What are the typical failure modes of the PFD simulated during training?
In reply to:
What I’m getting at is if the normal way for the instructor to fail the PFD is to simply turn it off (or cover it up), that presents a very different situation from the PFD showing bad attitude info. (As Steve D. noted, if an instructor turns off or covers the PFD there’s no question that it’s bad and should be ignored.) Is there any way for an instructor to fail the attitude source alone so the pilot under instruction will get a true feel for what it would be like in the real world?
Joe,
You bring up some really good questions that in the near future are going to apply to pretty much every aircraft rolling off the production line as manufactures move away from the old “six-pack” to PFD’s. I still have the six-pack in 4AG, so I am not an authority on this subject, but I don’t think that there is a way to just gradually fail attitude information on the Avidyne (anybody correct me if I am wrong). This kind of training is where flight-simulators really begin to shine (although a training mode in the actual aircraft could be useful, as well). SimTrain is working on putting together a network of full-motion Cirrus flight-simulators that will be great in which to practice this type of scenario.
In reply to:
I know this isn’t a problem unique to the PFD (the same applies to an AI), but an inaccurate PFD would probably “scream” at your senses even more loudly than a failed AI.
I presume you’re taught to turn the PFD off altogether once you’ve lost confidence in it, particularly if the attitude is suspect. Right? Or are you taught to keep it on for the info that might still be good (heading, airspeed, etc.)? (I doubt it, but I had to ask.)
I’ll have to let the PFD users comment on that. [:)]
The “inflight sim” mode you describe would be useful, but I expect there would be some huge certification issues.
As to the PFD with bad attitude info, I can imagine how hard it would be to try to focus on good airspeed, altitude, and VSI info superimposed on a bad horizon! I doubt anyone could cope with that! That’s why I presume you’d turn it off altogether if the attitude was bad and rely on the bolster instruments and wet compass/GPS’s.
At least with a sixpack you can cover the AI alone if it acts up, and still have the rest – all still in their familiar places with their familiar faces.
I don’t think that there is a way to just gradually fail attitude information on the Avidyne
I believe that the actual attitude sensors used are quite cheap, derived from automotive use.
PFD makers should have however many sensors necessary so the software can constantly compare them (cross check) to detect that something is wrong (conflicting values), and X out the indicator.
On such a critical, digital, software driven device, there should never be erroneous, erratic, or gradually failing indications; if they occur, there’s no excuse for it, IMHO.
The “inflight sim” mode you describe would be useful, but I expect there would be some huge certification issues.
Joe,
This is a very good point. Add potential liability issues to the mix as well.
In reply to:
As to the PFD with bad attitude info, I can imagine how hard it would be to try to focus on good airspeed, altitude, and VSI info superimposed on a bad horizon! I doubt anyone could cope with that! That’s why I presume you’d turn it off altogether if the attitude was bad and rely on the bolster instruments and wet compass/GPS’s.
At least with a sixpack you can cover the AI alone if it acts up, and still have the rest – all still in their familiar places with their familiar faces.
All very good points, as well. I think there are different failure modes on the PFD, so maybe the attitude indicator is addressed somehow (although I can’t imagine how) without having to “deep-six” the entire screen. I look forward to hearing from PFD users. [:)]
I believe that the actual attitude sensors used are quite cheap, derived from automotive use.
PFD makers should have however many sensors necessary so the software can constantly compare them (cross check) to detect that something is wrong (conflicting values), and X out the indicator.
On such a critical, digital, software driven device, there should never be erroneous, erratic, or gradually failing indications; if they occur, there’s no excuse for it, IMHO.
Warren,
It would be really interesting to see a specific parts cost breakdown to reveal how the Avidyne PFD can be sold for such a small fraction of the price that somewhat comparable Honeywell and Collins displays are sold for. Granted, there are more features in the latter brands, but certainly not enough features to justify selling for 10 or more times as much. I think you are on the right track with sensor costs.