Simulators Are For Real

The advent of better visual simulation may well find the Australian simulators being used for daylight ‘flights’, but most Australian customers say they find the daylight visuals still too Disneyesque, too crude for sophisticated use.

Be this latter point as it may, the dusk/night picture on today’s simulators is very real, thanks to CGI (computer-generated imagery) systems, made possible in turn by micro-circuitry.

The entire concept of simulators has come a long way in 16 years; in those days, most simulators depended on a massive scale model of an airfield and its environs, a TV camera slaved to the simulator’s calculated position and attitude- and a ton of power to provide lighting and cooling.

Qantas had one of the earlier models- a 747 simulator using a Singer-Link simulator system wedded to a GP-4B computer, online as far back as 1970.

The system is now mothballed and up for sale, according to Qantas. And there is apparently a ready market for even old Singer-Link systems, with Third World customers among the leading potential buyers.

Another 747 simulator which has served Qantas faithfully since it came online in 1975 is still in operational use, working up to 16 hours each day. This one has a Singer-Link system wired to a PDP11/45 computer. Both old stagers use Vital visuals and stand on six-axis legs.

Today’s Qantas simulators are just about the last word in dusk/night visuals and systems. The airline is taking delivery of a WADS (Wide Angle Display System) this year, enhancing an already excellent five-panel view from the cockpit.

A WADS installation will improve out-the-window visibility by up to 30 per cent, say Qantas officials.

The two new simulators and associated systems, custom tailored to Qantas requirements by Singer-Link, are already proving their worth. Airline spokesmen tell AUSTRALIAN AVIATION that real savings run in the percentage region of $500/hour for simulator usage compared with about $6000/hour for an actual 747 operation.

Up to 75 abnormal procedures can be carried out in the new simulators, and the airline’s 480 pilots and 170 Flight Engineers use the systems four times a year, giving an average utilisation of a very significant 5000 hours a year for each 747 simulator.

Air New Zealand and MAS crews rent time on the systems, helping to bring up the high annual utilisation total.

The motion system for Qantas’ new simulators gives a high realism to a ‘flight’; with ten tonnes of cockpit simulator sitting on those six legs, delicate movement in all six axes- roll, pitch, yaw, heave, surge and sway – is infinitely possible.

The visuals are faithful- and remarkable. With the Novoview SP1 using more than 6000 light points and 400 shaded surfaces to produce the dusk/night imagery, the outside picture is reproduced on five panels in the Qantas simulator.

During our evaluation of the new system, the fifth panel was switched from right to left window to give a better representation of the circuit we were flying.

With the WADS installed, Qantas simulator users will get an increased vertical field of view on each individual display monitor.

(In the conventional SP series, each display module consists of a display monitor, beam-splitter and collimating mirror. Wide angle fields of view up to 150 degrees are achieved using multiple displays. )

Qantas is determined to keep at the forefront of simulator technology and is holding out for the best before committing itself to advanced daylight visuals in the SP series, made by Rediffusion.

Meanwhile, it is getting the ultimate out of its dusk/night presentations by installing a $40,000 spectroradiometer to check colour faithfulness in its visuals.

The instructor’s station on board the Singer-Link installation is a lesson in itself, of advanced technology. A development in the MAGGS series, it features a microprocessor-based facility with VDUs giving a visual representation of up to 6000 data items in a library of the aircraft and environment parameters, as well as data inputs and outputs from the flight deck.

Slaved VDUs and controls are installed in a separate Control Room, which doubles as a ground station voice link.

An on-board graphic facility is installed in the Qantas simulators for aircraft track and approach plotting, and the instructor’s station generates all malfunction, initial condition and radio aid data- it can also provide custom-made lesson plans, and can feed in all sorts of conditions ranging from wind shear to final approach ‘waffle’.

Weather can be fed into the system which, combined with aural effects, gives reality too close for comfort in many climatic conditions!

Lightning flashes are seen and the sound synchronised in the pilots’ earphones as static.

During our evaluation, we were treated to a fine display of turbulence which felt real enough for a ‘passenger’ on the flight deck to ask it to be turned off.

All this adds up to a $7 million system which now caters for up to 90 per cent of all Qantas pilot training.

We flew a takeoff, circuit and landing on a Hong Kong visual, to get the feel of this all-moving simulator, with Flight Simulator Superintendent Captain George Lindeman in the righthand seat, Flight Engineer Superintendent John Muir at the F/E’s station, and Simulator Co-ordinator Jim Jones out there somewhere.

Qantas’ Ray Page had earlier briefed the AUSTRALIAN AVIATION team on the simulator’s technicalities.

George first performed an aborted takeoff in Qantas 29 (even simulators have to answer to a name. . . ) before we took the lefthand seat.

It was all too real. We were rapidly approaching V1 when the bells went off, the red light came on and we identified the No. 1 engine as the ailing motor.

One couldn’t do that in a real aircraft without either causing real damage or playing the whole thing back afterwards to check the correct drills had been carried out.

Our takeoff, circuit and curling approach to Hong Kong’s unique airport was performed with all the realism one would expect from a genuine aircraft. Takeoff was performed in a night mode, with George calling off V1, Rotate and V2.

Control forces were indistinguishable from the flying type aircraft, and we climbed out with undercarriage and flap coming up, cueing in the right amount of pitch.

The night image given by the SP1 picture showed those hills through which most pilots have to thread a delicate needle. Out in the bay as we curved around were tiny pinpricks of junks’ riding lights; the flight deck environment was faithful to the point where we could eyeball out the lefthand window, thanks to that fifth panel.

Approach and landing were authentic, right down to that little irritating waffle at about 100 feet from the threshold.

And just for kicks, the situation was taken back to a spot about 600 feet, with the runway lying out in front, to do the same thing all over again from long finals. . . Try doing that with a real aircraft.

Before leaving the flight deck somewhat reluctantly and emerging into the unreal light of day, we were told that Qantas, who have gone right through the spectrum of simulator technology since the airline took delivery of its first system years ago, is now consulted by the manufacturers as a technical adviser of the system.

A few days earlier, in Melbourne, we had our first taste of TAA’s new A300 simulator, which is another magic system. . .

TAA laid down the technical aspects of the simulator manufacture to Thomson-CSF (late LMT- LeMaterielTelephonique), married it to a SEL 32/77 in the best traditions of updated computers and used the Novoview SP1 visual system as a base mark.

The airline didn’t waste any time getting the A300 simulator on stream; with a 59-part shipment and four boxes crammed with tools arriving from France, along with ten Thomson-CSF specialists to glue it all together and train airline technicians, the system was assembled in just over three weeks.

The system arrived in knock-down form in May last year, according to TAA spokesmen in the Essendon Flight Training Centre.

The entire system was ready, sitting on its six-axis legs and looking at every kilo of its 10 tonnes and $7 million, in July last year. . .

TAA reckons it will save up to a million gallons of fuel a year by using the A300 simulator, which runs at an economical $400-$500/hour compared with an actual A300’s $4000-$5000/ hour and the gulping of about six-plus tonnes of fuel every 60 minutes.

Crews normally spend eight four-hour periods _inside the simulator, operating their ‘aircraft’ in a dusk/night environment. Like Qantas, TAA will not apparently go to Level 5 operations at this stage but intends to sit it out until adequately faithful daylight visuals are a reality.

Two Systems 32/77s are normally used, with another as a hot spare, under normal conditions.

A big CRP 24 takes care of control loadings, while a Texas Instruments 980 controls the visuals. All this adds up to what the airline’s PR system conservatively calls computerised systems which are designed to allow full simulation of flight, aircraft system performance and flight deck environment.

TAA estimates that they will get about 2000 hours a year utilisation from their new A300 simulator, which includes conversion training.

Their older simulators – a Boeing 727 Redifon using an SEL 32 with a four-panel SP1 visual display which entered service in 1979, and a DC-9 Redifon simulator using a DDP 124 host computer and 6000 two-panel display on a three-axis system circa 1966 – are still doing yeoman service.

According to Captain Len Heard, Flight Ops. Manager of TAA’s Flight Operations Division, and Captain Dave Wiltshire, Flight Captain Research, the 727 simulator has ‘flown’ 3000 hours each year; the DC-9 has racked up the same annual figure with a total of 41,000 hours of operation logged since it came on stream in 1968.

The A300 simulator justifies all the claims made by its manufacturers, in terms of sharpening flight crew night work with an emphasis on approach and landing phases.

AUSTRALIAN AVIATION flew a test circuit of the simulator, using a Tullamarine visual, and got a glimpse of Sydney operations plus a Cat. 2 approach in startlingly realistic conditions fed into the visual by an instructor (Dave Wiltshire) from his panel at the flight decks after the end.

Before that, we were treated to a dramatic display of a malfunction at V1, an engine flame-out.

Flight deck drill was routine as we lined up on Tullamarine’s main runway, digesting ATIS information of QNH 1013 and outside air temperature of + 15.

At V1, 141 knots indicated, the flame-out occurred, with the normal visual cues of fire bells and lights. The ram air unit was deployed after confirmation that the starboard engine had malfunctioned, and we went on to compute Vr of 153 kts.

Back to Square One – easy to do when flying a black box instead of a bucking aircraft – we were shown another malfunction, a compressor stall which came in on cue after takeoff giving a series of explosive and solid bangs with accompanying tremors.

Yet another unusual situation was fed into the system, to add more spice to a dramatic display of what one can do while sitting in the same place; an explosive decompression when the forward cargo door parted company with its parent fuselage.

The only thing missing was the painful result in one’s ears. . .

Our circuit was as routine as if we had really been pushed back, taxied out and arrived at the sharp end of the duty runway.

With a computed weight of 141 tonnes, V1 was 133, Vr 138 and V2 143 indicated.

Len called out the cues and we arose from the simulated runway, leaving the tiny runway lights and the track of our landing lights way astern.

We aimed respectably far into the duty runway on arrival over the fence, but the A300 took it without a murmur, with moderate to heavy braking and reverse thrust taking care of the stopping process.

The Cat. 2 landing was as dramatic and startling as the real thing, with a decision height of about 60 feet.

After all this sophistication, we emerged from the A300 simulator to find the old world safety measures hanging outside the all-enclosed simulator – two ropes and a fire screen, to facilitate evacuation from the simulator in the unlikely event of a fire on board.

Meanwhile, Ansett Airlines of Australia (with five Boeing 767s on order and 12 Boeing 737-200s and 16 Boeing 727-200s in the fleet), is awaiting delivery of its two new simulators – for the 767 and 737-200, costing a total $11 million.

The airline has been operating a 727 Singer-Link simulator since 1975, using a Vital 111-6000 visual display on two panels and a three-axis motion system, with a host computer of GP-4 series. In contrast, the ordered Ansett 767 simulator will have a Novoview SP1 six-panel WADS (Wide Angle Display System) on which to display visuals giving as much outside view as that on board a flying Boeing 767.

Ansett tells AUSTRALIAN AVIATION that Redifon visuals will be fitted to the new AST simulators and provide more realistic training for a large percentage of the airline’s 800 pilots.

With the Redifons they will be using training techniques and scenarios not currently available to airlines in Australia, adds an Ansett spokesman.

Ansett plans to eventually use a simulator capable of training pilots with zero flight time – a Phase 3 FAA draft requirement and one not so far used by many other nations’ simulator customers.

However, that standard will not be met for about five years until advanced colour daylight visuals are more developed.

Each of the simulator visuals will use all six panels in the new display, to give the aircrew as uninterrupted a field of vision as that experienced in the Boeing 767 – almost double the field of vision in most other simulators, according to Ansett.

Full advantage will be taken of the Novoview CGI system, giving a full range of variable weather, visibility and cloud conditions.

Effects generated by the aircraft, such as landing lights, anticollision beacon and rain removal, are programmed in to provide a new level of realism.

Surfaces can be used to represent three-dimensional structures and terrain features, which correctly occult other solid objects and light points.

Adverse weather simulation is achieved by combinations of features, such as scud, ground fog/mist, in-cloud glare and low visibility halo effects.

In addition to the normal runway and approach lighting systems, special light-point types such as T-Vasis, airport beacons, flashing beacons, stars and moon are used to enhance the ‘real world’ environment.

Curved light strings provide the database designer with a powerful technique with which to create metropolitan and urban features, say Ansett spokesmen.

One presumes that Ansett will be making some of its own visuals.