According to several of its operators, the Mitsubishi MU-2B is one of the
fastest, most fuel efficient, strongest and most responsive handling general
aviation twin turboprops yet built. With flaps retracted, it has virtually
the highest wing loading of any popular twin turboprop, providing a smooth
ride in turbulence and minimal frontal area for reduced drag and efficient
cruising. Fitted with Honeywell (Garrett) TPE331-10 turboprop engines as
standard or retrofit equipment -- the prototype MU-2, or A model, was fitted
with twin Astazous but did not go into production -- the short-body MU-2B
models, including the Solitaire, will cruise as fast as 315 KTAS at 20,000
feet. The long-body models, including the Marquise, can fly faster than 305
knots. The MU-2B, without a doubt, provides more speed for the dollar than
any other general aviation turboprop on the resale market.
It's also able to fly slowly in general aviation airport traffic patterns
because it's fitted with full-span, double-slotted Fowler flaps that
increase overall wing area by almost one-quarter when extended. The MU-2B
can slow down to 120 KIAS to sequence with other general aviation aircraft
on downwind. Vref speeds on final range from 100 to 110 KIAS, depending upon
landing weight. In many ways, the MU-2B flies like a current-production
light jet with turboprop engines.
Dick Allan, president of Internet Jet Sales, a well-known MU-2B broker in
the Northeast United States, says the aircraft's jet-like performance has
special appeal to a special set of pilots, ones who are notably different
from those who fly more matronly turboprops. He likens the aircraft to "a
stanine test that separates fighter jocks from bomber pilots."
But Allan points to another part of the MU-2B population. He believes the
aircraft's popularity with low-budget canceled check transporters and
all-night air freight operations is problematic since they can afford to buy
the airplane, but don't have the money (or the will) to invest in rigorous
pilot training or top-notch maintenance.
That view is shared by FAA insiders. "The business model has changed. It's
now migrated to the bottom-feeders in the air freight industry," said one
FAA source. "We need to bring up pilot qualifications to a level that the
aircraft demands. And we may see some recommended Service Bulletin and
factory maintenance procedures made into ADs."
The MU-2B has been a dream-machine for the plaintiffs' bar for several
years. Of the 800 or so aircraft that were built between 1967 and 1985, more
than 200 have been involved in incidents or accidents, according to NTSB
statistics. The accident rate was particularly bad in 2004 and 2005, with a
dozen-plus crashes and 13 fatalities. Overall, its five-year accident rate
from 2000 to 2004 was 3.17 per 100,000 flight hours, compared to 1.73
accidents per 100,000 flight hours for that time frame among other popular
turboprops, according to Robert E. Breiling Associates. During the same
five-year period, the Mitsubishi's fatal accident rate was 1.66 per 100,000
flight hours, or more than triple that of popular turboprops, Breiling
asserts. Since the airplane's entry into service 37 years ago, more than 200
people have been killed in MU-2 accidents, trial lawyers say. Those
statistics also focused the attention of the FAA on the aircraft. About 400
MU-2B aircraft are still in active service.
To understand how that history impacts fleet values, see this month's
20/Twenty on page 112.
So, what's wrong, if anything, with the MU-2? The answer depends upon whom
you ask. Few aircraft incite such polarized opinions as does the MU-2. The
two camps are divided primarily into folks who have flown it or in it for
years without incident, and those who have never gotten in one and don't
intend to do so under any circumstances.
Among the current group of MU-2B foes are several members of the U.S.
Congress who have received letters from constituents with relatives who have
perished in recent Mitsubishi crashes. Many of these letters request that
the aircraft be grounded immediately by the FAA and that a full
investigation be launched as to why it's so accident prone. Some of those
letters look very much like documents written by trial lawyers because of
their precise accident analyses, bar graphs and inclusion of pilots',
passengers' and even pets' names for each crash. It's rare that heirs of
crash victims independently undertake such thorough accident history
research and have access to so many personal details of other crash victims,
numerous sources told B&CA.
The letter writing campaign nonetheless has been effective. Members of
Congress are demanding action from the FAA to ground the aircraft or at
least mandate stringent standards for pilot training. The FAA, under
pressure from these lawmakers, recently concluded its fourth look at the
MU-2B, as we'll discuss later in this report.
"This is a dangerous bird, with an extraordinarily high crash incidence
rate. We are not convinced that it's ever been fully tested. I think there's
a design flaw in the aircraft," said Terry Van Keuren, constituent advocate
for Rep. Tom Tancredo (R-Colo.). Van Keuren said he's had several letters
from constituents who are heirs and relatives of people killed in MU-2
accidents in Colorado. He said he doesn't believe the FAA properly evaluated
the Japanese-designed and -built aircraft during the initial 1965 type
certification, the follow-on 1984 Special Certification Review or the 1997
Fact Finding Focus Special Certification Review related to approval for
flight into known icing.
"It has a very thin, high-performance wing. It's the only aircraft in its
class with spoilers [for roll control]. It has an inherently wrong fuel
system. And a Mitsubishi test pilot was told to lie about its icing
vulnerability [sic] in the Rickert [v. Mitsubishi Heavy Industries] case,"
he said. "It has a disturbing history of icing problems, of prop problems
and there's a lot of sneaky stuff going on. Too many things just don't pass
the sniff test."
Van Keuren also said the engines have a tendency to "burp" (lose power) and
the NTSB is looking into aircraft problems related to power interruption.
Notably, NTSB accident analyses don't appear to indicate that the aircraft
has an abnormally high, unexplained engine failure rate.
He also claims the aircraft provides very little stall warning. "There's
even a warning in the Flight Manual," Van Keuren pointed out, in reference
to Mitsubishi's caveat about fully stalling the aircraft with one engine
providing "lift producing thrust" that could cause a "rapid rolling and
yawing motion." He did not mention that this would be an unsafe maneuver in
virtually any other high-performance twin turboprop. It's useful to note
that the same section of the Pilots Operating Manual also states that the
aircraft's stall characteristics are "conventional in all configurations"
and that a stick shaker warns of the impending stall four to nine knots
above stall speed.
"The airplane is just unsafe. Rep. Tancredo wants the aircraft grounded or
for the FAA to follow Mitsubishi's recommendations for pilot training," said
Although a former U.S. Air Force B-52 pilot with extensive flying
experience, Van Keuren admitted he's never flown the MU-2. So when
questioned on technical details, rather than respond, he suggested B&CA
contact Robert Cadwalader, an 11,000-hour former Part 135 pilot and
columnist for the Atlantic Flyer, a regional general aviation publication.
Van Keuren claimed that Calwalader was a leading industry expert on the
airplane. However, when we contacted him, Cadwalader said he'd not flown the
aircraft extensively and later clarified that to say all of his MU-2B time
was accumulated in an approved flight simulator. Nonetheless, Cadwalader is
cited by some trial lawyers as an authority on the aircraft. They quote from
his numerous columns in which he has criticized the aircraft extensively.
"It simply doesn't matter how good a pilot is -- if he loses power at low
altitude [in an MU-2] he is going to crash," Cadwalader asserted in a recent
issue of Atlantic Flyer. "An MU-2 with one engine out is a very, very
dangerous airplane and it can go out of control without the pilot being able
to stop it."
When pressed for specifics, Cadwalader told B&CA that his accounts of the
aircraft were "apocryphal" rather than statistically based. He asserts, for
instance, that the NTSB mentions engine failure or power loss due to
undetermined reasons. Engine failure indeed has been a contributing factor
in many fatal MU-2B crashes, but the NTSB has named it as the probable cause
only in a handful of incidents. In most engine failure events that led to
fatal accidents, the NTSB has placed primary blame on pilot error, according
to our review of NTSB accident statistics. The Safety Board investigator
leading the review of the December 2004 crash of an MU-2B-60 at Denver's
Centennial Airport, for instance, told B&CA, "Certainly, that aircraft is
very capable of flying single-engine. Granted, you need to be on top of the
airplane and very attentive to airspeed [control]."
Cadwalader also claimed the FAA never retested the aircraft adequately
during the 1984 Special Certification Review. He said the FAA's pilots never
flew it outside of the flight envelope published in the Approved Flight
Manual but offered no proof of that assertion.
He also claimed that the original FAA certification of the MU-2 was tainted
by the State Department's putting political pressure on the FAA on behalf of
the Japanese, who became a strong U.S. ally during the early years of the
Cold War. That assertion was hotly contested by FAA insiders who consider
that statement an indictment of their professionalism.
Cadwalader further notes that plaintiffs' representatives are not allowed
"to witness or be involved in" MU-2 accident investigations. But as a matter
of practice, the NTSB and FAA only invite representatives of the accident
aircraft's airframe and engine manufacturers, plus a few other select
parties with a vested interest in the aircraft or operation, to participate
in the investigations. The NTSB often requests assistance from manufacturers
to expedite fact finding but limits their participation and normally bars
their contact with witnesses. Cadwalader views this exclusion of trial
lawyers and plaintiffs as a cover-up.
During the interview Cadwalader made several references to observations by
Donald Kennedy, Ph.D., a retired aerodynamics professor from the University
of Colorado at Boulder -- and now based in Kihei, Maui, Hawaii. As it turns
out, Kennedy is frequently called as an expert witness in aviation
litigation cases on behalf of the plaintiffs' bar.
In one expert opinion letter about the MU-2B written for a Denver-based
trial lawyer, Kennedy cites 37 reference documents and comes to several
adverse conclusions about the aircraft. One opinion is that the aircraft
fails to meet FAA standards for controllability, based upon its high
accident rate and wing loading, which, he wrote, is "well beyond the prudent
standards of aircraft design . . ." Kennedy also opined that the "choice of
spoilers for roll control in a light aircraft is a defective design and an
unusual application." Yet a third opinion is "the choice of airfoils for
wing and horizontal tail surfaces were chosen to reduce drag at the expense
of poor stall characteristics in icing conditions. . . ."
These opinions are in direct contrast with the FAA's conclusions in the
original and two subsequent special certification reviews of the aircraft.
Agency officials said that there's "nothing wrong" with the MU-2B's
fundamental design and that it meets or exceeds all type certification
standards that were in effect in 1965 when it was undergoing its
certification trials. One former NTSB investigator said Kennedy was trying
"to rewrite the laws of physics."
However, trial lawyers have prevailed in numerous MU-2B product liability
Generally, they question the FAA's competence in approving the MU-2B and
insist that the FAA's initial certification and subsequent certification
reviews were inadequate. NTSB conclusions, according to them, are equally
suspect. And they hold that any caveat in the Approved Flight or Pilots
Operating Manuals warning pilots not to venture outside the published flight
envelope is a tacit admission by Mitsubishi Heavy Industries that the MU-2B
is inherently unsafe.
When interviewing the aircraft's critics, we heard lots of hyperbole such as
"when you use a [roll] spoiler, you lose all lift on that wing;" and "the
MU-2 lacks the controllability at slow speed during single-engine
operations;" and "below 153 knots, you can't maintain directional stability
if you lose an engine. . . ." We figured we'd find out for ourselves.
But we also concluded that there's no way to change the minds of MU-2B foes
if they won't accept a basic level of competency and honesty on the part of
the FAA and NTSB in the first place.
We spoke with several experienced MU-2B pilots who strongly dispute the
opinions expressed by the aircraft's detractors. None believe exceptional
piloting skill is required to fly it safely. But none had any illusions
about the need for vigilance in the cockpit, the mandate to operate the
aircraft within the flight envelope and the critical need for comprehensive
recurrent training and good maintenance.
John S. "Jack" Broome of Oxnard, Calif.-based Broome Ranches has been flying
MU-2B aircraft for more than three decades, accumulating more than 2,700
flight hours in type. He first soloed in 1935, later became a military pilot
and airline captain, and he served on the NBAA's board of directors for 20
years. He owned and operated Beech 18s for 27 years before buying his first
MU-2B in 1973. He hasn't had an accident in 70 years of flying.
He has forceful opinions about the MU-2B allegedly being unsafe. "Going back
years ago, folks said the same thing about the V-tail Bonanza, Learjets and
Aerostars." He puts prime importance on being a competent, well-trained
pilot. "People can get out of a Cessna 310, legally step into an MU-2B and
say 'I don't need any training.' Then they get themselves into trouble."
That results in accidents, which in turn, drive down resale prices and drive
up insurance premiums.
"There are lots of good airplanes out there, but this one will do 300-plus
knots on 80 gallons per hour. It will also slow down to 120 KIAS on
approach. There's just no way to get into trouble with the airplane if you
fly it by the numbers. For example, I don't use full flaps until I'm over
the fence and slowing to final landing speed. I don't want any surprises;
I'm too old for surprises," he said.
Broome insists on undergoing rigorous recurrent training and has high praise
for Mitsubishi's Pilot Review of Proficiency (PROP) training program. "That
makes the best pilots out of MU-2 pilots," he said.
A veteran pilot with thousands of hours in DC-3s, DC-4s and Beech 18s,
Broome claims that each of those aircraft was considerably more difficult to
fly than the MU-2B. "I've had 'memorable landings' in all those aircraft,"
he said, "but I've had no memorable landings in the MU-2."
He thinks the MU-2B is a "pussycat' and he wouldn't sell the aircraft unless
he couldn't fly it. Now 88, Broome still flies his current MU-2B regularly.
Col. Frank Borman, the former Apollo astronaut, flew three models of MU-2B
aircraft during a 15-year period, accumulating a total of 3,500 hours flight
"They run very, very well and they're most robust, built like military
airplanes," he told B&CA. "It's a solid, honest airplane."
Borman said he underwent FlightSafety recurrent training yearly while
operating the MU-2 and attended the Mitsubishi-sponsored, three-day PROP
course. He believes other MU-2B pilots should do the same, but that most
just don't get enough training. "If you're a relatively new multiengine
pilot, if you lose an engine you'll have your hands full."
The Las Cruces, N.M., resident believes that the recent pressure from
certain congressmen to force grounding of the aircraft is totally
unwarranted. "I don't understand the criticisms of the aircraft. It went
through two FAA certification procedures, plus the most extensive flight
into known icing approval I can remember."
Jack Jaax, an experienced, former Part 135 MU-2B charter operation owner and
chief pilot, echoed these comments. "It's a great airplane, mechanically and
it has a high build quality. But you need to fly it by the numbers." Jaax
flew the MU-2B on air ambulance and charter missions for several years in
the Southern California area. Well-known in San Diego, Jaax recently sold
his MU-2B charter operation and now flies a locally based Beech King Air F90
for its owner.
Don Taylor, vice president of training at Eclipse Aviation, owns an MU2B and
says he "really likes the airplane. It's fast, extremely well built and the
pressure vessel is really tight. It's 15 knots faster than a Twin Commander
with the same engines."
Having logged about 325 hours in the aircraft in less two years, Taylor
admits it flies differently than airplanes with aileron roll control. He
said it's a "little disconcerting" in that it has a slight tendency to keep
rolling in a turn, lacking the spiral stability of some other aircraft he's
"I think the main issue is training," said another experienced MU-2B pilot.
"So many accidents have occurred in the Part 135 community among those who
haven't had formal training. It's all done in house. You need frequent
proficiency training in this aircraft and it needs to be standardized."
This pilot said he lost an engine on climb-out leaving a Philadelphia area
airport. "It was a non-event. The engine went 'whoosh' and NTS [negative
torque sensing] reduced the prop pitch. I checked the torque gauge and
feathered the engine."
The same pilot said the aircraft has somewhat of a split personality.
"It flies like a Patriot missile with the flaps up and a Cessna 172 with the
flaps down." At MTOW, for instance, the Marquise stalls at 105 KIAS in the
clean configuration, but only 87 KIAS with flaps set to 20 degrees. Its
64000-series NACA airfoil also doesn't provide much aerodynamic stall
warning, thus the need for a stall warning stick shaker, a safety device
commonly fitted to many jet aircraft.
Experienced MU-2B pilots emphasize the need to stay ahead of the aircraft,
particularly in the Window of Risk associated with takeoff and landing.
"Fly it with your left hand and think with your right hand," one cautioned.
"On departure, don't make any turns until you get it cleaned up." Gaining
altitude is critical, the old pros say. Several cautioned to keep the flaps
extended until reaching 400 feet agl and never touching the flap switch in a
turn. They also said that it's very important to get the gear up as soon as
a safe landing cannot be made and you've established a positive rate of
Flying stabilized approaches is equally important. Allow yourself ample
distance to get configured and stable, they recommended. Once you get the
gear down and flaps to 20 degrees on approach, 120 KIAS comes up quickly.
You can't afford to get distracted, especially on "black hole" approaches.
Concentrate on the basics. Fixed shaft turboprop engines can cause some
inexperienced MU-2B pilots to confuse the sound of high prop speed with high
power, according to Broome. Airspeed control is especially critical.
The consensus was clear from all with whom we spoke. Fly the MU-2B as you
would a jet, they advised. Use the same operational protocols, fly it by the
numbers and fly it with discipline.
"Amateurs should not be flying the airplane, " said William Seaman, chief
pilot at Flightpath Aviation.
And yet, Dick Allan of Internet Jet Sales says the MU-2B's jet-like
performance pulls them in, attracting "people who don't belong in it." He
describes this group as "a very odd set of pilots -- the fire eaters, the
sword swallowers and the lion tamers. All of them are like bent nails in a
can in your garage. When you need some, you look for the least bent ones and
then try to straighten them out before you can use them."
We Fly the MU-2B-60
In late December 2005, Tom Berscheidt, president of Dallas-based Turbine
Aircraft Services (TAS), invited B&CA to fly a 1980 MU-2B-60 Marquise, s.n.
794, accompanied by chief pilot Pat Cannon. TAS is under contract to
Mitsubishi Heavy Industries to help support the MU-2 and specializes in
sales, service and parts for the aircraft. It is closely associated with
both Mitsubishi and Simcom and is a co-sponsor of the PROP seminars.
Our mission profile consisted of a normal VFR takeoff from San
Diego-Montgomery Field, a climb to 7,500 feet for air work over Borrego
Springs and then a series of normal and simulated one-engine-inoperative
(OEI) takeoffs and landings at Thermal's Jackie Cochran Field.
Cannon started the preflight briefing by showing us a number of inflight
video clips that help dispel common misconceptions regarding the aircraft.
First, MU-2B foes often say that Mitsubishi's use of spoilers for roll
control causes the entire aircraft to drop when one deploys, rolling the
aircraft about the centerline of the upwing tiptank. The video, though,
shows clearly that the aircraft rolls crisply about its longitudinal access
with virtually no adverse yaw.
Another video clip disproves the naysayers' assertions that the aircraft is
uncontrollable at low speed with flaps extended and OEI. The sequence was
shot at 5,000 feet agl, with the landing gear down, flaps extended to 20
degrees and aircraft stabilized and trimmed at 120 KIAS. One engine then is
suddenly shut down. This is accompanied by a momentary yaw and a slight roll
into the dead engine. The pilot them trims and stabilizes the aircraft while
increasing power on the operating engine to 100 percent torque. The aircraft
remains fully under control. A gradual climb is achieved at 125 KIAS at a
After landing gear retraction is initiated, the opening of the gear doors
causes a 50- to 100-fpm decrease in climb rate. Once the gear are fully
retracted, though, the climb rate increases substantially and the aircraft
begins to accelerate. The flaps are retracted to five degrees at 140 KIAS,
then fully retracted as the aircraft reaches 150 KIAS. The pilot then
accelerates to 154 KIAS, the best rate of climb speed at MTOW.
Video clips also show the aircraft is fully controllable during symmetric
power stalls, with no tendency for wing roll-off at the stall break.
Cannon explained that we would repeat some of these maneuvers during the
demonstration flight, but at no time would he allow the aircraft to be
operated outside of the approved flight envelope. With that we readily
The external preflight of the MU-2B is conventional. Along with the usual
fluids, pressures and integrity checks, though, it's essential to check that
the props are set at zero pitch, frozen in position on the start locks. This
assures minimum drag on the fixed-shaft engines during start. It's also
important to extend the flaps and check the rigging. The MU-2B has one flap
motor and a series of interconnected shafts and flex cables that drive flap
jack worm screws. To ensure proper operation, the flaps must extend
symmetrically and exhibit no evidence of binding.
Checking the fuel caps can be a challenge. Tall pilots, flying the long body
models, can step on the main doorsill and peer over the wing to make sure
the caps are secure. We prefer, however, to use a short ladder to get a
close-up view and tactile confirmation of the fuel caps' being secured.
Folks flying short body models must use a ladder because the entry door is
under the wing.
Cannon mentioned that it's also important to check operation of the tiptank
recognition lights, if the aircraft will be flown at night and in icing
conditions. The light shields collect ice quickly if icing conditions are
encountered. The tiptank lights make it easy to see the ice accumulation,
thereby warning the crew in time to activate ice protection systems and
execute an exit strategy from the adverse weather. The aircraft we flew also
was equipped with an optional ice detection system, an addition we strongly
recommend on any business aircraft operated in icing conditions.
After a rash of icing-related accidents in the early 1990s, the MU-2B
underwent a thorough Fact Finding Focused Special Certification Review of
its approval for flight into known icing (FIKI) conditions. After a full
series of ice shape tests and also supercooled large droplet (SLD) icing
tests flown behind a water-spray tanker, the MU-2B was shown to be fully
qualified for FIKI by the FAA. However, Mitsubishi added caveats to the
flight manual that actually apply to any aircraft flown in icing conditions:
Maintain minimum recommended speed, ask for priority handling by ATC to exit
the conditions without delay, avoid abrupt control movements, don't lower
the flaps and don't use the autopilot.
Serial number 794 had an empty operating weight of 8,260 pounds. With two
crewmembers, the BOW was 8,660 pounds. Filled with 1,900 pounds of fuel, the
ramp weight was 10,560 and our computed takeoff weight was 10,500 pounds.
Using the flaps 20 degrees takeoff configuration, the all-engine takeoff
distance was 3,000 feet based upon the airport's 423-foot elevation,
ISA+3ııC OAT, 30.02 altimeter and calm winds. Cannon recommended using a
102 KIAS rotation speed, three knots above Vmca. This was two knots faster
than the book value for Vr.
Many useful charts have been eliminated from newer MU-2B manuals. The
original books provided accelerate-stop distance charts, OEI takeoff
distance charts for flaps five degrees and 20 degrees, and maximum takeoff
weight limited by OEI climb requirements. The FAA directed those charts be
removed from the manuals because they did not conform to the GAMA standard
and involved "demonstrated" data as opposed to "approved" data. In our
opinion, those charts should be restored to the AFM. They provide useful
information to pilots, even though they're not required for certification of
this class of aircraft.
The new manuals, though, do provide OEI climb performance data for the gear
up, flaps retracted configuration. Assuming the same conditions for our
departure from San Diego-Montgomery Field, our climb rate at 10,500 pounds
would have been 650 fpm on one engine with gear and flaps retracted.
Experienced MU-2B pilots told B&CA that the aircraft will climb
satisfactorily at flaps 20 degrees under those conditions, but only if the
landing gear are retracted. If an engine fails at rotation, "and continued
flight is not possible," the AFM advises pilots to "land straight ahead."
In lieu of providing such OEI takeoff data, the AFM now warns "continued
climb performance is not assured unless the landing gear are completely
retracted, the gear doors are closed and the flaps are at five degrees or
less." Experienced pilots told B&CA that this caveat is very conservative.
They said that if the aircraft will climb at 400 to 500 fpm in the clean
configuration it will also climb satisfactorily at flaps five degrees or 20
degrees, assuming the landing gear are retracted.
Pre-start and pre-taxi procedures in the MU-2B aren't as simple as they are
in most modern light jets. The MU-2B cockpit is very busy. This is a Learjet
23-era design airplane and its systems are characteristic of the mid-1960s.
The instrument and side panels seemingly are stuffed with as many switches,
buttons, indicator lights and gauges as a 1960s vintage military airplane.
The voltage of each battery, for instance, must be checked individually by
using isolation switches. Nickel cadmium batteries are standard and they're
recommended for cold weather operations because of their superior starting
power. The aircraft may be fitted with lead-acid batteries for operations in
more temperate conditions.
AC inverter power is required for fuel and oil pressure indications, fuel
quantity indication and several analog avionics functions, so one inverter
must be operating prior to engine start. Stall warning systems must be
checked, along with fuel low level and empty aux tank indicators, boost and
transfer pumps, prop feather valve and NTS functions. It's important to note
that most of these checks must be performed in any TPE331-powered turboprop
of that era, so the MU-2B isn't much more procedure intensive than a Cessna
441 Conquest, Merlin, Cheyenne 400LS or Turbo Commander.
We'll skip most of the preflight details, but it's reasonable to assume that
pilots new to the MU-2B will spend several minutes in the chocks running the
various start and pre-taxi checks. The Mitsubishi is a noisesome machine
externally and internally (also characteristic of its era), and active noise
attenuating headsets are recommended.
With the condition levers in the taxi (minimum rpm) position, prop speed is
about 72 to 74 percent of maximum. Before taxiing, the condition levers must
be set to maximum rpm and the power levers must be pulled aft from ground
idle toward reverse pitch to release the prop blade start locks. This allows
the props to increase pitch and produce forward thrust. Off the start locks,
the condition levers are returned to taxi. A properly set-up MU-2B won't
produce enough forward thrust at ground idle to move out of the chocks, so
the power levers must be positioned between ground and flight idle to roll,
even at comparatively light taxi weights.
The MU-2B has excellent wheel brakes and somewhat sensitive, direct-link
nosewheel steering. The steering design allows the aircraft to be taxied on
one engine, if necessary, for repositioning on the ramp.
Once cleared for our VFR departure, we advanced the condition levers to the
takeoff/landing [maximum rpm] position and switched on auto ignition to
assure a relight in the event of a non-mechanical engine failure. We
advanced the power levers to 90-percent torque. Ram rise during takeoff roll
increases torque to 100 percent. We noted that P-factor induced yaw is
opposite most other turboprops. The engines turn clockwise, but the props
turn counter-clockwise resulting in a right yawing moment with increasing
thrust. It takes very little pressure on the rudder pedals to counter this
yaw because of the effective nosewheel steering. But it takes some practice
to reverse old rudder-vs.-thrust-change habit patterns.
Rotation forces at 102 KIAS were considerably heftier than in some
turboprops because the main landing gear are well aft of the center of
gravity. Minimum rotation speed in the Marquise is never less than 100 KIAS
because Vmca is 99 KIAS. Initial pitch force is much heavier in short body
models because the horizontal tail is closer to the c.g. and the main gear
are much farther aft of the c.g. With weight off the wheels, pitch force was
much lighter and we had to take care not to over rotate beyond the
recommended eight-degree nose-up attitude. The aircraft quickly accelerated
through 120 KIAS.
With a positive rate of climb, we retracted the landing gear. That's a
comparatively long process. It takes about 14 seconds for the first 10, and
the drag from the gear is considerable. After the gear were fully retracted,
the aircraft rapidly accelerated to 150 KIAS and we retracted the flaps. As
the large area Fowler flaps retract, it's necessary to increase pitch
attitude by four degrees as the flaps move from the 20 degrees to five
degrees position. Then pitch attitude must be increased by another four to
five degrees as the flaps retract from five degrees to flaps up. Think
Falcon 10, in terms of flap position vs. pitch attitude characteristics.
We began an eastward VFR departure and reduced power without delay to
maintain 200 KIAS below the floor of San Diego's Class B airspace. Once
clear to the east, we climbed to 7,500 feet and rapidly accelerated to 250
KIAS, the aircraft's Vmo.
Checking the aircraft's handling qualities, we were surprised by the almost
total lack of rudder input needed to maintain coordinated flight when
rolling left and right, even with full control wheel deflection. The
spoilers prevent virtually all adverse yaw, but lateral control forces are
considerably heavier than in most turboprops that use ailerons for roll
We slowed to 180 KIAS, the Simcom recommended speed for steep turns. Setting
about 60 percent torque, we rolled into a 45-degree left turn, then a
45-degree right turn. The pitch force required to maintain altitude was
The demo profile next called for an approach-to-stall series. We again noted
a conservative warning note in the AFM that says "Up to 560 feet of altitude
loss can be expected in recovery from a full stall." We reduced torque to 20
percent. As we slowed the aircraft in the clean configuration at a weight of
10,200 pounds, we trimmed to about 130 KIAS and then just maintained
altitude with increasing back pressure on the yoke, decelerating at about
one knot per second. At 100 KIAS, the stick shaker activated. We added
thrust and recovered with no loss of control along with very little altitude
We then set up for an approach turn stall by extending the flaps 20 degrees
and trimming to 120 KIAS. Again we noted the need for eight to 10-degrees
nose-down pitch change as the flaps moved from clean to 20 degrees. We
rolled into a 20-degree left bank, reduced power and slowly decelerated.
When the stick shaker fired at 83 KIAS, we leveled the wings, added thrust
and flew out of the maneuver. Again the aircraft suffered no loss of
composure and lost just 50 to 100 feet of altitude.
So what about that caveat about losing 560 feet when recovering from a full
stall? Later in the flight, we slowed the aircraft in the clean
configuration until the stick shaker fired and then just continued to
increase back pressure on the yoke. As one expects in an aircraft with a
64000-series wing, there was very light airframe buffet prior to the full
stall, highlighting the need for the artificial stall warning system. At the
full stall, the nose started to rock and then fall. We persisted in holding
back on the yoke. There was a little wing roll that was easily countered
with opposite roll spoiler. We held back on the yoke and kept the aircraft
fully stalled for several seconds, maintaining wings level with roll spoiler
control alone, but being careful to keep the ball in the center with rudder
The aircraft's stall behavior reminded us of a Falcon or a Learjet 45. We
could not make it lose its composure during the maneuver. However, when
we've attempted the same maneuver in some other popular, aileron-equipped
turboprops in the past, we experienced considerably more exciting results.
Abusing aircraft with NACA 23000-series wings during stalls just invites the
onset of a spin.
Recovery from this maneuver in the MU-2B, in contrast, consisted of reducing
angle of attack and adding thrust. And, yes, we lost 500-plus feet during
the recovery because of the abuse we heaped on the airplane. But it never
bit back with a nasty surprise. It's too bad that we didn't have time for a
full stall series with the flaps extended to various positions.
Cannon then demonstrated the effects of losing an engine in flight. While we
flew the aircraft at 150 KIAS, he switched the run-crank switch to the off
position. The negative torque system responded by reducing prop pitch on the
affected engine until the prop was slowly windmilling. There was momentary
yawing as NTS caught up with the power failure. Cannon pulled the condition
lever to feather, which caused the blades to streamline. He also moved the
power lever to maximum, the procedure called out by the AFM. This ensures
that all oil pressure is relieved from the prop controller and that the
spring in the hub fully feathers the blades.
We trimmed the aircraft hands off first by using the trim ailerons to
neutralize the need for roll spoilers and then by using the rudder trim to
eliminate the need for asymmetric rudder pedal pressure. It's important to
trim the aircraft in roll because a deployed roll spoiler will reduce climb
performance by 100 to 150 fpm. It's also important to keep the ball in the
center, maintaining coordinated flight. Failure to do so results in
substantial roll into the dead engine. Cross controlling the aircraft with
roll spoiler and not enough rudder during engine-out maneuvers just ruins
its climb performance.
We slowed the aircraft to 140 KIAS and performed a series of steep turns
both away from and into the dead engine. We added sufficient thrust to
maintain speed in a 45-degree bank turn into the dead engine and found no
loss of composure, no controllability difficulties.
We then headed to Thermal for pattern work and Cannon restarted the right
engine. Cannon positioned the stop-run-crank switch to "run" to arm
auto-ignition and fuel flow. Air start must be done by windmilling the
engine because the electric starter could never overcome the airloads of a
feathered prop on a fixed shaft engine. The condition lever, when moved from
"emergency stop [feather]" to "taxi," has no effect on prop pitch because
without the engine turning, there is no oil pressure for prop control.
TPE331 engines, as a result, have an "unfeather" function that requires use
of an auxiliary electric oil pump. When activated, the pump ports oil
pressure to the prop, thereby reducing pitch and causing it to windmill,
assuming the condition lever is in the "taxi" to "takeoff" range. The
unfeather switch must be held until the engine windmills to 30 percent rpm.
As light-off occurs, the engine will continue to accelerate, start to
generate oil pressure and drive the prop to the desired position. As thrust
is restored, again trimming the aircraft in roll and yaw is imperative. The
MU-2B is a trim-intensive airplane, one that rewards precise pilot technique
and one that doesn't tolerate sloppy airmanship.
We set up for a left base and straight-in approach to Runway 35 at Thermal
at a weight of 10,000 pounds. Based on using flaps 20 degrees, the computing
landing distance was 2,750 feet, assuming a 1.3 Vs landing speed. The first
landing was flown at 20-degrees flaps at a speed of 120 KIAS on approach,
using about 20 percent torque, until we approached the airport boundary. We
slowed to the 105 KIAS Vref over the threshold and reduced power to flight
idle. The engine fuel flows and prop pitch of this airplane were fine-tuned
by Tulsa-based Intercontinental Jet Corp. and they were spot on. Each engine
stabilized at 17 degrees torque, very close to zero thrust. The aircraft
settled down to the runway with the aplomb of a light jet.
At touchdown, we increased back pressure on the control wheel smartly to
prevent the nose from slamming down, moved the power levers to ground idle
and flew the nosewheel down to the surface. At that point, we moved the
power levers to reverse and the aircraft slowed to taxi speed.
Taxiing back to Runway 35, we flew a second circuit to a full stop, this
time using flaps 40 degrees for landing. The aircraft pitch attitude is
considerably more nose down using this flap configuration and, ironically,
AFM approach speeds actually are higher than at flaps 20 degrees because the
book requires use of a 1.5 Vs landing speed. The net result is virtually the
same landing distance because the additional drag slows the aircraft quickly
in the landing flare with power to flight idle.
Our next takeoff was at flaps five degrees. Cannon said we would "lose" an
engine on departure. On takeoff, we rotated at 105 KIAS, about four knots
above the recommended AFM speed and accelerated to 120 KIAS. At about 100
feet agl, about 10 seconds into gear retraction and with all gear doors
open, Cannon pulled back the right engine power lever to flight idle. We
responded by pushing hard on the left rudder and countering the wing roll
with spoiler. OEI climb performance indeed was sluggish until the landing
gear fully retracted. It then improved to 300 to 400 fpm while we used
differential trim aileron to eliminate the need for roll spoiler input and
put in plenty of rudder trim to counter pedal pressure. After the aircraft
was fully trimmed, it climbed at 400 to 500 fpm and continued to accelerate.
At 150 KIAS, we retracted the flaps completely and continued to accelerate
to the 154 KIAS blue line, best OEI climb speed. Climb rate exceeded 650 fpm
at that point. Meanwhile, we were quite busy retrimming the airplane in
pitch, roll and yaw during the level-off and subsequent asymmetric power
Continuing with the simulated OEI emergency, we flew downwind, delaying
extension of gear and flaps to 20 degrees until we were on extended base
leg. We turned to final, slowed to 120 KIAS and made the commitment to land.
Landing technique was almost identical to the all-engine landing, as we
slowed to 104 KIAS over the threshold. Light use of prop reverse and plenty
of differential braking and rudder to counter the resulting yaw moment kept
us near centerline as we slowed to taxi speed.
Cannon positioned the flaps to 20 degrees for another simulated OEI
departure. We used the same takeoff technique and speeds. Cannon pulled back
on the right engine power lever at 100 feet agl about 10 seconds into the
gear retraction cycle. OEI climb performance was lackluster, but
controllability was excellent. Once the landing gear were fully retracted,
the aircraft climbed at 200 to 300 fpm as we accelerated to 140 KIAS. At
that point we retracted the flaps to five degrees, rotated four more degrees
nose up and climb performance increased to 500 fpm. At 150 KIAS, we cleaned
the wing and accelerated to blue line.
With all engine power restored, we headed to San Diego-Montgomery at
redline. SOCAL approach gave us priority over other arriving aircraft
because of our speed advantage. We touched down one hour, 37 minutes after
departing Montgomery Field.
Risky Airplane or Risky Pilots
Few multiengine airplanes we've flown demand more skill and proficiency than
the MU-2B. The Learjet 23, Citation X, CRJ700 and Saab 2000 readily come to
mind as similarly demanding aircraft, but to fly as PIC in any of them one
needs type ratings and annual proficiency checks.
While the MU-2B is more demanding to fly than most business aviation
turboprops, we encountered no nasty surprises or untoward handling qualities
in any part of the low-speed flight envelope, with one or both engines
operating, during our brief demo flight. We never had to assume the role of
fire-eater, sword swallower or lion tamer -- or experimental test pilot --
to keep the MU-2B-60 under control. We concentrated on directional control,
airspeed control and trim control while closely following AFM procedures,
under the watchful eye of Cannon.
Aircraft performance, though, was severely degraded when flown out of trim
during simulated OEI operations. Trim this machine in all three axes in any
regime, so that it will fly hands off, and it will perform better than most
general aviation turboprops at both low and high speeds, on one or two
It's critical that the aircraft be properly maintained, specifically
regarding rigging of the engines, props and NTS functions. If NTS fails to
function properly during an engine failure after liftoff, the aircraft could
be quite a handful to control until the condition lever of the affected
engine is moved to the emergency stop (feather) position. This is another
reason why the pilot must perform the NTS ground checks every day before the
Single-engine takeoff and climb performance in the MU-2B is naturally
limited by weight, altitude and temperature. Assuming standard day
conditions, if you depart at MTOW, using either flaps five degrees or 20
degrees and lose an engine at 125-plus KIAS with the gear retracted, you're
virtually guaranteed a satisfactory OEI climb rate while the aircraft
accelerates to the 150 KIAS flap retraction speed.
Below 125 KIAS and with gear down, you may have to pull back the power, slow
to 105 KIAS and land straight ahead. Old pros say you'll need 4,500 to 5,000
feet of runway to stay on the pavement if you lose an engine just after
liftoff and abort the takeoff.
Departing B&CA's 5,000-foot elevation, ISA+20ııC airport, we recommend
loading the Marquise to no more than 10,250 pounds and using the flaps five
degrees configuration to assure a positive OEI rate of climb once the gear
are retracted. Under these conditions, the MU-2B should climb at 500 fpm on
one engine after accelerating to 150 KIAS and retracting the flaps,
according to the AFM. But if the aircraft suffers an engine failure with the
gear down or in transit, you may not be able to maintain a positive rate of
climb, especially since it takes 14 seconds for the gear to retract
completely. Again, plan on maintaining directional control and land straight
ahead if the aircraft won't climb on one engine.
With all engines operating, gaining altitude is more important than
accelerating, according to experienced MU-2B pilots. Get the landing gear up
with a positive climb rate and when you're out of runway, they say. The AFM
recommends climbing at best all-engine rate of climb speed, using 120 KIAS
for flaps five degrees and 113 KIAS for flaps 20 degrees.
Vyse speeds are increased respectively to 140 KIAS and 135 KIAS, according
to the AFM. If an engine fails after you've reached 400 feet agl, it's a lot
easier to trade a little altitude for airspeed to accelerate to the 140 KIAS
minimum flap retraction speed, the MU-2B pilots suggested. Some MU-2B pilots
recommend climbing to 1,000 feet agl before accelerating and retracting the
flaps. They also say they never move the flap switch in a turn to guard
against flap asymmetry.
The MU-2B has had two Airworthiness Directives related to flight into known
icing conditions, but they're not focused on the aircraft. Rather, they're
aimed at the pilots. AD 2003-22-07, along with AD 97-20-14, which it
supersedes, requires specific pilot training before they fly the aircraft
into known icing conditions. The latest AD requires pilots to view a video
that contains critical information on how to recognize the onset of severe
icing conditions and how to use ice protection systems effectively. The ADs
were issued because of "an increased chance of icing-related incidents or
accidents of the MU-2B series airplanes due to pilot error" [emphasis
added]. Recurrent flight into known icing conditions training is required at
Once you've neared your destination, we advise flying approaches in the
MU-2B at typical light jet speeds, not typical turboprop speeds. Fly no
slower than 150 KIAS with a clean wing. With flaps extended to five degrees,
140 KIAS is the recommended speed. At flaps 20 degrees, use 125 KIAS in the
turns and no slower than 110 KIAS on final approach until you cross the
In flight, when the power levers are pulled back to flight idle, the
aircraft shouldn't suffer a sudden loss of lift from flat pitch or
asymmetric drag. If it does, the props and fuel flows are not rigged
Pilots must truly take command of the MU-2B, especially when it comes to
grounding it for maintenance discrepancies. This aircraft can be most
unforgiving to those who defer squawks. If you can't afford to fix it,
Upcoming Mandate for Formal Training
Mitsubishi Heavy Industries has been requesting that the FAA mandate
training in the MU-2B since the early 1990s, including advocating the need
for an MU-2B type rating. But until the recent spate of crashes and
resulting pressure from U.S. Congress, the agency has been reluctant to
increase the regulatory burden on operators. A recent analysis of MU-2B
fatal crashes, though, indicated that only one or two of the pilots killed
successfully completed formal, third-party training.
That's about to change. The most recent review of accidents by the FAA found
nothing wrong with the aircraft. Pilot error, in contrast, pointed to a
glaring lack of standardized, formalized training. As a result, the FAA
convened a Flight Standards Board, under the direction of Johnathon Vetter,
in the Wichita Aircraft Evaluation Group, to evaluate the need for better
Vetter's FSB team published its draft recommendations on Dec. 16, 2005,
using the regulatory guidance of AC120-53, "Crew Qualification and Pilot
Type Rating Requirements for Transport Category Aircraft Operated Under FAR
Part 121." The upshot is that the FSB recommended airline-quality Level E
initial training, such as that normally required for a type rating, for
pilots new to the MU-2B. Annual Level C recurrent training, using a specific
syllabus, also is recommended. Requalification Level C training should be
undertaken by pilots who have flown the aircraft in the past two years, but
haven't undergone recurrent training. Differences training will be required
when transitioning from certain models of short- and long-body models.
The FSB defines specific requirements for both ground and flight training
syllabi, along with areas of special interest and emphasis, such as stall
recognition, crosswind landing technique, single-engine operations and
flight into known icing conditions.
The FAA's top brass in Washington has made the upgrade of MU-2B training a
priority issue. While they're adamant about not making a knee-jerk reaction
to the recent pressure from Congress, they do plan to issue a formalized
training plan for MU-2B operators by the end of the first quarter of this
year. It's doubtful that a type rating will be required for the MU-2B, but
most of Mitsubishi Heavy Industries' training syllabus could become
mandatory. This one-level-of-training mandate would even out differences
between third party training providers such as Reese Howell and Simcom, and
in-house training provided by Part 135 operators and some private
contractors. Watch the Federal Register for a Notice of Proposed Rule Making
in the next 30 to 60 days.
The folks at the FAA's Aircraft Certification Service also reviewed the
aircraft in the past several months, but they're convinced that there's
nothing faulty with the basic design, especially after three rounds of
Already, some Part 91 MU-2B operators are calling the FSB recommendations
"overkill," anticipating a significant increase in training expense and
hassle because of the Part 121 approach to pilot training. But just look at
the accident statistics.
Dick Allan says of the MU-2B, "It's so easy to fly that it breeds
complacency." He believes that this higher level of pilot training will
slash MU-2B mishaps in half. We agree. The primary reason that so many MU-2B
aircraft crash can be found directly behind the control yoke, in our
We're looking forward to undergoing a complete Level E initial training
course in the MU-2B and then getting more experience in the aircraft. The
Marquise reminded us a little of our first experience in the Learjet 23 in
the late 1970s. It was both exhilarating and illuminating. We could hardly
wait to fly it, but only after thorough training. B&CA
QUALITY AIRPLANE MAINTENANCE IS CRITICAL
Bob Kidd, head of Tulsa-based Intercontinental Jet Corp., has no illusions
about the painful learning curve associated with the MU-2B's entry into
service decades ago. It had multiple propeller and prop-coupler failures, a
resonant vibration that cracked prop blades and plenty of engine failures.
But now the aircraft is a mature design and it's as reliable as any general
aviation turboprop -- if it's properly maintained.
Fine-tuning the flight idle fuel flow and prop governor performance is high
on Kidd's list of must-do maintenance items. He believes that Honeywell (nee
Garrett Aviation) Service Information Letter P331-141 that describes
adjusting the functioning of the prop governor in relation to both the
underspeed and overspeed fuel governors needs to be made an Airworthiness
Directive. Every time work is performed on a propeller, prop governor,
engine or fuel control, a post-maintenance flight check should be done to
check proper flight idle fuel flow and torque, along with prop blade angles.
Kidd also believes the FAA should issue an AD requiring a specific procedure
to rig the flaps and check for proper flex shaft and worm-gear box
functioning. If the system is out of adjustment, the flaps can bind and/or
extend/retract asymmetrically. That can degrade lateral control to the point
at which the aircraft becomes dangerous.
Honeywell issued a mandatory Service Bulletin requiring that prop shaft
couplers be replaced with modified versions that are more robust. Kidd
believes this SB should be made an AD to prevent the possibility of prop
coupler failure, an event that has caused some fatal accidents according to
Early aircraft have three-piece wings. The outer sections are attached with
hardware fittings, including barrel nuts that should be inspected, if not
replaced, at 7,500-hour intervals. This maintenance requirement is often
overlooked by some MU-2B operators, according to Kidd.
There's also a one-time inspection requirement to check for indications of
tip-tank bracket overload. Kidd claims that two-thirds of the fleet has never
undergone this critical inspection.
But he was quick to add he believes that the vast majority of MU-2B
accidents are caused by pilot error, especially if folks don't abide by AFM