Pure curiosity, but how are the assorted V_speeds derived?
Some are obviously just measured (stall, best rate of climb etc.), but how are things like Va, Vfe and Vne worked out - I kind of doubt you chaps are up there going 100 kts, full control deflection, wings still there ... 105 kts, full control deflection, ah, there they go... scribble down Va, jump out...
Genghis the Engineer
30th Nov 2003, 06:45
I'll start off simplistic Evo, if you want me to elaborate on anything, please say.
The aircraft is designed, through theoretical analysis with g-limits and a maximum speed, Vd. The loads on every part of the airframe are calculated at Vd, and at the g-limits, at a variety of speeds and gust loadings (ditto the various Vf values, at the prescribed g-limits with flaps selected). From this, the maximum loads the airframe should ever see at these limits are calculated. These loads are called limit loads.
Now a safety factor is applied to this structure - this factor will depend upon the level of safety needed, the type of material used, what structures are included - the list is quite long and complex. You can end up with some bits of even a non-aerobatic aeroplane being checked at 36g. These new loads are called the ultimate loads.
Then we set out to prove that the structure will take all of the limit loads without ever permanently deforming, and the ultimate loads (for a fixed period of time, usually around 10 seconds) without a catastrophic failure. This may be done by analysis (particularly for very large and expensive aeroplanes) or may be done by rigorous testing on the ground using various means of loading from sandbags to hydraulic rams (mostly for smaller cheaper aeroplanes, but also for the areas on a bigger aeroplane where the analysis shows it only just meets the requirements).
Then, and only then, is the aircraft flown. At the high speed end, the design limit Vd is approached very carefully flying a whole series of tests to ensure no flutter and adequate stability. If it is reached then all well and good, if it isn't safely reachable for any reason then a lower limit called Vdf is declared. Vne is then declared at a lower speed, which is (almost) universally no greater than 90% of Vdf.
So that's given Vne. A check is then made against Vh (the maximum speed in level flight), because there's always a safety margin needed (whose value varies) between Vh and Vd - basically to ensure that the aircraft can't exceed safe speeds too easily.
Va and Vf are more straightforward, they are based almost entirely upon structural analysis. That said, they are functions of the stall speeds in various configurations, so careful checking of stall speeds is needed during flight testing, and revision of Va and Vf can be necessary if the stall speeds aren't as predicted.
Having said that, it's worth mentioning the LIMIT loads calculated at Va and Vne. At Va they include all primary controls at full deflection, and at Vne they include all primary controls at 1/3 deflection. At Vf the flaps are at full deflection, but often a safety factor is applied later so that the Vf given to the pilot is 90% or less of the Vf calculated by the designer, to give an extra safety factor.
When it gets to flight testing, the real interest is in the aircraft's performance and handling at these corners of the envelope, and of-course that all systems function correctly. If anything actually fails then somebody has screwed up VERY VERY badly long before the aeroplane flew. More likely (but again pretty rare) is that some aspect of the controls fails to mechanise properly within the structural envelope because of aeroelastic deformation, or something starts fluttering (probably the single hardest aspect of structural analysis to get right).
G
Thank you Genghis - very interesting (although "simplistic" still required some careful thinking!) :)
Nice answer.
I read how the SJ30 crashed due to uncontrolled turn to the left from a Vmo test.
Interesting note that many jets while structurely sound seem to be limited to aerodynamic forces that act on the aircraft well before design load limitations are reached.
In the SJ30 case, preliminary reports indicate the pilot put the nose over and the plane at a certain speed was using full scale aeleron deflection to keep the aircraft straight. Any faster and the plane keeps rolling left.
I guess most jets have the same problem. In the Citation SII that I fly, when I get lower(thicker air), in a decent, at Vmo, on the autopilot, once I take it off the autopilot I have to retrim the aelerons to the right.
Food for thought.