Fidelity of older flight simulators causing unfair assessment of pilot competency

[A37575]

Level Five full flight simulators are normally credited as zero flight time simulators which includes credits for landings. Below those Levels are various older technology simulators with credits only for certain defined sequences including instrument approaches - for example to Cat 1 ILS minima - but not for landings.

If the simulator (say Level 3 or 4 or sometimes known as Category B or C) fidelity is acceptable for instrument rating purposes - let us say for example acceptable to 200ft agl on ILS - but is not accredited for flare, touch down and roll out, then at what point in the final approach path below say 200ft does the fidelity stop?

Only in the landing accredited simulators (Level Five or Level D) is the interface between the visual system and the computer generated instrumentation judged to reflect real world conditions. That is why landing credits are not allowed in the lower level simulators (below Level Five or D) and as far as the regulator is concerned the pilot is deemed incompetent to land the real aircraft until he has completed circuit training in the real thing.
Pilots under training in non-landing credit simulators are occasionally criticised for failure to "fly" the simulator accurately to smooth touch down after visually acquiring the runway environment (approach lights followed by runway lights).

Sometimes the simulator will "land" heavily and the pilot is frowned upon in his training record as "failure to flare properly, etc".
Often the design lack of fidelity visual aspect of the runway causes a flare and float or apparent drift error by the pilot - simply he has no choice, because he is relying on false cues provided by simulator design which of course is why landing credits are not approved by the regulator.
It then becomes a case of trick flying applicable to that specific simulator to obtain a "good" landing. And that is not what good training is about...In fact a "good" landing in a non-landing accredited simulator and its learned technique could paradoxically lead to a bad landing in the real aircraft.

But if the simulator is not technically capable (lacks fidelity) of displaying the correct visual cues (which are present on a Level Five or higher category) then pilot competency cannot be honestly and fairly assessed.
You get what you pay for. Some older simulator such as F28 and B737-100/200 do not have landing credits and hence regulations require certain training including take off and landings performed in the actual aircraft. Typically either 2 or even 4 hours of circuit training before the type rating is given. These old simulators are naturally cheaper and lack the fidelity than the seriously expensive Level Five.

Having bumbled my way about those introductory paragraphs I'll try and get to the nitty gritty. For a non landing credits simulator, if pilot under test or under training experiences obvious difficulties in the last part of the ILS and landing from below the simulator fidelity cut-off point - or when especially on a visual approach where he relies on visual cues as well as inside the cockpit, then perhaps an accurate assessment of his flying proficiency in this zone is not feasible and should not be attempted simply because of the poor level of fidelity in these older simulators. In other words all assessment bets are off once the runway is acquired visually at any stage on final because the simulator fidelity is unavailable on that class of simulator.
Your comments would be appreciated.

[Shawn Coyle]

Not sure if this answers the question directly.
We were in the final stages of certifying a simulator to FNPT II standards, and found some 'holes' in the way the sim flew.
Rather than try to fix the problem (which was way outside our capability), we invoked the phraseology of the requirment that said something to the effect of 'for the training required'. We clearly stated the limitations of the device, and the maneuvers which were not considered to be part of the training.
This presupposes that those who use the sim get to read the certification reports, or that those who do the flight ops part of the sim approval read the engineering reports. But at least we could say 'this device is not to be used for xxx'.
(which was much better taught in the real aircraft anyway, due to the peripheral vision and propriocetive cues needed.)

[John Farley]

Pilots under training in non-landing credit simulators are occasionally criticised for failure to "fly" the simulator accurately to smooth touch down after visually acquiring the runway environment
I fear that probably says more about the attitude (or experience) of the trainers and possible personality conflicts with students than the simulator hardware.

I agree that learing a 'trick way' to land a simulator could be bad training - especially for junior students. Probably does no harm to old hands who may well be able to differentiate in their heads between the simulated and real experience quite easily with no adverse cross over.

In any educational setting life suggests the quality of the instructor/examiner has a big effect in determining the outcome for the student and flight simulators are no exception to this experience.

I have long held the view that training simulators should auto freeze as soon as they are flown outside their certificated limits (which can never be outside the certificated aircraft envelope and may indeed be inside as you discuss).

I feel this way because if the limits were being exceeded accidentally then the freeze becomes a good wake up call and if the exceedence was deliberate then the simulator should not aid and abet 'finding out what it will do' because the experience may not be valid.

[A37575]

John Farley. Your points are taken. During simulator training on the non-landing credits trainers, there are pilots who are able to consistently execute good landings on the same machine and others who appear (for example) to fly straight into the deck due failure to flare. The problem for the simulator instructor is to decide whether to disregard either technique because he knows the simulator was never designed to have landing fidelity. It would be wrong for the instructor to say "a bit late on the flare Bloggs" when Bloggs could quite understandably come back with "Excuse me Sir, but forgive me if I am wrong, but isn't this a non-landing credits flight simulator and the landings do not represent the behaviour of the real aircraft at the flare - so why are you commenting?" And the way I see it F/O or Captain Bloggs has a good point.

Put another way, a lawyer would rip the instructor apart if Bloggs was to complain that his landings were being unfairly hammered by the simulator instructor. Hence my original question of where does simulator fidelity deemed to be unsound following the break out on ILS final to complete a visual path below the ILS decision height?

[ZFT]

Actually, the (JAA) regulations are not quite that straightforward or even logical. It is possible to be checked/trained for landings & take offs for recurency training on a device as low as Level B. This is why DLH and Ryanair have both ordered specific Level B 737NG simulators recently.
Conversion take off & landings require Level C/D

The following is from JAR-STD1A.030

Level A - Suitable for: – Crew procedures training. – Instrument flight training. – Transition/conversion training, testing and checking except for take off and landing manoeuvres. – Recurrent training, checking and testing (type and instrument rating renewal/revalidation)

Level B - As for Level A plus: – Recency of experience (three take-offs and landings in 90 days). – Transition/conversion training for take-off and landing manoeuvres – Transition/conversion testing and checking except for take-offs and landings.

Level C - As for Level B plus: – Transition/conversion testing and checking of take- offs and landings for flight crewmembers whose minimum experience level is defined by the Authority.

Level D - As for Level C plus: – Transition/conversion testing and checking of take-off and landings for flight crews, who may be required to meet a minimum experience level defined by the Authority.

Additionally the DGAC in isolation from the JAA/EASA have recently approved an FTD for full type ratings including all manual control handling skills including take off and landings.

At a time when experience levels (especially within the developing world) are not what the used to be, an ‘interesting’ and potentially far reaching development.

At a time when the regulators should be striving for tougher regulations it appears that at least one will now accept an easier solution.

[A37575]

Thank you all, for the thought provoking information so far.
A37575.

[AngloPepper]

The key thing here is that JAR 1A, 1H, AC 120-45, -63 and FAR part 60 are technical documents, addressing the technical capabilities of Flight Training Devices. While Level D devices may be capable of zero flight time transition training, the final judgement is down to what courseware the FTO has had approved. If the courseware says that CAT1 is approved down to DH, then the FTO is wrong to assess pilots on landings. On the other hand, if CAT 1 approaches are approved to be taught all the way to landing, the FTO can carry on as before; they have the authority on their side.

In the helo world, this is a common problem. Many sims are level D certified but are not capable of zero flight time training. It all comes down to what training delivery the FTO has pursued during the certification process.

One further thought. Level D simulators are certified using 3 designated airports. The device is only guaranteed to perform flawlessly at these airports. Many simulators have other airports added, but there is no guarantee that the scene content, lights and runway textures are up to the same standard as the 3 airports used during certification. Needless to say, any loss of visual fidelity (in this case, runway texture particularly) will impact the pilot's ability to judge the flare.