How to detect the approaching stall ?

[Genghis the Engineer]

I am in the second semester of my final year of a Bsc(Hons)degree in industrial design. I have chosen a project for my final year concerning flight safety in general aviation.

Good for you, and good luck

I have been working with some great people over the last few months developing ideas that may improve safety and help build confidence and to date have produced to similar concepts that primarily deal with how the pilot is presented information in flight and a form of virtual co pilot or instructor to watch over the pilot.

The main pre-stall clues are airframe buffet, stick buffet, back-stick force, rear-stick position, audible stall warning (if fitted), pitch attitude, reduced primary control responsiveness, reduced airspeed. The stall itself is marked (in piloting terms) by the point at which full control is no longer available over the aircraft - this could be pitch break, loss of nose-up pitch authority, wing drop, wing rock, roll inverted without asking, incipient spin, or operation of some artificial advice such as a stick pusher. I've seen all of these, bur prefer the first two !

What I am looking at now is how to collect the flight information traditionally presented via the 6 primary instruments.

A domestic camcorder, mounted between the pilots heads, looking at the relevant instruments, ideally with stick position and visual horizon in the peripheral view. Record intercom using a magnetic pickup (£3.50 from Maplins last I looked, sold for recording phone calls) within one of the headsets. Oh yes, and make sure the pilots fly with the sun on one wingtip, into or down sun the picture isn't very good. A good test pilot will give running commentary that you can record, and ultimately only he can define the point of stall. Don't use anybody but a good test pilot, you'll be wasting your time and money; there are many excellent pilots out there who simply haven't had the training to do this.

I propose to use a combination of GPS information and a separate gyro to provide attitude information.

GPS is of no use to you in this context except possibly to aid the subsequent accident investigation once you screw up. If you are using a certified aircraft there should already be a PEC curve available, which is the only thing GPS has any relevance to in this exercise. If you've already got an attitude gyro, just film that and save money and possible inconsistencies between the two systems.

With these two inputs providing a three-dimensional model of the aircraft in flight.

3-dimensional position in space is irrelevant unless you have local wind data. Also, stalling tests are primarily relevant only to the pilots perception, so you need to use what he or she will perceive, not create a separate set of data that will cost you time and money to obtain, and also need calibrating. Additionally, GPS update rates are not good enough for stall testing.


I hope you are still with me?

Of course, but are you with me?

Could anyone, pilots or engineers help me work out exactly what the computer would need to know in order to detect the approaching stall? The basis of the system is a computer that interprets the incoming values and compares them to type specific aircraft figures. This allows it to monitor the various factors that relate to the aircraft being flown safely. Much as your instructor watching over you would.

See my notes above. Stall warning systems are invariably based upon AoA. The stall will always be at the same AoA for the same flap&slat setting. Also, AoA will be increasing as the stall is approached, so fit a cubic spline to the AoA .v. time curve, and you can predict the onset of stall continuously - say set it to trigger at 3 second or less

I am aware there are many factors that relate to the stall not just airspeed or angle of attack but don’t know exactly what they all are !

Depends on how you model it. For a given flap, slat & gear setting the stall may be reasonably regarded as always at the same AoA. For a given flap setting, and a rigid wing, the stall speed in CAS will be proportional to the square root of weight. The relationship between CAS and IAS will be in the POH, or failing that can be determined in an hour or three using a GPS, trailing static, calibrated chase-plane or a range course method (all in the standard textbooks). Note that pilot perceived stalling speed will be a function of CG also - reducing with aft CG, and both perceived and aerodynamic stall will occur at a lower speed / higher AoA with a greater deceleration rate.

Any help at all would be hugely appreciated with regard to any of the points I have raised here.

Hope this helped a bit. Try also the following references (all on the web somewhere): FAA AC23-8, BMAA form BMAA/AW/010a and BMAA/AW/043, various stuff on the websites of SETP, SFTE, and NASA - look at some of the links in tech log.

G


Note to moderator, if our man doesn't get any more response than mine, this post might bear moving to either tech-log or better still flight test where a lot more specialists in the field are likely to be found?