Test pilot Michael Gerzanics put the A321neo through its paces for FlightGlobal, and here delivers his verdict
When Airbus launched the original A320 in the 1980s, it faced entrenched competition in the narrowbody segment from Boeing and McDonnell Douglas. To answer and perhaps better these rivals, Airbus chose to offer an aircraft with a larger-diameter fuselage and advanced fly-by-wire (FBW) technology. Since then, Airbus has stretched (A321), shrunk (A319) and re-shrunk (A318) its innovative design. And, with the exception of the A318, these models have enjoyed huge success in the market.
Since the A320's entry into commercial service with Air France in 1988, Airbus has made continual refinements to the series. After initially being delivered with CFM56-5 engines, Airbus quickly introduced a second engine option – the International Aero Engines V2500 – and production of the baseline A320 family continues with the latest versions of both those engines. The stretched A321 arrived in 1994 and Airbus incorporated double-slotted flaps to help low-speed performance of this larger variant. The first major aerodynamic change across the family came in 2012 when the small wing-tip fences were replaced with a 2.4m high blended winglet, dubbed "sharklet", promising fuel savings of as much as 4% on longer routes.
Of late, the narrowbody market has evolved, with carriers desiring to offer service on so-called long, thin routes. Narrowbody transatlantic and Hawaii services from the continental USA are now a reality.
The sharklets were a good first step, but further extension of the A321's range would require a more involved and innovative solution. An aircraft's wing loading and thrust-to-weight ratio are two critical parameters in determining its capabilities. The A321's wing area, while increased 2.15m2 by the sharklets, was in effect the limit on available gross weight. How else could the range be increased? While there was no margin for added fuel uplift, a radical increase in fuel efficiency could reduce the total fuel required.
Both CFM and Pratt & Whitney were developing engines that promised a 15% reduction in fuel consumption, enough to give the A321 the added range it needed. Versions of both those engines are applied to the A319, A320 and A321 variants, with the new iterations suffixed "neo" (new engine option).
The way that CFM and P&W have attained the new efficiency is markedly different. CFM chose a path of incremental improvements and refinements to the CFM56 to develop its Leap turbofan, designated the Leap-1A for the Airbus single-aisles.
P&W choose to develop a whole new series of geared turbofan (GTF) engines, dubbed the Pure Power PW1000G family, specifically the PW1100G-JM for the A320neo models. As well as offering significantly increased efficiency, both engines have about 1,000lb (4.5kN) more thrust than their predecessors.
More efficient engines and sharklets gained the range desired by operators, but the circle had yet to be completely squared. Airbus also wanted its more capable offering to have the same or better runway performance as its predecessor. The new engines and associated airframe modifications added 1.8t to empty weight, forcing Airbus to the drawing table for a solution.
The A320neo family's sharklets did have a positive effect on take-off performance as they effectively increased the wing's aspect ratio, which directly improved second-segment climb performance. The ground-roll performance for the A321, at all except low-gross-weight conditions, is limited by its layout. Its main gear height and aft fuselage length limits the attainable pitch attitude for take-off. In general, the slower the lift-off speed (Vlof), the shorter the take-off roll. Correspondingly, lowering Vlof lowers the rotation speed Vr.
One critical parameter for determining Vr in a geometry-limited aircraft like the A321 is its minimum unstick speed (Vmu). Vmu is the slowest speed the aircraft can lift off the runway and fly away from ground effect. In short, lowering Vmu lowers VLof and Vr, yielding a shorter take-off roll.
Further complicating Airbus's task were new requirements levied by certification authorities. Certification speeds would be determined while accounting for average pilot skills, not those of steely-eyed test pilots. To this end, Airbus made two modifications to the pitch flight control axis for the take-off condition to lower the critical Vmu parameter.
Unlike Airbus's widebody aircraft, the A320 series has no specific rotation flight control scheme. In legacy A320s, elevator movement is proportional to stick displacement. To ease the task of attaining and maintaining the right pitch attitude, Airbus made pitch response to stick input a rate-command one for take-off. The new scheme should allow pilots to hit the desired 3°/sec rotation rate and easily capture the target pitch attitude.
A321s, whether the current engine option (Ceo) or Neo version, have some inherent tail-strike protection due to ground effect as the tailplane and elevator approach the runway surface. As in legacy aircraft, tail-strike avoidance cues are also provided in the Neo's primary flight display.
To further enhance safety, Airbus also added a tail-strike protection feature to the A321neo flight control system (FCS) rotation law. Coined an "electronic tail bumper", the FCS system actively prevents a tail-strike as long as the stick is displaced less than three-quarters of the way aft. While full aft stick deflection can still cause a tail-strike, the modifications to the FCS should make real world take-off performance more closely match certification data.
At first blush it might seem the flight-test programme for the Neo models could be fairly cursory, as they share 95% parts commonality with their predecessor models. But this was not the case, because the addition of the new heavier, higher-thrust engines necessitated completion of 75% of the test points a new design would have required.
Airbus has flown around 4,000h in its certification effort for both powerplant versions of the A320neo and A321neo. Testing of the Leap-powered A319 was under way when FlightGlobal was invited to fly the Pratt-powered A321neo.
Late last year, I was fortunate to fly Bombardier's CS300, also equipped with Pratt & Whitney's GTF. Needless to say, I was quite keen to see how the A321neo compared and fared in the skies over southern France.
Our preview aircraft – Airbus's Pratt & Whitney-powered A321neo test aircraft (D-AVXA) – was parked at the manufacturer's flight-test centre at Toulouse-Blagnac airport. Before boarding, I did a quick walk around and found the GTF's large-diameter fan most impressive, with the A321's tall main landing gear allowing ample ground clearance for the 2.23m cowling.
The cabin was configured for flight test, water barrels to serve as ballast and centre-of-gravity control installed. Midway in the cabin was a two-place flight-test engineer station. On the flightdeck, Etienne Miche de Malleray – Airbus experimental test pilot and head of flight test product design – had already programmed the flight management system.
As I strapped into the left seat I found the cockpit to be familiar. One of Airbus's strengths is that its commercial flightdecks have as much in common with each other as possible. With the exception of LCD displays in place of CRTs and an optional HUD, the Neo's flightdeck was essentially unchanged from the A318 I had flown 14 years earlier.
After completion of the pre-start flows, I noted a “COOLING” time of approximately 2min 40s displayed on the engine-indicating and crew-alerting system display for each engine. Each would have to be dry-motored for that span before fuel could be applied and the start sequence continued.
Much has been written about the long start-cycle times of P&W's GTF engines. Longer start times not only affect the starting aircraft; at crowded airports they can delay other aircraft trying to push back or park at their gates.
On our flight, the start sequence took just over 7.5min. It should be noted that these were developmental engines and not production representative. Airbus and P&W have made efforts to reduce total time to operationally acceptable levels.
While not installed on our aircraft, Airbus has developed a "dual motor" feature for the start sequence. When the first engine is dry-motored, the second engine will also motor and work off its cooling time.
For its part, P&W has added a cubic boron nitride coating to the tips of the engine's 11 integrally bladed rotors. The coating creates a better seal and reduces cooling time by about 1min. These efforts are welcome, and according to P&W start times for production engines will be on par with those of the V2500-powered A321s.
INCREASED RUDDER AUTHORITY
Prior to leaving the chocks, Miche de Malleray set "flaps 2". During taxi to runway 32L for take-off I reacquainted myself with the A321's thrust reversers, handy for keeping a slow taxi speed in wet conditions on our light 74,400kg aircraft (18,500kg of fuel, 14 occupants and 5,000kg of flight-test kit).
Once cleared by ATC for take-off, I advanced the thrust levers to the FLX/MCT detent. Power on both engines ramped up symmetrically and quickly, allowing me to keep the Neo dead on centreline. The 1,000lb of additional thrust offered by the new engines, when combined with slower rotation and lift-off speeds, required more rudder authority. To this end, Airbus increased the rudder's maximum deflection from 25° to 30°, with pedal range of motion unchanged.
Once airborne the gear and flaps were retracted for transit to our working area west of Toulouse. During the climb to an intermediate altitude I hand-flew the Neo to re-familiarise myself with the FBW aircraft's handling qualities. Once in the working area I did a series of high angle-of-bank steep turns, up to the normal envelope limit of 67°. I appreciated the inherent turn co-ordination of the flight control system, as these manoeuvres were flown feet on the floor. Simply placing the primary flight display's flightpath vector on the horizon allowed me to maintain level flight, with the responsive engines manually modulated to maintain speed.
Flight envelope protections are unchanged, and to many are a strength of Airbus's commercial aircraft. With normal flight-control laws, three angle-of-attack values are computed. The highest, AlphaPROT, is shown on the primary flight display at the top of yellow hash marks to the right of the speed tape.
If AlphaPROT is reached, the autopilot will disconnect and speedbrakes retract, if extended. The stick stops commanding g and now commands angle-of-attack directly. Continued aft stick pressure is required to slow the aircraft further, just like a conventional flight-control system.
The next slower speed is AlphaFLOOR, which is not displayed to the pilot. Before reaching AlphaFLOOR, a low-energy warning is triggered, causing an audible "SPEED, SPEED, SPEED" to be sounded. At AlphaFLOOR, the autothrust system will engage and advance both engines to the take-off/go-around (TOGA) power setting regardless of the actual thrust lever position. In no case will the aircraft slow below AlphaMAX, a speed just above the actual stall angle-of-attack which is indicated by a red band on the speed tape.
Miche de Malleray next had me slow for a demonstration of these envelope protection features. The demonstrations were performed in two configurations: one clean, and the other with gear down and flaps "3". I slowed twice in each configuration. The first time, when AlphaFLOOR was reached, Miche de Malleray disconnected the autothrust, allowing me to stabilise at AlphaMAX.
In both configurations the aircraft was stable as a rock at AlphaMAX, with no wing rock or meandering in side slip. Roll control was still precise, with turns at 20° angle-of-bank turns to a heading easily accomplished.
For the second deceleration in each configuration the autothrust was allowed to engage when AlphaFLOOR was reached. In both instances, TOGA power was automatically commanded and the aircraft powered out of the slow-speed state.
To highlight the Neo's capabilities in an emergency situation, we simulated a close encounter with the ground. Level at FL140 while still configured with gear deployed and flaps "3", I used manual thrust to hold 150KIAS (knots, indicated airspeed) – equivalent to Vls+10 for our configuration and weight.
Miche de Malleray then bellowed "Terrain!" to simulate a GP&WS warning. With my hands off the thrust levers, I rapidly pulled full aft on stick. At AlphaFLOOR the autothrust engaged and selected TOGA thrust. The aircraft reached a nose-high attitude of over 20° as the airspeed stabilised at 140kt (260km/h).
We climbed away from the simulated terrain at a rate of 2,400ft/min, with no aircraft exceedances. The FBW control system made the escape manoeuvre a no-brainer, helping even the least proficient pilot avoid controlled flight into terrain.
For the last low-speed manoeuvre we added a lateral component to the above GP&WS escape exercise. At the same conditions as detailed above, Miche de Malleray directed me to roll the aircraft to the right to avoid rising terrain. While holding full aft pressure, I planted the stick in the aft right corner. The nose initially dropped a few degrees and the aircraft crisply rolled to 45° AoB. The nose tracked smoothly to the right as we calmly climbed away from the simulated terrain.
SIMULATED ENGINE FAILURE
While still configured with gear-down and flaps "3", I lowered the nose and recovered to level flight. Once in a level attitude, I advanced both thrust levers to TOGA to simulate a take-off. As has been pointed out in previous flight-test reports, a fly-by-wire Airbus's initial response to an engine failure is similar to an aircraft with conventional flight control: yawing and a wing drop. The design of the Airbus flight-control system allows for recognition of the failure, but limits the aircraft's dynamic response so that things don't quickly get out of hand.
Climbing in a 12° nose-high attitude at about 145KIAS, Miche de Malleray rapidly retarded the right thrust lever to idle to simulate an engine failure. Initially I let the aircraft respond to the asymmetric thrust condition, to observe its response. As expected, the nose yawed to the right with a corresponding wing drop of about 5°. The magnitude of the drop was on par with that I had experienced when I flew the A350 several years ago. This came somewhat as a surprise, as I had expected a bigger wing drop.
While developing the A320neo series, Airbus refined its flight control laws, endeavouring to make them handle more like their widebody stablemates in crosswind landings. They have a more wings-level attitude when they de-crab for touchdown. With landing flap settings (3 or 4), the A320neo's roll due to yaw (phi to beta ratio/dihedral effect) has been reduced by half compared with the A320ceo series.
Once stabilised in the TOGA power engine-out climb, less than 20kg of left-pedal input was needed to center the PFD's BETA target. To optimise climb performance some side slip is allowed, with the Neo not being in full co-ordinated flight. The low pedal forces and newly reduced apparent dihedral effect combined to make the Neo's response to an engine failure a somewhat docile event.
CRUISING BACK TO TOULOUSE
After completing the medium-altitude evaluations, another journalist pilot climbed the aircraft to FL330 for two brief cruise-performance points. The first was at Mach 0.76, simulating a long-range cruise condition. On the ISA -2°C day the 67t Neo trued out at 501kt with a total fuel flow of only 2,200kg/h. Next, Mach 0.80 was held to simulate a high-speed cruise condition. Fuel flow increased to 2,440kg/h as the Neo trued out at 527kt.
While by no means definitive, these snapshot figures show the Neo can maintain a respectable pace at altitude while gulping less fuel than its predecessor.
Midway through the descent back to Toulouse, I sampled how the A321neo handled in the traffic pattern. ATC provided radar vectors for an ILS approach to runway 32L. I hand-flew the approach to 200ft above ground level (AGL), where Miche de Malleray levelled the aircraft off to demonstrate the airborne part of the optional runway-end overrun warning/runway-overrun prevention system (ROW/ROPS).
ROW is active at 500ft AGL and aims to prevent long landings and subsequent runway departures by alerting the crew to the necessity of a go-around. Configured for landing, Miche de Malleray flew the A321neo at 140KIAS down the runway. The ROW uses real-time position and energy data (speed and altitude) to determine if a landing can be made in the available runway remaining. About a third of the way down 32L, the message "IF WET: RWY TOO SHORT" appeared on the primary flight display. A little further down the runway a repetitive aural, "RWY TOO SHORT", was sounded with the terse "RWY TOO SHORT" message displayed.
At the end of the runway, the gear was retracted and Miche de Malleray started a climbing right turn to the crosswind pattern leg. On downwind, I took control for a hand-flown visual approach to 32R with a sidestep to the land on 32L at 1,000ft AGL. The initial plan was modified when low clouds forced us to fly the ILS final segment to 32L.
With the gear down and flaps "4", I manually held an approach speed of 129KIAS. We broke out at 1,600ft AGL and visually acquired both 32L and 32R. I hastily aligned with 32R, about 200m to the southeast. I was well stabilised at 1,000ft AGL when Miche de Malleray directed a sidestep to 32L for our full-stop landing.
Stabilised on the ILS, Gerzanics flies the A321neo towards Toulouse's runway 32L
Whilst still in a 3° descent, I crisply rolled the aircraft to the left in a 25° bank to align with the left runway. At 500ft AGL, I was aligned with 32L as judged by the localiser and on the ILS glidepath. At 20ft radio-altimeter (RA) I started the round-out portion of the flare manoeuvre, retarding the thrust levers to idle at 10ft RA.
After a smooth touchdown, the aircraft was initially slowed by thrust reversers alone, light wheel-braking applied for runway turn-off at taxiway S10. Once clear of the runway I taxied to the hold point for RWY32L.
When Airbus offered FlightGlobal the chance to fly the A321neo, my first thought was: how could an engine swap on a proven airframe be that big a deal? Compared with the current engine option, the A321neo can carry the same passenger load, out of the same or shorter runways, 490nm (900km) farther. That's London to Berlin as the crow flies – a big deal indeed.
While both CFM and P&W are to be applauded for bringing dramatically more fuel efficient engines to the table, the success of the A321neo will not be a reflection of their efforts alone. Airbus engineers and flight-test personnel played critical roles in tweaking the A321's aerodynamics and flight controls to put these fuel-sipping engines to good use. With the A321neo I can wholeheartedly agree with the marketing hyperbole. It really is new and improved.