. The
rotation was limited, but Dr. Morris felt it was limited by geometric
constraints due to external and mechanical damage rather than heat damage to
the balls or the races.
[89]
That limited external examination of the balls led him to conclude the surface
damage was limited and some parts of the ball surfaces were relatively unworn.
He concluded this damage was inconsistent with the dramatic damage usually
observed as a result of oil starvation.
[90]
Dr. Morris says the inner diameter of the bearing, where there
would have been maximum heat accumulation according to the Heath report, was
not severely worn or scored, nor was there a metal smear upon its surface. He
concluded that the pinion shaft did not rub on a stationary inner bearing race.
[91]
Dr. Morris says the thermal damage to the duplex bearing was local,
and was only really severe near the aft surface of the bearing set. The
temperature indicator strips on the outer coupling, which ought to be triggered
by heat in the range of 250 degrees centigrade, had not been triggered. He
concluded that the metal temperature immediately behind the duplex bearing set
never reached even what he refers to as this moderate value.
[92]
He believes the duplex bearing was functional during and after the shaft
failure. He says that scoring marks aft of the duplex bearing were made by a
turning pinion shaft, after the failure of the tail rotor drive shaft.
[93]
Dr. Morris second report, dated September 4, 2011, describes the
more extensive damage to the duplex bearing removed from the Cobra helicopter
following the 1991 accident. That duplex bearing had disintegrated through
thermal distress, while the pinion shaft remained intact. Examining the
photographs of the 1991 failure, Dr. Morris describes the damage as
pervasive. He notes:
The bearing balls are deformed
and distorted, the bearing cages are broken, melted and re-solidified, and the
bearing races are heavily scored and distorted.
On the other hand, the quill
shaft is bent but intact. In contrast the 2007 accident did not cause
significant damage to the duplex bearing.
[94]
Dr. Morris is of the opinion that the Advantage duplex bearing continued
to turn after the pinion shaft broke, but the Cobra duplex bearing could not
have done so, because of its disintegration. Dr. Morris concludes:
it seems clear that the 2007
failure initiated in the quill shaft while the 1991 failure initiated in the
duplex bearing set.
[95]
This opinion was unaltered following destructive testing and more
detailed examination of the duplex bearing that preceded Dr. Morriss
third report. That examination, in his view, showed none of the expected
symptoms of progressive failure prior to the fracture of the quill pinion shaft.
If there had been lubrication failure, according to Dr. Morris, one would
expect more significant degradation of the bearing.
[96]
Dr. Morris conclusion that there was not a bearing failure
in this case stems, in part, from his examination of the bearing itself, and in
part from his view that the damage at the pinion/bearing interface is not
significant. Dr. Morris does not believe there is evidence that the pinion
shaft was spinning against the inner races of the bearing before its failure. Mr. Rupert
and Mr. Heath say the pinion failed because the roller bearing became
deformed and stopped rotating, causing its inner ring to rotate against the
pinion. That metal-on-metal contact is said to have caused excessive heat,
deformation of the pinion and, eventually, its failure. Dr. Morris is critical of the Health/Rupert theory as he says that
theory requires the shaft to rotate under high force against the inner rim of
the bearing but he says there is no evidence of that.
[97]
Dr. Morris describes what he
refers to as very light scoring of the inner rim of the aft bearing race which could
have been caused by motion under fairly low load but doesnt show severe
scoring, the smeared metal, the evidence of overheating, or contact on the
surface that would be expected if the inner ring of the bearing seized on the
shaft and the shaft was trying to rotate out of the seizure (I should note that
is not an entirely accurate description of Mr. Heaths views; it would be
more correct to say that Mr. Heath believes the bearing ceased to function
efficiently and resistance overcame the pressure fitting between the inner race
and the pinion).
[98]
Dr. Morris says the only part
of the pinion that appears to have been heated significantly is the part that
was actually subjected to high friction loads: the side that wore against the
bearing and the nut. Dr. Morris finds the duplex bearing to have
suffered little thermal damage but significant mechanical damage. The balls are not severely deformed except in isolated
places where there appears to have been skidding, they have retained their
spherical shape and look like they should look. The mechanical damage to the
bearing is said to be suggestive of an impact.
[99]
Dr. Morris says the
metallurgical evidence is inconsistent with the plaintiffs hypothesis. When
operating temperatures exceed 250°C the steel balls and races begin to anneal
and soften. When they soften, the balls will lose their geometry, become
inadequately lubricated and become further distorted. That process doesnt
happen at all below 250°C. Toward the end of the process the balls and races
become so hot that bearing material is smeared onto the races, resulting in
local melting and deformation. Eventually the metal welds the races together,
causing the bearing to seize. If the bearing seizes, the shaft must fracture
immediately, rotate the rigid bearing within its case, or rotate inside the
seized inner race. If the shaft separates from the inner race there should be
severe scoring and local wear.
[100] Dr. Morris says the bearings in this case did not
disintegrate. He says the balls show no evidence of severe long-term wear,
softening, creep, distortion, progressive smearing or welding. In comparison
with those taken from the Cobra helicopter, the balls were in relatively good
condition. There is metallurgical evidence that the aft duplex bearing was
fixed on and rotating with the pinion shaft for some time after the
pinion shaft broke. The damage to the inner race and the sleeve nut of the
bearing showed opposite rotations. Dr. Morris says we can be confident
that both the damage to the inner ring and the damage to the sleeve nut were made
by the aft segment of the broken shaft after the fracture.
[101] Dr. Morris believes the pinion shaft was deformed
by damage to the drive shaft aft of the roller bearing set. He says a load
appears to have been applied from the rear down the tail rotor shaft. He postulates
that the initiating event of the failure sequence was a fracture of the drive
shaft that caused cocking or misalignment of the pinion and then a misalignment
of the bearing which, ultimately, caused it to fail.
[102] It is Dr. Morris view that the limited visible
scouring of the aft inner race that is present occurred after the
fracture of the tail rotor drive shaft. The inner race was still turning with
the shaft under power for at least some time before it became frozen.
[103] He explains the mechanical damage to the bearing as
follows: mechanical contact pushed the aft bearing back against the forward
bearing; as it was twisted and heated, the pinion became thin and turned, with
very little force, on the inner ring. He concedes that the pinion would not
have bent or deformed significantly before it became heated. He acknowledges,
given the evidence that the brass in the duplex bearing cages was found to be
melted and adhering to the ball bearings that the temperature of the bearings
probably rose to between 800-1,000°C. When asked whether he determined the
melting range for the alloy of brass that was used for the cages, he said that
he saw no reason to be concerned about that issue.
[104]
Dr. Morris clung tenaciously to the view that the duplex bearing
continued to function as a bearing, even after the detailed destructive
testing. In his testimony at trial Dr. Morris was asked:
Q We now know, although well get into that
later, that having looked at them, theres marks of skidding and other problems
with the ball bearings; correct?
A Yes, but they
turn freely today
[105] That opinion is at odds with the physical evidence at
trial and inconsistent with the evidence of Dr. Cepu, the plaintiffs expert witness, which I accept, that
only slight movement of the duplex bearing can be demonstrated now if the inner
ring is forced to turn. As that occurs, each of the balls is stuck in the
impression it has made in the metal of the race. The balls skid or scrape
against the races; it is not a functional bearing. The bearing cannot be made
to turn and cannot be said to turn freely.
[106] Dr. Morris denied that his theory of causation
rested upon establishing a means by which the force of an impact could be
transferred forward into the quill assembly and the duplex bearing. He did,
however, describe impact damage to the bearing.
[107] In his last report, prepared following the destructive
testing and examination of the duplex bearing, Dr. Morris says there is
clear evidence of mechanical damage to the bearing, particularly evidence of
the movement of the balls of the bearing onto the shoulder of the races and,
what he refers to as highly polished regions, brinelling of the surfaces,
caused by vibration of the balls in a fixed position. He also notes that the
bearing has gone through a substantial failure that clearly did involve a
significant amount of heat but, on page 11 of his last report, he says:
The aft face
of the inner ring was only lightly damaged. (Figure 18), excepting one local
impact mark (figure 19).
[Emphasis
added.]
[108] He referred to this mark in his re-examination in
chief on March 20, 2013, as a mechanical impression and gave the following
description of the photograph, marked as figure 30 in his last report:
There is a region around the
inner race where there are no balls, and that region corresponds to a
deformation that you cannot see in this figure that is on the inner ring on the
inside that was the impact that apparently caused this distortion to happen.
[109] Despite this testimony to the effect that the bearing
was deformed by an impact, he later testified in cross-examination, on March
21, 2013, that he did not believe the quill assembly was affected by an impact.
He said it was damaged as a result of the mechanical load from the rear
triggered by an unknown incident. When it was suggested to him that Dr.Cepu, testifying for the
plaintiffs, understood Dr. Morris to
have attributed damage to the duplex bearing to impact originating from the
rear, the following exchange occurred:
Q. So youre distinguishing between Dr. Cepu interpretation of that as an
impact to the rear of the unit, and youre saying a load from the rear is not
necessarily an impact? Is that —
A Of course not.
Q — is that your point?
A In fact, I just did a word search on that
first report for the word impact. And according to the word search, it nowhere
appears. Perhaps the word search is wrong, but thats what happened.
Q All right. So do you say the load from the
rear, then, was not initiated by an impact?
A I dont believe so. It appears to have been
initiated by as I go through in these reports, by a buckle in the tail
rotor shaft.
Q So we can discount impact as being part of
your hypothesis?
A Sir, I believe I
said I dont know precisely what caused that buckle. There could have been some
impact involved. I really dont know. We have no evidence.
[110] Finally, in relation to damage to the duplex bearing, Dr. Morris
says the chip detector should have been activated by debris if there was some
lubrication to the duplex bearing but insufficient lubrication to prevent
overheating. That could only be so, of course, if the metal in the duplex
bearing began to degenerate while lubrication continued to flow through the
bearing.
Damage to the Coupling
[111] The shaft
of the tail rotor drive quill pinion gear was said by Dr. Morris to be
fractured by a torsional force. He concluded that it was being driven by the
main transmission against a resistance that interfered with the rotation of the
rear section when it fractured. Dr. Morris considered damage to the outer
coupling to be evidence that the pinion gear was attempting to drive the
coupling against strong resistance while the pinion was cocked. These
indentation marks on the coupling are said by Dr. Morris to be evidence
that the coupling was mechanically stressed while the pinion gear was still
rotating with force.
[112] Dr. Morris
notes that over-temperature indicators on the outer coupling had not become
discoloured. He suggests that one was completely
immersed in hot lubricant following the accident, while the other was not (as
noted above, that explanation is not disputed by Mr. Rupert). Dr. Morris
says the fact that one indicator did not change color is evidence that the
coupling, even though it was close to the failure, remained relatively cool
through the whole event. He concludes that the thermal distress that
accompanied the failure, while intense, was very local and short-lived. In a
short period of time the heat could not be conducted through the surrounding
thermal mass.
[113]
Both experts agree that there was sufficient heat generated at the
bearing/pinion interface to result in the steel pinion becoming so malleable
that it could be twisted and break in torsion. Both experts agree that local
heat was not intense or long lasting enough to raise the heat of the outer
coupling sufficiently to activate both temperature indicators. That evidence,
however, is not particularly helpful in addressing causation, because the
experts agree there was very significant heat generated at the interface
between the pinion and the duplex bearing. The issue is whether that heat was a
cause or effect of the drive shaft failure.
[114] Dr. Morris
believes the evidence of a significant tensile load on the drive shaft while it
was in rotation is supportive of his theory that there was a misalignment at
the coupling, resulting in severe mechanical damage. Dr. Morris has
concluded that the drive shaft fractured in a sequence that first left certain
marks and gouges; following that there was a second fracture that left marks at
a greater angle; and last, a fracture that left the small stub of the drive
shaft completely free and spinning at 4300 rpms with enough force to heat and
fracture the pinion.
[115] Dr. Morris says the damage to the sleeve nut is
evidence that the pinion fragment made a few revolutions after it broke. The
photographs of the aft end of the drive quill unit are said to demonstrate asymmetric
damage to the nut. Severe scoring goes round the full 360° circle; this,
according to Dr. Morris, is evidence that the aft segment made at least
one complete rotation, and perhaps several, rubbing against the stationary sleeve
nut. Further, Dr. Morris says the fracture at the rear of the bearing set
caused the lubricant contained within the bearing to spray out to the rear. The
primary spray pattern was radial, due to the rapid rotation of the pinion and
bearings. There is lubricant seen all over the transmission bay.
[116] Dr. Morris says there was moderate mechanical
damage and evidence of thermal distress at the splined interface of the pinion
gear and the inner coupling. Damage to the teeth is obvious in photographs. Discrete
indentations are visible on the outside surface of the coupling. There are deep,
local, mechanical impressions over one half of the external circumference of
the inner coupling. These indicate a significant load between the two coupling
surfaces. When these indentations were made, in Dr. Morris view, large
forces were attempting to push this coupling relatively forward against the
rear of the drive quill. It is also severely cocked with respect to the drive
quill, which has wider indentations in one small segment of the surface. There
must have been severe cocking in order to bring the teeth in contact with the
surface in this pattern. Dr. Morris believes the indentations were made as
the broken part of the shaft was being forced against the back of the unit.
[117] There are two discontinuous bands of damage, so in Dr. Morris
view, at least two separate impact events occurred, perhaps three, because there
is also damage to the interior of the splines. The outer coupling had to tilt
substantially in order for its teeth to come into contact with the inner
coupling. The outer coupling teeth were severely overloaded in an attempt to
turn the coupling in a clockwise direction.
[118] In my opinion, the evidence of damage to the coupling
and the sleeve nut does not strongly support either theory of causation. There
is certainly evidence that the pinion continued to rotate under power after the
tail rotor drive shaft fractured. However, the plaintiffs experts do not dispute
that was the case. Evidence that pressure was exerted on the pinion from the
rear of the helicopter would weigh in favor of the opinion of Dr. Morris but,
in my view, Dr. Morris evidence in support of the conclusion that
pressure was exerted from the rear is meagre.
Installation of the Cork
[119] Much of
the most detailed evidence in this case was introduced for the purpose of
establishing that the cork was seated in the position in which it was found
after the accident at the time of the overhaul or, in the alternative, that the
cork could, and did move within the centre of the pinion as result of the
sequence of events that led to the crash.
[120] Dr. Morris
says there is some evidence the cork was originally placed in an appropriate
location in the pinion. He says there is reason to believe the cork fractured
at its edges and was pushed forward in the accident. He says the limited adhesion
of the cork to the wall of the pinion in the position in which was found after
the accident was a result of the re-bonding of the epoxy. He observed what he
thought was organic material on the aft wall of the bore of the pinion where
the cork would have been mounted and glued if it had been positioned correctly
at the time of the overhaul by Heliponents. He suggests this is a remnant of
the original placement, most of which was obliterated by extreme heat.
[121] Dr. Morris
says there is some evidence the cork moved to the location at which it was found
after the accident as a result of the forces exerted upon it at the time of the
failure of the pinion. When he ultimately examined the cork, Dr. Morris
observed what he considered to be signs of both thermal and mechanical stress.
He says the epoxy on the cork would have flowed over the surface and produced
smooth surfaces and a meniscus bonding to the wall at the time the cork was
installed. What is seen in the photographs taken after the accident is said to
be a rippled surface that has the appearance of an epoxy that has liquefied
and then again solidified. He says the epoxy probably flowed in high temperature
creep. In certain photographs it is said to be evident that the epoxy has
drawn back from the cork and the epoxy has partly de-wet from the cork surface.
He says the epoxy at the front end of the cork has the appearance of having
been pushed up to its present position. That coating, previously in contact
with the inside wall of the bore, has become detached and is dangling in free
space. Dr. Morris regards this as essentially definitive proof that this
cork, with the epoxy on it, was pushed forward mechanically at some point.
[122] If the cork was pushed forward when the accident
occurred, as is described by Dr. Morris, it could have moved through a
relatively wider section of the bore of the pinion to the point where it would
become trapped at the front end of the pinion where the bore slightly narrows
to a lip, the position in which it was found after the accident.
[123] In
Dr. Morriss opinion, there was less damage to the cork than would have been
expected at the extremely high temperatures that ought to have been present if
the plaintiffs theory is right. The cork appeared
to be fractured or broken away at the edge; it was charred but not burnt.
Burning would not leave certain of the features he says are present on the
surface of the cork. There was no epoxy on the fractured surface beyond a small
residue that extruded between the cork and the inside surface of the pinion
bore.
[124] In support
of his theory that the epoxy used to hold the cork in position can re-bond
after creeping at moderate temperature under load, Dr. Morris performed
an experiment at home. He had no difficulty getting the epoxy to creep and
re-bond. This experiment was not mentioned in his report; it was not conducted in
a scientific fashion; and it was not measured, documented, and reported in such
a manner as to be tested under cross-examination or by the plaintiffs expert.
He heated a quantity of epoxy after it had cured in a toaster oven and clamped
a small threaded steel pipe on top of it. After heating it for an hour he found
the steel had bonded to the epoxy that had previously set. I consider this
experiment to be some evidence, albeit evidence of little weight, because of
the unscientific nature of the exercise, in support of Dr. Morris theory
of causation. It is evidence to suggest that the epoxy can be caused to bond
under pressure with another surface after it has cured if it is heated to a
sufficient temperature. The experiment is not a complete response to the
evidence of Dr. Cepu
that if the epoxy is heated enough to cause its existing bonds to fail (which Dr. Morris
did not do) pyrolysis and thermochemical breakdown will prevent the epoxy from
again functioning as an adhesive. In Dr. Cepu opinion, if the cork was heated to such a
temperature that it could become freed from the walls of the pinion then the
epoxy could not have later re-bonded with the interior walls of the pinion.
Failure Sequence
[125]
Dr. Morris described the Heath hypothesis as follows:
In summary, he concludes that the
cork seal was misplaced at the time of overhaul, 560 hours before the failure.
The misplaced seal resulted in a dearth of lubricant, which led to the gradual
deterioration of the duplex ball bearing set. It eventually seized, that is,
the deteriorating balls welded the inner and outer races, turning the bearing
into a rigid ring. The pinion shaft continued to rotate within the frozen ring,
driven by the torque on the bevel gear. The severe friction between the
unlubricated surfaces caused the pinion shaft to overheat, extrude and
fracture. The failure then induced a simultaneous fracture of the tail rotor
drive shaft. … The central element of the Heath hypothesis is, of course, the
lubrication failure of the duplex bearing set, leading to seizure and the
spinning of the pinion gear against the bearing inner race, which he identifies
as the proximate cause of the shaft failure.
[126] This may
embody a misapprehension of the plaintiffs theory of causation which is a theory
of sequential failure: first, overheating and weakening of the pinion, leading
to misalignment of the tail rotor drive shaft; then, fracture of the drive
shaft; and last, following shortly upon that, total failure of the pinion.
[127] Dr. Morris
says the sequence of events described by Mr. Heath should have resulted in
extensive thermal damage to the parts surrounding the bearing as well as to the
bearing itself. He says the tail rotor drive shaft appeared from the
photographs to have fractured in at least two places: at the point where it
passed through the aperture between the transmission bay and the engine bay;
and in the engine bay beneath the engine exhaust shroud. This second fracture
was referred to by Dr. Morris as very important. Relying only upon the
photographs depicting the broken drive shaft in this bay he says there is
evidence of a buckling failure in the drive shaft. The photographs depict a
jagged fracture and a segment of the shaft that Dr. Morris says is
severely scored. The shaft was rotating when the scores were made, presumably
under power. Dr. Morris believes this damage, well inside the engine bay, is
evidence that the drive shaft was being driven after it fractured. Hence, he
says, the drive shaft fractured before the pinion failed entirely.
[128] Dr. Morris considers the evidence of damage to
the tail rotor in the engine bay to be inconsistent with the Heath theory, and
proof of its falsity, because he understood Mr. Heath to say that the
pinion failed before misalignment of the drive shaft. In fact, the plaintiffs
theory is consistent with the drive shaft becoming misaligned while it was
being driven, before failure of the pinion. It is also consistent with
continued rotation of the drive shaft for a short period following the fracture
of the pinion.
[129] Dr. Morris acknowledges some of the similarities
between the Cobra accident and the Advantage accident. In the Cobra accident the
pinion shaft overheated at the same location and was distorted in the same
manner as the pinion shaft in the Advantage quill assembly. The twisting of the
Cobra pinion apparently caused the tail rotor drive quill to rub against the aperture
in the firewall between the transmission bay and the engine bay, or against the
tail rotor drive shaft tunnel, leading to failure of the drive shaft. It was
suggested to Dr. Morris that shaft whirl would result from twisting of
the pinion and because of the stabilizing effect of the hanger by which the
drive shaft is suspended, the tail rotor drive shaft would have fractured at
the forward end of the first drive shaft segment. Dr. Morris disagreed
with the suggestion that the drive shaft fractures were similar in the two
incidents. He says, based upon examination of the photographs, that in this
case the scoring of the drive shaft is unilateral, suggesting the drive shaft
was tilted at the point where it passed through the aperture. The Cobra drive
shaft, on the other hand, was straight, because there appeared to be scoring
360° around the shaft.
[130] It was suggested to Dr. Morris that his theory
was inconsistent with significant heat damage evident in this case: if the loss
of authority was triggered by mechanical failure, there would be very little
time for the quill assembly to heat up to cause bearing damage after the pilot
noted loss of tail rotor authority and rolled off the throttle. Dr. Morris
responded in a manner suggesting he agreed that the extent of the heat damage
was inconsistent with a purely mechanical cause of the accident. He said: Well,
keep in mind we dont know exactly what caused this fracture, so there may have
been some preceding problem that was affecting the quill in some way. I just
dont know. I cant talk about that. I might be able to do if I had the parts.
He added, however, that he was confident that a couple of hundred revolutions
of the pinion shaft after an initial bending moment would cause the temperature
at the pinion/duplex bearing interface to rise above 820°C.
Time in Service
[131] In his second
report, dated September 4, 2011, prepared following his receipt of the 1992
report into the Cobra accident, Dr. Morris concluded that the 1991
failure, while involving the same drive quill assembly as that involved in the
Advantage accident, differed in critical details from the Advantage accident
and almost certainly had different causes.
[132] The
service times before failure in the two incidents were said by Dr. Morris
to differ by two orders of magnitude (from 6 hours in the Cobra case, to 560
hours in the Advantage case). In his opinion, there is no adequate explanation
why the small amount of oil that would travel through to the duplex bearing
from the roller bearing would have been sufficient to permit the duplex bearing
to operate over 500 hours of service and suddenly become inadequate at 560
hours. He thought the duplex bearing would fail in a very short time if it
were deprived entirely of lubricant. If anything is going to operate for
hundreds of hours, in Dr. Morris opinion, it must be properly lubricated.
There had to have been some cause of a lubrication failure after this period of
operation, other than an error at overhaul.
Adequacy of Investigation
[133] Dr. Morris
is critical of the manner in which the accident investigation was conducted.
He says there is evidence of multiple failures in the tail rotor drive shaft of
the helicopter but no effort was made to examine the drive shaft and
investigate the nature and cause of the fractures.
[134] Having
said that, Dr. Morris remained of the view that there is metallurgical
evidence sufficient to state that the Rupert/Heath hypothesis of the cause is almost
certainly wrong. He remained of the view that the parts that have been examined
strongly suggest that the failure was traumatic and primarily mechanical,
rather than progressive and thermal. He is of the opinion the pinion shaft
became misaligned by some traumatic event aft of the quill assembly, and that
the misalignment caused deformation, heating, and failure. The damage to the
duplex bearing was a result, not a cause, of the damage to the tail rotor drive
shaft, and Dr. Morris remained of the view that the cork seal played no
causal role in the failure.
The Responsive Opinion of Dr. Elvis Cepu
[135] Dr. Cepu
is a mechanical engineer with a doctorate
in materials engineering. He prepared a report for the plaintiffs counsel on
December 31, 2012, in response to the opinions of Dr. Morris.
Damage to the Duplex Bearing and Pinion
[136] Dr. Cepu
says that on his examination of the
duplex bearing there was sufficient damage to render the bearing inoperable.
Small movements of the sort that could be elicited by Dr. Morris do not
mean that the bearing has not seized. Dr. Cepu says that when he examined it, the forward bearing in the duplex set turned
with roughness and such friction that it could not be considered operational,
and that the aft bearing would not turn.
[137] He says the visible balls all had skid marks and flat
spots and were covered in melted-on brass. The balls were not spherical. There were
clear striations on the balls from skidding. Examination of the surface of the
balls under energy dispersive spectroscopy established that they were covered
in brass/copper alloy, not rust, as suggested by Dr. Morris. Dr. Cepu says that for Dr. Morris to refer to the ball bearings in this case
as relatively pristine is preposterous.
[138] Dr. Cepu
points out obvious deformation of the
pinion and notes that that deformation would not have been possible unless the
temperature of the pinion at the point of the deformation was extremely high.
[139] Dr. Cepu
says destructive analysis of components
has proven that significant thermal softening occurred; the balls in the aft duplex
bearing were almost 50% softer than normal due to overheating from prolonged
exposure to high temperature. The significant softening of the balls speaks to
thermal, rather than mechanical damage. Not only was there significant damage
to the aft bearing but the forward bearing also demonstrated substantial
softening due to overheating. That overheating, as expected, first occurred in
the bearing furthest from the source of lubrication, the aft bearing, but also
affected the forward bearing. The thermal damage, in Dr. Cepu view, is consistent with lack of lubrication. Dr. Cepu disagrees with the view that there are mechanical damage signatures. Surprisingly,
despite the fact he received the results of the thermal testing, Dr. Morris
admitted he had not studied the test results and he did not dispute the extent
to which they were evidence of substantial heating of the balls in the bearing.
[140] Dr. Cepu
expresses the view that the deformation
of the pinion in this case exceeded what the material is capable of at normal
operating temperatures. The deformation could only have occurred if the pinion was
hot and had a torsional load applied at the same time. He says the pinion
experienced prolonged exposure to elevated temperatures that could not have
occurred due to an impact event or torsion alone.
[141] The evidence set out in Dr. Cepu report is detailed, thorough, and compelling. In particular there is
thorough documentation of the significant observations made during the
destructive testing and microscopic examination of the components of the quill
assembly. These provide evidence of thermal damage to the duplex bearing,
coking of lubricant on the components (evidence of thermally degraded oil),
brass melting onto the ball bearings from the disintegrating cages, flat spots,
power transfer to the pinion during overheating, and damage caused by rotation
of the bevel gear against stationary inner races.
[142] Microscopic analysis of the surface of the bearings was
indicative of oil starvation, overheating and high friction skidding. The aft
cage had disintegrated and was incapable of maintaining the required spacing
and positioning of the balls. There was extensive evidence of the
disintegration of the cages of the bearing as a result of the development of
heat; this would have occurred over what is referred to by Dr. Cepu as some non-transient period of time, which was explained by him to mean
a period longer than that which would have occurred in the interval between the
fracture of the tail rotor drive shaft and impact with the ground.
Damage to the Coupling
[143] No damage away from the quill or the tail rotor drive
shaft exhibited any torsional resistance that would have caused a torsional
failure of the quill. The quill, however, exhibited numerous thermal signatures
that indicate it ran hot for some time and that the ultimate reason for failure
was overheating of the pinion while torque continued to be applied to it.
[144] There is also evidence in Dr. Cepu report supporting of the conclusion that as the failure occurred there
was aftward pressure on the components of the assembly, contradicting the
suggestion the accident was a result of a load exerted on the quill assembly
from the rear.
[145] Dr. Cepu
addresses, in detail, the theory that an
axial force was somehow transmitted forward to the quill assembly. Not only is
it difficult to explain how this could have occurred, but there is an absence of
damage that would support this theory.
[146] Dr. Cepu
says the damage referred to by Dr. Morris
as splaying of the inner race is from off axis rotation of a deformed bevel
gear or pinion and not from impact. Impact would not have left circumferential
lines in the direction of rotation.
Installation of the Cork
[147] In response to Dr. Morris theory that the cork
in the pinion shifted during the incident and was forced forward, Dr. Cepu says there are no forces that would have moved the cork, either forward
or backward, in the event of an accident. Those forces that are present are
too small to overcome cork friction with and adhesion to the bevel gear
(pinion) inner surface. There was lengthy and detailed evidence with respect
to the coefficient of friction between the cork and the wall of the pinion, and
detailed evidence with respect to the adhesive qualities of the epoxy that
might have been used. I am satisfied that if the cork had been installed in the
appropriate location at the aft end of the pinion shaft it would not have
easily been dislodged. While Mr. Cepu was
extensively cross-examined on his evidence with respect to the characteristics
of cork, the fundamental proposition that the cork would have been firmly seated
in the pinion if it had been correctly installed at its aft end was not
challenged by the defendant.
[148] Dr. Cepu disputes the suggestion there was any
significant organic residue on the wall of the pinion at the location in which
the cork ought to have been installed. The small amount of material on the wall
in that location, in Dr. Cepu opinion, is burnt-on oil. There is, on the
other hand, significant organic residue on the walls of the pinion at the
location at which the cork was ultimately found after the accident. There was
no similar-looking debris towards the splined end of the pinion where the cork
should have been.
[149] Dr. Morris
testified that on that microscopic examination of the cork it was possible to
see signs of chatter marks or displacement cracks from friction. The presence
of these marks is not inconsistent with the plaintiffs theory, which is that
there was some minimal movement of the cork.
[150] Dr. Cepu
says the amount of creep that can occur
in epoxy is in the order of fractions of a millimeter and the epoxy would
recover back to normal once temperature returned to normal. He says the
discoloration of the epoxy in this case suggests that its temperature increased
to 50% to100% above the temperatures under which it normally operated. He
described this as intense overheating. He says it runs counter to fundamental
principles of thermoplastics to suggest that an epoxy could be heated to such a
high temperatures to lose its adhesive qualities and then regain those adhesive
qualities, so as to re-bond to the pinion wall.
Failure Sequence
[151] On examining photographs of the damage to the
helicopter taken at the scene of the accident, Dr. Cepu noted evidence that the tail boom suffered severe ground impact damage.
It moved upward enough to crush the turbine outlet shroud, located above the
tail boom, aft of the transmission. This impact was clearly sufficient to have
broken components of the tail rotor drive shaft. He concluded that some of the
damage to the drive shaft was consequential, rather than causative of the
crash.
[152] Dr. Cepu
says that in the Cobra accident, the
pinion welded itself to the inner race of the aft duplex bearing, thereby forcing
the inner race and bearing to continue to rotate and experience more intense bearing
damage. He points to evidence that the Advantage pinion was momentarily welded
to the inner race before it broke away and continued to rotate.
[153] Dr. Cepu
says a gap developed between the inner
race and the pinion as it was deformed. This explains not only the limited
destruction of the duplex bearing, in comparison with the damage sustained in
the Cobra case, but also the passage of oil through the pinion after the pinion
failure. Dr. Cepu says that once
the gap developed oil from the roller bearing could have flowed freely through
the pinion. The presence of a spray of
oil aft of the quill assembly is not surprising in the circumstances. The oil
is noted to have had the appearance of thermally damaged oil, suggesting that
it has passed through an area of intense heat.
[154] Dr. Cepu did a
rough calculation of the force that would have to be exerted against the pinion
in order to overload the duplex bearing. That exercise was not particularly
helpful, in part because of the many assumptions that had to be made in order
to do even the most basic calculation. Further, however, both the plaintiffs
and the defendant suggest that the unrestrained rotation of the broken end of
the drive shaft ultimately caused the pinion to fracture. The plaintiffs say
that could only have occurred if the pinion had been weakened by extreme heat
and extruded from the quill assembly. Dr. Morris believes the pinion
fractured as result of a combination of torsion exerted by the broken drive
shaft and overheating due to friction between the pinion and at least the aft
ring of the duplex bearing.
Time in Service
[155]
Dr. Cepu says when the Cobra pinion was tested after the 1991 accident the
adhesive and cork in the pinion maintained an airtight seal, whereas the
adhesive and cork in the Advantage helicopter pinion did not. Dr. Cepu suggests that leakage of oil into the pinion through this aperture may
explain the disparity in the time-to-failure in the two accidents. Additionally,
he says, it is possible that the tolerances in the Advantage helicopter may
have been slightly greater than those in the Cobra helicopter, allowing the
duplex bearing to have operated with lower stresses for longer period of time
and to get more lubricant before it ultimately failed. In his view: Its
logical to conclude that some level of lubrication between ideal and
non-existent will result in a lifespan somewhere between normal and very
short.
Analysis
Spoliation of Evidence
[156] The
defendant argues that the plaintiffs took steps to make evidence unavailable to
Heliponents. It says the court must fashion a remedy that reflects the gravity
of the plaintiffs actions and the significance of its adverse impact on the
defendant. The defendant, citing St. Louis v. Canada (1896), 25 S.C.R.
649, invites the court to make a finding of spoliation against the plaintiffs and
says such a finding, carries a presumption that the evidence destroyed would
have been unfavourable to the party who destroyed it, but that presumption may
be rebutted.
[157]
The defendant relies upon the description of the necessary elements in
support of finding spoliation in Chow-Hidasi v. Hidasi, 2013 BCCA 73,
and draws the courts attention to para. 29 of the judgment of the
majority in that case. In my view that paragraph should be seen in the context
of the following discussion of the evidentiary presumption:
[27]
I turn to the plaintiffs argument that ICBCs
(apparent) destruction of the Jeep effectively destroyed her ability to
challenge the theory of mechanical failure, and that the court below should
therefore have inferred that an examination of the vehicle would have shown no
mechanical failure.
[28] I have considerable sympathy for the plaintiffs
position, but in my view the presumption she seeks may not be drawn in the
circumstances of this case. First, the evidence as to the conditions under
which the Jeep was destroyed is negligible: there is only the defendants
hearsay evidence that he was told that it had been destroyed. Most importantly,
there is no evidence as to whether ICBC was aware the plaintiff would be making
a claim or if she made any effort to advise them or have the vehicle examined
before it was destroyed. (It was Mr. Hidasi who requested that the vehicle
not be destroyed.)
[29] On the present state of the law, it is clear that
spoliation requires intentional conduct: see St. Louis v. Canada
(1896), 25 S.C.R. 649; McDougall v. Black & Decker Canada Inc.,
2008 ABCA 353 at para. 29; Endean v. Canadian Red Cross Society
(1998) 157 D.L.R. (4th) 465 (B.C.C.A.); Dawes v. Jajcaj, 1999 BCCA 237
at para. 68; and the discussion in Holland v. Marshall, 2008 BCCA
468 at paras. 70-2. (I understand intentional to mean with the
knowledge that the evidence would be required for litigation purposes.)
As stated in McDougall v. Black & Decker, When the destruction is
not intentional, it is not possible to draw the inference that the evidence
would tell against the person who has destroyed it. (Para. 24).
[30] The Court observed in McDougall that where
evidence has been destroyed unintentionally, a court of law may fashion a civil
remedy to assist in ensuring the fairness of a trial. A costs award may be
made, or evidence may be excluded. We were not referred to any case binding on
us, however, that would indicate that such remedies would include the drawing
of an adverse inference such as that sought in this case by Ms. Chow-Hidasi.
(See McDougall, para. 25, British Columbia Law Institute, Report
on Spoliation of Evidence (2004), at 10-20.)
[31] In
my view, neither the state of the law nor the evidence as presented in this
case could support the drawing of an adverse inference that an examination
would have shown no mechanical failure in the brakes or steering wheel of the
Jeep. Like all litigants, the plaintiff was required to prove her case on the
evidence available to her at the time of trial. I would therefore dismiss this
ground of appeal.
[Emphasis
in original.]
[158]
In Holland v. Marshall, 2008 BCCA 468, the court held:
[59] In a legal context, the
term spoliation refers to the destruction, mutilation, alteration or
concealment of evidence. The harm to the trial process that spoliation can
cause is well-recognized. The more difficult problem is finding an appropriate
remedy for spoliation. The sanctions or remedies available to litigants who
suffer due to spoliation include procedural remedies, evidentiary presumptions,
contempt proceedings and costs orders. Preventive measures may also be taken
through preservation orders.
[159] It is
clear that in this case the plaintiffs failed to preserve physical evidence that
might have been useful in determining causation with greater certainty. It is
also clear, in my view, that the destruction of that evidence was not intended
to preclude the defendant from having an opportunity to examine it. The
defendants inability to examine the evidence was, in part, due to its failure to
promptly seek to examine the components of the helicopter when it had notice of
the accident and was advised of the presumptive cause of the accident. The
defendant failed to ask that the parts be preserved, visit the accident scene,
travel to the location where the parts were held, or inspect the parts
themselves. Heliponents principals indifference to the claim was reflected in
the fact they did not speak with their own employees to investigate the claim
or preserve what they could of their memories of the overhaul work.
[160] The
components of the helicopter were available for inspection long after the
accident but appear to have been discarded because no one expressed an interest
in them. In my view, the destruction of those parts was not intended to conceal
evidence and its destruction did not occur, in the words used in Chow-Hidasi
with knowledge that the evidence would be required for litigation.
[161] Nevertheless,
the components were destroyed by the plaintiffs or their agents. The
destruction of evidence without notice to adversarial parties in the course of
litigation should not be condoned. In the circumstances it would be
inappropriate to discount any theory of causation advanced by the defendant
that would rely, for its proof, upon access to those portions of the helicopter
that have been destroyed simply because that theory could not be substantiated
with physical evidence.
[162] I
cannot accede to the defendants argument that the court should find there was
intentional spoliation of evidence by the plaintiffs, nor should I draw any
inference in favor of the defendant. It is enough, in my view, to account for
the limited physical evidence available to the defendant by taking into account
the maxim described by Lord Mansfield in Blatch
v. Archer (1774), 98 E.R. 969:
It is certainly a maxim that all
evidence is to be weighed according to the proof which it was in the power of
one side to have produced, and in the power of the other to have contradicted.
The Quality of the Overhaul.
[163] The
defendant says the plaintiffs have not shown that Heliponents incorrectly
placed the cork plug during overhaul or that incorrect plug placement caused
the accident. While in its written argument the defendant says the plaintiffs
have not proven the applicable standard of care and cannot prove the defendant
has breached that standard, it acknowledges that if the cork was placed in the
wrong location at the time of overhaul, that would amount to a breach of the
appropriate standard of care.
[164] The
defendant expends considerable effort addressing the plaintiffs allegations
with respect to inadequacy in the documents, specifically the lack of directions
for the insertion of the cork, or description of the method of affixing the
cork and the manner in which the worksheets were completed. These allegations,
in themselves, are not the foundation of the plaintiffs claims. These
complaints were made as proof of poor practice, said to be supportive of the plaintiffs
argument that the manner in which the work was done at Heliponents was so
slipshod that there can be little confidence that Mr. Mayhugh knew what he
was doing, or did it properly.
[165] The
evidence adduced by the plaintiff leaves me with no doubt that there is a
substantial possibility that the cork was misplaced in the pinion at the time
of the transmission overhaul at Heliponents. Heliponents used an outdated
maintenance manual, the wrong worksheet, and the wrong overhaul set. Mr. Mayhugh
had little training or experience, perhaps no prior experience overhauling a
Bell tail rotor drive quill. There is no evidence his work was inspected
periodically as it ought to have been, according to Heliponents own documented
procedures. The circumstances were rife with possibilities for error. There is
reason to believe supervision may not have been sufficient to avoid error: correspondence
from Heliponents to Bell Helicopter after the accident is evidence of
Heliponents unfamiliarity with the installation of the cork in the pinion of
the tail rotor drive quill assembly of a Bell 204 helicopter. Heliponents did
little to address the plaintiffs concerns with respect to the overhaul. It did
not conduct its own investigation to determine whether it erred in the course
of completing the overhaul or attempt to collect information from its employees
with respect to the work they had done on this equipment.
[166] I have no
doubt that if the cork was not displaced by the accident but, rather, was
inappropriately positioned before the accident occurred, that positioning error
originated in the Heliponents workshop and was the result of negligence on the
part of the mechanic and his supervisors. In my view, the carelessness was
amply established by the examination for discovery of Heliponents
representatives and the testimony at trial.
Lay Witnesses
[167] The
defendant makes some submissions with respect to the credibility and demeanor
of the plaintiffs witnesses. In my view, the lay witnesses were all credible. They
all made reasonable efforts to recall accurately their observations and
involvement in the case. Mr. Mayhugh had little to offer, but was honest
in admitting that was the case. In my view, both Mr. Mizera and Mr. Partridge
appeared to testify in an honest and forthright manner.
[168] The
defendant did not call some witnesses who might have testified with respect to
overhaul procedures, documents, and instruction given to Mr. Mayhugh. The
failure to call those witnesses is not particularly significant, but I can and
do draw the inference that their testimony would not have assisted the
defendant in establishing that Mr. Mayhugh received specific instruction
with respect to the installation of the cork or that Heliponents employees performed
any quality assurance procedures to which he did not testify.
[169] I do not
draw the adverse inference sought by the defendant from the plaintiffs failure
to call Mr. Heath as a witness. His report was in evidence for some
purposes despite the fact was not called as a witness. While I can place no weight
on the opinions in his report, neither can I draw the inference that he would
have said something at odds with the report. In this regard, experts ought not
to be treated like lay witnesses who may have evidence that cannot be obtained
elsewhere. I can infer that Mr. Heaths opinion would not have been more
convincing than the opinions of Mr. Rupert or Dr. Cepu. That inference is not
particularly adverse to the plaintiffs.
Expert Evidence
[170] The two
most significant questions in this case: whether the cork was inappropriately
positioned at the time of the overhaul and, if so, whether that caused the
accident, are questions that must be answered with the assistance of experts.
The defendant suggested in its cross-examination of the plaintiffs expert witnesses
that the first impression, based upon what was acknowledged to be a cursory
examination of the helicopter at the accident scene, that the accident was the result
of overheating of the pinion due to oil starvation led them to inappropriately
discount other potential causes.
[171] The
defendant says that the plaintiffs failure theory was the beginning and the
end of the plaintiffs efforts to inquire into the actual cause of the
accident. It is true that the plaintiffs thought the cause of the accident to be
so obvious that only certain parts had to be retained and examined. It is fair
to say the plaintiffs should have conducted more searching analysis. The
significance of that failure hinges, however, on the strength of the case now
presented, on the extent to which they were correct in saying that the cause of
the accident was glaringly obvious.
[172] The
defendant says the initial conclusion arrived at by Mr. Mizera led to a failure
to fully examine the wreckage and the failure to retain material parts. They
say Mr. Mizeras opinions inappropriately predisposed other experts to share
his conclusion. Those experts, in turn, are said to have influenced each other.
Mr. Ward affected Mr. Heath and Mr. Rupert and they, in turn,
affected Dr. Cepu.
There is no doubt that the summary conclusion of Mr. Mizera did affect the
nature of the initial investigation. I am of the opinion, however, that the
subsequent consideration of the case by other experts was not unduly affected
by Mr. Mizeras opinion, although the scope of their investigations were limited
to an examination of the material that was collected and made available to
them.
[173] It is
frequently the case that experts are retained after the parties have arrived at
a preliminary opinion as to the principal questions in issue. It is not
uncommon for those preliminary opinions to be shared with the experts. In this
case, it would have been unusual for Mr. Mizera, a person with some
acknowledged expertise, to have withheld his opinion from the insurance
adjuster retained by his insurer to respond to the claim, or from the accident
reconstruction experts who then took over the investigation. The court relies
upon experts to come to the case dispassionately and to subject their clients
theories to appropriate scrutiny. There is no doubt that in this case both the
plaintiffs experts and the defendants experts knew of their clients
interests and their preliminary views.
[174] Having
considered Mr. Ruperts evidence, I reject the defendants suggestion there
was intentional concealment of the Cobra accident and the Cobra report. I
attribute no inappropriate motive to any of the experts.
[175] The
defendants expert, Dr. Morris, made some significant technical errors. Dr. Morris
acknowledged that the statement at p. 8 of his December 6, 2011, report that
The forward member of the duplex bearing set remains attached to the drive
quill pinion and rotated with it at the time of my last inspection is wrong.
He asked that it be crossed out. That is not a minor or insignificant error. It
is contained in his conclusions and immediately follows the statement: In
fact, there is persuasive metallurgical evidence that the duplex bearing set
continued to rotate for some brief time after the pinion shaft failed.
[176]
The error is repeated on p. 13 of the report:
The forward member of the duplex
ball bearing set is mechanically damaged, with some associated thermal damage.
It can be seen through the oil drain hole in the sleeve (#22 in Fig. 7) of the
tail rotor drive quill (Fig.9). The bearing balls show severe mechanical
distress; they have been flattened and have penetrated the outer bearing race
and the duplex bearing seal (#10 in Fig.6). However, the forward roller bearing
is still attached to the drive quill shaft and rotates with the shaft (Fig. 37).
It appears that the bearing outer race spun against the sleeve while the inner
race remained attached to the pinion gear shaft. There appears to be residual
lubricant on the bearing.
[177]
Dr. Morris acknowledged that this paragraph contains worst mistake
in this analysis. Dr. Morris testified that in this section of his
report he was attempting to see if the shaft could be rotated in the forward
bearing of the duplex set. The conclusion that the forward roller bearing
is still attached to the drive quill shaft is intended to refer to the forward
ball bearing in the duplex set. But Dr. Morris was not looking at the
forward bearing in the duplex set. He mistook a spacer for the cage of the
bearing. Figure 37, to which he refers, depicts a spacer, not a bearing. He
could not see the bearing balls. He said, referring to his mistaken
understanding of the appearance of the spacer:
In my haste, I – -I mistook that
for a bearing cage on the aft bearing and interpreted these pictures to be
rolling elements that had been sheared off or deformed off in that aft bearing.
That was incorrect.
[178]
Dr. Morris suggested that that it would be appropriate to remove
the whole paragraph in the middle of page 13 from his report. It is difficult
to see how this error could be inconsequential, given the weight he afforded to
these observations in the report itself. There were repeated references to his
erroneous conclusion. At p. 23 of his report, he notes:
As both Mr. Rupert and I
have noted, the pinion shaft can be turned easily and smoothly to this day. As
documented in Fig. 37, the forward member of the duplex bearing set is still
attached to the pinion shaft, and rotates with it.
[179]
At the bottom of the same page, summarizing his conclusion that the bearings
in this case did not disintegrate, he notes:
Finally, I should comment on the
damage to the forward member of the duplex bearing set. As documented in Fig. 37,
this bearing remains attached to the pinion shaft and rotates with it. The
balls and outer race are, however, significantly damaged, primarily by a relatively
clean wear that forced the bearing outer race onto the balls and wore the balls
flat where they penetrated the outer race. Much of this damage certainly
occurred after the failure of the pinion shaft since the unit would continue to
turn.
[180] When he
acknowledged that all references in this report to the forward member of the
duplex bearing set are in error, and asked that they be removed, Dr. Morris
did not explain the cause of the error other than to say that it was made in
haste.
[181] That
significant error, coupled with Dr. Morris repeated reference to the
duplex ball bearing set as the roller bearing set suggests some unfamiliarity
with the quill assembly components.
[182] Unfamiliarity
with the mechanics of the helicopter was also reflected in the fact Dr. Morris
did not know that the tail rotor drive shaft was contained within a tunnel in the
engine bay. That was surprising, given his specific and detailed testimony with
respect to the cause of the fracture of the drive shaft in that bay. Dr. Morris,
while an expert in metallurgy, appeared to be relatively unfamiliar with the design
of helicopter engines and components, and their maintenance.
[183] I accept
that the plaintiffs experts were, at times, careless with respect to
speculation as to why the Advantage helicopter was capable of functioning for
so long without adequate lubrication.
[184] The
plaintiffs experts were occasionally adversarial in their approach to the
case. The defendant fairly suggested that Dr. Cepu was occasionally an advocate. However, Dr. Morris
equally engaged in advocacy, as illustrated by his December 29, 2012,
correspondence with counsel in which he suggests that Dr. Cepu does not make much of a
point against us and says will have to look at the report, which
presumably will say things adversarial to us. [Emphasis added.]
[185] When
testifying in relation to alternate theories of causation, in particular, Dr. Morris
appeared to be adamant when there was no cause to be adamant and appeared to be
reluctant to agree with some propositions put to him by plaintiffs counsel
that seemed to flow logically from his own testimony. When it was suggested to Dr. Morris
that his theory was inconsistent with the evidence of the pilot he suggested
the pilot suffered a severe loss of memory after the accident. He noted, one
learns very quickly to take reports or recollections of events during an
accident as a piece of evidence, but certainly not an infallible piece of
evidence. He then referred to the pilots loss of memory after entering auto
rotation and said: So he blacked out at this point, and hes talking about
events that happened just before he blacked out.
[186] Dr. Morris
testimony with respect to the pilots loss of memory was entirely speculative
and unfounded. When the pilot, Mr. Batty, testified he appeared to have a
clear and reliable memory of events until seconds before impact. The quality of
his memory was not questioned on cross-examination and I believe him to have
given very accurate account of events and in particular a very accurate account
of the onset of problems he experienced flying the helicopter.
[187] Both
experts, at times, expressed opinions without having sufficiently investigated
the evidence in support of those opinions. On critical points, however, in my
view, Dr. Cepu
conducted a more specific and scientific investigation. For example, in order
to test his theory that oil could have leaked past the cork into the pinion, he
performed a vacuum test, which was videotaped. In order to test his theory that
the balls in the duplex bearing had been subjected to prolonged and extreme
heat he conducted a hardness testing. Like Dr. Morris, he examined images
of the surface of the metal components with an electron microscope. Dr. Morris,
in contrast, tested one of his significant theories, the ability of the epoxy
to re-bond by conducting an undocumented test in a toaster oven at his home.
[188] Notwithstanding
the frailties of the expert evidence, it is essential in this case that I rely
upon what I can fairly take from the opinions of the experts to assist me in
coming to a conclusion on the critical questions.
Pre-Accident Location of the Cork in the Pinion.
[189] In my
view, the starting point of the analysis in this case must be a weighing of the
evidence with respect to the location of the cork in the tail rotor drive quill
assembly prior to the accident.
[190] The
defendant argued, at great length, that the court should find that Mr. Mayhugh
followed appropriate procedure in the placement of the cork in the pinion. The
defendant points to Mr. Mayhughs evidence that he learned how to perform
transmission overhauls under the instruction of Patrick Felice. There is,
however, no suggestion that Mr. Felice instructed Mr. Mayhugh on
appropriate procedure before he worked on the Advantage helicopter. The
plaintiffs refer to a record of inspection of Mr. Mayhughs work by Mr. Felice,
but, as noted above, the inspection record is not reliable. Mr. Mayhugh
could not testify with respect to the completion of the inspection record and Mr. Felice
was not called to do so.
[191] The
defendant says the natural inference that should be drawn from his evidence is
that Mr. Mayhugh would have followed a standard practice, consulted
manuals as required, and, if he had any questions he would have consulted with
a more senior mechanic. I cannot draw such inferences from Mr. Mayhughs
evidence. He had not yet developed a practice. He does not claim to have consulted
manuals; the manuals available to him were not helpful. He did not testify that
he consulted a senior mechanic with respect to this job. No senior mechanic
testified. His testimony, in short, in my view, amounts to saying that he knew
what he was doing and if he did not, he would have consulted someone who did.
That evidence is of little value in this case
[192] The
starting point of the defence argument is that the Cobra accident is evidence
that if the cork is improperly placed the duplex bearings will not be properly
lubricated and will fail within hours. The defendant says, given the long life
of the tail rotor drive quill before failure, the plaintiffs must demonstrate
how sufficient oil could have reached the duplex bearing to permit it to
operate for over 500 hours, and explain why that oil suddenly became inadequate
on the day of the accident. This is said to be an insuperable hurdle.
[193] The
defendant argues there should be no shift of the evidentiary burden in this
case. I agree. The question in this case is whether the evidence led by the
plaintiffs may result in an inference being drawn adverse to the defendant.
That is a matter of weighing all of the evidence: Snell v. Farrell,
[1990] 2 S.C.R. 311.
[194] In responding
to questions at trial, Dr. Morris took the view there should be a default
assumption that the cork was displaced by the accident; and the conclusion that
the cork was in the wrong position prior to the accident should only be drawn
if the cork could not move or if there was probative evidence that it did not
move as a consequence of the accident.
[195]
In Dr. Morris last report, he appears to start from the assumption
that the cork had been displaced as result of the accident, because he thought
there was no evidence of inadequate lubrication of the duplex bearing. The
presumption that the cork moved is reflected in the statement made on p. 28
of his last report, where Dr. Morris writes:
Since the duplex bearing set did
not seize, and shows no obvious sign of inadequate lubrication, the only reason
I can identify to suppose that the cork was incorrectly placed is the simple
observation that it was incorrectly positioned after the accident.
[196] The
approach taken by Dr. Morris was reflected in his testimony at trial when
he said, referring to his theory that the cork was pushed forward in the pinion
as result of the forces that were present at the time of the failure of the
drive shaft: So I would argue that that is the default interpretation of what
we see, and that someone who wishes to make another interpretation has a burden
of proof to show that theyre putting forward a reasonable interpretation.
[197] This is an
adversarial, rather than a scientific position. I do not believe that the
experts looking at the evidence in this case ought to have arrived at a
default interpretation and required those challenging that interpretation to
discharge the burden of proof. I do not regard it as an appropriate starting
point for scientific inquiry. The experts should not have started with
assumptions with respect to whether the cork was correctly or incorrectly
installed, and looked for reasons to displace those assumptions. They should,
rather, have weighed evidence in support of the conclusion the cork was
installed where it was found against evidence that it moved after
installation.
[198] The Court comes
to this analysis with no presumption with respect to the location of the cork.
Had there been some good, reliable evidence that the cork was placed in the
appropriate location at the time of the overhaul, I might have been prepared to
begin this analysis with the assumption that the cork, like other broken or
out-of-place equipment was probably displaced by the accident. There being no
record or recollection of it being installed, I do not start with that
presumption.
[199] There is
no evidence the cork moved or was dislodged between the date of the overhaul
and the date of the accident. There is no evidence of intermediate repair. The
experts have addressed only two possibilities: the possibility that the cork
was improperly installed during overhaul and the possibility it became
dislodged and moved during the accident. Dr. Morris acknowledges that the
cork could not have moved late in the sequence of events, after the pinion
fractured. If it had been installed in the appropriate location, it would have
been in the pinion shaft at a point aft of the location of the fracture. Dr. Morris
therefore postulates the existence of some mechanical force pushing the cork
forward in the pinion before the fracturing of the shaft. It is necessary to
suppose a very specific and difficult to explain sequence of events in order to
account for the movement of the cork in a manner consistent with Dr. Morris
theory. The cork would have to be heated, to loosen the seal that held it in
its initial position. It would have had to pop, in Dr. Morris words,
and move forward into the wider section of the middle of the pinion while being
forced further forward by torsion as the pinion twisted. It must initially have
moved as result of forces sufficient to cause what he considered to be a
fracture of the walls of the cork.
[200] Until
destructive testing was conducted in this case, the cork remained in position
at the narrow neck of the pinion close to the beveled gear at the front of the
shaft. Although there was some evidence with respect to damage to the sides and
the aft end of the cork it was more or less intact. Epoxy covered its forward
surface, and some of the epoxy adhered to the inside wall of the pinion,
holding the cork in position. The theory that the cork moved to this position
as a consequence of the failure that led to the crash would have to explain the
movement of the cork through the pinion, the preservation of most of the
substance of the cork and the adhesion of the epoxy to the side of the pinion
at its forward edge.
[201] Dr. Morris
suggests that the cork might have moved easily because it had been operating in
hot oil for 500 hours and what that does to this particular epoxy is a
question that, so far as I know, no one has accurately addressed. The
suggestion by Dr. Morris that over time the cork in the pinion might
become movable seems to disregard the fact that it is designed to remain fixed
in place by that epoxy in that environment. He says, in my view without
apparent justification, referring to the cork in the pinion: It does not have
to have any structural integrity to seal the pinion.
[202] Dr. Morris
suggests that the cork might have moved easily because it might have been
undersized; it might have relaxed over time; and heat over a long period might
have caused the epoxy to lose its structural integrity. Dr. Morris was
quick to speculate with respect to potential reasons why the cork might not
have adhered to the wall of the pinion and too quick to discount the obvious
reasons why cork would not easily have moved against pressure from the position
in which it was initially installed, if it had been installed correctly.
[203] Dr. Morris
acknowledges there was some bonding of the epoxy to the wall of the pinion in
the position in which was found after the accident. He says, however, that was
very weak re-bonding and that can happen at high temperature. That is the point
of the experiment he performed at his home.
[204] The plaintiffs
concede there is some evidence of slight movement of the cork. There is some evidence,
in Dr. Morris experiment, to suggest the epoxy might re-bond to adjacent
surfaces in a very hot environment (although in my view, as I have noted, that
did not establish that the epoxy could re-bond after being heated to a point
where it lost its adhesive character). They say, however, it is unlikely the
cork could have travelled the distance suggested by Dr. Morris and come to
rest at the position in which it was found and re-bond, at it appears to have
done, to the wall of the pinion bore. There is no good explanation that would account
for the movement of the cork within the core of the pinion as result of the
forces that might have been exerted in the accident. There is very little
evidence to suggest that any significant force could be exerted upon the pinion
emanating from the rear of the helicopter. Dr. Morris suggests that the
deformation of the shaft was the result of a torsional force and the squeezing
of the pinion would have tended to push the cork forward. This is conjecture.
There was some clear deformation of the pinion at the point where it came into
contact with the duplex bearing and aft of that point. There was no evidence to
suggest squeezing, torsional deformation of the pinion forward of that point.
[205] In my view
it is unlikely that the cork moved forward in the pinion while oil was being
forced under pressure against its forward face. There is no reason to conclude
that large mechanical forces were exerted upon the cork, and in weighing the opinion
of Dr. Morris, I must bear in mind that the cork was designed to remain
fixed in place within the shaft of the pinion in normal, high operating
temperatures.
[206] Having
considered the evidence of the condition of the cork, the surface of the
pinion, the appearance, color, and edges of the epoxy surface of the forward
edge of the cork, I am of
the view that the cork in the pinion of this helicopter was improperly
installed at or very close to the location which was found after the accident when
the overhaul was conducted by Heliponents.
[207] This
conclusion is, in my view, borne out by the evidence with respect to the manner
in which the accident occurred, and the damage to the components of the
helicopter discussed below.
Failure Sequence
[208] There are
two competing theories to account for the loss of tail rotor authority and the
crash in this case. The first theory, that advanced by the plaintiffs, is that
improper placement of the cork led to oil starvation, overheating, and failure
of the pinion at the point where it rubbed against the duplex bearing. The defence
theory is that some event aft of the tail rotor quill fractured the tail rotor
drive shaft leading to uncontrolled flailing of a broken section of the drive
shaft behind the tail rotor drive quill. Dr. Morris says a mechanical load
was applied from the rear down the tail rotor shaft causing it to bear severely
against the corner of the bearing set, splaying its inner race. He says that
torsional load caused the shaft to heat and extrude along its length, weakening
it until it broke. In Dr. Morris view, the damage to the components of
the duplex bearing was initially mechanical rather than a result of fiction or
heat.
[209]
The plaintiffs say there are no means by which the tail rotor drive
shaft can displace the axis of revolution of the pinion. The drive shaft is
hollow and made of an aluminum copper alloy. The pinion shaft is steel and much
stronger than the rotor drive shaft. When it was suggested to Dr. Morris
that it is more likely that off-axis rotation of the pinion caused damage to
the drive shaft, than a problem originating in the drive shaft affecting the
pinion, he said:
Well, lets be clear here. Its
my position that the quill was off axis because there was a problem in the tail
rotor drive shaft. Now, if youre asking me to assume that there was no problem
in the tail rotor drive shaft and the quill is slightly off axis for some other
reason, then how much wobble would go into the drive shaft really depends on
how much of it is accommodated by the splines in the elastomeric seal between
the inner and outer coupling, which is intended to sort of insulate those two
pieces from one another.
[210]
Dr. Morris suggested that off-axis rotation of the lighter drive
shaft, which has to pass through the splines and couplings and should be
accommodated by them, can cause the pinion to rotate off-axis. However, as
indicated in the above passage and the following exchange, he seemed reluctant
to concede that damage to the tail rotor drive shaft could be caused by
deformation and wobbling of the pinion:
Q: And if you assume that at that point the
pinion was cocked, there would be an off axis centre of rotation in the pinion
that would create a sympathetic misalignment with the tail rotor drive shaft
aft of that point, would it not?
A: Im not certain
thats true, because you have the inner and outer couplings and the elastomeric
seal to kind of cushion that alignment. The thing is designed so it can get
slightly out of line without causing any problem. Thats part of the design of
that unit. The wobble back in this unit, in my opinion, was due to the fact it
was already broken, and I think thats clear from the evidence. So its not at
that point so much that the pinion is out of line as that the tail rotor drive
shaft is broken and, therefore, flopping around.
[211] Dr. Morris
was reluctant to agree that the splined coupling is designed to absorb movement
and pressure. Initially he said it was not designed to absorb radial movements;
when pressed he said was designed to resist but not to absorb radial movement.
Initially he indicated that it has limited tolerance to absorb radial movement
but he did not argue when it was suggested to him that the stop is at quite an
extreme position. He did not attempt to analyze the forces that might be
necessary to cause misalignment of the drive shaft pinion.
[212] It does
not stand to reason that a force applied from the rear down the tail rotor
shaft could cause the evident damage to the pinion and drive quill assembly
that occurred in this case. On this point, I accept the expert opinion of Dr. Cepu that a force applied down the
drive shaft to the pinion would not transmit a load to the duplex bearing
because there is a splined connection between the drive shaft and the pinion.
Those splines will transmit torque but not axial load.
[213] No plausible
potential cause of the accident, other than overheating of the duplex bearing,
was specifically identified. Although there is some evidence that the
helicopter had been used to lift logs of a weight that would cause it to exceed
its maximum lift capacity, the evidence does not establish that doing so caused
the accident in this case. The experts did not testify to any theory that would
attribute the failure of the tail rotor drive shaft or the quill assembly to the
lifting of a heavy load.
[214]
The defendant notes that there is no record of the performance of an
engine-to-transmission main drive shaft alignment check having been performed
when the transmission was installed in June 2007. Mr. Partridge has no
specific recollection of such an alignment check being conducted. Mr. Rupert
and Mr. Partridge in cross-examination, acknowledged that the failure to
perform an alignment might result in damage to the input drive shaft. It was
not suggested such damage had occurred in this case, nor was there any evidence
with respect to the consequences of damage to the input drive shaft. If that
was a significant potential cause, I expect it would have been addressed and
considered by Dr. Morris. The only apparent reference by Dr. Morris
to misalignment appears on p. 17 of his report:
The failure appears to have
resulted from a tensile load that broke the shroud covering the driveshaft and
created a misalignment at the coupling between the driveshaft and the main
gearbox, resulting in severe mechanical damage at the joint.
[215]
When examined in chief at trial with respect to misalignment of the
coupling between the drive shaft and the main gearbox, he said:
In analyzing that damage, it
looks like the shaft kind of came apart and then, rotating, went back together.
So I photographed this and included it in the report. It did not play a major
part in my analysis, because were concerned about the drive quill and going
back from the drive quill. But, you know, its at least possible that this
played some role in the initiation of the accident. I dont know.
[216] It was not
further explored.
[217] While I
recognize there is no onus upon the defendant to establish any competing cause
of the accident to that proposed by the plaintiffs, acceptance or rejection of
the plaintiffs hypothesis may be a matter of weighing competing hypotheses.
[218] Both
theories of causation have weaknesses. The defence says there is no explanation
why the lubrication to the duplex bearing should have become suddenly
inadequate. The defence expert says that long, slow degeneration of the
lubrication should have resulted in signs of metal fatigue, rather than sudden
failure. The plaintiffs say there is no explanation for the sudden failure of
the tail rotor drive shaft that is at the heart of the defence theory of the
accident.
[219] The
plaintiffs theory is consistent with and supported by the evidence of the
pilot. Mr. Battys evidence was that he noted a burning smell, suggestive
of significant unusual heat, before he noted the loss of tail rotor control.
That failure sequence describes a rapid accumulation of significant heat
followed by an incident that disrupted the tail rotor drive shaft. That suggests
that there was a problem that preceded the fracture of the tail rotor drive
shaft rather than a problem emanating from a fracture of the tail rotor drive
shaft, the theory propounded by Dr. Morris. It is significant that the
pilot experienced no loss of control or power before noting signs of heat or
burning. It is also significant, in my view, that there were no indicator
lights or alarms before the loss of tail rotor control. This is supportive of the
theory that the incident occurred because of a lack of lubrication fluid.
[220] The
strength of the plaintiffs theory of causation, in my view, was reflected by
the difficulty Dr. Morris had in addressing the pilots evidence and instead
suggesting that the pilots evidence might not be reliable. The plaintiffs say,
and I agree, it is significant that, when Dr. Morris developed his
hypothesis, he had no knowledge of the pilots description of the sequence of
events that led to the crash.
Damage to the Duplex Bearing and Pinion
[221] There are
conflicting descriptions of the extent of the damage to the duplex bearing and
the pinion. Dr. Morris testified that the bearing has become more
difficult to rotate over the years since he first examined it. He says it was
initially easy to move the balls with finger pressure but that cannot now be
done. Even initially, however, the rotation was limited. Dr. Morris
attributed this difficulty on initial examination to geometric constraints due
to external and mechanical damage rather than heat damage to the balls or the
races. Even on initial examination the bearing was not functional. That damage was
sufficient to cause the inner race of the bearing to break free of its
compression fitting on the pinion and to rotate against the surface of the
pinion. Dr. Morris agrees there was some separation and friction between
the pinion and the aft inner ring of the duplex bearing. That friction clearly
contributed to the development of significant heat at the location of the
pinion fracture. The broken segment of the pinion was so hot when the fracture
occurred, that it scorched the floor of the fan bay. The question is not
whether that occured but whether it was a precursor to or consequence of the
failure of the tail rotor drive assembly.
[222] Dr. Morris
evidence with respect to heat damage to the duplex bearing is qualitative and
vague. In response, the plaintiffs have adduced quantitative evidence of the
temperatures to which the bearing components were exposed. There is good,
reliable evidence of the bearing being exposed to very significant heat. The
microscopic examination of the balls demonstrates evidence the failure of the
bearing was due to heat. Using an electron microscope, Dr. Cepu identified significant damage
to the balls in the duplex bearing; all of them have skid-marks, there are
numerous flat spots, and there is a scalloping of the surface of the balls.
Electron spectroscopy demonstrates the presence of copper, the source of which
must have been the bearing cage, smeared on the surface of the balls in the
bearing. Hardness testing demonstrates that the balls were very significantly
heated, to a temperature Dr. Cepu
estimates at 2000°C. There was clearly disintegration of the cage.
[223] There is
no doubt there was accumulation of sufficient heat at the bearing/pinion
interface to cause the pinion to overheat to the point where the steel pinion
became malleable and it was deformed. There is no doubt the pinion was deformed
at the point where it came into contact with the inner race of the duplex
bearing. There is a dispute on the evidence with respect to the extent of the
damage to the pinion at that point, but there is some evidence of damage due to
abrasion of the pinion by the inner race of the bearing. Dr. Cepu points to the presence of
portions of the inner race that appear to have broken off the race and become
welded to the pinion. In my view that evidence is persuasive proof the
bearing/pinion interface became so hot that the inner ring was at least
momentarily welded to the pinion.
[224] Dr. Morris agrees that the inner ring of the aft
bearing in the duplex set has been scoured out by one part of the pinion. This
could only have occurred if it was held in a fixed position while the pinion
was rotating. Dr. Morris says the scouring occurred after the fracture.
The inner race was still turning with the shaft under power for at least some
time before it became frozen. Dr. Morriss conclusion that the shaft and
inner race do not show signs of rotation against each other for a lengthy
period of time is inconsistent with his evidence that the pinion was extruded
and became thin as result of the overheating of the inner ring of the aft
member of the duplex bearing.
[225]
In cross-examination of Dr. Morris
there was a long exchange with respect to how the scouring incurred on the
inside edge of the aft bearing in the duplex set. It was suggested to Dr. Morris
that the most probable cause of the scouring was rotation of the pinion against
the bearing after it had become fixed. In response Dr. Morris asked: What
are you hypothesizing made it stationary? The balls didnt. It rotates to this
day. That response was argumentative and is inconsistent with Dr. Morriss
own evidence that, at some point, as a result of the balls riding up on the
races, the bearing overheated, stopped functioning and separated from the pinion.
[226]
In summary, there is evidence of the accumulation of sufficient heat on
the surfaces of the duplex bearing to cause erosion of and damage to the balls
and to cause the duplex bearing to cease functioning as a bearing. There is evidence
of heat damage to the cork, but more significantly to the races and balls in
the duplex bearing and, particularly significantly, to the steel pinion itself.
I find persuasive the plaintiffs argument that mechanical forces alone or
primarily mechanical forces exerted by the rotation of a broken section of the
drive shaft are unlikely to have caused the significant distortion to the steel
of the pinion if the pinion had not been weakened and deformed by heat before
the fracture of the tail rotor drive shaft.
[227] Dr. Cepu, in my view, adequately explains the evidence of
damage aft of the quill assembly as damage caused by the rotating components
before and after failure of the pinion.
[228] On the
other hand, the defendants expert had difficulty explaining how a fracture of
any segment of the tail rotor drive shaft could produce forces of such
significance as to twist off the steel pinion at the point at which it
fractured and to cause the evidence damage to the duplex bearing. It stands to
reason, in my view, that the off-kilter rotation of the pinion caused the tail
rotor drive shaft to rub against the walls of the aperture and firewall and
fracture. It is difficult to conceive of a means by which off kilter rotation
of the tail rotor drive shaft can in any way had been transmitted through the
assembly in such a manner as to cause the obvious damage to the aft duplex
bearing and, in particular, the much more limited damage to the forward bearing
that occurred in this case.
[229]
It should be borne in mind that Dr. Morris posits a sequence of
events that is not very dissimilar from the plaintiffs theory. In
cross-examination he put it as follows:
Bearing failures tend to have
rather similar features, whatever their source was, in general
If youre
talking about pinion failure, the pinion failed, largely because the bearing
failed. A failure sequence was initiated, which resulted in the fracture of the
pinion and damage to the bearing
The nature of that failure appears to be a
misalignment of the pinion, which caused it, and the bearing to come out of
misalignment, and then they failed together
So — and you really cant
separate the pinion failure and the bearing failure
when you characterize what
I am saying, you have to be a bit specific on what you mean by failure. The
initiating event of the failure sequence appears to be some cocking or
misalignment of the pinion which caused a misalignment of the bearing, and then
caused it to fail
So the initiating event was the misalignment of the pinion,
right, but not the fracture of the pinion. That came later.
[230] Damage to
the drive shaft aft of the quill, in Dr. Morriss view, caused off-axis
rotation of the pinion. That led to accumulation of heat at the interface
between the pinion and the duplex bearing. The heated pinion fractured under
torsional load. According to this scenario, there was a period during which
excessive heat was generated in the duplex bearing. At a certain point, the
bearing stopped functioning normally and the pressure fitting between the
bearing and the pinion failed. For a very brief period of time, before the
fracture of the pinion, a broken piece of the tail rotor drive shaft would have
flailed about in the transmission bay.
[231]
The plaintiffs theory is that the sequence of events leading to failure
began with overheating of the duplex bearing. When it stopped functioning
normally, the pressure fitting between the bearing and the pinion was broken.
That led to the generation of excessive heat, and the extrusion and wobbling of
the pinion. That in turn led to the fracture of the tail rotor drive shaft,
which would have flailed about in the transmission bay, briefly, before the crash.
Where Dr. Morris theory departs substantially from the plaintiffs theory
there is, however, little support for Dr. Morris view. The principal
frailties in this theory, in my view, are:
a)
it is inconsistent with the accident sequence as described by the pilot;
b)
it leaves unresolved the question of what type of event caused the
initial failure of the drive shaft; and
c)
there is no apparent means by which what he describes as impact damage
was inflicted upon the duplex bearing, and no accounting for the force that
drove the cork forward in the pinion.
[232] Dr. Morris
did not address these frailties in his theory. He
agreed that he formed the opinion that the duplex bearing was unable to
withstand the radial and the moment forces that were exerted on the splined end
of the pinion without doing any calculation of those forces, and without
looking at the load rating of the bearing. He agreed with the proposition that
the only way one can transfer the bending moment to the duplex bearing in this
assembly is if the angle of the off axis rotation exceeds the ability of the coupling
assembly to absorb some off axis rotation in the aft segments of the tail rotor
shaft. At some point in the failure sequence, the keys of the splines and the
outer coupling assembly bore against the steel shaft quite severely. This was a
hypothesis that was not subjected to any particular engineering analysis by Dr. Morris.
Criticism of the Plaintiffs Theory
[233] The
plaintiffs, on the other hand, did attempt to grapple with the principal
criticism of their theory of causation, the argument that the helicopter could
not have flown for over 500 hours with an inadequately lubricated duplex
bearing. The plaintiffs response to this criticism is that the failure
sequence is complex and will differ from case to case. Helicopter components
are manufactured with various degrees of tolerance and some equipment may be
more forgiving of a lack of lubrication than other equipment. There is some
supply of lubrication to the duplex bearing through the roller bearing, even
where the cork is in the wrong location. Last, the plaintiffs say it is
probable that there was some leakage of oil through the improperly placed cork.
[234] Some
effort was made to confirm this last theory by testing the cork after the
accident to see if it properly sealed the pinion shaft. Water was able to leak past
the cork, suggesting that some oil may have traveled to the duplex bearing past
its incomplete epoxy seal. The theory was further validated by the vacuum test
conducted by Dr. Cepu. I
accept the evidence of the plaintiffs experts to the effect that insufficient,
rather than nonexistent, lubrication caused this accident and that their theory
is not inconsistent with the helicopter operating for significant period of
time before the accident.
[235] The
evidence and opinions expressed in relation to the Cobra accident are, to a
certain extent, supportive of the plaintiffs theory. The differences between
the accidents are not such as to convince me that the Advantage accident had a
different cause. The Cobra accident is cogent evidence of the fact that inappropriate
placement of the cork in this helicopters tail rotor drive quill can lead to
overheating of the duplex bearing, failure of the pinion, and loss of tail
rotor control. It is noteworthy that the pinion was distorted at the same
location and similarly twisted in both the Cobra and Advantage helicopters.
[236] It was
clear from the photographs that the Advantage bearing was not as significantly
damaged as the bearing removed from the Cobra helicopter after the 1991
accident. However the Advantage pinion was more significantly deformed. In my
view, Dr. Cepu explanation
of the difference is plausible. The failure sequence in the case at bar clearly
differed from the failure sequence in the Cobra case. There is evidence in this
case that the pinion was only briefly welded to the inner race of the bearing. Dr. Cepu says the pinion appears to have
overheated quickly and extruded. Once it lost contact with the inner race, the
heat source was removed from the bearing and the heat destruction of the duplex
bearing came to an early end. Dr. Cepu
points out that the pinion was welded to the inner race in the Cobra
accident and the heat damage appears to have been greater because the duration
of exposure to heat must have been longer.
Summary
[237] In my
view, the plaintiffs theory of causation is most consistent with the failure
sequence described by the pilot. That theory is supported by physical evidence,
particularly the position of the cork, the damage to the duplex bearing, the
indication that the bearing became welded to the pinion, and the nature of the
damage to the pinion. The theory is not inconsistent with any significant
physical evidence and further, it occurred in a fashion that bears some
significant similarity to a previous incident arising from the same cause.
[238] The
plaintiffs, in my view, have adequately addressed objections to their theory,
principally the lengthy time it took for the problem to develop after the
transmission overhaul and the differences in the physical damage that occurred
in the Cobra accident and the damage that occurred in this case.
[239] Last, I
must note that there is little in the evidence of Heliponents to cause me to believe
it was unlikely that the cork was misplaced at the time of the overhaul of the
transmission. I am of the view that it is highly improbable the cork moved to
the location at which was found as result of the accident.
[240] Despite
the complex evidence in this case, the causation analysis ultimately amounted
to a choice between two competing explanations of the physical evidence. The
loss of tail rotor authority might have been a result of damage to the tail
rotor drive quill that resulted in off axis rotation of the tail rotor drive
shaft, leading to breakage of the shaft and loss of power to the tail rotor.
Alternatively, overheating of the duplex bearing and then the pinion leading,
in turn, to its distortion might have led to serial fractures of the shaft,
cocking of the remaining stub end of the shaft in the transmission bay, and
ultimately the failure of the pinion. There is an evident cause of the pinion malfunction,
the presence of a cork at the forward end of the pinion shaft. There is
evidence, the Cobra accident, that placement of a cork in that position can
lead to overheating of the bearing and the failure, bearing the hallmarks that are
seen in this accident. On the other hand, there is no apparent explanation for
a sudden fracture of the drive shaft as an initiating cause of the accident and
an obvious explanation of how that damage could have occurred as a result of off-axis
rotation of the pinion. Further, that theory of causation is inconsistent with
the pilots evidence.
[241] I am
satisfied that the plaintiffs have established on the balance of probabilities
that this accident occurred as result of the misplacement of the cork in the
shaft of the pinion of the tail rotor drive quill when the helicopter was
serviced at the Heliponents facility in May 2007. It is acknowledged that this placement
could only have occurred as a result of a breach of the standard of care owed
by the defendant to the plaintiffs. I find the defendant liable to the
plaintiffs for that negligence.
[242]
The action is allowed; there will be judgment in favor of the plaintiffs.
The parties may speak to any matters that are not been addressed in these
reasons.
P. Willcock J.
The Honourable Mr. Justice P. Willcock
APPENDIX
A
APPENDIX
B
