|| Tutorials home | Safety tour | Emergencies | Meteorology | Flight Theory | Navigation | Communications | Builders guide ||
Decreasing your exposure to aerodynamic risk
2. Don't fly real fast!
Rev. 14 — page content was last expanded 12 July 2014
Speed seems to hold some kind of fascination for us. We derive a thrill from flying at top speed but the freedom of the air has its dangers; what are the dangers of flying too fast?
Here is an extract of an email from Italy to an RA-Aus member discussing the crash of a Flight Design CT recreational light aircraft:
"The airfield has been closed due to a fatal accident with a CT that went over Vne and (broke up) just 50 metres above the airfield. I don't know if you ever experienced the CT, never fly it with pilots that want to show you how fast it is. I saw ... what happens if you feel greater than aircraft limits."The theme common to the difficulties encountered in that and similar disasters, is when moving really fast there will be little warning and no time to do anything. The results are invariably fatal.
The airspeed that constitutes 'too fast' can change depending on the load carried and how it is distributedThere are limits to the payload an aircraft may carry safely and load must be distributed so that the aircraft's balance — the position of the aircraft's centre of gravity — is contained within defined limits. In addition there is a maximum safe operating weight permitted by the aircraft designer for all flight conditions. So, before you load a 90 kg passenger and 30 kg of gear into your light sport aircraft with full tanks, check the weight and balance charts in your Pilot's Operating Handbook or aircraft flight manual.
Weight and balance affect control and stability at all speeds. Excess weight reduces the designed structural load limit factors. Cg positions outside the designated fore-and-aft limits may enhance elasticity reactions to aerodynamic loads, or reduce controllability because of moment arm changes, or delay (even prevent) recovery from unusual and/or high-speed situations. FAR Part 23 section 23.335 requires Vd — the 'design diving speed' — to be not less than 1.4 times the design cruise speed for a normal category aircraft. To receive type certification, it must be demonstrated, possibly by analytical methods, that at Vd, the propeller, engine, engine mount, and airframe will be free from over-speeding, severe vibration, buffeting, control reversal and most importantly flutter and divergence. To provide some safety margin, Vne (the IAS that should never be exceeded in level flight, descent or manoeuvre) is set at 90% of the lower of Vd or Vdf. Vdf is a diving speed that has been worked up to by a test pilot and demonstrated to be without problem — though without pulling heavy airframe loads — and which must be lower than or equal to Vd. Vd is not required to be demonstrated in an in-flight test.
Vne is always specified in the pilot's documentation as an indicated airspeed and should be marked on the ASI (the red line) but unlike the performance airspeeds (also specified as indicated airspeeds and perhaps marked on the ASI), Vne is related to those structural characteristics and limitations associated with bending, twisting and elasticity, and which affect stability, control and even structural integrity. Limiting speeds are also associated with structural reaction to pilot-induced loads and to gust-induced loads. Limiting speeds could also be associated with other potential problems; for example, suction effects at particular speeds and attitudes might lead to canopy departure, or door or cowling security problems , but normally Vne is limited by the critical flutter speed. dynamic pressure (½roVis²) no matter what the aircraft's altitude might be. But there is a qualification in that one of the conditions that limits maximum safe airspeed is the onset of flutter, which is a function of the speed of the airflow, the elasticity of the airframe, the balance of the control surfaces and the condition of the flight control actuating systems, rather than just the dynamic pressure. See 'Aerodynamic reactions to flight at excessive speed' but also read 'You can also induce structural damage at moderate speeds!'
For most sport and recreational aviation light aeroplanes only one Vne is specified in the Pilot's Operating Handbook or aircraft flight manual. That value is possibly conservative and applicable for operations below 10 000 feet amsl. The designers of most piston-engined GA aircraft specify one fixed-value Vne for operations up to the aircraft's service ceiling; that value is represented by the fixed red line on the ASI. (Note that some EFIS displays may be able to continually adjust the redline display so that it reflects the IAS proportional to a Vne expressed as a true airspeed.) The flight testing program would have determined that the aircraft has no potential flutter problems at Vdf up to the service ceiling. However, some GA aircraft have supplementary lower-value IAS Vne for operations in altitude bands above a stated altitude — perhaps above 10 000 or 15 000 feet. This approach to Vne specification is normal with sailplanes and powered sailplanes whose long, elastic, high aspect ratio wings are likely to develop flutter problems at high true airspeeds, also long wings vibrate less rapidly than short wings, i.e. have lower natural frequencies.
Random inflight airflow perturbations may cause wings to bend upward and downward i.e. flap, the degree of oscillation or flapping is more pronounced with high aspect ratio wings. While bending upward the wing adds a vertical velocity* to its forward velocity — the true airspeed — which results in a decreasing angle of attack reducing the lift of the up-moving wing and thus producing an aerodynamic damping of the flapping motion. Similarly the downward flapping motion results in an increasing angle of attack, increasing the lift of the down-moving wing and again producing an aerodynamic damping of the flapping oscillation.
*A somewhat similar resultant velocity calculation to a vertical gust encounter.
As true airspeed increases the vertical component of a flapping wing's velocity becomes increasingly less significant, so the aerodynamic damping of the flapping oscillation lessens as the aircraft's TAS increases; consequently increasing TAS will increase flutter potential.
Where the aircraft designer selects a true air speed value as a limiting airspeed applicable when flying above a nominated altitude, FAR Part 23.1545 (c) requires that "If Vne varies with altitude, there must be means to indicate to the pilot the appropriate limitations throughout the operating altitude range". The 'means' is normally a placard next to the ASI. So, in such circumstances, designers must specify a series of Vne IAS values, corresponding with all possible operating altitude bands. For example, the RA-Aus registered Pipistrel Sinus has the altitude capability of 28 500 feet and can build speed rapidly even in a shallow descent. The Pipistrel designers have deemed it wise to limit maximum speed to a particular TAS at all altitudes. The following table reflects the Sinus flight manual and the ASI placard; the maximum true airspeed target is 122 knots. Note that density altitude rather than altimeter-indicated altitude, is specified — which is significant in Australian climatic conditions.
If there is insufficient manufacturer's information available for the aircraft you fly — and you are uncertain about the appropriate Vne for an operating altitude — then multiply the density altitude, in thousands of feet, by a factor of 1.5 to get the percentage decrease to apply to the specified Vne to establish a safe Vne appropriate to the altitude. For example if density altitude is 8000 feet and specified Vne is 100 knots then 8 × 1.5 = 12%. Corrected Vne = 88% of 100 = 88 knots IAS/CAS.
2.2 I like flying my aircraft fast. If I stay below Vne, I won't have to worry about structural failure, right?Vne is assessed at or near MTOW, with the cg at, or within, the fore and aft limits for the aircraft's specified category; it does not apply if weight, manoeuvring loads or cg positions are outside the specified limits. As a maximum airspeed it applies only in reasonably smooth atmospheric conditions for moderate, smoothly applied control movements and symmetrical aerodynamic loads. Even gusts associated with mild turbulence or sudden control surface movements greater than perhaps several degrees travel might lead to some unpleasant surprises, if operating close to but below Vne. Remember that dynamic pressure increases with the square of the true airspeed. At high speed the controls should be quite stiff and are very effective, with a probability of over-control applying extreme loads to the structures. Asymmetrical aerodynamic loads, such as combined rolling and pitching, reduce the maximum allowable airframe load by perhaps 30%. Take care because some aircraft control systems provide inadequate feedback of the load being exerted; i.e. a high load can be applied at high speed with a relatively low stick force, see 'The stick force gradient'.
(The effect of gust loads is expanded in the section on wind shear and turbulence.)
If an aircraft is operated within its specified manoeuvring and gust envelopes and weight and balance limits — observing the limiting accelerations and control movements, and maintaining airspeeds commensurate with atmospheric conditions — then the only possibilities of inflight control or structural failure relate to:
Be aware: deliberately exceeding Vne is the realm of the test pilot — who always wears a parachute! The following text is an extract from an RA-Aus accident investigation report illustrating aileron flutter and wing divergence:
"(Witnesses) observed the aircraft in a steep dive at what appeared to be full power. The port wing appeared to detach from the aircraft …
Design regulationsAircraft structures, engineered by aeronautical professionals, are designed for adequate strength and stiffness while being as lightweight as possible. To receive type certification the design of an aircraft must conform with certain standards, including the in-flight manoeuvring loads plus the turbulence or gust-induced loads that the structure must be able to sustain for the category in which the aircraft may be operated. Even if not seeking certification the designer would still conform to the standards, but this may not apply to those aircraft not designed/engineered by professionals.
Even design by professionals may not provide a guarantee that the aircraft is safe. Read this United States Federal Aviation Administration special review team report [pdf format] which identified issues with a LSA category aircraft's wing structure, flutter characteristics, stick force gradients, airspeed indicator calibration, and operating limitations.
FAR Part 23 is a recognised world standard for light aircraft certification and the following is an extract:
"... limit loads [are] the maximum loads to be expected in service [i.e. the highest load expected in normal operations] and ultimate loads [are] limit loads multiplied by [a safety factor of 1.5]. [FAR Sec. 23.301] … The structure must be able to support limit loads without detrimental, permanent deformation. At any load up to limit loads, the deformation may not interfere with safe operation. … The structure must be able to support ultimate loads without failure for at least three seconds …" [FAR Sec. 23.305]
In FAR Part 23 the minimum limit load factors that an aircraft must be designed to withstand at maximum take-off weight are:
There is an increasing risk of structural failure when exceeding the limit load factors, and each instance of excessive loading will accumulate airframe strain and add to the failure risk. See 'Metal fatigue in airframes'. We use load factors in terms of both g and total wing loading.There is an amplification of this relating to gust-induced loads, rather than manoeuvring loads, in the article 'Wind shear and turbulence'.
Remember that aerodynamic forces increase with the square of the airspeed; i.e. dynamic pressure = ½roVis². An increase in IAS from 125 knots to 150 knots represents a 44% increase in dynamic pressure.
Notes: 1. Uncertificated minimum recreational light aircraft, even with their low wing loading, can readily be overstressed just by flying straight and level at maximum speed and increasing load in a pull-up (positive g) or a full push-over (negative g).
2. Many GA aircraft are type certificated in both normal and utility category, and some are certificated in those plus the acrobatic category. In such cases the MTOW, cg limits and limit load factors are not fixed values but vary according to the intended flight operating category.
Airframe elasticityAll aircraft structures exhibit some degree of elasticity. That is, they deflect a little, changing shape — flexing, bending and/or twisting — under applied aerodynamic loads. Those short-lasting structural distortions also contribute to a change in the aerodynamic forces, so the distortions and forces are mutually dependent. This is particularly so with the wings, tailplane and control surfaces. However, structures usually spring back to the normal position when the load is removed. Aeroelasticity may lead to some problems at high speed, but reducing elasticity means increasing rigidity, which perhaps involves an unwarranted increase in structural weight. So, aircraft structural engineering must be a compromise between rigidity and elasticity. See the notes on 'stress and strain' in the 'Builders guide to safe aircraft materials'.
The 'clean' (i.e. flaps/spoilers stowed) three-axis aircraft can be flown within the speed and load limits of the light green region at any time. It is only possible to manoeuvre a light aircraft in the reduced-g band between +1g and –1g (light blue) for seconds rather than minutes. Controlled flight at speeds less than the Vs1g stall speed may be accomplished with any manoeuvre that 'unloads' the wings; for example, 'push-overs', which reduce apparent weight (make you feel light in the seat). Aerobatic aircraft can be pulled into a full-power vertical climb where the aoa is held close to the zero lift (zero load) aoa until the airspeed drops close to or below Vs1, then rudder and the slipstream energy is used to cartwheel the aircraft through a 180° hammerhead turn into a vertical descent. And of course an aerobatic aircraft would be able to sustain 1g negative (i.e. inverted) flight for a period.
The aircraft will stall if flight is attempted outside those aerodynamic load/speed limits defined by the curved lines. It can be operated above the Va manoeuvring speed without limits on smooth control use, and within the olive-green area in light to moderate turbulence, but it should not be operated above Vno (in the yellow area) except in a reasonably smooth atmosphere.
If operating in the region outside the structural load limits, or at velocities greater than Vne/Vd, structural distortion then failure may result.
All the foregoing is only applicable to a totally airworthy aircraft. If the airframe is not always properly maintained then the design manoeuvring flight envelope is not applicable; nor is it applicable if aerobatics are performed in an aircraft that is not certified for aerobatics.
The following are extracts from a report concerning certain engineering aspects of a fatal accident involving a Skyfox CA22. The aircraft had taken off from an airfield some 20 km from the accident site. The aircraft was seen to break-up in flight while overflying the pilot's house. The port aileron (or a portion of it) and the port wing were seen to detach from the aircraft and descend separately and relatively slowly. The fuselage with the starboard wing attached struck stony ground at high speed.
Conclusions: "The most probable primary cause of failure was exceedance of the aircraft's structural design envelope, primarily in regard to speed in conjunction with negative load factor due to a gust, leading to compression failure of the forward strut. Aileron flutter, due to an out-of-balance condition, may have been a factor. It seems probable that the aircraft was flying close to, or above, its Vne of 93 knots. The permissable flight envelope is very small, and would not be at all difficult to exceed inadvertently, especially in a shallow descent."
Also, read the Coroner's findings in regard to a double fatality following the inflight structural failure of a Drifter aircraft.
FlutterWing structures are akin to a very-low-frequency tuning fork extending from the fuselage. When a tuning fork is tapped, the fork vibrates at a particular frequency; the stiffer the structure, the higher its natural frequency. The natural frequency of a wing or tailplane structure may apply another limiting airspeed to flight operations related to a self-exciting interaction between elastic, aerodynamic and inertia forces that result in 'flutter' of control surfaces and the structure to which the surface is attached.
For example, when the airflow around a wing, tailplane or control surface is disturbed (by aerodynamic reactions, turbulence or pilot inputs) the structure's elastic reactions – twisting and bending – may combine as an oscillation or vibration of the structure that will quickly damp itself out at normal cruise speeds because of the structure's resistance. It is possible that the oscillation does not damp out but is sustained at a constant amplitude (perhaps felt in the airframe as a low-frequency buzz) that is not, in itself, dangerous but may contribute to structural fatigue. At some higher airspeed — the critical flutter speed, where the oscillations are in phase with the natural frequency of the structure — the oscillations will not damp out but will become resonant, rapidly increasing in amplitude. (Pushing a child on a swing is an example of phase relationships and amplification.) This flutter condition is very dangerous, and unless airspeed is very quickly reduced, the increasing aerodynamic forces will cause control surface (or even wing) separation within a very few seconds.
For more information on flutter see 'Aerodynamic reactions to flight at excessive speed'.
Twisting the wings off!Wing divergence refers to a state where, at the very low angles of attack at high speed where the nose-down pitching moment is already very high, pressure centres develop pushing the front portion of the wing downward and the rear portion upward. This aerodynamic twisting action on the wing structure — while the rest of the aircraft is following a flight path — further decreases the aoa and compounds the problem; finally exceeding the capability of the wing/strut structure to resist the torsional stress and causing the wing to separate from the airframe — with no warning! This could be brought about if turbulence is encountered at high speed.
Many of the uncertified minimum ultralights, and perhaps some of the certificated aircraft, have low torsional wing rigidity that will not only make the ailerons increasingly ineffective with speed (and prone to flutter), but also will place very low limits on Vne and allowable wing loadings. Vne may be so low that it can be readily achieved in a shallow descent at normal cruise power. The problem is that in some home-built aircraft Vne may not be known and could be unexpectedly low!
Wing washout: handy at low speed, not so good at high speed!Wings incorporating geometric washout have a significantly lower aoa towards the wing tips. At high speed when the wing is flying at low aoa there are high aerodynamic loads over the wings. But, the outer sections could well be flying at a negative aoa and the reversed load in that area — or just a badly distributed load due to the wing shape — will bend the outer wings down, possibly leading to outer spar fracture.
Note: it can happen to certified GA aircraft; two recent (Victoria 2007 and Tasmania 2004) high-speed crashes of Shrike Commander aircraft both exhibited simultaneous negative load failure of both main wing spars at their outer splice joints. This aircraft incorporates 6.5° washout.
The atmosphere will demonstrate how puny you are: vertical gust shear and gust loadsThe effective angle of attack of an aircraft encountering an atmospheric gust with a significant component parallel to the aircraft's normal axis (updrafts, thermals, down-currents, downdrafts, microbursts, macrobursts and lee waves) will be momentarily increased if the air movement is upward relative to the aircraft's flight path, or momentarily decreased if the air movement is downward. Thus an updraft will increase CL and lift causing an upwards acceleration of the aircraft, the magnitude of which is largely determined by the aoa change, the aircraft speed (higher speed — greater g load), the design wing loading and the aspect ratio. The higher the aspect ratio, the greater the change in load factor for a given increase in aoa and the easier it is to overstress the wings at high speed.
The effects of shear and gust loads are expanded in the section on wind shear and turbulence.
And there are other effects to think about!It is not just the preceding items that may provide problems at high speed. The maximum speed may be limited by the ability of the fuselage to withstand the bending moments caused by the loads on the tailplane necessary to counter the wing's high nose-down pitching moment at very low aoa.
Also when nearing the zero-lift angle of attack in a high-speed descent, many cambered wings suddenly experience a very strong nose-down pitching moment and the aircraft will 'tuck under' or 'bunt' rapidly. This instability will certainly make any pilot wish they had not been flying so fast. The symmetrical aerofoil wings often used in powered aerobatic aeroplanes don't have this problem.
Also the possibility of a runaway propeller in a fast dive is always there for those aircraft with a constant-speed propeller governor. There is nothing much you can do about that except close the throttle and reduce to minimum flight speed by easing the nose up. Once everything is settled down fly slowly, consistent with the default fine pitch stop blade setting, to a suitable airfield using minimum throttle movements.
Another problem is the possibility for extreme loads to be applied in a high-speed pull-up.
You can also induce structural damage at moderate speeds!Excessive speed is not always a factor in an aircraft structural failure. In Britain, June 2007, a 900-hour Europa Classic (a type that is represented in the RA-Aus aircraft register) suffered an in-flight break-up. Witnesses said the aircraft had been flying normally but then the tailplane started to make significant up-and-down movements. Then the horizontal stabilisers detached from the rear fuselage, and the wings folded up before separating from the aircraft. The engine stopped and the fuselage plummeted to the ground. The primary cause was probably tailplane flutter, possibly initiated by excessive play developing between the stabiliser torque tube and a mass balance arm.
Also, for example, mishandled manoeuvring of weight-shift aircraft can lead to a very fast-acting and uncontrollable pitch autorotation or tumble that imposes extreme transient loads on the structure. The following is a condensed version of an Australian Transport Safety Bureau 'Technical Analysis Occurrence Report' into three fatal accidents.
Note: the Coroner's findings in relation to the fatal accident near Atherton does not support any view that the accident was caused by pilot mishandling; rather, the Coroner's "preference is towards port side wing tip separation as a consequence of the un-airworthy state of the aircraft ..."
An Airborne Edge trike impacted terrain near Atherton, Qld during a 2005 flight. In 2006 a similar Airborne Edge aircraft impacted terrain at Cessnock, NSW. In both instances, RA-Aus initiated safety investigations during which similarities in the structural failures of both aircraft were observed. In addition, a third accident occurring in 1996 near Hexham involving an HGFA registered Airborne aircraft with similar structural failure was identified.
Recovery from an erect diveGenerally excessive speed can only build up in a dive, though just a shallow descent can build speed — and rate of descent — quite quickly, particularly in aircraft with high L/D.
Table 2 is a calculation of the rate of descent after a few seconds at dive angles of 10°, 30° and 45° — for a moderately slippery light aircraft. It can be seen that even a non-turbocharged aircraft entering a 10° or 15° descent from 8000 feet or so could quickly be exceeding Vne.
Recovery from an inadvertent venture into the realm of flight near, or even beyond, Vne is quite straight-forward but requires pilot thought and restraint when initiating the recovery procedure, particularly so if the aircraft is turning whilst diving. Considerable height loss will occur during recovery, so restraint is required when the ground is rapidly expanding in the windscreen.
A problem with this procedure is that most light aircraft do not have an accelerometer fitted, so it is difficult to judge the g being pulled. However, if properly executed 60° steep turns are practised then you can gain some idea of the 2g load on your own physiology. At the higher end, the average fit person will probably start feeling the symptoms of greyout by 4g.
Recovery from a spiral diveIn a well-developed steep spiral dive, the lift being generated by the wings (and thus the aerodynamic loading) to provide the centripetal force for the high-speed diving turn, is very high. The pilot must be very careful in the recovery from such a dive, or excessive structural loads will be imposed. If back elevator force is applied to pull the nose up while the aircraft is turning, the result will be a tightening of the turn and rapid increase in rate of descent — thus further increasing the aerodynamic loading or possibly prompting a very high speed stall.
Reducing power and levelling the wings must start first, with the rudder and elevators held in the neutral position. As the wings become level with the aircraft still diving at high speed, all the lift that was providing the centripetal force may now be directed vertically (relative to the horizon) and if up elevator is applied the aircraft may start a rapid high g pitch-up — even into a half loop. Thus to prevent this the pilot must hold the elevators in the neutral position while rolling level and even be prepared to start applying forward stick pressure even before the wings become level.
Remember: the theme common to all problems encountered when moving at very high speed is that there is little or no warning, and little time to do anything about it! The ONLY safe procedure is not to push the high end of the envelope at any height, don't exceed Vno if the atmosphere is exhibiting any other than light to moderate turbulence, and keep the aircraft airworthy.
The next article in this series discusses the stall/spin accident.
'Decreasing your exposure to risk' articles
| Introduction and contents | Recent RA-Aus accident history | Don't fly real fast | Don't stall and spin in from a turn |
| Don't land too fast in an emergency | Engine failure after take-off | The turn back: possible or impossible — or just unwise? |
| Wind shear and turbulence |
Copyright © 2007–2014 John Brandon [contact information]