DAVID HUGHES/YUMA, ARIZ.

On the McDonnell Douglas MD-11, computers perform

everything from checklist tasks to stall recovery in an aircraft

where the flight engineer's duties are accomplished by a bank of

automatic controllers that run the aircraft's systems.

The cockpit design distills the experience of 19 years and more

than 16 million hours of DC-10 commercial airline operation into

computerized system controllers that operate hydraulic, electrical,

air (pneumatic) and fuel systems. These aircraft system

controllers, as they are called, run in parallel rather than in series

for both normal and emergency procedures. This means, for

example, that the fuel system configuration changes to the proper

status without waiting for steps to be taken to reconfigure the

hydraulic, electrical or air systems. Each system is run by two

computers and each one can revert to manual operation if

necessary.

In addition, the automatic flight system (AFS) includes

augmentation in pitch, yaw and roll. Roll control wheel steering is

optional. The autothrottle portion of AFS keeps the pilot from

unintentionally flying too fast or too slow in a particular

configuration. A flight management system provides automatic

navigation in both vertical and lateral dimensions. It also provides

the optimum speeds and altitudes to achieve the most efficient fuel

consumption on a particular route. The two dual-channel flight

control computers direct the throttles, the ground spoilers, the

elevators, ailerons, rudder, elevator feel, flap limiter and stab trim.

This AVIATION WEEK & SPACE TECHNOLOGY pilot

evaluated the MD-11 recently from the left seat in test aircraft

No. 4. The aircraft, which is powered by General Electric

CF6-80C2 engines, is scheduled for delivery to American

Airlines later this year.

Capt. John Miller, chief of flight operations for the MD-11

program, occupied the right seat for the demonstration. The flight,

which was part of the regular test program, lasted 5.8 hr. and

more than 40 test procedure cards were accomplished. A crew

of technicians manned computer and video consoles in the back

of the aircraft to monitor the tests in progress, which included

numerous checks on the accuracy of the Honeywell flight

management system.

The exterior of the MD-11 looks very much like a DC-10

because the airframe is stretched only 18 ft. 6 in. The MD-11,

however, is most recognizable by its winglets. The outer portion

of the wings have a blunt trailing edge that varies in thickness from

.5 in. near the wingtip to 1.5 in. at the inboard edge of the outer

flap panel. The inlet for the No. 2 engine is the same as that used

on the DC-10 Series 40. This is the case even though the

MD-11's 61,500-lb.-thrust General Electric CF6-80C2 and

60,000-lb.-thrust Pratt & Whitney PW4460 engines require

more airflow than the 52,500-lb.-thrust Pratt & Whitney

JT9D-59A series engines on the DC-10-40. McDonnell Douglas

engineers were concerned about whether the opening would be

large enough, but flight tests have demonstrated that the air mass

movement is sufficient. The third engine in the program is the

65,000-lb.-thrust Rolls-Royce Trent 665. Composite material is

now used in the No. 2 inlet structure.

Once we were in the seat, Miller turned on the inertial systems

and called up route F-150 in the flight management system

(FMS) memory. F-150 is a round-robin route that goes from

Yuma to San Diego, Los Angeles and Palmdale in California and

then on to Boulder, Colo. This route would be used to test the

FMS navigation accuracy. Miller entered a zero cost index to

govern FMS calculations. An index can be selected between 0

and 999 depending on whether fuel economy is the primary

objective or whether reducing time en route is more important.

The next entry in the FMS control display unit was our gross

weight of 460,900 lb. and total fuel of 186,400 lb. with a zero

fuel weight center of gravity of 23%. At this point the FMS

calculated our c. g. with full fuel.

Miller filed to climb to a cruising altitude of flight level 260 and the

FMS computer told us that flight level 340 was optimum for this

fuel weight and flight level 366 was the maximum level. One of

the features of the MD-11 FMS that international operators will

find useful is the system's ability to calculate a series of step

climbs with up to six steps. This allows for more accurate fuel

planning on long-range flights.

The aircraft system controllers are programmed to perform

self-tests. The fuel system, for example, completes its test when

the refueling door is closed. Miller checked with the scanner on

the ground that the flight controls were clear before initiating an

automated hydraulic system test that would move control

surfaces. If any problems are discovered they are annunciated to

the crew.

A check of the fuel system synoptic page, a diagram with tanks

and pumps and engines depicted, showed that the No. 1 main

tank contained 41,100 lb., the No. 2 main tank 64,600 lb., the

No. 3 main 40,900 lb. and the auxiliary tank 39,800 lb. The No.

2 tank is really two tanks in the inboard sections of the left and

right wings, and the auxiliary tank is in the center wing. This

diagram is a dynamic one that tracks fuel levels and notes with

changing colors when fuel pumps are on or off. The fuel quantities

are measured by a computerized system that relies on several

probes in each tank. The tail tank, which has a 2,000-gal.

capacity, was empty but the fuel system controller would route

fuel into it after takeoff to achieve the desired aft center of gravity.

A check of the configuration page on the Systems Display

cathode ray tube (CRT) showed that the brakes were at about

27C and the tire pressure was about 190 psi. A check of the

overall system status page showed that there were no alerts with

consequences.

Miller depressed the APU start button, and the APU doors

opened and the power unit began to turn over.

He then called up the fuel status page to show that a fuel pump

had been turned on automatically to feed the APU and the fuel

system configuration was changing. He called up the electrical

system status page next and showed that the APU was

automatically powering the main buses and the electrical system

was reconfiguring itself to draw power from the APU. External

power was turned off and the air cycle machines began operating

automatically.

MANUAL-AUTOMATIC SHIFTS

Miller demonstrated that it is possible to intervene at any point

and convert the fuel, electrical, air or hydraulic system to manual

operation if the pilot so desires. Reverting to automatic operation

is as simple as depressing a button.

The crew entrance door was closed and the FMS calculated all

of our takeoff speeds, which appeared as markers on the speed

tape. V (takeoff decision speed) was 134 kt., Vr (rotation speed)

was 150 kt. and V (takeoff safety speed) was 163 kt. These

markers would begin moving down from the top of the display as

we accelerated and approached the speeds involved.

Miller planned to have me fly the takeoff with the autopilot off

and the auto-throttles engaged. We planned to take off on

Runway 3L, which is 13,299-ft. long, and fly on runway heading

before turning left to Bard, which is the VOR located near Yuma.

Miller pushed the No. 3 ignition switch, and the air system

reconfigured itself by shutting off the packs to provide bleed air

from the APU for engine start. He then pulled out the start switch.

N2 started to accelerate and when it reached a blue line on the

engine and alerting display at about 15%, Miller pulled out the

electrical fuel switch to turn it on. A line, which represents the

starting limit, appeared on the exhaust gas temperature gauge on

the engine and alert CRT. This line goes away after engine start is

complete. Light-off occurred and the engine accelerated to about

45% of N2. The air, hydraulic, fuel and electrical systems were

automatically reconfiguring as the start sequence progressed.

After starting engines 1 and 2, Miller said, ''The electrical system

has picked up on the buses and the tie bus. The air system is now

giving us air conditioning. The fuel system is transferring fuel.'' I

then started the No. 2 engine, and Miller shut down the APU.

Miller pointed out one feature of the fuel and hydraulic synoptic

pages that serves as a good reminder to the pilots later in the

flight. At the time of engine start, a blue line appears at the top of

the three hydraulic system reservoir diagrams to show where the

quantity is at engine start. Should fluid be lost during the flight, the

discrepancy is noted by the difference between the fixed blue line

and declining fluid level marked in solid gray. The same sort of

blue line also appears on the oil quantity diagrams. Miller used an

abbreviated checklist that fits on one side of a laminated card to

double-check critical items.

I released the brake and began taxiing the aircraft. The nosewheel

steering control on the left side was easy to use; however, I had

to work on my taxi speed and rate of turn to come around a

corner smoothly because I was unfamiliar with the aircraft.

Nosewheel steering provides 67 deg. of authority left and right,

and rudder pedal steering provides 12 deg. of authority left and

right. One feature on the primary flight display that was helpful

was a readout of taxi speed. I kept the MD-11 moving between

10 and 20 kt.

We were cleared into position on Runway 3L with the winds at

060 deg. and 8 kt. The takeoff would be made with bleeds off

and air cycle machines off. We were cleared to fly runway

heading to a point 3 naut. mi. past the airfield boundary before

turning left to Bard. Our assigned altitude was 4,000 ft. and we

were told by ATC to expect clearance to flight level 260 10 min.

after departure. We used a call sign of DACO 450, which was

derived from Douglas Aircraft Co. aircraft number 450.

We were cleared for takeoff. The c. g. was 23.1% and the flap

setting was 15 deg. for the derated thrust takeoff. The takeoff run

began at 11:15 a. m. as I began advancing the throttles. With

autothrottles on, the automatic system takes control of the

throttles when the engines reach 60% of N1 and sets takeoff

thrust.

The takeoff thrust target is marked with a ''V'' which rests on the

outer edge of the three round-dial N1 diagrams on the engine and

alert CRT. A line that ends in a T moves up as power is

advanced, and when the desired setting is reached the T fits

inside the V to provide the pilot with a simple visual cue to show

that the power setting is correct. The pilot can always override

the autothrottles and push the power up to maximum rated thrust

at the forward physical stop. The throttles can be pushed past this

gate in an emergency with a 30-lb. force to achieve the maximum

available thrust limited to engine red line.

The aircraft reached 80 kt. 7 sec. after the final thrust setting was

made and the autothrottles entered CLAMP mode, which meant

the throttles were fixed for takeoff and could not roll back. Nine

seconds later we reached the V speed of 134 kt. at which point

we were committed to the takeoff. Four seconds later I pulled

back on the yoke and the nose rotated smoothly into the air as I

began following the rising pitch V bar. We had used about 4,000

ft. of runway on the takeoff roll. Above the flight director V bar

was a pitch limit indicator that marked the not-to-exceed angle of

attack. This symbol would change from blue to amber if we

approached within 2 deg. of the stick shaker angle of attack, and

it would turn red if we were about to reach the stick shaker angle

itself.

A pitch limiter is a standard feature for wind shear escape

maneuvers, but the MD-11 can display it during normal

operations as well. Miller noted the positive rate of climb and put

the gear handle up. As each V speed was reached the associated

marker would come down the airspeed scale.

The aircraft climbed rapidly as it had plenty of excess power

considering it was nearly 160,000 lb. below the 618,000-lb.

maximum allowable takeoff gross weight.

I rotated the nose up to about 25-deg. to follow the V bar, which

was calling for the maximum pitch the flight guidance system ever

commands. The aircraft continued to accelerate to V and then to

173 kt., or V +10.

At 400 ft. above the ground, Miller engaged the autoflight system

and the profile VNAV mode and the autopilot continued the

climbout as I monitored the controls. At 1,500 ft. above the

ground, the autothrottle system reduced the power to a climbout

setting without any further action from me. At 3,000 ft. above the

ground, the aircraft nosed over and Miller retracted the flaps as

we began to accelerate from 173 kt. to 250 kt. for the climb

profile to 10,000 ft. The MD-11 will fly this portion of the profile

at 1.3 Vstall +5 kt. if that happens to be higher than 250 kt. for a

given weight and set of conditions.

At 213 kt. (V +50 kt.), Miller retracted the slats. With the lateral

NAV mode engaged, the automatic system had already initiated a

left turn to proceed to Bard VOR and we were cleared up to

7,000 ft. Setting a new altitude limit into the FMS is a simple

procedure using the controls on the glare shield.

The flight control panel on the glare shield allows the pilot to

make changes in heading, airspeed and altitude and to alter these

values in the flight management system computer without going

through the two multifunction control display units on the

pedestal. The pilot simply changes the preselect value in a

window and pulls a knob out to activate a new heading, airspeed

or altitude. To hold the current speed, heading or altitude, the

pilot simply pushes in on the same knob. FMS push buttons

under the altitude, speed and heading knobs allow the pilot to

turn the knobs and then send these changes to the FMS for use in

its calculations. If the pilot depresses one of the FMS push

buttons without turning the knob, the value in the FMS program

will be retrieved for use.

''You don't have to get down to the control display unit and fiddle

around because there is a great deal of interface between the

glare shield and the FMS,'' Miller said. The glare shield system

directs the changes to take place regardless of the FMS or

automatic flight mode engaged.

Five minutes after takeoff, Miller called up the fuel system

synoptic page on the systems CRT and I noticed that the fuel

system had automatically pumped 3,900 lb. of fuel into the tail

tank and that our c. g. had moved aft to 25.6%. The MD-11 is

designed to be flown at an aft center of gravity in cruise to reduce

drag. After passing 10,000 ft., the system established the best

economy climb speed of 347 kt.

After passing Imperial VOR as we were flying west to Kumba

intersection, we were told by ATC to maintain our present

heading. This was accomplished by pushing the heading knob on

the glare shield to maintain the heading, and the deviation to the

left of our intended course was depicted clearly on the navigation

display.

Later, as we returned to course and passed over Julian, the

autopilot smoothly banked 10 deg. to turn to a heading of 294

deg. The navigation display provided a wealth of data. All six

CRTs were easily readable in the strong sunlight, and the color

legends made it easy to interpret data.

All speeds, headings and altitudes specified by the pilot are

displayed in white, for example, while all speeds, headings and

altitudes derived from the FMS are shown in magenta. This helps

the pilot remember when he has intervened in the automatic

operation and when the FMS is navigating. It is possible to call

up a standby flight plan on the navigation display in a second

color for planning purposes.

The FMS automatically tunes the appropriate navaids for the

route being flown. The system relies on three Honeywell ring laser

gyros as well as two scanning DMEs with five channels each --

one for VOR, one for ILS and three for the FMS to use in

precision navigation. The pilot can take control of the VOR/DME

or the ILS/DME if he prefers. The MD-11 has two VORs, two

ILSs, two automatic direction finders and provisions for two

microwave landing system receivers as well. Global Positioning

System capability can be added in the future by inserting a circuit

card in the inertial reference unit.

We passed over Catalina Island at flight level 260 and Mach

.808 and were cleared up to flight level 270 as we turned north

toward Los Angeles. Miller entered the new altitude assignment

into the FMS from the flight control panel on the glare shield. The

AFS vertical alert warned us that the system was initiating a climb

to flight level 270, so we could override it if we wanted. We

burned an average of 18,000 lb. of fuel an hour during the en

route cruise portions of our flight.

The MD-11 has a range shortfall owing to lower than expected

fuel efficiency on the General Electric and Pratt & Whitney

engines and higher than expected aircraft weight. The engine

manufacturers are working to correct the 4-5% specific fuel

consumption shortfall, while Douglas is working to decrease the

aircraft's empty weight (AW&ST Aug. 6, p. 70). Douglas may

also achieve some small improvements by fine tuning the

aerodynamics, according to Miller.

The empty weight has been reduced by 1,700 lb. so far, and the

maximum allowable takeoff gross weight being offered as an

option has been increased as part of the effort by Douglas to

meet various payload guarantees over defined routes. Douglas

just added another 3,000 lb. to bring this optional weight to

618,000 lb. At this weight, the aircraft will be able to fly nearly

8,000 stat. mi., and some airlines are expected to opt for the

higher weight. The standard maximum takeoff weight remains

602,500 lb. At this point, other Douglas test pilots on the flight

took turns in the left seat to perform additional tests on the FMS

during the round robin back to Yuma. The MD-11 completed the

circuit to Yuma and headed back to California and out over

Mission Bay to Warning Area Whiskey 291 over the Pacific

Ocean.

After getting back into the left seat, I glanced at one of the CRTs

to update myself quickly on the status of the flight. The progress

page, as it is called, consolidates on one CRT display all of the

data needed for an International Civil Aviation Organization

(ICAO) position report, plus a lot of other useful information. In

addition to the last position, time and altitude, it gives the next

position, estimated time of arrival and altitude and the position

after that. The page also records the outside air temperature,

wind, fuel remaining, distance to go to destination and the fuel that

will be remaining at the time of arrival. It also gives a distance to

the alternate airport, estimated arrival time and fuel that will be

remaining there. ''This is an example of how we have designed the

system to serve the pilot,'' Miller said. In earlier generation

aircraft pilots had to hunt for this type of data before making a

position report.

I noted our fuel on board was 113,300 lb. and we would have

99,100 lb. left on arrival at Yuma, 322 naut. mi. away at 4:05 p.

m. To keep track of our position within the warning area, Miller

called up a flight plan loaded with the points on the area's

boundary. When this flight plan was displayed on the navigation

display unit, the connect-the-dot pattern depicted the Whiskey

area and how close we were flying to the area's borders.

I disconnected the autopilot and an autopilot-off message

surrounded by a box that was flashing appeared on the primary

flight display. A touch of a button acknowledged that I was

aware the autopilot was off and the flashing stopped. I began to

fly the aircraft at 17,500 ft., and when the aircraft was 150 ft.

below the designated altitude in the FMS a chime sounded and a

computer voice said ''altitude.'' The MD-11 offers a variety of

voice warnings that customers can select including 1,000 ft.

above a level-off. If the aircraft is climbing or descending too

rapidly as it approaches a level-off, a voice warning will be

activated.

The autothrottles were holding the speed at 220 kt. I

disconnected that system and an autothrottles-off message

appeared in a flashing box. A speed bug on the speed tape

provided an easy peripheral visual reference as to whether I was

flying the desired speed +- 5-10 kt.

As the aircraft decelerated slightly, a green bar extending down

appeared, predicting the speed I would be flying in 10 sec. I

added a little power to maintain speed, and Miller turned off the

longitudinal stability augmentation system (LSAS), which uses +-

5-deg. of elevator deflection to augment longitudinal control. With

LSAS on, the pilot sets a pitch attitude and the LSAS system

holds it with elevator inputs. When LSAS is turned off, the

aircraft still flies smoothly, but I needed to trim the aircraft to

maintain the selected attitude. ''It's just a regular airplane now,''

Miller said.

I experienced no difficulty flying the aircraft at an aft center of

gravity and a slow speed.

Miller reengaged the LSAS to demonstrate its speed protection

features as I began to slow the aircraft in a clean configuration

toward stickshaker speed. Miller suggested that I attempt to

close the throttles and maintain a 10-deg. pitch attitude. The first

protection feature to be activated was the autothrottle system,

which started to advance the throttles to maintain Vmin. This

process would have continued until the autothrottles reached

maximum continuous thrust. To override this, I held the throttles

back against pressure.

The LSAS then activated its Vmin protection and began pushing

the yoke forward. ''There are two protections for every

condition,'' Miller said. To slow enough to reach the stickshaker

speed of 150-160 kt. in a clean configuation, I had to hold the

yoke back against substantial pressure as well.

The pitch limit indicator turned amber and Miller estimated the

LSAS was providing 50 lb. of forward pressure on the yoke by

this time. The PLI turned red as the stickshaker began to vibrate

the yoke.

By releasing the yoke and throttles, I allowed the aircraft to

recover on its own. The LSAS system lowered the nose and the

roll control wheel steering kept our bank angle under 5 deg.

during the recovery as the autothrottle system selected maximum

continuous thrust. The slats also extended automatically as noted

by a legend on one CRT, and we felt a little bit of shuddering.

Our altitude loss was negligible. The slats retracted when a safe

speed was achieved. Miller later explained that if the maximum

speed allowable is about to be exceeded, the protection systems

throttle back and raise the aircraft's nose.

Miller then demonstrated how the speed tape keeps the pilot

informed of his changing minimum and maximum speeds,

depending on the aircraft's configuration. Two amber-colored

areas marked on the top and bottom of the tape move whenever

the configuration changes to show where the speed limits are

located. The lower amber region, for example, starts at 1.3 Vs

for the current configuration and extends down to the speed

marked in red where the stickshaker would activate. The speed

command bug cannot be moved any closer than 5 kt. above the

minimum or 5 kt. below the maximum speed, even if the pilot

inadvertently tries to select an inappropriate speed.

Miller showed how the top of the slow-speed amber region

dropped from 197 kt. to 160 kt. as he extended the slats and

lowered the aircraft's 1.3 Vs speed. Another marker showed the

speed at which we could safely retract the slats. As I slowed the

aircraft, Miller extended the flaps to 28 deg. and the top of the

slow-speed amber region moved down again to 153 kt. He then

extended the flaps to 35 deg., and I slowed the aircraft to 120 kt.

When the stick began to shake, I released the controls and the

aircraft automatically recovered.

Following this, Miller extended full flaps to 50 deg. and the

maximum speed for leaving the flaps out was marked on the tape

at 175 kt. As I accelerated toward this speed, the flaps began to

retract automatically to 40 deg. When Miller raised the flaps to

28 deg., the maximum speed marker for leaving the flaps out

moved up to 210 kt.

Following this slow-flight sequence, we engaged the flight

management system to take us toward San Diego climbing back

to flight level 270 at a speed of 325 kt., the economy speed for

our weight.

As we headed east back to Imperial, we encountered a line of

thunderstorms and the weather radar painted heavy rain shown as

red on the navigation display CRT. The storms were ahead and

to our left. Miller later explained that the intensity of the weather

presented on the navigation CRT can be adjusted independently

from the intensity of the navigation map display even though the

two appear together.

Weather can be overlaid on an HSI or any other navigation

display. As the tops were above our altitude, we deviated to the

southeast. We were approaching the top of descent as calculated

by the FMS.

We began a descent to flight level 240 until ATC held us at flight

level 250 for traffic. Miller said the top of descent points used in

test flights had worked out well, but in this situation ATC was

keeping us higher than we wanted to be. As we proceeded on at

flight level 250, an altitude message appeared to indicate we

would not be able to reach our desired altitude of 2,500 ft. at

Bard VOR, our initial approach fix. Miller began slowing the

aircraft below 300 kt. and then a speed error message appeared

indicating we would not be able to slow to our speed target of

180 kt. by Bard.

Nearing Imperial we were permitted to descend and Miller

deployed the speed brakes, consisting of 30 deg. of spoiler

extension, and picked up a 2,000-fpm. rate of descent. He noted

that it is possible to use both slats and spoilers at the same time.

The aircraft could descend at 4,500 fpm. at about 335 kt., but

we were nearing 10,000 ft. at this point and we would have to

slow to 250 kt. anyway. We slowed to 209 kt. at 11,000 ft. and

Miller lowered the landing gear to increase drag.

The navigation display of our approach into Yuma showed a right

turn to the downwind, a left base and the final approach leg. A

blue arrow along that route of flight showed where we would

reach 5,000 ft., our next altitude limit. The display also showed

distance and time to waypoints on the approach. We did not

reach 2,500 ft. by Bard but we did descend to 1,700 ft. on

downwind, so we managed to get down in time without additional

off-course maneuvering.

As we passed the outer marker inbound on the approach, I

disconnected the autopilot and began flying an ILS approach to

Runway 3L with a Vmin +5 kt. approach speed of 140 kt. With

LSAS engaged, it was easy to hand fly the approach. Weather

was not a factor because the sky was clear and the winds were

relatively calm.

I went a little high on glideslope but corrected back. At 200 ft.

above the runway, a computer voice began counting off the

altitude based on a combination of radar and barometric altimeter

inputs. The voice started at 200 ft. then read out 100 ft., 50 ft.,

40 ft., 30 ft., 20 ft. and 10 ft. It took about 20 sec. to go from

200 ft. to touchdown. Miller said Douglas pilots can tell from the

pacing of the readouts whether the touchdown will be a good

one. The touchdown was smooth but I was slow in lowering the

nose gear to the runway. I depressed the go-around button on

the throttles as Miller reset the flaps, and then I advanced the

throttles. We rotated and climbed about 18 sec. after touchdown,

with autothrottles engaged. Miller raised the gear and had to

remind me to follow the single-cue flight director which was

providing takeoff guidance. He engaged the autopilot before we

reached our level-off altitude.

We planned a full stop and a taxi back for another takeoff on our

next approach.

When I clicked off the auto flight system on final, Miller also

disengaged the LSAS system. Flying the same ILS approach with

LSAS off was only a little more difficult than it was the first time

around with LSAS. With a little more practice, the LSAS

disengagement would be hardly noticeable and certainly not a

factor of any consequence.

Our gross weight at this point was 366,000 lb. with 89,800 lb. of

fuel remaining and an aft c. g. of 30%. Miller said a c. g. for a

normal landing would be about 26%, but owing to some flight test

experimentation going on during this part of the flight, we were

carrying some fuel in the tail tank. I noticed no problems flying the

approach with the c. g. farther aft than it will be in airline

operations.

After touchdown, I lowered the nose and applied the brakes with

Miller backing me up as I reversed thrust. Auto spoilers deployed

to 30 deg. on main gear spin-up and to 60 deg. on nose gear

spin-up. We turned off to the right and taxied back for another

takeoff.

An option in the test aircraft let me switch from a single-cue V

bar to split bars, which I am more comfortable with. Airlines will

be able to select one or the other display option. After the

takeoff, I flew another touch-and-go and a full-stop landing as we

terminated the flight. Our fuel was down to 79,000 lb. on the final

landing with a gross weight of 353,000 lb. and a c. g. of 28.6%.

After our final landing at 4:55 p. m., I braked so we could make

an intersection turnoff. The carbon brakes provided smooth

deceleration. From the touchdown zone 2,000 ft. down the

runway to the turnoff into the Douglas facility we rolled about

5,000 ft. A check of the brake temperatures showed them to

vary from 140 to 220C, which Miller said was not very high for

the amount of braking I did on the final full stop. The temperature

probes are mounted directly on the brakes, rather than on the

axles as they are on some transport aircraft. We met the tug,

which took us back to parking. Miller said the MD-11 passed a

demanding 100% refused takeoff test in its certification trials in

which brakes that were 90% worn away were used to stop an

aircraft that weighed 613,500 lb. (AW&ST June 18, p. 32).

Miller shut down the No. 1 and 3 engines first, and the aircraft

hydraulic, electrical, fuel and air systems automatically

reconfigured themselves. After some additional checks, we shut

down No. 2 as well.

Miller departed for a debriefing on the test cards accomplished

during the flight as the MD-11 test program moved nearly 6 hr.

Developed by Joint Venture

DAVID HUGHES/YUMA, ARIZ.

The MD-11 avionics suite developed in a joint design effort by

McDonnell Douglas and the Honeywell Air Transport Systems

Div. breaks some new ground in automation but is also designed

to keep the pilot in the loop.

In an unusual arrangement, Honeywell is supplying and integrating

all of the avionics for Douglas and is even a financial risk-sharing

partner in the MD-11 program.

The philosophy for the design of the MD-11 is to take care of

routine tasks and even emergency procedures automatically to

assist the pilot and reduce workload (AW&ST May 11, 1987, p.

147). Rather than present the pilot with a series of tasks to

perform in a certain order, the automatic controllers for the fuel,

air, hydraulic and electrical systems simply accomplish the steps

for him in the desired order. ''We wanted to find another way to

make it happen the way we intended it to happen without loading

the pilot up with the responsibility to remember or find out,''

Capt. John Miller, chief of flight operations for the MD-11

program, said.

The pilot receives feedback in the MD-11 automatic flight system

by having the yoke move when the autopilot is moving control

surfaces and the throttles move whenever a power setting is made

by the autothrottle subsystem. The pilot can override the

autothrottle by moving the throttles at any time, and he can push

the throttles past a gate to go all the way to engine red line. He

can override the autopilot just as easily.

Paul Oldale, chief design engineer for the MD-11 system

controllers, said the idea is to keep the pilot informed on the root

causes of a malfunction without overloading him. Designers

wanted to tap into the understanding of how to operate DC-10

systems that an experienced DC-10 flight engineer would have.

Incorporating this knowledge in the software captures ''all the

guile and cunning of a flight engineer,'' according to Miller.

Exactly how line pilots will react to the system will be seen when

the MD-11 enters commercial service later this year. Pilots from

37 airlines participated in the cockpit development. ''The cockpit

design was driven by pilots,'' Miller said. Automated systems can

always be overruled by the pilot, who remains the ''final authority''

on the MD-11 flight deck.

Douglas pilots brought more than just test flight experience to

bear when they participated in the design process. Most of them

have logged a substantial number of hours as captains in

command of Douglas aircraft on line trips for customer airlines.

Miller, for example, has flown many hours in Europe, Africa and

Asia, and his experience there provides insight into the needs of

international operators.

Miller and Douglas test pilot Phil Battaglia briefed this

AVIATION WEEK & SPACE TECHNOLOGY pilot on the

cockpit layout prior to a recent demonstration flight. Battaglia

pointed out that the aircraft system controllers on the overhead

panel are organized logically. Beneath the row of three engine

emergency fire shutdown ''T'' handles is the hydraulic system

control panel followed by the electrical system, the air system and

the fuel system panel. Each of these systems is normally

controlled automatically by two computers, but the pilot can take

over manually at any time. All the controls for the No. 1, 2 and 3

hydraulic, electrical, air and fuel systems are located directly

beneath the related No. 1, 2 and 3 engine fire handles. This

allows the pilot to easily scan all the systems that would be

affected by an engine shutdown. ''This is an integrated system, not

a jigsaw arrangement of little control panels,'' Miller said.

If a pilot selects manual operation, the system controller reverts to

the safest possible operation. The fuel controller, for example,

turns on all fuel pumps and transfers fuel in one direction only --

toward the engines. (in a normally operating elec system, that is)

During normal operations, the pilot can keep track of the status of

the various systems without looking overhead by monitoring the

system display CRTs on the front panel.

The electronic instrument system (EIS) 8 8-in. CRTs are located

six abreast. In normal operation, the outer tubes serve as primary

flight displays (PFDs), which incorporate attitude director

indicators (ADIs) with a speed tape on the left and an altitude

tape on the right and a partial compass rose with heading

information at the bottom. Inboard of the PFDs on both sides are

the navigation displays, which allow the pilot to select a horizontal

situation indicator (HSI) or a map, plan, VOR and approach

mode. The aircraft's flight management system computes the

aircraft's trajectory and provides guidance for vertical and lateral

navigation.

The two display units in the center of the six-tube lineup include

the engine and alert display on the left with N1, N2, EGT and fuel

flow. The system display on the right presents 12 different

displays, including a secondary engine page with oil pressure,

temperature and quantity as well as gross weight, center of

gravity and stabilizer trim setting. The pilots can also call up

synoptic displays showing simplified views of the hydraulic,

electric, fuel and air system configurations.

The synoptic displays are dynamic and the color of hydraulic

pumps change to green to show when they are in operation. The

color of the electrical buses changes from white to green to show

when they are powered or to amber when an electrical fault

occurs. On the air page the temperature in an aircraft

compartment such as the cockpit is shown in white next to the

commanded temperature the air conditioning system is trying to

meet in blue.

A configuration page shows an MD-11 in outline form viewed

from the rear with control surface positions noted. In a glance a

pilot can check the position of the flaps, slats, ailerons, elevator,

rudder, spoilers and landing gear. The diagram also notes the

temperature for each brake and the air pressure for each tire.

The fuel synoptic page shows the number of pounds of fuel being

burned by each engine per hour as well as the number of pounds

of fuel remaining in the three main tanks, the auxiliary tank and the

tail tank. The horizontal stabilizer on the MD-11 is about 30% smaller

than the one found on a DC-10. It contains a tail fuel tank with a

2,000-gal. capacity which allows the center of gravity to be maintained

well aft to reduce drag.

If any of the CRTs should malfunction, the pilot simply turns the

faulty unit off and the EIS system automatically reconfigures itself.

There are no compressed display formats; instead the full-size

displays are presented on the CRTs that are still working

properly. Avoiding special display formats reduces pilot training

requirements.

The dark cockpit concept is used, so normally there are no lights

illuminated during routine operations when there are no system

discrepancies. There is a heirarchy in the types of system alerts

made to the pilots of the MD-11 in the lower third of the engine

and alert display. A red master caution light warns of a Level 3

alert while a master caution light in amber alerts the pilot to a

Level 1 or 2 alert. The master caution directs the pilot's attention

to the engine and alert display where a box around an alert

message means that action is required.

When the pilot calls up the synoptic page for the hydraulic system

to investigate a warning that the No. 1 hydraulic system has

failed, he will see the status of the system and a list of

consequences from the failure. He will be reminded among other

things that he will only be able to use 35-deg. of flaps, that

autopilot No. 2 is inoperative and that flight control effect is

reduced.

In case of hydraulic line damage in the tail, the MD-11 has the

same fast-acting shutoff valve to protect hydraulic system No. 3,

which was added to DC-10s following the accident in Sioux

City, Iowa, last year.

The MD-11 also has the standard DC-10 rudder and elevator

trim, yoke and nose-wheel steering system among other systems.

The standby electromechanical altimeter and attitude direction

indicator are located below the landing gear handle on the

pedestal where they are easy to see. The left-side primary flight

display and the engine and alert display are retained when the

battery is powering a single emergency bus. With the air-driven

generator deployed into the airstream, power is supplied to five

out of six EIS display units. In test flights the MD-11 has

operated for 25 min. on battery power and for up to 40 min. on

the air-driven generator without any problems.

One example of how simplified things have become on the

MD-11 compared with the DC-10 is that the 28 switch actions

required by the flight engineer to jettison fuel have been reduced

on the MD-11 to one pilot action. The system automatically

jettisons fuel at 5,000 lb. per min. to the aircraft's maximum

landing weight unless the pilot selects a different figure.

In a complex emergency situation involving multiple failures, the

computerized systems would become the troubleshooter and

reconstruct the aircraft to get all systems back on line if possible.

The pilot can always check the status of the systems or override

the automatics if necessary. And pilot action is always required to

take an irrevocable step such as to jettison fuel or to disconnect a

generator.

Analysis of one incident involving an uncontained engine failure on

a DC-10 aircraft in the early 1970s found that the flight engineer

was faced with having to perform 110 steps to complete all his

checklists. On the MD-11, the four system controllers would

perform all but five of these steps for the pilots.

McDonnell Douglas recognizes the key role of computer

technology in the program and plans to place launch teams in the

field with sufficient expertise and specialized tools to handle any

fine-tuning that may be required in the computer systems.

A portable Sundstrand optical disk will be used to troubleshoot

any problems that develop, according to Joseph R. Ornelas, a

manager for MD-11 test Ship No. 4. It can take data from 12

digital ARINC 429 buses on the aircraft, each of which can

supply data on 240 different parameters. Once the information is

recorded, it can be converted to engineering units and played

back on the ground for analysis using a personal computer.

The importance of software in the program is demonstrated by

the fact that Douglas and its subcontractors have employed about

1,500 software engineers for two years during the peak of

MD-11 design work. The teams working on the FMS and the

autopilot each had 100-150 software engineers involved as did

the EFIS and the automated fuel, hydraulic, air and electrical

systems.

By building optional features into the software, such as CRT

displays in pounds or kilograms and flight directors with a single

cue or split cue, Douglas keeps all line replacable units in the

avionics suite identical. If the aircraft are ever sold to another

airline, reconfiguring the instrumentation to make it compatible

with the new operator's fleet will simply involve the movement of

three plugs on each of the two flight guidance computers.

The fact that so many avionics features are contained in the

software means that service bulletin updates will be made by

shipping each customer a 3.5-in. floppy disk with the latest

program. This can occur much faster than when a service bulletin

requires that a box be removed and worked on or replaced. In

fact, Douglas officials hope to devise a method that will be

acceptable to civil authorities of sending service bulletin software

over the telephone by modem. The key is to make sure that no