Minimum control speeds: how testing generates the manual figures

Bord de piste avec caméra de suivi sur trépied et mât météo portable dans la brume du soir

The takeoff data sheet outputs a V1 speed that nothing in the available runway length justifies: the aircraft is light, the runway is long, the day is cool, and yet the decision speed remains stuck to a floor that no longer moves. This floor has a name, VMCG, and it comes from a test campaign conducted years earlier on a test runway, with a pilot cutting a full-throttle engine and only allowed the rudder pedals. Between that runway and your speed chart, there is a chain of measurements, extrapolations, and corrections that the flight manual never shows. It is this chain that is worth reconstructing, because it explains why the numbers behave as they do.

Three tests, three success criteria

The family has unity only in the acronym. Three speeds bear the name of minimum control speed, and each answers a distinct question, in a distinct configuration, with a success criterion that is not of the same nature.

VMCG addresses ground roll. After a sudden and complete failure of the most unfavorable engine, the crew must keep the aircraft on the runway using only the rudder: steering the nosewheel is prohibited, differential braking is also, and no contribution from tire friction is expected. The criterion is not a balance, it is a deviation: the trajectory must not stray from the centerline by more than a bounded value, thirty feet in the specifications applicable to transport aircraft. The rudder pedal effort is also capped by the text: it is forbidden to compensate for a lack of rudder authority with the pilot’s strength, otherwise the demonstrated speed would depend on the crew’s stature more than on the aircraft.

VMCA addresses free flight. Out of ground effect, gear retracted, in takeoff configuration and at maximum available thrust, the aircraft must maintain a straight and level flight with a bank angle not exceeding five degrees toward the operating engine. The text adds a ceiling relative to the reference stall speed to prevent an aircraft well-equipped with a vertical stabilizer from having a VMCA higher than its useful lift speed. This is supplemented by a dynamic demonstration: after a sudden cutoff, recovery is made without a dangerous attitude and without letting the heading slip beyond a few degrees.

VMCL shifts the problem to the approach and go-around. Landing configuration, go-around thrust on the remaining engine(s), and a criterion that is no longer static at all: the aircraft must be able to bank twenty degrees toward the failed engine side in five seconds or less. It is no longer the yaw that commands the result, it is the roll. At low speed, once the rudder is fully used to maintain symmetry, the residual roll authority — ailerons and spoilers combined — becomes the limiting factor.

The three demonstrations and what they bound
Vitesse Configuration d’essai Pass criterion What it limits in the flight manual
VMCG On the ground, take-off run, rudder only, nosewheel free Lateral deviation from the centreline bounded, rudder pedal force capped V1 floor
VMCA In flight, out of ground effect, take-off configuration, maximum thrust Stabilised straight flight, bank angle no more than five degrees; dynamic recovery without excessive heading change VR and V2 floors
VMCL Landing configuration, go-around thrust Ability to roll twenty degrees away from the failed engine within a set time Floors for approach and approach climb speeds

A nearly constant moment against a moment in the square of the speed

The mechanism behind the rule lies in a comparison of two yaw moments. On the asymmetry side, we find the net thrust of the remaining engine(s) multiplied by its leverage arm, increased by the drag of the lost engine — propeller windmilling or idling fan, plus installation drag. This sum depends little on speed: it decreases slowly as speed increases, due to the effect of ram drag.

On the authority side, we find the moment produced by the rudder: dynamic pressure, vertical stabilizer area, leverage arm, and deflection effectiveness. This term is proportional to air density and the square of true airspeed, hence, in practice, to dynamic pressure. A nearly horizontal line on one side, a parabola on the other: their intersection is the minimum control speed. Everything else follows from this.

Intersection of the asymmetric moment and the control momentTwo curves plotted against speed. The asymmetric moment due to thrust forms an almost horizontal line. The moment available from the rudder increases with the square of the speed. Their intersection defines the minimum control speed: below it, the rudder can no longer balance the asymmetry.yawing momentvitesseasymmetry (thrust)rudder, in V²VMC
The minimum control speed is an intersection point, not a structural limit. The solid line, in speed squared, is what the fin can produce; the almost flat dashed line is what the asymmetry demands. Move either one and the point moves: that is the whole sensitivity of VMC to conditions.

Imagine two forces fighting for control of the plane after an engine fails. One force comes from the remaining engine’s thrust trying to turn the plane, and the other comes from the rudder trying to keep it straight. The minimum control speed is the slowest speed where the rudder can still win this fight. If you fly slower than this, the rudder can’t overcome the engine’s thrust, and the plane will start turning dangerously.

On a Boeing 737-800 with one engine failed at takeoff, the VMCG might be around 60 knots. If you’re trying to take off from a long runway on a cool day with a light aircraft, the optimal V1 speed might be 55 knots. But since 55 is below the 60-knot VMCG floor, the aircraft’s computer will raise V1 to 60 knots to ensure you can still control the plane if the other engine fails.

This reading explains at once several behaviors that are observed without always linking them. The minimum control speed varies as the square root of thrust: that is why reducing the thrust of the live engine is the fastest way to move the boundary, and why reduced takeoff thrust lowers VMC. It degrades with an aft center of gravity, which shortens the leverage arm of the vertical stabilizer. It depends on the bank angle, because a few degrees toward the live engine engages a lateral component of weight that relieves the rudder — hence the care with which the text limits this bank angle to five degrees, otherwise the published figure would have no comparative meaning.

The critical engine, finally, is not always the one you think. On a symmetric twinjet, the two engines are in principle equivalent: the demonstration is then done in both directions, and the most penalizing value is retained, with any differences due to installation details or systems rather than aerodynamics. On a conventional turboprop, the helical slipstream and the descending blade effect make a failure significantly more penalizing than the other, and both cases are demonstrated.

On the ground, it is not a balance: it is a trajectory

This is the point that is most often overlooked. VMCA is a balance problem: we seek the speed at which the rudder, at its stop or effort limit, exactly balances the asymmetry. VMCG is not of this nature. The criterion is on a cumulative lateral deviation, i.e., an integral: the aircraft drifts while the pilot recognizes the failure, then while applying the rudder, then while the yaw nullifies and the trajectory straightens. The maximum deviation reached therefore depends as much on the assumed reaction time and the speed of rudder application as on the available authority.

Lateral deviation after an engine failure on the take-off runLateral deviation from the runway centreline as a function of time, after an engine cut. An initial plateau corresponds to the recognition delay, then the deviation grows, reaches a maximum and is recovered. At the minimum control speed on the ground, the maximum just touches the permitted limit. At a lower speed, it exceeds it.lateral deviationtempscoupure moteurfull rudderpermitted deviationpath at VMCGbelow VMCG
VMCG is read from a trajectory, not from a balance of forces. The initial plateau is the recognition delay: everything that builds up during it is lost for what follows. That is why the value used depends as much on the reaction time convention as on the fin area.

The practical consequence is twofold. First, a modification that speeds up recognition — a better-placed alarm, an autopilot that was not there — does not change the aerodynamics but can change the number. Second, the demonstration is inherently dispersive: two pilots, two rudder applications, two deviations. This is why modern campaigns rely on a simulation model validated by a limited number of real test points, rather than the direct search for the limiting speed on the runway, which would involve deliberately leaving the centerline at low speed to find the edge.

In flight: what the static demonstration leaves out

The published VMCA is obtained in steady state, aircraft already in stabilized sideslip, rudder at stop or effort limit, bank angle at five degrees toward the live engine. It is an existence boundary: below it, no straight and level equilibrium state exists. It says nothing about the ease with which it is reached or what is needed to return to it.

The transient is significantly more demanding. At the moment of the failure, the aircraft yaws, the sideslip sets in, the induced roll follows, and it is necessary to both stop the movement and rebuild the balance. The speed at which this recovery remains manageable is higher than the static value, and the text implicitly recognizes this by imposing a behavior criterion at the sudden cutoff in addition to the balance criterion. The margin between the two is not published in the flight manual: it is absorbed by the ratios imposed between VMCA and operational speeds.

The trap: treating the VMCA in the manual as a safety speed. It was demonstrated with five degrees of bank toward the live engine, in steady state, out of ground effect, and at unfavorable center of gravity. Wings level, the same situation requires more rudder: the boundary rises. A flight director that commands wings level, or a pilot who corrects roll as a priority, thus approaches the edge without the airspeed indicator signaling it. The remedy is not to fly faster afterward, it is to maintain the compensation bank angle prescribed by the operator’s procedure.

Measuring without going to the stop: extrapolation

VMCA is not determined by flying until loss of control. The method consists of constructing a family of asymmetric equilibrium points in a safe domain, then properly extrapolating. At fixed thrust and for several successive speeds, the aircraft is stabilized in straight flight at controlled bank angle and the required rudder deflection and pedal effort are recorded. Since the rudder moment varies with dynamic pressure and the asymmetry varies little, the required deflection is very close to an affine function of the inverse of the square of the speed. The curve thus straightens into a line, and the line is extended to the rudder stop or effort limit.

Extrapolation of rudder deflection to the stopRequired rudder deflection plotted against the inverse of the square of the speed. The points measured in flight line up on a straight line. Extending this line meets the deflection stop: the abscissa of this point gives the minimum control speed without having to reach it in flight.required deflection1 / V²measured pointsextrapolationdeflection stopVMC
The published speed is most often a calculated intersection, not a flown point. Plotting deflection against the inverse square of the speed gives a straight line that is extended to the stop. A few dynamic points, closer to the edge, are then used to check that the extrapolation is not misleading.

The instrumentation that makes this maneuver usable is heavier than it seems: actual rudder position and not pedal command, effort measured at the pedal head, angle of attack and sideslip measured by the test probe, attitudes and bank angles from the inertial platform, and above all a thrust reconstruction. Because thrust is not measured in flight: it is calculated, from the recorded engine parameters and air intake conditions, using a model engine calibrated on the test bench. The quality of VMC does not exceed the quality of this model.

Bringing back to reference conditions

No test point is flown exactly in the conditions that the manual must cover. The reduction work therefore consists of transporting each measurement to a set of unfavorable reference conditions, then publishing only the result of this transport. This is where the figures are actually made.

A pilot in a helmet and headset is seated in the cockpit of an aircraft, holding the controls and looking out at a runway during low-light conditions.
Testing ensures safe minimum control speeds for aircraft.
What shifts the minimum control speed, and in which direction
Parameter Effect on VMC Mechanism Treatment in data reduction
Thrust of the live engine Increases with thrust Asymmetric moment proportional to net thrust Corrected to the maximum thrust of the certified rating considered
Centrage Increases as it moves aft Shorter fin moment arm Corrected to the aft CG limit
Inclinaison Decreases when banking towards the live engine Lateral component of weight relieving the rudder Corrected to the reference bank angle in the regulation
Mass (at a given bank angle) Decreases as mass increases Larger lateral component of weight Corrected to a penalising reference mass
Altitude and temperature Roughly constant while thrust is flat-rated, then decreasing VMC in calibrated airspeed follows the square root of available thrust Taken at the most penalising condition of the envelope
Wind and runway slope (on the ground) Bias on the measured lateral deviation Crosswind component and natural drift Tests in reciprocal directions, average of both

The case of altitude and temperature deserves to be seen in image, because it contradicts intuition. One expects a minimum speed to increase with altitude, as does a true airspeed stall speed. However, expressed in indicated airspeed, VMC follows only thrust: the dynamic pressure required to balance a given moment does not depend on air density, only on the moment to be balanced. As long as the engine is at its rated thrust, VMC does not move. As soon as the thrust decreases, VMC decreases with it.

Effect of thrust flat-rating on the minimum control speedMinimum control speed expressed in calibrated airspeed, as a function of outside air temperature. While thrust remains flat-rated at its nominal value, the curve forms a horizontal plateau. Beyond the kink, available thrust decreases and the minimum control speed decreases with it. Without flat-rating, it would be higher in cold weather.VMC conventionnellesans plafonnementpublished VMCdecreasing thrustflat-rating kinkoutside air temperature
The plateau comes from the engine, not from aerodynamics. A flat-rated engine delivers the same thrust over the whole cold range: the minimum control speed stays constant there in calibrated airspeed. The kink in the flight manual table is therefore the signature of the engine rating, which explains why it moves from one certified rating to another.

Two other corrections deserve to be known. On the ground, the points are flown in reverse, in both runway directions, and averaged: this is the only honest way to eliminate the residual crosswind component and slope. In flight, the demonstration is done out of ground effect, because the proximity of the ground modifies the flow over the vertical stabilizer and would make the result dependent on the exact height. The manual does not publish what was measured, but what was measured and then transported to the worst corner of the authorized domain.

From the plot to the speed chart

Once the minimum control speeds are established, the text imposes ratios that operational speeds must respect. V1 cannot be less than VMCG, otherwise the continuation of the takeoff would assume a lateral control that has not been demonstrated. VR has at least five percent margin over VMCA, and V2 at least ten percent: these margins precisely cover the gap between the demonstrated static boundary and the actual handling of a transient. On the landing side, approach and missed approach speeds maintain a fixed margin above VMCL.

It is through this mechanism that VMCG appears in a routine takeoff data sheet. At low weight and high thrust — long runway, cool day, almost empty aircraft — the optimal decision speed, which equalizes acceleration-stop distance and acceleration-go-around distance, falls below the VMCG floor. The computer then has no choice: it raises V1 to the floor, which lengthens the acceleration-stop distance and may, on a short runway, cost takeoff mass even though the aircraft was light.

The VMCG floor moves the decision speed away from the optimumRunway length required as a function of decision speed. The accelerate-stop distance increases with this speed, the accelerate-go distance decreases, and their intersection gives the balanced speed. When the floor imposed by the minimum control speed on the ground lies to the right of this intersection, the decision speed must be raised and the runway length required increases.runway lengthV1 prohibited below VMCGplancher VMCGaccelerate-stopaccelerate-goadditional runwaybalanced V1V1
The lateral control floor takes precedence over the performance optimum. When VMCG falls to the right of the crossing of the two distances, the decision speed is raised as a matter of course and the runway required increases. This is the typical situation for light take-offs at full thrust in cold weather.

The chosen thrust regime changes everything, and this is where the two lightening techniques diverge. Reduced thrust obtained by fictitious temperature leaves the possibility of restoring full thrust at any time: the applicable minimum control speeds remain those of the fully certified regime. A certified reduced regime, on the other hand, constitutes a separate engine regime, with its own set of minimum control speeds, lower, and its own conditions of use. Both tables exist in the documentation and do not mix.

The right reflex: before reading a value in a minimum control speed table, first read its three headings — the thrust regime to which it is linked, the air intake configuration retained, and the speed unit used. Then check in the takeoff data sheet if the computer has flagged V1 as limited by VMCG. When this flag is raised, the stopping margin is no longer what the runway length suggests, and a mass reduction will improve nothing as long as the thrust regime does not change.

What remains to be verified in your own documentation

Three checks take half an hour and avoid false reasoning. The first: does your manual give tabulated minimum control speeds in altitude and temperature, or a single conservative value valid over the entire domain? In the second case, part of the performance has been traded for simplicity, and this is especially noticeable on hot or high-altitude runways, where the available thrust would have justified a lower figure.

The second: what air intakes does the table assume? Cutting off the intakes increases the available thrust, hence the potential asymmetry, hence the minimum control speed. An operating policy that allows takeoffs without intakes must be based on the corresponding table, not the other.

The third touches a gray area. The ground demonstration assumes that the only available lateral action is the rudder, which in principle makes it indifferent to the surface condition. But the lateral deviation allowed was observed on a dry runway, with normal lateral landing gear forces; on a contaminated runway, the way the aircraft follows the residual yaw is not exactly that of the demonstration. The operating data on contaminated surfaces fall under a different set of assumptions, and they cannot be recomposed on one’s own.

For those preparing a type rating, the point to remember is not the list of definitions but their nature: two of the three speeds are existence boundaries obtained by extrapolation, and the third is the result of a timed trajectory. The examiner who asks why V1 is blocked at the bottom of the table does not expect a definition, he expects this reasoning. The operational preparation modules of formations.fl330.com address this link between certification data and takeoff data sheet.

In case of discrepancy between the above and your reference documents, the applicable certification specifications, the approved flight manual of your aircraft, and the operating documentation validated by your authority prevail: this article exposes a mechanism, it does not replace any approved data.

Commentaires

Leave a Reply

Découvrez l’écosystème FL330

FL330, c’est toute l’aviation professionnelle — explorez nos autres espaces.