![a shows V 30 ft values of 97 kt at 0.2 s, 103 kt at 0.4 s, 111 kt at 0.6 s, and 124 kt at 0.8 s (NWS off). With NWS on (Fig. 6b), these values are, respectively, 82, 88, 98, and 117 kt. The model confirms that the reaction time is of great influence on V 30 ft. For evaluation of the effect of environmental conditions, the effective reaction time in the model was set to 0.5 s ([1] Par. II.H). The V 30 ft value at 0.5 s. in Fig. 6a is 107 kt, which is the certified V mcg value for the Boeing 737 − 300∕20 k.](https://www.researchgate.net/profile/Erik-Jan-Huijbrechts/publication/330722384/figure/fig2/AS:720589073420288@1548813231405/a-shows-V-30-ft-values-of-97-kt-at-02-s-103-kt-at-04-s-111-kt-at-06-s-and-124-kt-at_Q320.jpg)
Content uploaded by Erik-Jan Huijbrechts
Author content
All content in this area was uploaded by Erik-Jan Huijbrechts on Oct 01, 2019
Content may be subject to copyright.
Using Vmcg-Limited V1, Controllability Issues
on Contaminated Runways and in Crosswind
Erik-Jan A. M. Huijbrechts,∗Herman J. Koolstra,†and J. A. Mulder‡
Delft University of Technology, 2600 GB Delft, The Netherlands
DOI: 10.2514/1.C035222
Vmcg, or ground minimum control speed, is established by aircraft manufacturers during the aircraft
certificatio n process. Vmcg is used as a limi ting speed for V1(decision spe ed) when performing takeoff pe rformance
calculations. Performance calculations on contaminated and slippery runways will result in a V1speed equal to
Vmcg-limited V1for a wide range of takeoff weights when using aircraft manufacturer procedures in a flight crew
operations ma nual or computer calc ulations based on the V1−min policy. In thi s paper, it will be shown tha t Vmcg will
not be a safe speed to continue a takeoff after an engine failure in strong crosswind or reduced runway surface
friction conditions. A model is used to determine the effect of these environmental conditions on lateral deviation.
Apart from the continued takeoff, the lateral deviation in the rejected takeoff after an engine failure was also
calculated under different environmental conditions. This resulted in advice for the use of a differential braking
technique to prevent a runway excursion if a runway is not dry. A method to mitigate the risk of runway excursion
on contaminated and slippery runways is presented. An evaluation, conclusions, and subjects for further research
are also presented.
Nomenclature
er = right engine
F= force, N
gl = (main) left gear
gr = (main) right gear
Hdg = heading, deg
M= moment, N∕m
ng = nose gear
q= dynamic air pressure, N∕m2
r= yaw rate, deg ∕s
Trk = track
V= speed, kt∕m∕s
Vef = engine fail speed, kt
Vg= ground speed, kt
Vmcg = minimum control speed ground, kt
VR= rotation speed, kt
V1= decision speed, kt
V1MCG=Vmcg-limited V1,kt
V2= one-engine-inoperative climb speed, kt
V30 ft = engine failure speed that will result in a 30 ft
deviation from runway centerline, kt
W= weight, ton
β= aerodynamic sideslip angle, deg
βg= ground sideslip angle, deg
βt= tire sideslip angle, deg
μ= friction coefficient
μs= side force friction coefficient
ρ= air density, kg∕m3
ω= rotational speed of tire, deg ∕s
I. Introduction
FLYING in winter, a pilot will encounter situations with reduced
runway friction. Not all airfields clear their runways or, if they
do, freshly fallen precipitation will result in a contaminated runway.
On a contaminated (in particular, slippery) runway, perfor-
mance calculations using flight crew operations manual (FCOM)
procedures or computer performance based on the V1min policy (an
understandable choice for an operator regarding the recommenda-
tions from the industry) will result in a V1equal to Vmcg -limited V1
for the greater part of the takeoff weight. Computer performance
output will result in a V1even below the Vmcg-limited V1speed in the
FCOM and shows that, in most cases, there is excess performance
available.
The question arises as to whether a V1equal or close to the Vmcg-
limited V1is safe to continue a takeoff on a contaminated runway.
Vmcg is established with favorable environmental conditions (dry
runway, and no crosswind) with an uncoupled nosewheel steering
during the certification process of an aircraft.
A pilot will reject his takeoff after V1if a runway excursion is
imminent by continuing the takeoff. In these cases, a higher V1would
have been a better choice. Accident investigators are reluctant to
blame procedures. If a rejected takeoff initiated after V1leads to a
runway excursion, most probably, the crew will get the blame for not
complying with procedures. The causal procedure fault may remain
hidden.
Environmental conditions do affect Vmcg, as demonstrated by
simulated certification tests using a model developed in Ref. [1]. The
present procedures using Vmcg-limited V1as V1contain a part of the
takeoff roll in which an unacceptable lateral deviation or runway
excursion will occur after an engine failure. Also, the simulations
of rejected takeoff resulted in some interesting discoveries. The
simulation findings are summarized in the Evaluation section
(Sec. VI). Some recommendations to improve safety for takeoff on
contaminated runways are also presented.
II. Takeoff
A. Takeoff Performance and Speeds
Takeoff performance calculations must assure the aircraft will
reach a certain screen height at the end of the runway and clear
obstacles lying ahead of the runway when an engine failure is
experienced. They must also assure the aircraft can be stopped on the
runway in case the takeoff is rejected because of an engine or other
failure.
Received 21 August 2018; revision received 28 November 2018; accepted
for publication 2 December 2018; published online Open Access 24 January
2019. Copyright © 2018 by Delft University of Technology. Published by the
American Institute of Aeronautics and Astronautics, Inc., with permission.
All requests for copying and permission to reprint should be submitted to CCC
at www.copyright.com; employ the eISSN 1533-3868 to initiate your request.
See also AIAA Rights and Permissions www.aiaa.org/randp.
*Captain Boeing 737, KLM Royal Dutch Airlines, P.O.Box 7700, 1117 ZL
Schiphol; erik.huijbrechts@klm4u.com.
†Experimental Test Pilot/Researcher, Faculty of Aerospace Engineering,
Control and Simulation Division, P.O. Box 5058; H.J.Koolstra@TUDelft.nl.
‡Professor, Faculty of Aerospace Engineering, Control and Simulation
Division, P.O. Box 5058; J.A.Mulder@tudelft.nl.
1342
JOURNAL OF AIRCRAFT
Vol. 56, No. 4, July–August 2019
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
The procedures used by flight crew are incorporated in the flight
crew operations manual. Calculation methods are based on runway
surface condition. Different methods are used for dry, wet, and
contaminated runways. For dry or wet runways, a reduced thrust
setting can be selected. For contaminated runways, rated (or fixed
derated) thrust must be used. Factors that are considered are weather,
aircraft, and runway related: e.g., air pressure, anti-ice systems used,
and accelerate–stop distance available. The head wind or tailwind is
accounted for; the crosswind, however, is not considered for
performance or takeoff speeds [2].
Performance calculations result in a performance-limited
takeoff weight (PLTOW), an assumed temperature (if reduced
thrust is allowed), and takeoff speeds at the selected takeoff
weight (TOW).
The takeoff speeds are 1) V1(decision speed), 2) VR(rotation
speed), and 3) V2(initial climb speed with one engine inoperative).
In the past, “paper”calculations were made using takeoff weight
limitations: tables for the runway concerned, and tables to calculate
the speeds. Nowadays, most performance calculations are performed
by onboard (electronic flight bag) or home-based (accessible via an
aircraft communications addressing and reporting system or satellite
communications) computer programs.
Performance on dry runways is based on certified performance data
delivered by the manufacturer in the airplane flight manual (AFM).
Wet runway performance can be found in the performance engineers
manual (PEM) or, for newer aircraft, as certified performance in the
AFM or the AFM’s digital performance information (AFM-DPI).
Performance for aircraft on contaminated runways is based on
advisory information in the PEM or AFM. The research and
calculation methods for this information date back to the 1960s.
In 2006, European regulations had a major revision in the
acceptable means of compliance (AMC) with guidance material on
the calculation method for contaminated runway performance
including hydroplaning [3].
B. V1(Decision Speed)
Before every takeoff, a single value for V1is established. Below
V1, the takeoff can be rejected. At speeds higher than V1, takeoff must
be continued. A pilot will lift his hand from the thrust levers to the
yoke at V1to continue the takeoff. Elaborate discussions about the
go/no-go decision can be found ([4] Par. 2.2).
Decision speed is too short a description of V1. The Federal
Aviation Administration (FAA) definition of V1is as follows:
The maximum speed in the takeoff at which the pilot must
take the first action (e.g., apply brakes, reduce thrust, deploy
speed brakes) to stop the airplane within the accelerate–stop
distance. V1also means the minimum speed in the takeoff,
following a failure of the critical engine at Vef , at which the
pilot can continue the takeoff and achieve the required height
above the takeoff surface within the takeoff distance [5].
On the low side, V1speed is limited by the ground minimum
control speed Vmcg [6] CS25.107, [7] FAR25.107. On the high side,
V1is limited by VR(at TOW). Performance calculations for
continued and rejected takeoff limit V1as a function of TOW.
For dry and wet runways, V1is mostly calculated to match
accelerate–stop and accelerate–go distances. This is called the
balanced field principle ([2] Par. 3.1.5, [4] Par. 2.3.1.3). If excess
performance is available, reduced thrust can be used to save on engine
life. Dry runway calculations typically result in a V1equal, or close,
to VR. Wet runway calculations typically result in a lower V1with a
speed gap to VR.
Speed calculations on contaminated and slippery runways based
on manufacturer FCOM procedures will result in a V1speed equal to
Vmcg-limited V1for a wide range of takeoff weights, even if this is
not required to meet accelerate–stop performance. Standardized
computerized aircraft performance (SCAP) software offers the
option to calculate the minimum V1and maximum V1. Some
operators choose to use the V1−min option for their operation,
assuming these speeds are safe.
Figure 1 shows a typical diagram for a contaminated runway. At
the lower side, V1is limited by Vmcg; at the high side, it is limited by
accelerate–stop performance. The V1in which both curves intersect
is called the balanced V1.AV1within the range complies with AFM
performance requirements.
C. Runway Surface Condition and Braking Action
The relation between the runway surface condition used for
performance calculations and the reported or measured braking
action (runway surface friction coefficient) of a runway is not clear.
Following a runway excursion after landing in 2005, the Federal
Aviation Administration instated the Takeoff and Landing
Performance Assessment Aviation Rulemaking Committee. This
committee released Safety Alert for Operators 06012 [8] and, later,
the paved runway condition assessment table (Table A1 in the
Appendix) [9]. These publications give better guidance for operators
and pilots to assess what performance calculations to use in specific
runway contamination situations. Some operators have incorporated
this guidance in the FCOM. The FAA and the European Aviation
Safety Agency (EASA) have devoted a lot of effort to harmonize
runway surface condition reporting and the methods to establish and
report measured braking actions [10].
D. Takeoff Safety
In the 1990s, the FAA, together with the aviation industry,
published the Take-Off Safety Training Aid [4]. A continued takeoff
was considered safer than a rejected takeoff (RTO). Research showed
that, in 58% of RTO accidents, the takeoff was rejected at a speed
above V1. In 24% of the accidents, engine problems played a role and,
in at least one-third of the accidents, the runway was wet or
contaminated ([4] Par. 2.2.4). The Training Aid aimed at emphasizing
to pilots not to reject a takeoff after V1. To reduce the number of
RTOs, V1should be reduced as much as possible.
In preparation for the revision of AMC 25-1591 [3], the British
Civil Aviation Authorities (U.K. CAA) have questioned the use of
Vmcg for contaminated runways and in crosswind conditions. The
Joint Airworthiness Authorities Flight Study Group expected some
effect of crosswind and little effect of runway surface conditions but
recommended to seek an improvement in the knowledge necessary to
account for Vmcg and crosswind effects on controllability when
operating on contaminated surfaces [11]. The dissenting opinion of
the U.K. CAA has led to the following statement: ”The provision of
performance information for contaminated runways should not be
taken as implying that ground handling characteristics on these
surfaces will be as good as can be achieved on dry or wet runways, in
particular following engine failure, in crosswinds or when using
reverse thrust”([3] Par. 8.1.3). However, Vmcg as established under
test conditions (dry runway, no crosswind, and free castering
nosewheel) was preserved as the minimum speed for V1.
Fig. 1 Typical V1envelope for a contaminated runway.
HUIJBRECHTS, KOOLSTRA, AND MULDER 1343
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
III. Dynamics of an Engine Failure
Apart from aerodynamic forces, ground forces play a role in the
dynamics of the takeoff roll. Lateral forces on a rolling tire are, for
given conditions, a function of the vertical force and the slip angle of
the tire [12]. The slip angle of an individual tire is the vector sum of
the slip angle of the aircraft related to the ground, the yaw rate induced
slip angle, and (if applicable) the steering angle of the tire. The
maximum side force coefficient (μsFY∕FZ) is dependent on the
runway surface condition and the velocity of the tire (see Fig. 2).
A. Before Engine Failure
Before an engine failure occurs, an aircraft tracks the centerline of
the runway. If crosswind is present, a certain rudder deflection will be
applied to counteract the weathervane and side force effect. An
equilibrium of moments and lateral forces acting on the gears and
aerodynamic surfaces will exist.
B. Initial Reaction
The initial reaction of the aircraft after a left engine failure will be a
heading change toward the failed engine causing a slip angle βgto the
right (see Fig. 3). This slip angle causes the main gear to generate a
side force for a track change toward the failed engine. The yaw rate r
will cause an additional slip angle on the nose gear to the left. The
aerodynamic slip angle βwill be dependent on crosswind but, as
compared to the previous situation, will counteract the moment
generated by the asymmetric thrust, as will the yaw rate induced
aerodynamic sideslip angle at the tail. The moments generated by the
gears will also counteract the asymmetric thrust moment.
C. Rudder Application
The pilot will react on the heading change of the aircraft and apply
the rudder to steer back to the centerline (see Fig. 4). The moment
generated by the rudder will counteract that due to the asymmetric
thrust. If nosewheel steering is coupled, it will support this
counteracting moment. If these moments are strong enough, the yaw
rate will be reversed. This will cause the slip angle βgto shift to the
left. Forces on the main gear will be reversed, and the aircraft will
return to the centerline. The aerodynamic moment and the moment of
the main gear forces are in the opposite direction of the rudder
moment. As the main gears are close to the center of gravity, the
contribution to the total moment about the center of gravity is
relatively small.
Aerodynamic forces are proportional to the dynamic pressure
(q1∕2ρV2) and are the dominant forces at high speeds. At low
speeds, the gear forces are dominant in the dynamics during the
takeoff roll.
IV. Minimum Control Speed on the Ground Vmcg
To certify an aircraft, the manufacturer has to run ground tests to
establish a value for Vmcg.
The following is the definition of Vmcg [6] CS25.149:
Vmcg, the minimum control speed on the ground, is the calibrated
airspeed during the take-off run at which, when the critical engine
is suddenly made inoperative, it is possible to maintain control of
the aeroplane using the rudder control alone (without the use of
nosewheel steering), as limited by 667 N of force (150 lbf), and the
lateral control to the extent of keeping the wings level to enable
the take-off to be safely continued using normal piloting skill. In the
determination of Vmcg, assuming that the path of the aeroplane
accelerating with all engines operating is along the centreline of the
runway, its path from the point at which the critical engine is made
inoperative to the point at which recovery to a direction parallel to the
centreline is completed, may not deviate more than 9.1 m (30 ft)
laterally from the centreline at any point (see Fig. 5). Vmcg must be
established, with 1) The aeroplane in each take-off configuration
or, at the option of the applicant, in the most critical take-off
configuration; 2) Maximum available take-off power or thrust on the
operating engines; 3) The most unfavourable centre of gravity; 4) The
aeroplane trimmed for take-off; and 5) The most unfavourable weight
in the range of take-off weights.
The same certification basis can be derived from Ref. [7]
FAR25.149 in combination with the flight test guidance [13].
As there is no requirement on the runway surface condition or wind
conditions, the Vmcg certification test is conducted on a dry runway in
calm wind conditions. The flight test guidance suggests that not using
the nosewheel steering compensates for the effect of a wet runway
surface [13]. Major aircraft manufacturers have summarized their
experience with Vmcg certification tests [14].
Fig. 2 Forces on a rolling tire. Adapted from Ref. ([12] p. 91, fig. 3.2).
Fig. 3 Initial reaction of aircraft after engine failure. Fig. 4 Situation after rudder application.
1344 HUIJBRECHTS, KOOLSTRA, AND MULDER
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
Vmcg is called an aerodynamic speed. This means that it is
expressed in indicated air speed. It does not imply that it is fully
determined by aerodynamic forces alone.
The lateral deviation of the aircraft from the centerline is largely
affected by the side forces between the main-wheel tires and the
runway. In fact, major manufacturers say the test is highly influenced
by the ground-to-tire reaction, and they advise using new tires to
perform these tests ([14] p. 3, 7).
Crosswind is another factor with a high influence on the Vmcg
certification tests ([14] p. 16).
Aircraft manufacturers use the Vmcg value, obtained by this test, to
calculate the Vmcg-limited V1[V1MCG] by applying the engine
failure recognition time (1 s). V1MCGis presented in the PEM, the
AFM [15], or the AFM-DPI for performance calculations and the
FCOM [16] for use by crews.
V. Modeling Results
A model of a Boeing 737-300 with 20 klbf rated engines is used
to evaluate the influence of the pilot reaction time, nosewheel
steering, and environmental factors, such as runway surface
condition and crosswind, on V30 ft [1]. This model was validated
by reproducing the certified Vmcg value by simulating the
certification test.
The following definition was made up for the present research:
V30 ft is the engine failure speed that will result in a 30 ft deviation
from the runway centerline.
Most figures show lateral deviation as a function of the engine
failure speed Vef .
The width of most runways is 45 m (150 ft). As a rule of thumb, a
lateral deviation of 30 ft or less can be considered safe. A lateral
deviation of 60 ft or more, with regard to the position of the main
wheels, can be considered a runway excursion.
The Vmcg-limited V1value for the Boeing 737 −300∕20 k
(standard conditions) is 111 kt [15,16]. The corresponding Vmcg
value is 107 kt. This can be used as a reference value for Vef.
NWS on refers to nosewheel steering (NWS) coupled to the rudder
pedals. NWS off refers to an uncoupled nosewheel steering modeled
as the absence of lateral forces on the nose gear.
Continued takeoff (CTO) was evaluated first. To make sure a
rejected takeoff (RTO) would not result in a larger lateral deviation
than a continued takeoff, the RTO was also evaluated.
A. Continued Takeoff
1. Reaction Time
Figure 6a shows V30 ft values of 97 kt at 0.2 s, 103 kt at 0.4 s, 111 kt
at 0.6 s, and 124 kt at 0.8 s (NWS off). With NWS on (Fig. 6b), these
values are, respectively, 82, 88, 98, and 117 kt.
The model confirms that the reaction time is of great influence
on V30 ft.
For evaluation of the effect of environmental conditions, the
effective reaction time in the model was set to 0.5 s ([1] Par. II.H). The
V30 ft value at 0.5 s. in Fig. 6a is 107 kt, which is the certified Vmcg
value for the Boeing 737 −300∕20 k.
2. Runway Surface Condition
Based on NASATechnical Paper 1080 [17], three runway friction
models were developed from measurements on dry, damp, and
flooded concrete runway surfaces. These models are called
NASA dry, NASA damp, and NASA flooded. These models can be
linked to, respectively, dry, wet, and contaminated runway surface
conditions ([1] Par. II.F).
Figure 7 shows that V30 ft is 107 kt on a dry surface, 113 kt on
surface damp (wet) runway surface, and 114 kt on a flooded
(contaminated) surface (NWS off). With NWS–on, these figures are,
respectively, 93, 109, and 113 kt.
The effect of nosewheel steering is considerable (14 kt) on a dry
surface and negligible on a flooded (low friction) surface. Runway
surface friction has a considerable influence on V30 ft. The
simulations show that the adverse effect of a wet runway or slippery
runway is worse than the adverse effect of an uncoupled nosewheel
steering.
3. Crosswind
All crosswind-related figures are calculated with a crosswind from
the right and a right engine failure (critical engine). (This may not be
correct for a B737-300, see Sec. VI.J for an explanation.)
Figure 8 shows that V30 ft increases from 107 to 113 kt with a 10 kt
crosswind component, to 118 kt with a 20 kt crosswind component,
and to 125 kt with a 30 kt crosswind component (NWS off). With
NWS-on, the numbers are, respectively, 93, 100, 108, and 116 kt. The
effect of crosswind turns out to be considerable. The model shows
NWS may compensate for the adverse effect of almost 20 kt of the
crosswind component on a dry runway.
Fig. 5 Maximum permitted lateral deviation during Vmcg certification test.
Fig. 6 Lateral deviation as functionof Vef . CTO with different reaction times(in seconds): a) NWS off, andb) NWS on. (Runway frictionmodel: NASA dry.)
HUIJBRECHTS, KOOLSTRA, AND MULDER 1345
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
4. Crosswind and Runway Surface Condition
For the next figures, friction coefficients μswere assumed to be
independent of speed ([1] Par. II.F). At a 100 kt ground speed, a μs
value of about 0.5 can be linked to a dry runway surface, of about 0.1
to a damp/wet surface, and 0.05 to a flooded/contaminated surface
([17] table I, [18] table II).
Figure 9 shows, surprisingly, that the crosswind has more
influence at high μsvalues. This is caused by the larger gear moments
requiring more rudder input to correct for crosswind. Based on the
model, a margin of 10 kt on Vmcg (and thus on V1MCG) would be
sufficient to remain below V30 ft for crosswinds up to 25 kt for all μs
values (NWS on).
B. Rejected Takeoff
The model was also used to see if lateral deviation is acceptable
when rejecting takeoff after an engine failure. For a reaction time, a
rudder input of 0.5 s was used. The reaction time before closing the
operating engine was 1 s ([4] Par. 2.3.1.2), and a 1 s linear decay to
idle thru st for the operati ng engine was assum ed. The deployment of
lift dumpers and three braking modes (no braking, symmetrical
braking, and differential braking 0.2 s after full rudder deflection)
were modeled. The differential braking mode changed into
symmetrical braking when rudder deflection was reduced. No data
were available for combined braking and lateral forces on dry
surfaces [17]. So, lateral deviation is only calculated for damp and
flooded surfaces with braking. The use of reverse thrust was not
modeled.
Figure 10a clearly shows that NWS is required to prevent a runway
excursion after engine failure at low speed. With NWS on (Fig. 10b),
the lateral deviation will still result in a runway excursion when the
runway is not dry at low speeds. Figure 10c shows that symmetrical
braking will not prevent a runway excursion at lower speeds; it may
even aggravate the situation. Differential braking can keep the aircraft
on the runway when the runway is not dry (Fig. 10d).
Fig. 7 Lateral deviation as function of Vef . CTO with different runway friction models: a) NWS off, and b) NWS on.
Fig. 8 Lateral deviation as function of Vef . CTO for different crosswind components: a) NWS off, and b) NWS on. (Runway friction model: NASA dry.)
Fig. 9 V30 ft as function of μs. CTO with different crosswind values: a) NWS off, and b) NWS on.
1346 HUIJBRECHTS, KOOLSTRA, AND MULDER
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
The lateral deviation at speeds around Vmcg is smaller in the RTO
than in continued takeoff (see Fig. 7).
Figure 11 shows that, in crosswind conditions, the lateral deviation
at speeds around Vmcg is still smaller than in continued takeoff.
Figure 11a also shows that, in crosswind conditions, a runway
excursion is likely to happen at low speeds when the runway is not
dry. Differential braking, however, can still keep the aircraft on the
runway (Fig. 11b).
Figure 12 shows that, in stronger crosswind conditions, the lateral
deviation at speeds around Vmcg is still lowerthan in continued takeoff.
Figure 12b shows that, in stronger crosswind conditions, differential
braking may not prevent a runway excursion on a contaminated
runway. Operators use constraints on themaximum crosswind allowed
in case of reduced runway friction coefficients [19].
VI. Evaluation
In this section, findings from the research are discussed. Some
findings have resulted in conclusions; other findings have raised
questions that require further investigation. We have added the
findings we want to share with the aviation community as food for
thought.
A. V1Policy in FCOM Procedures for Contaminated Runways
Reviewing FCOM performance calculation procedures on
contaminated runways for different (Boeing) aircraft reveals that
V1is reduced to Vmcg-limited V1[V1MCG] for a wide range of
takeoff weights [for some types/variants up to maximum takeoff
weight (MTOW)], even if this is not required for accelerate–stop
performance. Apparently, a V1−min policy is applied in the FCOM
procedures.
B. Transition from Paper Performance Calculations to Computer
Performance Calculations
Pilots trust their performance calculations and the procedures they
use from the FCOM. Whereas, in paper calculations, numbers are
always conservative due to the simplified presentation; computers
Fig. 10 Lateral deviation as function of Vef. RTO with different runway friction models: a) NWS off with no braking; b) NWS on with no braking;
c) NWS on with symmetrical braking; and d) NWS on with differential braking.
Fig. 11 Lateral deviation as function of Vef. RTO, NWS on, and 10 kt crosswind with different runway friction models: a) no braking, and b) differential
braking.
HUIJBRECHTS, KOOLSTRA, AND MULDER 1347
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
can calculate numbers accurately to legal or certification limits.
Computer Vmcg-limited V1values will be lower than paper values
because weight corrections can be accurately applied with the
AFM-DPI performance data. PLTOW values have increased with
computer calculations.
When calculation methods get to be more accurate, it is
important that the underlying legislation and calculation methods
are correct. An accurate calculation of an unsafe value does not
improve safety.
The transition from paper performance calculations to computer
performance calculations has reduced safety margins in operation.
This requires an evaluation of established procedures to assess if
the safety standards are still acceptable.
C. Training Simulators and Pilot Expectations
Observations by the authors in training simulators (KDC-10,
B737, and B777) showed less (only a few feet) lateral deviation
and better controllability after an engine failure around Vmcg than
can be expected from the certification test. A training simulator is
not suitable to evaluate the influence of environmental conditions
on the resulting lateral deviation after an engine failure. The
aircraft behavior after an engine failure around Vmcg has no priority
when accepting a simulator from a simulator manufacturer
because it is not a part of the qualification process of a training
simulator.
An engine failure just after V1is an often a trained event in
training simulators. It serves as an examination topic for aircrew
proficiency checks. The handling qualities of the simulator provide
confidence to pilots that they can handle the engine failure in the
aircraft. Within airline companies, simulator instructors are often
considered to be experts in aircraft handling; and sometimes the
training simulators are used to solve handling questions. Pilots are
not used to experience a l ateral deviation of abo ut 30 ft. They will be
surprised by the aircraft behavior when a real engine failure occurs,
even if there are no adverse environmental conditions. The
benign reaction of training simulators has probably masked the
controllability problems at speeds around Vmcg after an engine
failure to pilots and instructors.
The quality of the ground model in training simulators should be
investigated in order to establish if the lateral deviations and
controllability on dry, wet, and contaminated runway surfaces are
realistic, also under crosswind conditions.
D. Runway Friction Coefficients and Ground Speed
The runway friction coefficients given in the paved runway
condition assessment table (Table A1; Appendix) are fixed values.
These values can be linked to friction measuring equipment. Pilots
use the paved runway assessment table to find a calculation method
for the takeoff ahead.
NASA reports [17,18,20] show a significant speed influence on
runway friction coefficients (braking actions) on damp and flooded
runway surfaces. The reports show that runway friction coefficients
dropfromaround0.6atlowspeedtoaround0.2at100kton
damp runway surfaces and 0.1 on flooded runway surfaces. Also,
AMC 1591 ([3] Par. 7.3.1) gives a speed-dependent runway friction
coefficient on wet runway surfaces to be used for performance
calculations. Braking actions on snow and ice do not show a high-
speed dependency, but they have a low value through the whole
speed range [20].
Measured lateral friction coefficients show the same speed
dependency on damp and flooded surfaces; they drop to around and
below 0.05 on flooded runway surfaces around 100 kt ([17] table I,
[18] table II). No data can be found for lateral friction coefficients on
snow- and ice-covered runway surfaces.
The conditions described in these NASA reports as damp are
linked to a wet runway surface and flooded to a contaminated runway
surface in the model.
E. Aircraft Tire Characteristics
Aircraft tires cannot be compared to car tires. Bias-ply tires are still
common, and the tire pressure is much higher than in car tires. Car
tires are mostly radial-ply tires, which are designed to cope with
lateral forces. Little information can be found on the lateral force
characteristics of aircraft tires at high speeds.
Additional research is required on the lateral force characteristics
of aircraft tires on dry, wet, and contaminated runway surface
conditions at operational speeds of aircraft in the takeoff roll.
F. Reaction Time
There are constraints on the reaction time that is used in
certification tests [13]. The reaction time is of great influence on
V30 ft. Major manufacturers have already pointed this out ([14] p. 15),
and it is confirmed by the modeling (Fig. 6). The reaction time for
rudder input was set to 0.5 s in the model. This is the average target
reaction time (0.4–0.6) as used in certification tests ([14] p. 15). Line
pilots will be surprised by an engine failure. A longer reaction time as
compared to that of a well-prepared test pilot can be expected. V30 ft
will increase due to the longer reaction time.
Manufacturers sometimes install systems to improve reaction time
with an automatic rudder input after sensing an engine failure [21].
These systems will have a positive effect on handling an engine
failure.
The rudder input to keep an aircraft near the centerline is skill-
based behavior. Rejected takeoff should be considered to be a rule-
based procedure with corresponding larger reaction times for
closing the operating engine [22]. Especially at low speeds, a short
reaction time is important to keep the aircraft on the runway in
a RTO.
G. Nosewheel Steering
Vmcg certification tests are conducted with a free castering
nosewheel or with the nosewheel lifted from the runway surface[13].
Fig. 12 Lateral deviation as function of Vef. RTO, NWS on, and 20 kt crosswind with different runway friction models: a) no braking, and b) differential
braking.
1348 HUIJBRECHTS, KOOLSTRA, AND MULDER
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
If nosewheel steering is coupled to the rudder, this will help to limit
the deviation from the centerline of the runway. The modeling shows
nosewheel steering is required at low speeds to keep the aircraft on the
runway after an engine failure. This is important, especially in a
rejected takeoff.
Some sources, including the FAA Flight Test Guidance [13],
suggest that uncoupling the nosewheel steering simulates the
runway surface condition of a wet runway. The simulations
show that the adverse effect of a wet runway or slippery runway is
worse than the adverse effect of an uncoupled nosewheel
steering (Fig. 7).
Experience from a Vmcg certification test with uncoupled
nosewheel steering shows considerable wear on nosewheel tires.§
This shows that lateral forces on the nosewheels will still have an
influence on the Vmcg certification test.
H. Runway Surface Condition
Observing Vmcg certification tests shows a high amount of strain on
the main wheels when dealing with the engine failure. When a
runway is not dry, the decreased lateral friction coefficient will affect
the amount of sideslip on the tires and the maximum achievable side
force. This maximum achievable side force is also reduced as speed
increases. Nosewheel steering will be less effective with lower lateral
friction coefficients.
The modeling confirms a considerable influence of runway surface
condition on V30 ft (Fig. 7b).
Vmcg, as certified by the manufacturer, is not a safe speed to
continue the takeoff after an engine failure on runways with reduced
runway friction coefficients (contaminated runways).
I. Crosswind
Major manufacturers confirm crosswind to have a high impact on
the Vmcg certification test. The effect of crosswind depends on the
design of the aircraft and is influenced by ground effect. An
accurate quantitative assessment of the influence of crosswind can
only be made if wind-tunnel data are available for the type
concerned.
When operating in crosswind conditions, a certain rudder input is
required to counteract the crosswind during the takeoff roll. In strong
crosswinds, large rudder inputs may be required. Any amount of
rudder input will decrease the remaining rudder deflection
available to counteract an engine failure. Crosswind will
increase V30 ft.
Aircraft are neither designed nor tested to cope with crosswind,
and an engine failure at the same time [6] CS25.149, [7] FAR25.149
does not require accounting for the effect of crosswind and [6]
CS25.237, [7] FAR25.237 does not require accounting for an
inoperative engine. The part of the takeoff roll after V1is a gray area
with respect to controllability in these crosswind conditions.
The modeli ng confirms crosswind is of con siderable influenc e on
V30 ft. With the model, NWS compensates for the effect of
crosswind to almost 20 kt of crosswind on a dry runway
(Fig. 8). A decrease in the runway surface friction coefficient will
decrease the influence of crosswind on V30 ft (Fig. 9). This is caused
by a higher ground slip angle when the runway surface friction
coefficient is lower, requiring less rudder input to counteract
crosswind.
Vmcg, as certified by the manufacturer, is not a safe speed to
continue the takeoff after an engine failure in strong crosswind
conditions.
J. Ground Effect
According to major manufacturers, a failure of the downwind
engine results in a higher lateral deviation in crosswind
conditions ([14] p. 16). In the model, the upwind engine is critical.
Large wing-mounted high-bypass engines close to the ground
have a more effective side area. This results in a higher side force
effect and a reduced weathervaning effect because the pressure
point shifts forward. The model parameters, derived from
DATCOM ([1] Par. II.C), do not account for this ground effect.
Observations showed training simulators have not incorporated this
ground effect either.
Through variation of relevant parameters, it is possible to
decrease the weathervaning effect of the model and make the side
force effect dominant. The downwind engine becomes critical,
and the impact of crosswind on lateral deviation at high μsvalues
(dry runway) (Figs. 8 and 9) decreases somewhat in the resulting
model. The qualitative effect of the reaction time and the runway
surface condition (Figs. 6 and 7) was checked to be similar in the
resulting model.
K. Is it Possible to Quantify the Effect of Environmental Conditions
on V30 ft?
The modeling showed a 6 kt margin on Vmcg (and thus on
V1MCG) would compensate for a slippery runway (Fig. 7). A 10 kt
margin would compensate for a crosswind component up to 25 kt,
even with a reduced runway fric tion coefficient (Fig. 9). The ground
model for the simulations is, however, based on very limited data
[17]; and the effect of crosswind on V30 ft depends on type-specific
aerodynamic properties. A margin on Vmcg, as established by the
manufacturer, will improve safety; but it is not possible to quantify
this margin.
L. Rejected Takeoff
The lateral deviation in the simulations is lower in the RTO than in
continued takeoff at speeds around Vmcg.
The simulations of rejected takeoff showed a quite violent reaction
of the aircraft after an engine failure at low speeds. This behavior is
confirmed to be similar on real aircraft by test pilots and can be
demonstrated in training simulators. Pilot response is critical in a low-
speed rejected takeoff. The reaction time for closing the operating
engine is of great influence on the resulting lateral deviation. The
modeling shows differential braking is necessary to keep the aircraft
on the runway if an engine failure is experienced at low speed and the
runway is not dry. The use of reverse thrust on the operating
engine may also help. More emphasis should be given to this in pilot
training.
The simulations showed that veeroffs in a rejected takeoff after
engine failure at low speed are likely to happen, especially in
crosswind and low runway friction conditions. The simulations also
showed that application of symmetrical braking at low speeds before
control toward the centerline is regained may aggravate the
situation. This is important to keep in mind when designing
autobrake systems.
This behavior may explain the runway excursion of the Iran Air
Airbus after an engine failure in Stockholm in 2010 [23]. This
incident also confirmed that a runway excursion at low speed does not
necessarily lead to a catastrophic or critical accident.
The low-speed RTO is recommended as a subject for further
research.
M. Incident/Accident Reports
Incident/accident reports sometimes lead to procedure changes.
After the Turkish Airlines crash [24] the stall recovery procedure
was changed. Accident investigators are, however, reluctant to
blame procedures. If a RTO, initiated after V1, leads to a runway
excursion, most probably the crew will get the blame for not
complying with procedures. In 58% of RTO accide nts, the RTO was
initiated after V1([4] Par. 2.2.4). A pilot will reject his takeoff after
V1if a runway excursion is imminent by continuing the takeoff. In
these cases, a higher V1would have been a better choice. Pilots may
perceive a lateral deviation of less than 30 ft as a loss of control
already. It is recommended to review the circumstances in accident
reports to establish if takeoffs were rejected after V1due to
controllability problems. Pilots may have saved their aircraft by
§Personal reference and fotos from Duke Ham, retired Performance
Engineer Fokker Aircraft, September 2017.
HUIJBRECHTS, KOOLSTRA, AND MULDER 1349
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
rejecting a takeoff after V1. In these cases, no evidence will be left to
feed statistics.
N. Overrun or Veeroff
A lot of effort is put in accelerate–stop performance to prevent an
overrun of the runway. Little effort is put in controllability issues to
prevent a veeroff. It is important to prevent runway excursions at
all sides.
O. Hydroplaning
The hydroplaning speed of modern aircraft tires [25] may be
lower than Vmcg for modern aircraft. When hydroplaning occurs,
the maximum achievable side force on the tires will drop to a
value at which we expect it is difficult to keep the aircraft on the
runway if an engine failure occurs and takeoff is either rejected or
continued.
It is recommended to investigate this with models of current
aircraft types.
P. What Can Aircraft Manufacturers Do to Improve Safety on
Contaminated Runways?
Aircraft manufacturers have access to more accurate aerodynamic
and inertial data of their aircraft. They can make an accurate
prediction of the Vmcg value before the test ([14] p. 7). Using the
manufacturer models to predict the Vmcg value, it is not difficult to
predict the V30 ft value at μs0. We can call this the aerodynamic
V30 ft. Aircraft manufacturers can calculate these values for their
types and make them available to the operators. An operator can then
choose if they want to stick to using Vmcg-limited V1as a minimum
speed for V1or use the aerodynamic V30 ft (corrected for engine
failure recognition time) in their SCAP modules to calculate takeoff
performance. Aircraft manufacturers can also make a better estimate
of the impact of crosswind on V30 ft on their types if wind-tunnel data
are available.
Q. What Can Regulatory Authorities Do to Improve Safety on
Contaminated Runways?
We share the concerns that were raised by the U.K. CAA in
preparation of the revision of AMC 25-1591 [11]. The modeling
provides scientific evidence to support this concern. We do not aim to
change the procedure to establish Vmcg during certification of an
aircraft. Any change in Certification Specifications [6]/Federal
Airworthiness Regulations [7] would not affect the operation for
earlier certified aircraft anyway. Rather, we want to change the way
Vmcg is used in operation.
The FAA and EASA can raise the issue in their safety committees.
If these committees agree with our point of view, a Safety Advice for
Operators can be issued to advise operators to replace their V1−min
policy by a policy that provides some margin on Vmcg. Regulatory
authorities can ask aircraft manufacturers to provide additional data
(e.g., an aerodynamic V30 ft) to operators to improve safety.
Regulatory authorities can discuss the need for a margin on Vmcg for
operation in crosswind.
VII. Risk Assessment
Modeling showed that Vmcg is not a safe speed to continue a takeoff
after an engine failure on either a runway with reduced friction
coefficients (contaminated runway) or in strong crosswind
conditions. The combination of either reduced runway friction
coefficients or strong crosswind conditions with a V1equal, or close
to, V1MCGwill lead to a situation with part of the takeoff roll in
which an engine failure, recognized after V1, will lead to an
unacceptable lateral deviation or runway excursion if takeoff is
continued.
Crosswind policies [19] prevent operation in strong crosswind
conditions on contaminated runways. The balanced takeoff
procedures on wet and dry runways usually result in a V1with a
considerable margin to V1MCG. The combination of strong
crosswind conditions with a V1equal, or close to, V1MCGis
remote.
On a contaminated (in particular, slippery) runway, however,
(Boeing) FCOM procedures will result in a V1equal, or close to,
V1MCGfor a wide range of takeoff weights. There is a (small) margin
in the V1MCGtables in the FCOM that is discarded when using
computer programs to calculate V1MCG. Using this lower value for
V1MCGas V1will increase the risk significantly because there is a
near-hyperbolic correlation between the engine failure speed and the
resulting lateral deviation (Figs. 6–8).
Operators are encouraged to choose for the V1−min policy
by the recommendations of the FAA’s Take-Off Safety Training Aid
([4] Par. 4.3.6.8).
VIII. Present Methods to Mitigate the Risk in
Contaminated and Slippery Runway Operations
Takeoffs on contaminated and slippery runways are conducted
with full rated (or fixed derated) power. Excess performance can be
used to increase the V1value if it is equal or close to V1MCG. This
way, a margin iscreated to compensate for the increase of V30 ft due to
the runway surface condition.
A. Method for Pilots
A pilot can find the V1value for the PLTOW, as provided by
computer performance output, in the FCOM for the runway surface
conditions/braking actions concerned.
This is an authorized value to use as long as it remains below VR.
As the actual TOW is lower, there will even remain a margin on the
accelerate–stop performance. In fact, a pilot can pick a single
value for V1from the range between the V1value on the computer
output and the V1value at the PLTOW (or VRif lower). All V1
speeds in this range will comply with AFM performance
requirements.
B. Method for Operators
SCAP modules of the present computer performance programs
offer operators a choice of one of the following options to use for their
contaminated runway operation:
1. V1Range
Both minimum and maximum values for V1are presented. A pilot
can choose a value from this range to use as V1. This requires an
explication of risks and benefits of low and high V1s to pilots. It will
leave the responsibility of choice to the pilot.
2. V1Mean
A single value for V1is presented, which is calculated as the
average of the minimum and maximum values for V1.
3. V1at Performance-Limited Takeoff Weight
A single value for V1is presented, which is calculated as the value
for V1at the PLTOW (or VRif lower).
IX. Additional Research
Additional research is required on the lateral force character-
istics of aircraft tires on dry, wet, and contaminated runway
surface conditions at operational speeds of aircraft in the
takeoff roll.
The quality of the ground model in training simulators should be
investigated in order to establish if the lateral deviations and
controllability on dry, wet, and contaminated runway surfaces are
realistic, also when including crosswind conditions.
It is recommended to review the circumstances in incident and
accident reports to establish if takeoffs were rejected after V1due to
controllability problems.
1350 HUIJBRECHTS, KOOLSTRA, AND MULDER
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
It would be interesting to analyze the controllability issues
involved with the Iran Air incident in Stockholm [23] and other
controllability-related runway excursions.
Through modeling, we may increase our knowledge of aircraft
behavior in a low-speed RTO.
Modeling can be used to assess the safety of operation of
modern aircraft (Vmcg ≈130 kt) in the takeoff roll when
hydroplaning occurs.
X. Conclusions
Vmcg, as certified by the manufacturer, is not a safe speed to
continue the takeoff after an engine failure on runways with reduced
runway friction coefficients (contaminated runways).
Vmcg, as certified by the manufacturer, is not a safe speed to
continue the takeoff after an engine failure in strong crosswind
conditions.
The transition from paper performance calculations to computer
performance calculations has reduced safety margins in operation.
Pilot reaction time is of great influence on V30 ft .
Runway surface friction is of considerable influence on V30 ft .
Crosswind is of considerable influence on V30 ft . The effect of
crosswind on V30 ft depends on the type-specific aerodynamic
properties and ground effect.
At low speeds, a coupled nosewheel steering is required to keep the
aircraft on the runway after an engine failure. This is important,
especiallyin a rejected takeoff.With the model, nosewheel steering can
compensatefor the adverse effect of crosswind on a dry runway upto a
considerable amount of crosswind. The effect of nosewheel steering is
negligible at low runway friction coefficients at high speeds.
Excess performance can be used to create a margin on top of Vmcg
to mitigate the risks in contaminated runway operation. A method for
pilots and for operators is presented in this paper.
The lateral deviation after engine failure is lower in the rejected
takeoff than in continued takeoff at speeds around Vmcg.
The modeling shows differential braking is necessary to keep the
aircraft on the runway if an engine failure is experienced at low speed
and the runway is not dry.
Appendix: Paved Runway Condition Assessment Table
Table A1 Paved runway condition assessment table [9]
Airport estimated runway condition assessment
Runway condition assessment: reported Downgrade assessment criteria
Code Runway description Mu, μDeceleration and directional control observation
Pilot reports (PIREPs)
provided to ATC and
flight dispatch
6 Any temperature:
1) Dry
—— —— Dry
5 Any temperature:
1) Wet (smooth, grooved, or porous friction course)
2) Frost
40 μor higher Braking deceleration is normal
for the wheel braking effort applied.
Directional control is normal.
Good
Any temperature with 1∕8in:or less of the following:
1) water,
2) slush,
3) dry snow, or
4) wet snow
4 At or colder than −13°Cat any depth:
1) Compacted snow
39 −36 μBraking deceleration and controllability is
between good and medium
Good to medium
3 Any temperature:
1) Wet (slippery)
35 −30 μBraking deceleration is noticeably reduced for the
wheel braking effort applied. Directional control
may be slightly reduced.
Medium
At or colder than −3°Cand greater than 1∕8in:of
the following:
1) Dry or wet snow
Warmer than −13°Cand at or colder than
−3°Cat any depth:
1) Compacted snow
2 Any temperature and greater than 1∕8in:of
the following:
1) Water
2) Slush
29 −26 μBraking deceleration and controllability is
between medium and poor. Potential
for hydroplaning exists.
Medium to poor
Warmer than −3°Cand greater than 1∕8in:of
the following:
1) Dry or wet snow
Warmer than −3°Cat any depth:
1) Compacted snow
1 At or colder than −3°Cat any depth of ice 35 −2l μBraking deceleration is significantly reduced
for the wheel braking effort applied. Directional
control may be significantly reduced.
Poor
0 Any temperature and any depth of the following:
1) Wet ice
2) Water on top of compacted snow
3) Dry or wet snow over ice
20 μor lower Braking deceleration is minimal to nonexistent
for the wheel braking effort applied.
Directional control may be uncertain.
Nil
Temperature warmer than −3°Cat any depth:
1) Ice
HUIJBRECHTS, KOOLSTRA, AND MULDER 1351
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
Acknowledgments
The Faculty of Aerospace Engineering of the Delft University of
Technology (the Control and Simulation section) facilitated this
research. There is no funding involved. We thank Duke Ham (retired
Performance Engineer, Fokker Aircraft B.V.), Wim Huson (retired
Certification Test Pilot, Fokker Aircraft B.V.), and Gerard Temme
(Certification Test Pilot, European Aviation Safety Agency) for
critical reading, hints, and tips.
References
[1] Koolstra, H. J., Huijbrechts, E. J. A. M., and Mulder, J. A., “Analysing
Aircraft Controllability After Engine Failure During Take-Off in
Adverse Weather Conditions,”Journal of Aircraft (to be published).
doi:10.2514/1.C035219
[2] “Takeoff,”Getting to Grips with Aircraft Performance, Airbus
Industries, Blagnac Cedex, France, 2002.
[3] “CS-25 Book 2,”Certification Specifications, and Acceptable Means of
Compliance for Large Aeroplanes, European Aviation Safety Agency,
Amendment 2 or Higher, AMC 25.1591.
[4] Take-Off Safety Training Aid: Federal Aviation Administration, 1993
(See. FAA AC No: 120-062).
[5] Electonic Code of Federal Regulations “Title 14, Subchap. A, Part. 1,
§1.2, https://www.gpo.gov/fdsys/pkg/CFR-2012-title14-vol1/pdf/CFR-
2012-title14-vol1-sec1-2.pdf [cited 9 June 2016].
[6] Certification Specifications and Acceptable Means of Compliance for
Large Aeroplanes European Aviation Safety Agency, CS-25 Book 1,
Subpart B.
[7] Federal Aviation Regulations Part 25 Airworthiness standards:
Transport category Airplanes, Federal Aviation Administration,
[8] Landing Performance Assessment at Time of Arrival, Safety Alert For
Operators 06012, Federal Aviation Administration, 2006
[9] Takeoff and Landing Performance Assessment Validation Effort of the
Runway Condition Assessment Matrix, Federal Aviation Administra-
tion, DOT/FAA/TC-TN13/22, 2013, Page 5: Paved Runway Condition
Assessment Table.
[10] RuFAB–Runway friction characteristics measurement and aircraft
braking Research Project EASA 2008/4,, European Aviation Safety
Agency, 2008.
[11] Notice of Proposed Amendment (NPA)N0 14/2004 Draft Decision of the
Executive Director of the Agency, on Certification Specifications for
Large Aeroplanes (CS-25) Operation on Contaminated Runways,
European Aviation Safety Agency, JAA NPA 25G-334 Comment-
Response Document: Par. 7.7.1 Comment NR 6,12,15,16,55,62,65,66,
2004, Sec. D.
[12] Pacejka, H., Tire and Vehicle Dynamics, 3rd ed., Elsevier Ltd., 2012.
[13] Flight Test Guidance (US FAA AC 25-7C), Federal Aviation
Administration, Minimum Control Speed § 25.149(e).
[14] Bolds-Moorehead, P. J., Chaney, V. G., Lutz, T. L., and Vaux, S.,
“Ground Minimum Control Speed Testing of Transport Aircraft,”AIAA
Flight Testing Conference, Dallas, TX, June 2015.
doi:10.2514/6.2015-3226
[15] Airplane Flight Manual B737-300, Minimum V1 for control on the
ground V1(MCG) The Boeing Company, Seattle WA, Doc. Nr. D6-
8730, p. 15, Sec. 4.7.
[16] “Performance Inflight V1(MCG),”Tire and Vehicle Dynamics, The
Boeing Company, Seattle WA.
[17] Yager, T. J., and McCarty, J. L., “Friction Characteristics of Tires
30 ×11.5–14.5, Type VIII, Aircraft Tires With Various Tread Groove
Patterns and Rubber Compounds,”NASA TP 1080, Dec. 1977.
[18] Dreher, R. C., and Tanner, J. A., “Experimental Investigation of the
Braking and Cornering Characteristics of 30 ×11.5–14.5, Type VIII,
Aircraft Tires With Different Tread Patterns,”NASA TN D-7743,
Oct. 1974.
[19] van Es, G. W. H., van der Geest, P. J., and Nieuwpoort, M. H., “Use of
Crosswind Limitations by Operators,”Safety Aspects of Aircraft
Operations in Crosswind, National Aerospace Lab./NLR TP 2001-217,
Amsterdam, 2001, Sec. 4.3.
[20] Yager, T. J., Vogler, W. A., and Baldasare, P., “Evaluation of Two
Transport Aircraft and Several Ground Test Vehicle Friction
Measurements Obtained for Various Runway Surface Types and
Conditions,”NASA Technical Paper 2917, 1990.
[21] Latimer, K., and Ross, J., C-17 Engine-Out Compensation System
Testing, Edwards AFB, CA, 1998, Paper 93524.
[22] van Es, G. W. H., “Rejecting a Take-Off After V1Why Does it (Still)
Happen,”National Aerospace Lab./NLR TP 2010-177, Amsterdam,
2010.
[23] “Serious Incident on 16 of January 2010 to Aircraft EP-IBB at
Stockholm/Arlanda Airport, Stockholm County, Sweden,”Swedish
Accident Investigation Authority Final Rept. RL 2012 21e, Stockholm,
2012.
[24] Crashed During Approach, Boeing 737-800, Near Amsterdam
Schiphol Airport, Feb. 2009, The Dutch Safety Board, The Hague,
The Netherlands, May 2010.
[25] van Es, G. W. H., “Hydroplaning of Modern Aircraft Tires,”National
Aerospace Lab./NLR TP 2001-242, Amsterdam, 2001.
1352 HUIJBRECHTS, KOOLSTRA, AND MULDER
Downloaded by 178.171.69.48 on September 30, 2019 | http://arc.aiaa.org | DOI: 10.2514/1.C035222
