Overspeeding

chrisN 27-Feb-13 01:12 #08 Flutter and vibration are not the same thing.##

A really nice explanation of this phenomenon. Worth bearing in mind that where structural vibration is concerned, what matters is the degree of coincidence between the natural frequency (2PiF=SqRoot of the controlling stiffness constraining the part involved, divided by the mass participating on the vibration 'mode') - and the exciting frequency.

So changing the mass or the stiffness will only increase or decrease the natural frequency of vibration of the structure, not eliminate it. This can be good or bad, depends on whether the change takes it closer to the 'exciting' frequency or not. In aeroelasticity, the airflow effectively becomes the spring device - storing energy from the vibrating airfoil.

Vibration can be reduced in amplitude by increasing the damping - the amount of energy in the vibration that is diverted into e.g heat energy, though this also tends to broaden the frequency range over which the natural frequency is at work.

Be in no doubt that real flutter of control surfaces can develop into catastrophic failure very quickly indeed . . . think Brrrp - - - gone

Do you mean how improve the trim hinge wear issue?

Assuming it is vibration that is the cause, triggered by prop wash and other things (there is always vibration throughout the airframe, I suspect), tighter tolerances would probably help. Different materials might help too, maybe including a liner which could perhaps absorb vibration. Redesign using ball or roller bearings might help, though there is a risk there – if it never moves, vibration can result in the balls in a bearing indenting the ball races. I believe it used to happen in car wheel bearings on cross channel hovercraft which had considerable vibration. (There are plenty of mechanical things working in vibration-saturated environments – think cars among others – which don’t repeatedly fail in the way you describe.)

All design is a compromise of many factors, including weight, development costs and ultimate product cost. I guess the airframe manufacturer does not want to do a redesign.

Regards –

Chris

In that case how would you go about improving that, Chris?

Tightening up the tolerances?

One needs to be careful with thermal expansion, though I notice that steel and phosphor-bronze (the best simple bearing materials) expand at virtually the same rate.

Then lubing becomes really important, too, and that is routinely neglected in GA maintenance.

Flutter and vibration are not the same thing.

Vibration can simply be excitation by turbulent airflow over an element. I would think that is the u/c door issue, and very likely the trim tab mechanism wear issue too.

Flutter occurs when a system moves from a stable state (vibrations would damp out in time) to unstable/positive feedback situation. Consider a slightly flexible wing with an aileron at the end. If it is unstable, slight disturbance of the wingtip upwards tends to leave the aileron behind, creating more lift and pushing the wingtip even further up from its resting position. This positive feedback continues until either the aileron flexes back into a neutral position, or the resistance of the wing to further bending is so great, or both, that the tip stops flexing upwards and starts instead to move down. Now, the aileron gets left behind, and in flexing upwards increases the downward forces on the tip, resulting in an eventual downward deflection even greater than the initial upward deflection. Etcetera.

Mass balancing the aileron helps avoid this. The stiffer everything is, the more resistant to flutter.

HTH.

Chris

This flutter stuff is very interesting.

Socata TB owners will know about the "elevator trim hinges issue" whereby the hinges wear at a rate which bear no relation to the fact that they almost never move. I reckon they wear about 100x faster than they should.

There is much debate about this but I have found (by getting very involved in the Annuals since I went N-reg in 2005) that about half the play comes from 3 or 4 bushes in the trim linkage (and on older planes possibly from wear in the jackscrew) and the other half comes from wear in the hinge pins (PTFE coated on the GT aircraft). The cost of replacing the hinges is about £1000 and it's a big job because Socata used screws on the half of the hinge which doesn't wear and used rivets on the half which does wear! A good halfway solution is to fix the linkage bushes (which are quite cheap) and replace just the two outermost hinge pins which are about £100 each.

Anyway, I've noticed quite a lot of dark residue coming out of the hinges and this is the aluminium of the hinge material. Yet the hinge almost never moves...

And exactly the same on the TB20/21 gear doors. Their hinges, and their linkages, get trashed at an alarming rate, again despite the fact that they hardly ever move intentionally.

So I think there is flutter going on.

In fact one can hear the gear doors rattle in the propwash during taxiing.

The Q is what one can do about it. Legally, nothing, because even if you rebuilt the trim tab assembly using brand new factory parts, you will still get almost the same amount of play - as some people who have done that have reported.

In case you have not seen it and are interested, here is a glider having flutter tests:

By the way, most if not all things that vibrate will have more than one mode of doing so. I think that for a glider wing, the first mode is simple bending from the roots, then the next is a harmonic with a single null point partway in from the tip, etc.

I have heard that on at least some gliders, the limiting flutter for VNE is the tailplane. I have also known of a glider with probably prior but unknown damage having the whole empennage go into a twisting mode about the fuselage axis, at less than VNE.

Dunno about power aircraft, but I suspect that similarly, there may be several different modes of imminent failure, depending on the model and perhaps its history. There was the well documented (and video’d) aircraft in the USA – a modified Mustang? - which lost part of the tail first, in a mixture of speed and G.

Chris N

PS:

http://en.wikipedia.org/wiki/2011RenoAirRacescrash

“The NTSB thoroughly investigated the extensive modifications made to the airplane. The modifications had made the aircraft lighter and reduced drag, but decreased stability. Leeward took the plane to 530 mph during the race, about 40 mph faster than he had ever gone before.[24] There was evidence of extreme stress on the airframe demonstrated by buckling of the fuselage aft of the wing and gaps appearing between the fuselage and the canopy during flight (clearly visible in high resolution photographs taken by spectators). However, the investigation (released in August 2012) found that probable cause of the crash was old reused locknuts in the left elevator trim tab system that loosened. This led to a fatigue crack in an attachment screw and allowed the trim tab to flutter. This flutter caused the trim tab link assembly to fail which led to loss of control of the aircraft. Untested and undocumented modifications to the airplane contributed to the accident. Particularly, the right trim tab had been fixed in place. Had both trim tabs been operational, the loss of the left trim tab may not have alone caused loss of control. When the trim tab failed, Leeward was exposed to 17Gs which quickly incapacitated him and likely rendered him unconscious.”

=============

While in IMC the A.P commanded the Comanche into a spiral dive, things did happened very quickly, while fighting for controll I calculated 2min. to the end.

Anyone who flies a CJ2 or CJ3 and says they haven't had the overspeed warning going off is probably being economical with the truth! These aircraft will go past the red line in the cruise at most levels unless it's very hot.