Originally Posted by PEI_3721
Few if any simulators have representative side force / lateral acceleration. The ‘ball’ might provide an indication, but the all-important body sense of going sideways is missing. Thus in a simulator the pilot lacks both a detecting sense for an engine failure, and a guide to the effectiveness of corrective action – how fast / much rudder application is required.
In part, this deficiency results in poorly matched lateral directional responses in the simulator. Also in some aircraft these aspects change with pitch rate/attitude and/or ground effect.
Once again PEI I have to admire the way you think. Not many of today’s aviators are particularly knowledgeable about the simulation they use day-in and day-out with respect to the fidelity to which they are exposed when IN their particular simulator – and I think that is one of the points you were making – and which I believe is accurate. However, I do have to offer corrections and explanations to your opening statement.
While it is true that there are quite a few simulators in use today, as anyone would recognize, the 6DOF simulators that have entered service, particularly those entering service within the past 10 – 12 years or so, are quite good – and do provide the kinds of side-force/lateral acceleration (as well as acceleration forces in all of the degrees of freedom characteristic of an airborne airplane) that you say is not very common. The correction I think is necessary comes from the fact that, in the US (and unfortunately I don’t have access to similar numbers for the rest of the world) of all the aircraft flight simulators (not flight training devices) in service today number just a bit more than 660 – where only something like 25 of them do not incorporate a 6DOF platform motion system. Surely, the more modern the simulator, the more technically capable it will be – and those placed into service within the past 10 – 12 years would be expected to be at the top of that list of most technically capable – but 635 out of 660 is certainly not “a few.”
Of course, anyone familiar with simulation will recognize that using a simulator that is mounted on the kinds of motion system bases currently in use cannot provide sustained “g-forces” (i.e., accelerations) in any of the degrees of freedom – but, even in the airplane, it is the “on-set” cues that provide the most significant cueing to the pilot at the controls. Once that “on-set cue” is recognized, there are other, equally important references that any pilot can (and should) use to verify the initial acceleration … outside visual references, instrument responses, sound cues, to name just a few. I can say this because in an airplane were a pilot to experience a recognizable and
continuous “g-force” (acceleration) in any axis, that he/she did not initiate and is not maintaining, that pilot should immediately recognize that an abnormal condition exists and should be taking immediate steps to correct the flight path. Some will argue that as long as “acceleration” continues, the associated “g-forces” will continue to be experienced, and from that, conclude that my statements here are “bogus.” To those persons I would recommend the following … if you have the availability of having an additional person in the cockpit or if you can suspend something from a non-objectionable location in the cockpit, which is observable by at least one of the pilots, where this object, having at least
some relative weight, can be suspended by a string or thread or similar, and allowed to swing freely during taxi-out … during the takeoff roll, you will see this object be displaced toward the rear of the airplane (displacement opposite the direction of aircraft movement and in the direction of acting force on stationary objects in the airplane) and as the acceleration continues, you will see this object begin to return to its original “straight-down” position. Depending on the smoothness of the runway surface and the timing to the initiation of the rotation for takeoff, the suspended object will return to its original, straight down position relative to other objects in the cockpit – because it too, will be accelerating at a rate that will eventually equal that of the airplane – and while the airplane continues to accelerate toward rotation speed, the suspended object will (until rotation is initiated – when acceleration forces are changed) retain that straight down position. You’ve recognized this exact sensation on every takeoff you’ve ever made. Initially, you are pressed into the back of your seat, but eventually (sooner, rather than later) you, just like the suspended object, just like the airplane and all of its contents, will continue to accelerate to match the acceleration of the airplane – and since there is no acceleration differential along that axis, all seems to return to just what it was just prior to brake release.
As a pilot you can confirm your continued acceleration by outside visual references, the movement of the airspeed indications, as well as various changes in sound – but the relative differences between you and the rest of the contents of the airplane (including the suspended object) will reach the acceleration of the airplane. It is these kinds of relationships – and their changing over time – that are the basis on which the effectiveness of simulation is accepted … i.e., “just like the airplane.” Of course, a simulator does not (it cannot) accelerate continuously – but it DOES provide an on-set acceleration cue – and then reduces that on-set cue value at a rate that is below the threshold of recognition to the occupants. Simulators that used to have extremely poor motion cueing, did not do that – at least not very well, and sometimes not at all. In fact, I recall the Navy putting it’s A-7 pilots on a “do-not-fly” restriction for 72 hours (it might have been more, but I don’t think so) after having competed a “recurrent simulator training session” for that airplane at the Jacksonville NAS, Florida, in the mid-1980s. In fact, that particular simulator had extreme latency problems in a lot of its systems but particularly so in its motions systems.
The goal of simulation is to provide as many of those “on-set” motion cues as possible … then reduce them at a rate that is not recognized by the human body – and relying on the other cueing provided (visual, feel, and sound cueing presented just as would be seen, felt, and heard in the airplane) to verify to the simulator occupants that the acceleration is continuing just as you would have expected it to continue. Simulators with very long motion actuator legs are better able to do this than simulators with shorter actuator legs – which may be required by the structure of the facility or other limitations to that particular simulator installation. Electric motion systems (verses the older, but still used, hydraulic motion systems) are gaining in popularity due to its more immediate response, the greater amount of control it provides, less messy areas (hydraulic systems leak – and electrons don’t accumulate under the machine!) and do so with a shorter required actuator movement – and at an astonishingly low maintenance cost of only 20% of what a hydraulic system costs.
I’m not saying that a pilot should not be able to tell the difference between being in a simulator and being in the airplane. But – and it’s a significant “but,” the instructor and the simulator – together – should be able to provide any pilot a sufficiently accurate on-set cueing system (particularly that of motion on-set – or acceleration on-set) that will allow that pilot to learn what to feel in the “seat of his/her britches,” in coordination with what he/she sees, hears, and feels in the controls, to make a direct connection with the same (or very closely approximated) experience in the airplane. If we don’t do this – as a MINIMUM – what we’ll be doing is training and checking – both initially and recurrently – in a simulator – and then allowing those pilots to go fly something that is significantly different from what he/she was used to working with … and hope that these pilots “learn” sufficiently well, how to fly the airplane and then shift gears to demonstrate his/her ability in flying the simulator – and “never the twain shall meet.” THAT is what I am seriously concerned about. And, I can point to airline after airline, at least in the US, where THIS is exactly what is happening … training pilots in reasonably good simulators – or partially training them in such equipment, and then finishing the training and getting “checked” in a flight training device (which is even less like the airplane than a relatively poor simulator) and then not only allowing pilots to go fly the airplane, but expecting them to do so professionally and competently – when the training has been structured so as to engender activity unlike what is necessary to safely and competently operate their airplanes.