Buffalo Crash

[Tom Goodrick]

Speculation at this point is meaningless. In a few months you'll be able to download a PDF with the info from the flight data recorder. You can study that and come to some reasonably accurate conclusions.

The info on the NASA study of tail icing would have been neat to see. Unfortunately even my new computer could not download and run the whole thing. It seems they intentionally sprayed the horizontal stabilizer to ice it up and then tested responses.

Maybe this accident will produce better deicing equipment.

[flaminghotsauce]

Actually, it's kind of neat what NASA did. They took casts of real actual ice and made a plastic model that can be placed on the horizontal stabilizer, or wherever, I suppose.

[hanspetter]

What we might discuss is "the tail is producing lift but downwards". If this makes sense I've got something to learn. I've been thinking of the tail as a second airfoil producing upward lift to the aft section of the aircraft. Sure, the elevator will be able to produce downwards "lift" but the basic function of the horizontal stabilizer would be to act as a second airfoil for the aircraft. Two separate airfoils are required for lateral stability just as two wheels on a bicycle. Flaming, are you saying that a horizontal stabilizer creates negative lift even when the elevator is in neutral position? Surely an aircraft rides on its tail as well as on its wings

[Ed Burke]

HP you raise an interesting point. Many aircraft, especially heavies, have a negative tailplane AOA. Some of these even have tailplane dihedral ! I recently mentioned, following the Hudson River landing, an incident where a Viscount had its tailplane destroyed by a bird (goose) strike and the aircraft immediatly bunted and disintegrated. However you set me an a search and there are many mentions of negative aoa in tailplanes. One of them is www.allstar.fiu.edu/aerojava/axes33.htm look under longitudinal stability I think.

Very interesting, thanks mate.

Ed

[Chris Ross]

Wait a minute here the whole aircraft creates lift otherwise it will fall out of the sky
The lift has to be balanced around the center of gravity
The tailplane adds lift and an arm of leverage If you add positive or negative lift to the tailplane you go up or down
If you are at an angle to straight and level ie turning you need a rudder to complement the elevators which have effectively become rudders ie youi circle while banked or go straight while banked
I hope this makes sense my eyes arent what they used to be and I find it hard to read and type at the same time besides Tom can explain this much better than I
Chris

[flaminghotsauce]

Chris, the CG of the entire airplane is usually just a bit toward the front of the wing, The tendency is to tip over forward, thus the necessity of having the horizontal stabilizer create downward lift. If you can watch the video I posted, it shows a graphic of the stabilizer zooming back to it's position, but clearly upside down. The main wing is where the maximum lift is created to hold the aircraft in the air, but it's unstable without the rear "feathers."

This was all explained in my Jepesen manuals as well. I'll have to go dig them out.

I've read before (never tried this myself ) that if you put some extra weight in the back of the aircraft, you need less downward lift to balance the aircraft, therefore less trim required, therefore less drag, causing an extra knot or two of available airspeed. It makes sense.

[Tom Goodrick]

Hans Petter you do have to adjust your thinking. Flaming is generally right but not about the CG being ahead of the wing.

We can idealize the airplane in the longitudinal (pitch mode) to the two points at which aerodynamic forces act (the wing and the tail) and the point at which gravity acts - the CG. Many people like to talk about the "aerodynamic center" of the wing or the 'center of lift". The best thing to do is talk about the 25% wing chord location which is 25% of the chord back from the leading edge of a straight wing. We engineers normaly assign all primary aerodynamic forces to act at the quarter chord point with a pitching moment also acting at this point. You can do some calculations to determine the point where the moment would be zero and only aerodynamic forces act. This is the pure aerodynamic center or "center of lift". But when compared to the potential moment produced by the tail, the wing pitching moment at the quarter-chord point is small. So we can talk about the quarter-chord point or 25%MAC as the point at which lift is applied. Now we normally keep the CG close to that position but it may be slightly ahead for best performance - perhaps at 5% to 20%. In this case, the wing lift (which is as large as the weight acting at the CG) is making a nose-down moment about the CG. The tail must produce a downward force to keep pushing the nose up for balance.

Only when the CG moves back beyond 25% must the tail produce an upward force to balance the airplane. In this case more lift is produced unless you reduce the wing angle of attack. More lift generally slows you down. But after trim adjustment, you might end up flying a little faster. You should be able to check this in FS9. Use my CG gauge to read the CG and test this. One thing FS9 is very good for is to check longitudinal stability.

There is also a matter of dynamic longitudinal stability. Any engineering text books say that a forward CG is best for good dynamic stability. This relates to the growth of the pitch oscillations. Having a total-airplane pitching moment that simply returns the aircraft toward its trim angle of attack is merely "static stability" which is what we talk about in "balancing" the aircraft. But there is always a short period oscillation when speed changes or there is a power or picth control input. This oscillation is quickly damped in a dynamically stable aircraft but can diverge and cause loss of control in a dynamically unstable aircraft. This generally happens with an aft CG. It is why an aircraft with an aft CG loading will often have a fatal crash. When the oscillation diverges, at some point the horizontal stabilizer gets loaded beyond its design limit and separates from the aircraft.

Again, FS9 is a great place to play with this problem. Use the CG gauge to load you favorite aircraft various ways. Make a takeoff, a brief cruise and a landing. Fly back and forth between two close airports about 60 nm apart so you have a period of cruise at normal cruise speed. With an aft CG you can have two problems: 1) the tail does not have enough power to level the aircraft, and 2) the tail levels the aircraft but then it gets into a divergent oscillation.

I'll do some tests like this today and let you know the results. Again, there is no point to this, or to any abnormal loading of an aircraft, if you don't have a gauge that tells you where the CG actually is.

[flaminghotsauce]

Hey, that's what I said! Just not so articulately....

;D

[hanspetter]

Thanks for enlightening me
I watched the movie to which Flaminghotsauce linked in a previous post and thought it all over once more. So, even the fixed part of the tail isn't a small second wing that provides lift to the aft section of the aircraft. The lift is provided by the wings (and the fuselage) only and a properly balanced aircraft would actually tend to pitch forward if it weren't for the slight "negative lift" of the horizontal tail.

The question remains, how would one deal with icing in terms of controlling the aircraft? I assume it's hard to tell whether wings or tail are most affected and the right move will not be the same in either case. It will always be a good idea to try and keep a healthy airspeed though and stay way beyond the normal stall limit.

[Tom Goodrick]

Check the threads now in the FS2004 section for icing examples you can fly in FS9.

I don't think we need to assign too much credit for lift to the fuselage. That seems fashionable these days. Only in the case of a flying boat does the lift carried by the fuselage become significant - rising to maybe 5%.

[flaminghotsauce]

Hans I would assume that normally the wing would collect the ice as fast as the horizontal stabilizer. I would think it would be an unusual situation to have the stabilizer stall, while the wing is clear enough not to stall. One would presume icing would mainly be normal wing icing, therefore keeping the speed up would be correct. But a situation could occur, like when icing equipment fails on the rearward leading edges while the wings still have effective anti-ice. I'll bet statistics of this sort would be very few and far between.

[Tom Goodrick]

If you guys can find your way over to the FS2004 section, you'll see that having those nice ideas about keeping airspeed high relative to stall may not be within your power as a pilot. You may just have to sit there and watch the airspeed decay to a point where you will stall if you remain level. So you descend until you contact the ground.

In several tries, I have managed to avoid crashing each time by climbing above the ice but the wrong plane in the bad conditions will make a crash inevitable.