In the 1960s, established manufacturers Boeing and Douglas dominated the lower end of the single-aisle mainline commercial aircraft market. Their original offerings – the 737-100 and DC-9-10 – seated fewer than 100 passengers. With the fielding of the 737-200 and DC-9-40, capacity increased to more than 120 passengers. This trend would continue, with newer variants offering increased passenger loads.
When Airbus joined the single-aisle fray with the A320 in the late 1980s, its typical two-class seating capacity was around 150. This upward trend left a void for regional jet manufacturers to field aircraft that seated approximately 50 passengers for short, thin routes. As with mainline manufacturers, regional aircraft passenger capacity increased over time. The upper limit for regional jets in the US market, however, was primarily set by pilot union scope clauses, which had prevented feeder carriers from operating aircraft with more than 100 seats.
In the 1990s, Bombardier sought to expand into the over 100-seat market, essentially abandoned by the mainline OEMs, with its BRJ-X programme. After much study, Bombardier cancelled the clean-sheet BRJ-X, and elected to stretch its successful CRJ100/200 50-seat model. Its larger CRJ700/900/1000 variants were fielded with maximum seating in the 70, 90 and 100 range, respectively. In 2005, the Canadian airframer reopened the books on a clean-sheet 100-plus seater, but this effort was cut short when it was put on hiatus in 2006. The programme was restarted the following year, with its prospects buoyed by Pratt & Whitney’s launch of the geared turbofan family of engines.
The viability of what was now Bombardier’s third effort to field a 100-plus seater hinged on lowering operating costs. While incremental improvements would be realised by refining aerodynamics and reducing structural weight, they alone could not justify the launch of a clean-sheet design. P&W’s PurePower PW1000G family promised fuel savings to the order of 15%: enough of a leap to justify relaunching the 100-plus-seater programme in July 2008.
Airbus and Boeing had offerings in this end of the market, but both the A318 and 737-600 were shrinks of larger models optimised to carry around 150-160 passengers. While they were typically offered at reduced prices, their direct operating costs were not proportionality lower than their reduced seat counts.
To tackle this segment, Bombardier’s CSeries has two variants: the CS100 and CS300. Both aircraft feature 99% parts commonality with a common cockpit and type rating. A 3.7m (12.1ft) stretch of the CS100 allows the CS300 to seat up to 160 people in a high-density layout: 25 more than its smaller sibling. The aircraft also share a common 112.3m2-area wing, with the CS300’s strengthened to carry the heavier load.
The CSeries aircraft were designed from the start and optimised for the 100- to 150-seat segment. The most prominent advance is the installation of PW1500G turbofans. The two-shaft PurePower engines have a large diameter fan that, through the use of a planetary gear drive, turns at one-third the rate of the low-pressure turbine. This scheme allows the fan and core to turn at more optimal speeds.
Aerodynamic refinements are highlighted by an efficient wing with integral winglets. Advanced materials play a role in reducing structural weight. By weight, 70% of the aircraft is constructed with advanced materials, including advanced composites and aluminum-lithium alloys. A three-axis fly-by-wire flight control system – a first for Bombardier – allows for further structural weight savings.
The interior of the aircraft has benefits for both the paying passengers and the crew. The spacious cabin has two- plus three-abreast economy seating, with the large diameter fuselage allowing for 47cm-wide seats. In a further nod to passenger comfort, the single column of centre seats are wider still, at 48cm. Large cabin windows help give an airy feel: at 27.8cm x 40.6cm, they are 26% and 50% larger than a 737’s and A320’s respectively, according to Bombardier. Finally, the overhead bins are large enough for every passenger to stow a roll-aboard bag. Equally as impressive is the flightdeck.
Just back from conducting route proving runs for launch CS300 operator Air Baltic, Bombardier offered FlightGlobal the chance to fly its second example, and see first-hand how the model compares with other 130-seat narrowbodies. I accompanied Andy Litavniks, Bombardier’s chief CS300 test pilot, on the pre-flight walk-around inspection.
As we approached the CS300, it reminded me of the 757, with its long fuselage, high landing gear and big engines. The inspection was straightforward, with nothing out of the ordinary for a transport aircraft. What was notable though was the quietness of the tail-mounted Honeywell 131-9C auxiliary power unit.
Entry into the aircraft was via the 81cm-wide, 1.88m-tall forward entry door. The flightdeck is the cleanest and most uncluttered I have seen on an airliner. The sizable cockpit has four fixed windows that offer an outstanding field of view. Sidestick controllers free up space and allow for the installation of slide-out tables, similar to those on an Airbus. A Rockwell Collins ProLine Fusion flight management system (FMS) is the star of the flightdeck. Five large, 15.1in LCDs offer plenty of real estate to display flight critical information. The overhead panel houses systems controls, which are logically arranged.
After strapping into the left seat I used the pillar-mounted eye alignment balls to set the proper seating height. The seats each have adjustable outboard armrests, similar to those on an Airbus. Centre pedestal-mounted thrust levers (TL) fell readily to hand. Outboard of the throttle quadrant are two multifunction keyboard panels and two rollerball cursor control devices (CCD).
Litavniks guided me through FMS initialisation. I found the process to be fairly intuitive, sequentially stepping through the appropriate pages. Pre-start flows consisted mainly of setting almost all overhead panel switches to their ON or AUTO positions. The full authority digital engine control (FADEC)-managed start process was initiated by placing each engine’s START SW to RUN. All I had to do was monitor each start. Total time from start to IDLE for both engines was 2min 50s. For the 737NG I normally fly, total start time is typically about 1min quicker. The longer start time for the CSeries is driven by rotor bowing and the need to motor the engines to stabilise temperatures. After landing we would have to operate each engine at 75% N1 or below for 3min to again thermally stabilise prior to shutdown. A major US airline requires only a 1min cooldown for the 737’s CFM International CFM56-7B engines. At most airports the extra cooldown time for the PurePower engines should not be an issue, but for small ones – Burbank, California, for example – it may slow down the pace of operations.
After Litavniks set the flaps to position 4 for our take-off, I released the parking brake. A slight bump of the TLs was all that was needed to start the CS300 rolling. En route to runway 19R for take-off, I found the tiller-controlled, fly-by-wire nose wheel steering allowed me to deftly negotiate the narrow taxiways at Bombardier’s facility at Wichita’s Mid-Continent airport in Kansas. While it was a short taxi, I did track our position on the multifunction display’s map view, which when scaled down to a mile or less offered a detailed airport diagram. The “Before Takeoff” checklist was accomplished using the standard electronic checklist: a smart one that automatically checks off sensed items, such as certain switch positions.
As we turned onto the runway I pushed the TL-mounted TOGA switch. Once cleared by the air traffic control tower I advanced the TLs, the auto-throttle (AT) engaging as N1 passed about 50%. Litavniks called “rotate” at 103KIAS, and I pulled aft on the sidestick. Stick forces were light as the nose crisply tracked up to 15º. Gear and flap retraction caused little noticeable change in stick forces as we accelerated to the initial 250KIAS climb speed. Passing 400ft I engaged the autopilot in the flight level change and heading modes.
During our climb to 28,000ft (FL280), Litavniks showed me some of the ProLine Fusion system’s capabilities. My primary flight display and horizontal situation display filled the entire left-hand outboard display unit. The large size made it very easy to read, even for ageing eyes. The inboard, left-hand display unit was configured as a map on one half and engine indicating and crew alerting system on the other. The FMS has a graphical interface capability. Using my CCD I could easily and intuitively change our routing, by going direct to a waypoint, for instance. One neat feature was the ability to display both terrain and radar information on the map at the same time: something a 737NG cannot do.
During the ascent to FL280 the AT maintained the Climb N1 power setting. The CSeries smoothly levelled off at FL280 just 11min after brake release. Once level, Mach 0.74 was set in the FMS to establish a long-range cruise condition. While well below optimum altitude, total fuel flow was 4,600lb/h, with a true airspeed of 447kt (830km/h). Next, I manually set a speed of M0.78 for a high-speed cruise point. True airspeed increased to 469kt, with a total fuel flow of 5,240lb/h. The CSeries' ability to cruise at M0.78 will allow it to seamlessly integrate with other mainline aircraft.
After our brief stay at the somewhat low cruise altitude, Litavniks suggested we exercise the CSeries' emergency descent mode (EDM) to descend to medium altitude for hands-on manoeuvring. EDM is an autopilot (AP) mode designed to expeditiously descend the aircraft to 15,000ft above mean sea level (MSL) in the event of a cabin depressurisation. It is automatically activated if cabin altitude exceeds 14,500ft or when the guarded EDM switch is pressed with the aircraft above 25,000ft. The AP and AT will engage, if not already, with the AP maintaining current heading and starting a near-maximum operating and mach operating speed (VMO/MMO) descent to 15,000ft. The TLs retard to IDLE to quicken the aircraft’s descent to a safer altitude. Once cleared by air traffic control to descend, I pushed the EDM button and deployed the speed brakes. During the emergency descent I clicked off the AP and hand-flew a portion of the descent. I found the CSeries was very stable at MMO and responsive to small control inputs. In less than 2min we were level at 15,000ft.
Before I discuss my in-flight observations, it is necessary to delve a little deeper into fly-by-wire controls in general and the CSeries’ in particular. Developing such a control system is no easy task, with each manufacturer having its own approach to flight control laws. From my perspective there are four areas where major differences are seen across the industry: pitch axis control scheme, lateral-directional coupling, envelope protections and engine out protections. Some readers may point out that I have neglected to mention whether a control yoke or sidestick is used, or if the rudder pedals are back-driven. While it is beyond the scope of this article to cover these topics in depth, the following comes to mind. The sidestick has effectively replaced the control yoke, and I would be surprised to see any future fly-by-wire aircraft employing a legacy yoke. One promising area with sidestick controllers is making them “active”.
An active sidestick can provide better artificial feel, as well as increasing crew situational awareness, by interconnecting their motion. Back-driven rudder pedals – those that displace to reflect actual rudder position – provide extra tactile feedback to the pilot, which also increases situational awareness. While non-back-driven pedals are safe and effective, as with AT, this pilot would prefer all control inceptors to be back driven.
In pitch the CSeries (while up and away) uses a C*U control scheme. The “C*” component is a blended pitch rate and g command. This is the baseline scheme for Airbus aircraft. The “U” is apparent speed stability. Boeing aircraft with fly-by-wire controls have apparent speed stability, much like a conventional aircraft. Interestingly, Embraer has chosen a pure g command system for up and away flight, while adding speed stability when in a landing configuration. In roll axis, roll-rate schemes would seem to be the industry standard when up and away. On the ground, most roll schemes are simply stated as “stick to surface”. Bombardier has adopted this methodology for the CSeries. Up and away, full sidestick deflection commands a 20º/s roll-rate, with commanded rate decreasing to 10º/sec at maximum angle of attack.
In lateral-directional coupling, the CS300’s roll and yaw axes are coupled. Pedal input displaces the rudder with the corresponding side-slip creating a roll response. Simply stated, you can bank the CSeries with rudder pedals alone, just like with a conventional control system. Not all fly-by-wire schemes do this, and this pilot prefers old school to the decoupling of the lateral and directional axes.
In terms of flight control law protections and limitations, the CSeries’ philosophy is somewhere in-between the positions of other manufacturers. In pitch, attitude is limited to a maximum of 30º nose up and 20º nose down. Additionally, Bombardier has developed a novel sidestick that has soft and hard stops. About 6.8kg (15lb) of force will displace the sidestick to the soft limit (15º displacement), where an additional step input of 4.5kg is needed to move past the soft limit. An additional 4.5kg of force (for a total of 16kg) is required to reach the hard limit. Pitch soft and hard stops limit both angle of attack and load factor (g). At the soft angle of attack limit the stick shaker activates and “speed” annunciates, with “stall” annunciated at the hard stop. Pulling or pushing to the soft stop will not over-g the aircraft, but pulling or pushing through it in perhaps what is a last-ditch manoeuvre could well do so. Limited over-speed protection is provided (gear and flaps up) by the application of nose-up bias to the horizontal stabiliser trim. Finally, on take-off a pitch damping function is available to counter overly aggressive rotation rates and lessen the chances of a tail strike.
In addition to the roll rate limitations, absolute angles of bank are also limited to 80º. At bank angles of up to 30º the CSeries exhibits neutral spiral stability. These banks, once established, are held with neutral lateral sidestick input. At bank angles greater than 30º, the aircraft exhibits positive spiral stability. Release of the sidestick in this region causes the aircraft to roll to and maintain a 30º angle of bank. Additionally, at bank angles up to 33º, automatic pitch trim maintains current pitch attitude, negating the need for aft sidestick pressure to keep the nose from dropping.
In yaw, rudder displacement is limited as a function of airspeed and configuration. Rudder pedal travel, however, is not affected. The full 18cm (+/-9cm) of travel is always maintained. Finally, partial engine out yaw compensation is provided; a feature that will be discussed more in depth later.
Theory aside, once we were level in the medium altitude environment, I had the opportunity to explore how the CSeries handles in the heart and at some of the extremes of its flight envelope. At speeds ranging from 250KIAS to 180KIAS, I did some bank-to-bank turns. I found aircraft response to lateral sidestick inputs was linear and allowed for accurate capture of desired bank angles. When established in a 20º bank, I released the sidestick and, as advertised, the CSeries maintained the bank and prior pitch attitude. I increased the bank to 45º and again released the sidestick. This time the aircraft rolled to and stabilised in a 30º bank, with the nose dropping.
Once rolled out with the wings level, I retarded the TLs to set up for a clean configuration slow speed investigation. During the deceleration I used the sidestick-mounted pitch trim switch to keep the aircraft in trim. As mentioned above, the CSeries has apparent speed stability in the pitch axis. The pitch trim sets the trim reference speed in the CSeries, just like in the C*U logic of the 777 and 787, and sets the horizontal stabiliser to maintain the reference airspeed. Unlike the Boeing approach, where there is no display of the trim reference speed, the CSeries places a trim speed bug, a cyan triangle, on the airspeed tape. With the sidestick at the soft stop the stick shaker actuated as airspeed stabilised at 141KIAS. Ignoring the “stall” aural I pulled through to the hard stop, the aircraft slowing to 136KIAS. At this slow speed I did several 30º-40º bank turns, with the aircraft smoothly rolling without any noticeable buffet. After levelling the wings I released sidestick back pressure and advanced the TLs to recover to normal flight conditions. Next, the gear was extended and flaps set to 4 in preparation for an aggressive landing configuration pitch manoeuvre. Slowing through about 120KIAS I snatched the sidestick full aft to the hard stop. Initially the CSeries shuddered as the pitch rapidly tracked up to and held 20º nose high. Airspeed stabilised at 101KIAS, where, with continued full aft sidestick, I rolled the aircraft to 20º left and right bank. I was impressed by the aircraft’s crisp roll response at such an aggravated condition.
Satisfied by the fly-by-wire protections offered by the CSeries at slow speed, the aircraft was cleaned up and turned towards the Wichita airport for some pattern work. En route at 270KIAS I mis-trimmed the aircraft, setting the trim speed bug to 300KIAS. Less than 3kg of forward force on the sidestick was needed to maintain the off-speed condition. Subjectively I found this force to be fairly light, but enough to alert the pilot to the mis-trimmed condition. Without retrimming, slowing to 250KIAS required very little aft sidestick pressure to maintain the off-speed condition. Pitch forces were trimmed off, and I did several rudder inputs to see if the lateral-directional axes were coupled. As was expected for this fly-by-wire control scheme the wing dropped with the pedal input, allowing me to change heading with my feet alone.
With the area manoeuvres complete we turned towards the Wichita airport for our recovery and some pattern work. I engaged the AP and AT while I loaded the area navigation Y approach to runway 19R in the FMS. Air traffic control provided vectors then cleared us direct to the CADAC waypoint; the closest initial approach fix. As with other ProLine Fusion systems, I found that the presentation of our own ship’s position on the actual approach chart greatly enhanced my situational awareness. Approaching CADAC I clicked off both the AP and AT to hand-fly the approach. The flight director guidance cue in the primary flight display was neither a split cue nor V-bar. It was a winged box with a curved “roof” that looked like it would have been more at home in a head-up display (HUD) than a primary flight display. HUDs are optional, and unfortunately our aircraft did not have them. That being said, the cue provided great lateral and vertical guidance.
As we slowed and descended on the approach, Litavniks configured the aircraft for our full stop landing. Flaps were set to 5 – their highest setting. Pitch trim changes while slowing and configuring were small and easily countered by pitch trim. I tracked a target and reference speed (Vtgt/Vref) of 112KIAS on final, the FADEC engines being very responsive to TL changes. As the CS300 is close in size to the 737-700, I flared about when I normally would passing about 30ft and retarded the TLs to IDLE. At our light weight, we floated for a bit in a 7.5º nose-up attitude before touchdown. After lowering the nose to the runway I applied moderate toe braking. The CSeries’ reverse thrust actuators are quite different than the 737’s, and by the time I had them unstowed it was the toe brakes alone that rapidly slowed the aircraft to a stop.
My final circuit in the CSeries would simulate an engine failure followed by a one-engine inoperative approach. The CSeries does have an “in-flight partial engine out compensation” feature. Unlike the 777’s and 787’s engine failure response logic, which allows an engine failure to be flown with feet on the floor, it is only designed to reduce the aircraft’s yaw and roll response to an engine failure.
After taxiing back to the approach end of runway 19R the flaps were set to 2. As there was a 1,000ft overcast layer, Bombardier safety protocol prohibited “failing” an engine below the low ceiling. Litavniks suggest we could still do the manoeuvre once in the clear above the clouds. After lifting off, we climbed out at V2 plus 10, or 132KIAS. At our light weight (42,526kg) the pitch attitude was 22º nose high. Breaking out of the clouds at 3,000ft MSL Litavniks rapidly retarded the right TL to IDLE. As the nose started yawing to the right, a little less than half of the available rudder and less than 15kg of pedal pressure was needed to maintain runway heading. Sensing the engine failure, the FMS displayed a Beta Index below the flight path marker in the primary flight display. While the Slip/Skid Indicator was still at the top of the primary flight display, following the Beta Index allowed some sideslip and improved climb performance on one engine. Although I would prefer an absolutely no-drama scheme like that found on the 777 and 787, the CSeries fly-by-wire scheme made the V2 cut manoeuvre a fairly benign event.
ATC provided vectors for an instrument landing system approach to runway 19R. On downwind I centred the rudder trim and used the rudder pedals to maintain co-ordinated flight. On final with flaps set to 4 and a Vtgt/Vref of 119KIAS, only about 10% of the available rudder was needed to keep things coordinated. After touchdown, thrust reverse on both engines and light toe braking was used to slow for runway turnoff. Taxi back to Bombardier’s ramp was uneventful, with post shutdown flows easily accomplished.
TRUE MAINLINE AIRCRAFT
After I flew the CRJ900 in 2002, I wrote: “While the CRJ900 can trace its lineage back to the pioneering CRJ100 regional jet, it felt every bit like a full-size airliner.”
From the pilot’s perspective that statement holds true, but perhaps less so for the passengers in the narrow tube. The CSeries is a clean-sheet design, which offers class-leading passenger space and comfort with industry-leading operational economics. The fly-by-wire flight control system was notable in that it was not really noticed during routine operations. Only at the extremes of the envelope were the various protections noticed and appreciated.
In my role as a flight test pilot for FlightGlobal I have had the opportunity to sample and report on a large number of different aircraft. For some of these – the Bombardier CL-415 water bomber, for instance – my points of reference were limited. The CS100 and CS300, however, were designed to fulfil a role that I am intimately familiar with. Having flown various models of the 737 over a span of 20 years and for nearly 10,000h, I can comfortably say that as a pilot I would rather fly the CS100 and CS300.
The CSeries has a larger and quieter flightdeck, along with fly-by-wire flight controls that allow for more precise control while mitigating flight envelope excursion risks. The only measurable way the CS100 or CS300 could be better is if Bombardier had forgone stretching its CRJ line and fielded the new product in the early 2000s.
Source: Flight International