Bombardier is adding the CRJ700 to its family of market-beating regional jets - but does it have what it takes to build on their success?
Bombardier's CRJ700 is the company's offering as a larger version of its best-selling CRJ200. Despite a busy programme expected to net type certification in Canada and the USA by the end of the year (Flight International, 5-11 December 2000), Flight International recently flew Prototype 10002 (registered C-FJFC), which was fitted with extensive flight test equipment.
Although 10002 was a development aircraft, its handling and performance were sufficiently close to a production CRJ700 to make the flight representative. In fact, its development status enabled the corners of the flight envelope to be explored more extensively than would probably have been possible with a production version of the aircraft.
In developing the CRJ700, the stretch has triggered a review of many current design features. As a result, the CRJ700 uses only 15% of unmodified CRJ200 airframe, says Bombardier experimental test pilot Chuck Ellis.
The CRJ700's General Electric CF34-8C1 engines are a new version of the CF34, produce a significant 50% increase in thrust, and flew for the first time on the CRJ700 prototype. Although the aircraft is a type derivative for certification purposes, the flight test programme has been extensive to embrace all the design changes.
The scale of airframe changes is significant: seat numbers have grown from 50 to 70-74 seats; the cabin floor has been lowered by 25mm; the cabin windows have been re-positioned and the fuselage frame dimensions have been reduced to increase the internal diameter of the cabin by 25mm, while maintaining the external width.
The wing has been increased in area and has been given leading edge slats to enhance field performance. The fin, rudder and tailplane are also larger in area to increase longitudinal and directional authority because 3.96m of the 4.72m stretch is forward of the wing, and the main undercarriage is new and stronger. When feasible, the new design features will be incorporated in the CRJ200 which will continue in production as a key member of Bombardier's regional aircraft family.
Despite so many differences, the design team has been at pains to preserve as much commonality as possible, from a pilot's perspective, between the CRJ200 and the CRJ700. Thus, even where a system is substantially different in operation, the pilot's control panel in the larger aircraft has been made as similar as possible to the earlier version.
This in turn has led to a cockpit that incorporates some compromise. For example, it shares the Rockwell Collins Pro Line W avionics suite with the CRJ200 rather than moving up-market to later, larger and fewer liquid crystal displays. The overhead panel is a mixture of push buttons and toggle switches.
None of this is a negative criticism of the CRJ700 cockpit - merely a reflection on changes that might have been considered had the aircraft been a "clean-sheet-of-paper" design. In fact, the cockpit is an extremely comfortable and practical environment in which to work.
Climbing into and out of the seats was not as easy as it could have been for a cockpit of this size. The centre console is wide and there is little space between it and each of the pilot's seats - I made similar criticism of the Bombardier Dash 8 Q400 (Flight International, 25 April-1 May 2000) - so the vast console seems to be a design feature of Bombardier regional aircraft.
This feature's benefit becomes clear once comfortably in the seat, however: it brings together a large number of the flight management controllers between the pilots where they come readily to hand and can be easily seen by both occupants. The seats themselves are comfortable, have five-point harnesses and are adjustable fore and aft, for height and recline. They also have fold-away adjustable arm rests. Rudder pedal reach can be altered via a small crank handle between the pilot's legs.
The cabin was untrimmed and included two flight test engineer stations, a water ballast tank for centre of gravity changes, and an airborne crew escape hatch in the forward floor. The cockpit had three small additional instrument panels for flight test purposes and a portable angle-of-attack (AoA) indicator.
The production cockpit has plentiful bag and document stowage outboard of each seat. The field of view is good, though impaired by one of the small instrument groups above the glare shield in this instance, and each pilot can just see the wingtip on his side, but only by pressing his head to the side window. The windows are curved and unopenable; the cockpit emergency escape route on the ground (for this and production aircraft) is via a cockpit roof hatch.
Chuck Ellis and his flight test colleagues took the flight profile I had suggested and turned it into a formal Bombardier flight test schedule with dedicated test points for longitudinal stability, directional damping and wind-up turns at altitude to assess the stick force per g characteristics and buffet boundaries. As a result, it is possible to gain an unusually broad impression of the CRJ700's handling characteristics at the end of a 2h 15min flight.
An early item on this flight test schedule is, not surprisingly, auxiliary power unit (APU) and engine start. The Honeywell RE220 APU is mounted in the tail cone, but because the cone also houses an anti-spin parachute in this particular airframe, the APU exhaust is directed out through the side of the cone instead of straight out of the rear.
The APU synoptic was selected on the right central cathode ray tube (CRT) display in order to monitor the APU start which, once initiated by a push button in the overhead panel, is automatic, and the subsequent flow of bleed air illustrated in the synoptic.
APU bleed air is then available for engine start. The CF34-8C1 engine is a high-bypass turbofan with dual-channel full-authority digital engine control (FADEC) for automatic engine start, fuel control and compressor airflow management but not, in this installation, auto-throttle. Engine start is initiated by pressing the START button in the roof panel and fuel is introduced during the start sequence by moving the thrust lever from SHUT OFF to IDLE. The start is thereafter monitored via the engine indicating and crew alerting system (EICAS) displayed on the left central CRT.
The engine indications fill the left half of this CRT and show N1 (% fan speed), ITT (inter-turbine temperature 0°C), N2 (% compressor speed), fuel flow, oil temperature and pressure, and fan vibration. The right half of the CRT shows crew alert messages, undercarriage (gear) position indicators, slats/flaps position and fuel.
Taxiing the CRJ700 is easy. Nosewheel steering commands from the cockpit are transmitted electronically to hydraulic actuators on the twin nose-wheel assembly. The rudder pedals will move the nosewheels up to 8° either side of centre, the tiller (on the captain's side only) up to 80° either side.
Steering by either means is smooth, progressive and predictable with good artificial feel. The steel disc wheel brakes are controlled via toe levers on the rudder pedals and are smooth and powerful. The parking brake is set by pressing the pedals and locking them by pulling up and twisting a T-handle on the captain's side of the centre console - the handle would have been less easy to use from the right hand seat.
There was a slightly ponderous first-flight-of-the-day check list to be completed before take-off but it does include several flight test requirements and should be briefer for line operations.
In this instance, the aircraft's take-off weight, including 6,800kg (15,000lb) of fuel, is 30,000kg (maximum all-up weight is 33,000kg). The centre of gravity is a little aft of the midpoint. Flap 20 was used. Runway 19 Right was in use at Wichita Mid Continent airport, which is about 1,300ft (400m) above mean sea level. The surface wind was light and the weather bright and sunny with a temperature of about 24°C (75°F).
Before departure, Ellis, who manages the flight from the right-hand seat, checked the arming of the anti-spin parachute, completed the take-off checks and briefed me on take-off procedure. Flexible power (Flex) - less than maximum fuel take-off power - is routinely used, when weather and runway conditions permit, to preserve engine life. To select Flex, an assumed ambient temperature, higher than the actual temperature, is entered via the performance menu page of the flight management system control display unit (FMS CDU). The FADEC then makes the Flex calculation and the setting appears in magenta on the N1 indicators. The thrust lever quadrant has what Bombardier calls "soft detents" and legends on the quadrant to determine thrust lever angles for IDLE, CLIMB, take-off and go-around (TOGA) and MAX POWER.
Blue legends on the EICAS screen also show the thrust lever positions. This approach worked well: the detents are sufficiently positive to be felt easily by the pilot, but also easily allow lever movement from the detent, and confirmation of the lever positions could be seen by glancing at the quadrant or the N1 indication.
When the thrust levers are moved to TOGA the appropriate power can be selected (full take-off or Flex) by the FADEC. Flex take-off N1 cannot be less than the climb N1 for the ambient temperature, so that retarding the levers from TOGA to CLIMB cannot cause a paradoxical increase in N1.
For take-off, Ellis sets TOGA and checks the Flex N1. As the engines stabilise, I keep the aircraft straight on the runway using rudder- pedal steering, with my right hand on the thrust levers until V1 (decision speed), and then rotate at VR.
The engines take three seconds to spool-up, after which the surge of thrust accelerates the aircraft impressively. The elevators are powerful at rotation and the control force light. Retraction of the undercarriage, flap and slats causes no discernible pitch trim change.
The relevant take-off speeds are V1 122kt (226km/h), VR 122kt, and V2 (safety speed) 132kt. If an engine fails after V1, the FADEC increases power on the live engine to APR (maximum). The APU is kept running throughout the flight, having provided bleed air for the environmental conditioning system during take-off. Transfer from APU bleeds to engine bleeds causes no pressurisation surge.
The CRJ700 has conventional control wheels and rudder pedals connected by cables and push rods to hydraulic power control units (PCUs) which move the ailerons, elevators and rudder. There is a total of eight spoilers (four per wing). The inboard four are ground spoilers and the outboard four provide additional roll control and airbrakes (speed brakes).
The spoilers are operated hydraulically but electrically controlled. Pitch trim is provided by varying the angle of incidence of the tailplane (horizontal stabiliser). All systems are double or triple redundant and provision is made to counter jammed surfaces. There is a flight control EICAS synoptic page for the primary controls.
The ailerons and rudder are particularly effective. The ailerons are sufficient to provide roll control during routine phases of flight. The only time I saw the spoilers deploy (on the synoptic display) was during a roll acceleration and rate of roll test point, when 45° of left bank to 45° of right bank take a respectably short 2s - their deployment is seamlessly blended with the ailerons.
The rudder is exceptionally powerful and has a travel limiter to prevent overstressing the fin. A rudder doublet (a right and left rudder application) at 250kt shows the CRJ700 to be lightly damped naturally, requiring five cycles to recover from the induced dutch roll, while, with the two yaw dampers switched on, the recovery is in half a cycle. The ailerons and rudder are well harmonised but the elevators are heavier, a trait similar to the Q400.
The elevator break-out forces are high (the pilot having to pull or push the control column relatively hard to initiate control surface deflection), although the stick force per g (the increasing force the pilot has to apply in a tightening constant speed turn) is as expected.
I found the CRJ700 initially difficult to fly and trim precisely to maintain a speed or altitude, although the task becomes easier with practice. Despite this minor criticism, the CRJ has excellent handling qualities for a 70-seat airliner and is more like a 20-seat business jet in this respect. The CRJ700 also has good longitudinal stability.
The handling at 35,000ft is equally good because of the Mach trim's effectiveness in countering a natural nose-down pitch tendency above Mach 0.4 (Mach tuck). The aircraft is accelerated in level flight to Mach 0.83, and the tailplane incidence is progressively and smoothly altered automatically by the Mach trim system which uses air data computer information to meet its task. The Mach trim is selected via a switch light on the centre console and its operation is shown on the STAB trim display.
During the descent from 35,000ft to 25,000ft the thrust levers were set to flight idle and the airspeed stabilised at 250kt. The spoilers were deployed fully, provoking only a momentary, slight pitch change and light aerodynamic burble. Their use increases the rate of descent by about 2,000ft/min (10m/s). The spoilers can be deployed incrementally if required.
The CRJ700 is impressive during configuration changes. The extension and retraction of flaps just below the relevant speed limiting prompts little pitch change.
The tailplane auto-trim responds quickly and well, making flap travel almost transparent to the pilot. At 15,000ft, with the aircraft weight at 28t, two stalls - the one wings-level, the other with power off - were flown to the point of stick shaker onset. When clean, the shaker activates at 146kt and an AoA of 15°. With flap and undercarriage set for landing, stick shake occurs at 109kt and 22° AoA. There is also some aerodynamic buffeting just before stick shake in this configuration.
Roll control remains good throughout with no tendency to drop a wing. The CRJ700 is reassuringly safe to fly near the stall and has good natural and artificial stall warning.
Unusally powerful rudder
Next, still at 15,000ft and 200kt, the right engine was shut down. Speed was reduced to V2 for take-off flap and undercarriage up. Here the unusually powerful rudder is evident. The rudder limiter is automatically de-activated when there is only one engine operating. Although 15,000ft is too high for assessing asymmetric handling, the rudder authority is so great that there is no doubt as to the ease with which the CRJ700 could be kept straight, wings level, if an engine should fail on take-off.
I find the foot force difficult to trim out because it is light and the rudder trimmer powerful, but the aircraft can still be flown easily with the pilot sustaining the foot load.
We flew briefly a coupled approach to the instrument landing system (ILS) at Wichita Mid Continent to assess the electronic flight instrument system and the autopilot.
The Pro Line W is well known and widely used. In the CRJ700 it has the standard six displays: two for each pilot, the primary flight display and the multifunction display, with the two central CRTs are for EICAS.
It is an excellent system. Later versions such as the large-format (200mm x 250mm) flat-panel liquid crystal display Pro Line 21 are better, however. The larger format allows the primary and multifunction displays to be combined, vertically, on one screen and presentation is less cluttered and crisper. There is often insufficient room to incorporate the larger panels in existing cockpits and there is also the need to maintain commonality with earlier models.
The CRJ700's EICAS, FMS and autopilot perform well. Airspeed is controlled by manual use of the thrust levers and rudder trim has to be adjusted by the pilot to compensate for power and speed changes. Altitude, heading, localiser and glide slope capture are smooth and accurate. Three roller (touch and go) and three full-stop landings were made.
As easy to land as to fly
The CRJ is easy to fly manually on the approach, into the flare and onto the runway, though the main undercarriage is uncommonly stiff and unforgiving. Even good touchdowns are greeted with a minor thump. Ellis says this is more noticeable in the cockpit than in the passenger cabin because of the gear's geometry.
Doubtless, when landing performance is not critical, power can be retained until after touchdown, and an acquired technique would help. The main undercarriage legs are, incidentally, longer than the CRJ200, while the nose leg remains shorter, so that the threshold of the forward cabin door remains below the regulation height requiring an escape slide. Consequently the CRJ700 has a rakish nose-down, wedge-shaped profile on the ground.
A feature of the CF34-8C1 engine is control of engine idle speed by the FADEC. On the approach, the idle speed (defined as a corrected N2 core speed) is kept high to allow the engine to accelerate to "go-around" thrust in an acceptable time. So the CRJ700 is limited to approach slopes of about 3.5°, and cannot cope with steep approaches such as at London City airport.
Despite this, the CRJ700 seems set to enjoy a similar success to the CRJ200. A bonus for airline crews will be the aircraft's handling qualities, which are head and shoulders above many earlier generations of regional airliners.