sensors

Article
Determining Wheel Forces and Moments on Aircraft
Landing Gear with a Dynamometer Sensor †
Jaroslaw Pytka 1 , Piotr Budzyński 1, * , Tomasz Łyszczyk 1 , Jerzy Józwik 1 ,
Joanna Michałowska 2 , Arkadiusz Tofil 2 , Dariusz Błażejczak 3 and Jan Laskowski 4
1 Faculty of Mechanical Engineering, Lublin University of Technology, Nadbystrzycka 36, 20-618 Lublin,
Poland; j.pytka@pollub.pl (J.P.); tomasz.lyszczyk@gmail.com (T.Ł.); j.jozwik@pollub.pl (J.J.)
2 The State School of Higher Education, The Institute of Technical Sciences and Aviation, 22-100 Chełm,
Poland; jmichalowska@pwsz.chelm.pl (J.M.); atofil@pwsz.chelm.pl (A.T.)
3 Department of Construction and Usage of Technical Devices, West Pomeranian University of Technology in
Szczecin, 70-310 Szczecin, Poland; Dariusz.Blazejczak@zut.edu.pl
4 Faculty of Management, Lublin University of Technology, Nadbystrzycka 36, 20-618 Lublin, Poland;
jlasko@wp.pl
* Correspondence: p.budzynski@pollub.pl; Tel.: +48-602-459-881
† This paper is an extended version of the conference paper: Pytka, J.; Budzyński, P.; Józwik, J.; Łyszczyk, T.;
Laskowski, J.; Gnapowski, E. Measurement of Forces and Moments Acting on Aircraft Landing Gear Wheel.
Presented at the 6th IEEE International Workshop on Metrology for Aerospace Metro Aerospace 2019, Totino,
Italy, 19–21 June 2019.




Received: 24 November 2019; Accepted: 25 December 2019; Published: 31 December 2019


Abstract: This paper describes airfield measurement of forces and moments that act on a landing
gear wheel. For the measurement, a wheel force sensor was used. The sensor was designed and built
based on strain gage technology and was embedded in the left landing gear wheel of a test aircraft.
The sensor is capable of measuring simultaneously three perpendicular forces and three moments
and sends data to a handheld device wirelessly. For the airfield tests, the sensor was installed on a
PZL 104 Wilga 35A multipurpose aircraft. The aircraft was towed at a “marching man” speed and the
measurements were performed at three driving modes: Free rolling, braking, and turning. The paper
contains results obtained in the field measurements performed on a grassy runway of the Rzeszów
Jasionka Aerodrome, Poland. Rolling resistance of aircraft tire, braking friction, as well as aligning
moment were analyzed and discussed with respect to surface conditions.

Keywords: airfield performance; landing gear; wheel; force and moment measurement; strain gage;
wheel force sensor; wireless data transfer; grassy airfield; GARFIELD System

1. Introduction
Airfield performance of an airplane plays an important role in the analysis of takeoff and landing
and determines, among others, ground roll distances. Forces and moments that act on aircraft landing
gear wheels are effects of gravitational acceleration, as well as surface reactions. These reactions are
easily obtainable on paved runways, but difficulties arise when an airplane operates on unpaved,
grassy, or gravel surfaces. One method is to monitor surface mechanical data, related to weather
conditions. The GARFIELD information system [1] is based on a wheel–grass model that includes
analysis of wheel–soil (wheel–grassy surface) interactions considering soil modeling with a special
caution to loads by aircraft tires [2]. The model will respect non-linear, dynamic effects such as
hyper-elastic tire deflection, rheological soil response to high rate deformation, and effects of grass
and roots. The wheel–grass model also employs the analysis of weather impact upon mechanical

Sensors 2020, 20, 227; doi:10.3390/s20010227 www.mdpi.com/journal/sensors

Sensors 2020, 20, 227 2 of 15

characterization of runway surface, which can be identified and verified by means of wheel–force
measurements in full-scale tests [3].
Researchers dealing with flight tests of aircraft, piloted or unmanned, pay significant attention to
the importance of ground tests, including measurements for the validation and verification of landing
gear systems [4–6].
Significant emphasis is placed on the development and research of new measuring technologies
and, as an example, a new solution in the field of strain gauge sensors can be cited. Pecora et al.
have presented an innovative strain gauge for monitoring of inflatable structures, suitable for an use
in aerospace [7]. Another interesting example is given by Petritoli et al. (2019-1), who investigated
uncertainty and measured errors of an inertial navigation systems for UAV (Unmanned Aerial
Vehicle) [8]. Petritoli et al. (2018) and Petritoli et al. (2019-2) targeted the problem of collocation of the
sensors to be installed in a UAV with a reduced space, together with the allocation of the payload [9,10].
This problem is vital for full size aircrafts, not only typically smaller UAVs, especially if a sensor
has to be embedded in an existing part of an aircraft. Eling et al. (2015) presented an innovative
georeferencing system for the guidance, navigation, and control of UAVs [11]. This is an example of
an inertial measurement unit (IMU)-based system, the method of a wide spectrum of applications
in aerospace.
Measurement of forces and moments that act on a road wheel is a well known practice in
automotive research and development. It is usually done either on laboratory test stands or on a
test vehicle. Either way, a device that provides wheel forces and moments is a wheel dynamometer.
A wheel dynamometer is a measuring device that enables to measure forces and moments acting
on a wheel in real time on a moving vehicle. The majority of wheel dynamometer solutions comes
from the automotive industry or research, and a typical solution has a form of a 6-element sensor that
measures three orthogonal forces, FX , FY , and FZ , together with corresponding moments, MX , MY ,
and MZ [12–15]. The measurement principle is based either on the strain gage or piezoelectric sensor
technology. Some selected works that describe the use of such dynamometer systems for measurement
of wheel forces and moments in various applications are included in the reference list [16–19].
The horizontal force, FX , measured by a wheel dynamometer can be correlated with bending or
shearing the grass plants, as well as soil deformation. The braking friction force can be correlated
with sliding over the grass. The vertical force, Fz, depends the grass compression and can be used
for coefficient calculations (braking friction, µ, and rolling resistance, kRR , coefficients, as well as
aligning moment, MZ ). This approach is alternative to commonly used laboratory drop tests of landing
gear [20–22]. The aim of this study was to apply a wheel force and moment sensor for measurement of
aircraft landing gear on grassy surface. The primary paper, entitled Measurement of Forces and Moments
Acting on Aircraft Landing Gear Wheel, was presented during the 2019 IEEE International Workshop
Metrology for Aerospace in Torino, Italy [23].

2. Materials and Methods

2.1. Wheel Dynamometer System

The wheel dynamometer system used in the experiment consisted of a sensor, modified wheel and
tire, wireless data transfer unit, and software. The sensor core was made of steel, and strain gages were
used as sensing elements. The instrumentation amplifiers, which prepare the outgoing signals for data
acquisition, were placed outside the sensor in a small box, together with a board of the microprocessor
data analysis and transfer system. Figure 1 shows a schematic of the sensor, but without the electronics.
This schematic also includes the orientation of forces and moments.
As mentioned earlier, the sensor was designed based on the strain gage measurement technique.
The sensing element was made of spring steel and linear type strain gages were glued onto it.
Sensors 2019, 19, x FOR PEER REVIEW 3 of 15

Sensors 2020, 20, 227 3 of 15

Sensors 2019, 19, x FOR PEER REVIEW 3 of 15

Figure 1. A schematic of the wheel sensor together with axes orientation and forces/moments system.
The X, Y, and Z axes are mutually orthogonal.

Figure 1. A schematic of the wheel sensor together with axes orientation and forces/moments system.
As mentioned
Figure earlier,
1. A schematic of the sensorsensor
the wheel was designed based
together with
The X, Y, and Z axes are mutually orthogonal.
onorientation
axes the strainand
gage measurementsystem.
forces/moments technique.
The X, Y,element
The sensing and Z axes
wasare mutually
made orthogonal.
of spring steel and linear type strain gages were glued onto it.
The The in-wheel
in-wheel transducer
transducer measures two
measures forces,FFZZand
two forces, andFXF,Xand
, andtwo moments,
two moments, MZ M and MX . M
Z and Due to to
X. Due
As
the mentioned
restricted earlier,
space the sensor
mentioned was
above, a designed
braking based
moment on the
sensor strain
was gage
made measurement
as
the restricted space mentioned above, a braking moment sensor was made as a separate, external a separate, technique.
external
The sensing
device.
device. Melement
The The Y momentwaswas
MY moment made
was of spring
measured
measured steel
with
with theand
the linear
strain
strain type
gages
gages strain
sealed
sealed gages
onon
the were
brake
the brake glued
jig,jig,
which onto
is
which is it.
bending
bending
proportionally
The in-wheel to braking moment.
transducer measures Figure
two 2 shows
forces, F the
andcomplete
F , and sensor
two system, embedded
moments, M and M in. an
Due to
proportionally to braking moment. Figure 2 shows the complete sensor system, embedded
Z X Z X in an
aircraft
the restrictedwheel.
aircraft wheel.space mentioned above, a braking moment sensor was made as a separate, external
device. The MY moment was measured with the strain gages sealed on the brake jig, which is bending
proportionally to braking moment. Figure 2 shows the complete sensor system, embedded in an
aircraft wheel.

Figure 2. The complete wheel force sensor.

Figure 2. The complete wheel force sensor.
The wireless data transfer unit was built as a microprocessor system (see Figure 3), with an
The wireless
8-channel A/D data transfer
converter and aunit was built
Bluetooth radioasdevice
a microprocessor systemdata
to transfer measured (seewirelessly.
Figure 3), with an 8-
Figure 2. The
channel A/D converter and a Bluetooth complete
radio devicewheel force sensor.
to transfer measured data wirelessly.

The wireless data transfer unit was built as a microprocessor system (see Figure 3), with an 8-
channel A/D converter and a Bluetooth radio device to transfer measured data wirelessly.
 Sensors
Sensors 2019, 19, x2020,
FOR 20,PEER
227 REVIEW 4 of 15 4 of 15

Figure 3. The heart of the microprocessor data transfer system, used in the wheel sensor system.
Figure 3. The heart of the microprocessor data transfer system, used in the wheel sensor system.
The complete and detailed description of the wheel force sensor system is given in the reference
Theby Pytka et al.
complete and[3].detailed description of the wheel force sensor system is given in the reference
by Pytka2.2.
et al. [3]. Used in the Experiment
Airplane
The PZL 104 Wilga 35 multipurpose airplane was used as a test aircraft. This is a light, four-place,
2.2. Airplane Used in the Experiment
multipurpose aircraft, powered by a 192 kW radial piston engine. The Wilga has a good reputation as
TheaPZL
STOL104 (Short
WilgaTakeoff
35 and Landing) or even
multipurpose the so-called
airplane was used“bush-plane”,
as a testthanks
aircraft.to itsThis
landing
is agear.
light, four-
The aircraft has a tail-dragger-type landing gear and the main gears are rocker type with oleo-pneumatic
place, multipurpose aircraft, powered by a 192 kW radial piston engine. The Wilga has a good
shock absorbers. The landing gear wheels have low-pressure 500 × 200 mm tires with hydraulic brakes.
reputation
Theas a STOL
wheels (Short and
are castored Takeoff
have aand Landing)
positive or even
rake angle of 18◦ ;the so-called
the axle offset is“bush-plane”,
400 mm. The airplanethanks to its
landing gear.
is 8.10 The aircraftlength,
m in overall has awith
tail-dragger-type
a wingspan of 11.12 landing
m, a winggearareaand themmain
of 15.5 2 , and gears
an emptyaremass,
rocker type
with oleo-pneumatic
equipped, of 900shockkg. In absorbers. The there
the landing tests, landing
were gear wheelsonhave
four persons boardlow-pressure
and the take-off500 × 200
weight wasmm tires
1150 kg. The test airplane is depicted in Figure 4.
with hydraulic brakes. The wheels are castored and have a positive rake angle of 18°; the axle offset
The wheel force sensor was installed on the test aircraft before measurements. Installation did
is 400 mm. The airplane is 8.10 m in overall length, with a wingspan of 11.12 m, a wing area of 15.5 m2,
not require any modification to the airframe or landing gear, since the sensor system was developed
and an empty mass,
specially for theequipped, of 900
given aircraft. kg.leveling
Simply, In the up landing tests,removing
the aircraft, there were four with
the wheel persons onand
its axle, board and
the take-off weight
placing was 1150
the sensor. Also, kg. The test
the brake airplane
hydraulic is depicted
system in Figure to
had to be re-switched 4. the sensor, since the FY
Thewheel
wheel force was
moment sensor was installed
measured by means of ona transducer
the test aircraft
embedded before
in themeasurements.
wheel brake. Installation did
not require anyAnother element oftothe
modification themeasuring
airframesystem was the
or landing electronic
gear, unit,sensor
since the which system
consistedwasof five
developed
differential signal amplifiers, each for one of the five channels (two forces and three moments), as
specially for the given aircraft. Simply, leveling up the aircraft, removing the wheel with its axle, and
well as of a microcontroller-based, wireless data transfer unit. The electronic box was mounted on the
placing the sensor. Also, the brake hydraulic system had to be re-switched to the sensor, since the FY
landing gear with glue tape and the sensor was connected with the electronics with signal cables. The
wheel moment was measured
whole system, by means
the wheel sensor, and theofelectronics
a transducer
power embedded
supply (7.2V in 800the
mAh wheel
NiMHbrake.
accumulator)
Another element
was placed in theof the measuring
cockpit. Figures 5 and system
6 showwas the electronic
the details unit, which
of the test airplane with theconsisted
installed of five
measuring equipment.
differential signal amplifiers, each for one of the five channels (two forces and three moments), as
well as of a microcontroller-based, wireless data transfer unit. The electronic box was mounted on
the landing gear with glue tape and the sensor was connected with the electronics with signal cables.
The whole system, the wheel sensor, and the electronics power supply (7.2V 800 mAh NiMH
accumulator) was placed in the cockpit. Figures 5 and 6 show the details of the test airplane with the
installed measuring equipment.
 Sensors
Sensors 2019,
2019, 19,
19, xx FOR
FOR PEER
PEER REVIEW
REVIEW 55 of
of 15
15
Sensors 2020, 20, 227 5 of 15

Figure
Figure4.4.
Figure The
4.The PZL
ThePZL 104 Wilga
PZL 104 Wilga 35A,
Wilga 35A, used
35A, usedfor
used forthe
for thetests.
the tests.
tests.

Figure 5. The wheel force sensor installed on the test airplane, ready for the tests.
Sensors 2019, 19, x FOR PEER REVIEW 6 of 15

Sensors 2020, 20, 227 6 of 15

Figure 5. The wheel force sensor installed on the test airplane, ready for the tests.

Figure 6. The electronics: Signal amplifiers and wireless data transfer system in the box, installed on
Figure 6. The electronics: Signal amplifiers and wireless data transfer system in the box, installed on
the landing gear leg (a silver tape was used for quick installation).
the landing gear leg (a silver tape was used for quick installation).
2.3. Grassy Airfield Conditions
2.3. Grassy Airfield Conditions
Measurements of wheel forces and moments were performed on a grassy surface of the Rzeszów
Measurements
Jasionka aerodrome,oflocated
wheel forces and moments
in southeast Poland.were
Theperformed on a grassy
measurements surface of
were carried outtheinRzeszów
the area
Jasionka
adjacent to the runway, with the surface and soil conditions identical to those on the grassy the
aerodrome, located in southeast Poland. The measurements were carried out in area
runway.
adjacent
The heightto of
thethe
runway, withapproximately
grass was the surface and10soil
cm conditions identical
each time of to those on the grassy runway.
the measurements.
The height of the grass was approximately 10 cm each time of the measurements.
The days in the autumn and spring months were selected, when there is a high probability of
The
increasingdays in the autumn
humidity and loweringand spring monthsproperties
the traction were selected,
of thewhen there due
grassland is a high
to theprobability
influence of
of
increasing
weather humidity
conditions and lowering
(rainfall, the traction
low temperature, lowproperties
sunshine).ofThey
the grassland
were on 17due to the
October, influence of5
5 November,
weather
December,conditions (rainfall,
and 13 March. Tablelow 1temperature,
gathers datalow sunshine).
describing They wereconditions
atmospheric on 17 October, 5 November,
prevailing on the
5days
December, and 13
of measurement. March. Table 1 gathers data describing atmospheric conditions prevailing on the
days of measurement.
Table 1. Weather and soil conditions on the days of tests.
Table 1. Weather and soil conditions on the days of tests.
Weather Air Temperature (◦ C) Soil Moisture (%) Soil Temperature (◦ C)
Air Temperature Soil Moisture Soil Temperature
Weather 17 October 2018
(°C) (%) (°C)
Sunshine, wind 5–7m/s S, SE 15 8.80 n.a.
17 October 2018
5 November 2018
Sunshine, wind 5–7m/s S, SE 15 8.80 n.a.
Sunshine, wind 10-12 m/s W 12 18.80 18.3
5 November 2018
5 December 2018
Sunshine, wind 10-12 m/s W 12 18.80 18.3
Cloudy, 7/8 Cu, wind 2–3m/s W, NW 2,5 23.67 4.6
5 December 2018
Cloudy, 7/8 Cu, wind 2–3m/s 13 March 2019
Cloudy, W,
withNW
break intervals 2,5
11 23.67
22.50 4.6
n.a.

13 March 2019
Cloudy, with
Before the break
test runs,intervals
we also measured soil11moisture using a handheld 22.50 TDR moisture n.a. meter. TDR is
an acronym for time-domain reflectometry, and this is a technique that utilizes the phenomenon of
Before
different the testpermeability
electrical runs, we alsoofmeasured soil moisture
porous, three-phase using
media, a handheld
such TDR
as soil. The moisture
TDR meter meter.
is very TDR
easy
is
toan acronym
use. for time-domain
Measurement reflectometry,
requires simply insertingandtest
thiselectrodes
is a technique thatand
into soil utilizes
the the phenomenon
instrument of
quickly
different electrical permeability of porous, three-phase media, such as soil. The TDR
gives results, which are collected in Table 1. Figure 7 shows the test engineer performing soil moisturemeter is very
easy to use. Measurement
measurement requires
on the grass field. simply
See also inserting by
the reference testPytka
electrodes into
et al. for soildetails
more and theon instrument
the use the
quickly givesinresults,
TDR sensor which arestudies
terramechanics collected in Table 1. Figure 7 shows the test engineer performing soil
[24].
moisture measurement on the grass field. See also the reference by Pytka et al. for more details on the
use the TDR sensor in terramechanics studies [24].
Sensors 2020, 20, 227 7 of 15
Sensors
Sensors 2019,
2019, 19,
19, x
x FOR
FOR PEER
PEER REVIEW
REVIEW 7
7 of
of 15
15

Measurement of
Figure 7. Measurement
Figure soil moisture with use
the usea of a handheld time-domain reflectometry
Figure 7.
7. Measurement of
of soil
soil moisture
moisture with
with the
the use of
of a handheld
handheld time-domain
time-domain reflectometry
reflectometry (TDR)
(TDR)
(TDR)
sensor. sensor.
sensor.
2.4. Field Procedures
2.4.
2.4. Field
Field Procedures
Procedures
Since the sensor system has not been certified for flight tests yet, we could only manage ground
Since the sensor system has not been
been certified for flight tests yet,
yet, we could only manage ground
tests.Since the
In this sensor
study, system
it was hasto
aimed not
determinecertified
wheelfor flight
forces tests
and moments weacting
couldon
only
themanage
aircraft’sground
wheel
tests.
tests. In
In this
this study,
study, it
it was
was aimed
aimed to
to determine
determine wheel
wheel forces
forces and
and moments
moments acting
acting on
on the
the aircraft’s
aircraft’s wheel
wheel
at low speed of taxiing over the grassy runway. We used an all-terrain vehicle (ATV) as a tractor vehicle
at
at low
low speed of taxiing over
over the
thetograssy runway. We
We used an
an all-terrain vehicle
vehicle (ATV) as aa tractor
and thespeed of was
aircraft taxiing
connected grassy runway.
this vehicle by means used all-terrain
of cables. As mentioned (ATV)
earlier,asfourtractor
adult
vehicle
vehicle and the aircraft was connected to this vehicle by means of cables. As mentioned earlier, four
persons were in the test aircraft during all tests. An approximate speed of taxiing was 5 km/h. four
and the aircraft was connected to this vehicle by means of cables. As mentioned earlier, The
adult
adult persons
persons were
werethein the
in the test aircraft
test aircraft during all tests. An approximate speed of taxiing was
during all tests. An approximate speed of taxiing was 5 km/h. 5 km/h.
aircraft performed following maneuvers:
The aircraft
The aircraft performed
performed the the following
following maneuvers:
maneuvers:
• Accelerating from stop
Accelerating
Accelerating from
from stop
to the
stop to
to the
taxi speed;
the taxi
taxi speed;
speed;
• Taxiing
Taxiing with a constant speed (about
Taxiing with a constant speed (about 20–30
with a constant speed (about 20–30 m);
20–30 m);
m);
• Turning left or right;
right;
Turning left or right;
• Braking to
Braking to aa full
full stop
stop or
or to
to block
block the
the wheels.
wheels.
We
We performed
performed aa minimum
minimum of of five
five repetitions
repetitions of
of all
all the
the maneuvers
the maneuvers for
maneuvers for every
for every measurement
every measurement time.
measurement time.
time.
Figure 8 shows a sample test run on the grassy surface.
Figure 8 shows a sample test run on the grassy surface.

Figure 8.
Figure 8. Starting
8. Startingaaatest
Starting test
test run.
run.
run. TheThe
The test
testtest airplane
airplane
airplane is pulled
is pulled
is pulled by
by anby an all-terrain
an all-terrain
all-terrain vehicle
vehiclevehicle (ATV).
(ATV).
(ATV). The The test
The test
test engineer
engineer
is walking
engineer is walking
is by by
the test
walking the test
byairplane, airplane, collecting
collectingcollecting
the test airplane, data wirelessly.
data wirelessly. Four
Four persons
data wirelessly. persons onboard
onboardonboard
Four persons the
the airplane.airplane.
the airplane.

The
The wireless
wireless data
data transfer
transfer system
system enabled
enabled that
that the
the test
test engineer
engineer could
could be
be anywhere,
anywhere, within
within aa 100
100
m
m range of the airplane. Since the measurements were performed with all four passengers on
range of the airplane. Since the measurements were performed with all four passengers on board,
board,
Sensors 2020, 20, 227 8 of 15

The
Sensors wireless
2019, 19, x FORdata
PEERtransfer
REVIEW system enabled that the test engineer could be anywhere, within a
8 of 15
100 m range of the airplane. Since the measurements were performed with all four passengers on
the test
board, theengineer was in
test engineer wasthe
in airplane during
the airplane measurements.
during measurements. WeWetried
triedtotocollect
collectdata
data on
on aa laptop
laptop
computer, but it was more comfortable to gather the results on a handheld
computer, but it was more comfortable to gather the results on a handheld smartphone. smartphone.

3. Results
3. Results
Resultsobtained
Results obtainedininthe
thefield
fieldexperiments
experiments include
include courses
courses of of wheel
wheel forces
forces andand moments
moments acting
acting on
on left
the the wheel
left wheel
of theofWilga
the Wilga airplane,
airplane, and weand we collected
collected data
data from from
four four
days days of measurements:
of measurements: 17
17 October,
October, 5 November, and 5 December 2018, as well as 13 March 2019. The following
5 November, and 5 December 2018, as well as 13 March 2019. The following is a presentation and is a presentation
and analysis
analysis of theofresults
the results obtained
obtained in thein the autumn–winter
autumn–winter part ofpart
the of the measurements
measurements campaign
campaign (a so-
(a so-called
calledseason”
“dead “dead season” in sport aviation)
in sport aviation) that was performed
that was performed as a full-season
as a full-season measurement measurement for the
for the verification
verification
of of theinteraction
the wheel–turf wheel–turfmodel
interaction
of themodel of the GARFIELD
GARFIELD informationinformation system [1,2,23,24].
system [1,2,23,24].

3.1.
3.1. A
A General
General Presentation
Presentation of
of the
the Results
Results
Figure
Figure 99shows
showstimetimecourses
coursesof ofthe verticalFFZZ and
thevertical horizontal FFXX forces
andhorizontal forces that
that act
act on
on the
thelanding
landing
gear
gearwheel. The FFZZ shows
wheel. The shows almost
almost constant
constant values
values with
with low
low fluctuations,
fluctuations, probably
probably caused
caused by byvertical
vertical
vibrations of the entire aircraft rolling over the grassy surface, and those values are close to
vibrations of the entire aircraft rolling over the grassy surface, and those values are close to the aircraft the aircraft
weight
weight component
component on the wheel.
on the wheel. TheThehorizontal forceFXFexhibits
horizontalforce X exhibits negative
negative values
values since
since thethe sensor
sensor has
has a defined
a defined orientation
orientation andand a force
a force of the
of the direction
direction backwards
backwards is indicated
is indicated as negative.
as negative. Moreover,
Moreover, the
the course of the horizontal force
course of the horizontal force FX reveals F X reveals effect of wheel function modes: Free rolling
effect of wheel function modes: Free rolling (the first second (the first
second of the course),
of the course), acceleration
acceleration increasing
increasing of the force’s
of the force’s valuesvalues
(between (between
1 and 1approximately
and approximately 4 s of 4thes
of the course),
course), braking braking tostop
to a full a full(5–9
stop
s),(5–9
and,s), and, acceleration
again, again, acceleration (from approximately
(from approximately 9 s of the9 scourse).
of the
course).
The peak The peak maximum
maximum value ofvalue of the horizontal
the horizontal force FXforce FX reaches
reaches 2.7 kN,2.7
butkN,
thebut the average
average value value
is much is
much
lower.lower.

Figure 9.
Figure 9. Sample
Sample result
resultof
ofthe
themeasurements:
measurements:Time courses
Time of wheel
courses forces
of wheel acting
forces on the
acting on landing gear
the landing
wheel.
gear wheel.

Anothersample
Another sampleresult
resultisisshown
shownininFigure
Figure10,10,where
wherewewecan
canseeseetime
timecourses
coursesof ofthe
thethree
threemoments
moments
actingon
acting onthe
theaircraft
aircraftwheel.
wheel. The character
character of these courses is very similar to those of Figure 9. 9. These is
isaa
shortperiod
short periodofoffree
free rolling,
rolling, then
then acceleration,
acceleration, braking
braking andand
stop,stop, and finally
and finally acceleration
acceleration again.again. The
The peak
peak maximum
maximum valuesvalues reach
reach 150 Nm150 forNm
the Mfor
Z the M
moment, Z moment,
280 Nm 280
for Nm
the Mfor
X , the
and M
250X, and
for the250
M Yfor the
moment.MY
moment.
The valuesTheof values of the moments
the moments acting onacting on the landing
the landing gear changes
gear wheel wheel changes theiratsign
their sign at a moment
a moment of a
of a change
change between
between two function
two function modes,
modes, for example,
for example, accelerating–braking.
accelerating–braking.
In the next sections, an analysis of the measured forces and moments is presented in order to
determine coefficients that describe wheel–surface interactions, which are of importance for the
analysis and modeling of airfield performance of an aircraft.
Sensors 2020, 20, 227 9 of 15
Sensors 2019, 19, x FOR PEER REVIEW 9 of 15

Figure10.
Figure 10.Sample
Sampleresult
resultofofthe
themeasurements:
measurements:Time
Timecourses
coursesofofthree
threemoments
momentsacting
actingon
onthe
thelanding
landing
gearwheel.
gear wheel.

3.2. In the next

Rolling sections,
Resistance an analysis of the measured forces and moments is presented in order to
Coefficient
determine coefficients that describe wheel–surface interactions, which are of importance for the analysis
Rolling resistance is a property of the tire–surface system and depends on many factors. In
and modeling of airfield performance of an aircraft.
automotive technology, rolling resistance is determined for a tire rolling on a drum test stand and its
coefficient
3.2. takes essentially
Rolling Resistance constant values, on the order of 0.010–0.012. In the case of a tire interacting
Coefficient
with a deformable surface, such as grassy surface, rolling resistance is difficult to determine, mainly
Rolling resistance is a property of the tire–surface system and depends on many factors.
due to the changing conditions of the surface, which was mentioned in the introduction. In addition,
In automotive technology, rolling resistance is determined for a tire rolling on a drum test stand and its
the values of the rolling resistance coefficient of the tire–grass surface system can vary by as much as
coefficient takes essentially constant values, on the order of 0.010–0.012. In the case of a tire interacting
an order of magnitude, taking into account, for example, changes in soil moisture. The presence of
with a deformable surface, such as grassy surface, rolling resistance is difficult to determine, mainly
vegetation (grass, roots), as well as the size and mass of green parts of vegetation, are also significant.
due to the changing conditions of the surface, which was mentioned in the introduction. In addition,
The rolling resistance coefficient, kRR, is defined as the ratio of horizontal force during free
the values of the rolling resistance coefficient of the tire–grass surface system can vary by as much as
turning to vertical force [25]:
an order of magnitude, taking into account, for example, changes in soil moisture. The presence of
vegetation (grass, roots), as well as the size andkmass
RR = FX
of/Fgreen
Z (1)
parts of vegetation, are also significant.
The rolling
Figure resistance
11 shows an example of khorizontal
coefficient, RR , is defined as the
force, FX, ratio
which of is
horizontal
input dataforce
for during free turning
calculations (upper
tograph)
vertical force [25]:
and the rolling resistance coefficient, kRR (lower graph), value determined using Equation (1).
Average values are marked on the graphs k(F = FX /FZN and kRR = 0.275). When determining the(1)
X = 1719
RR kRR
coefficient,
Figure 11 it shows
is important to correctly
an example select force,
of horizontal a portion of the isFXinput
FX , which forcedata
waveform. Namely,(upper
for calculations it is a
graph) and the rolling resistance coefficient, kRR (lower graph), value determined using Equation first
fragment where the waveform is determined (in Figure 11, up, this fragment begins after the (1).
second of
Average the wave).
values are marked on the graphs (FX = 1719 N and kRR = 0.275). When determining the
The exemplary
kRR coefficient, value of the
it is important kRR coefficient
to correctly selectwas determined
a portion of theto FXbeforce
0.275, which in comparison
waveform. Namely, it isto
the aforementioned value for a car tire on a paved road is almost 25 times
a fragment where the waveform is determined (in Figure 11, up, this fragment begins after [25]. This is the effect of a
the first
soft surface
second of the and
wave).a low-pressure aviation tire. The impact of weather conditions on the determined
values of the k RR coefficient will be presented in the next section.
The exemplary value of the k coefficient was determined to be 0.275, which in comparison to
RR
the aforementioned value for a car tire on a paved road is almost 25 times [25]. This is the effect of a
soft surface and a low-pressure aviation tire. The impact of weather conditions on the determined
values of the kRR coefficient will be presented in the next section.
 Sensors2020,
Sensors 2019,20,
19,227
x FOR PEER REVIEW 10of
10 of15
15

Figure 11. Sample

Figure 11. Sample courses
courses of
of the
the measured
measured horizontal force, FFXX,, and
horizontal force, and calculated
calculated rolling
rolling resistance
resistance
coefficient, kRR .
coefficient, kRR.
3.3. Braking Friction Coefficient
3.3. Braking Friction Coefficient
The braking friction coefficient µ is another important factor. It determines the course of the
brakingTheprocess
brakingandfriction coefficient
the landing μ isofanother
distance important
the aircraft. In the factor.
case of Itthe
determines
tire–pavedthe course
system of the
(asphalt,
braking process
concrete), and the
the friction landingusually
coefficient distance of the values
reaches aircraft. In the case
between of the
0.6–0.9. tire–paved
For system (asphalt,
unpaved surfaces such as
grass turf, the friction coefficient is usually lower and may be 0.1–0.5. Factors affectingsurfaces
concrete), the friction coefficient usually reaches values between 0.6–0.9. For unpaved such
the µ value
as grass turf, the friction coefficient is usually lower and may be 0.1–0.5. Factors
are grass length and humidity: The longer the grass and the more water it contains, the lower the affecting the μ value
are grass
friction length and
coefficient. Byhumidity:
definition,The
thelonger
frictionthe grass and
coefficient is the more water
determined it contains,
on the the following
basis of the lower the
friction coefficient.
relationship: By definition, the friction coefficient is determined on the basis of the following
relationship: µ = MY /FZ × rd (2)
μ = MY/FZ × rd (2)
where rd is the dynamic radius of the tire.
whereFigure
rd is 12
theshows
dynamican example
radius ofofthe braking moment, MY , course and the braking friction coefficient
thetire.
determined.
Figure As12 it can bean
shows seen, µ values
example of oscillate around
the braking 0.22. MY, course and the braking friction
moment,
coefficient determined. As it can be seen, μ values oscillate around 0.22.
Sensors 2020, 20, 227
Sensors 2019, 19, x FOR PEER REVIEW 11 of 15 11 of 15

Figure 12. Sample

Figure courses
12. Sample of the
courses of measured braking
the measured moment, MM
brakingmoment, Y,Y , and
and braking
braking friction
friction coefficient,
coefficient, μ. µ.

3.4. Aligning Moment

3.4. Aligning Moment

When driving on a curved

When driving track,
on a curved the the
track, road wheel
road wheelmoves
movesin
in such
such aawaywaythat
that
thethe longitudinal
longitudinal axis axis of
of the wheel is deflected from the direction of the instantaneous speed vector
the wheel is deflected from the direction of the instantaneous speed vector by an angle, called by an angle, called thethe side
side slip angle. The reaction of a pneumatic tire to side slip is to generate a moment that tends to
slip angle. The reaction of a pneumatic tire to side slip is to generate a moment that tends to reduce
reduce this angle. This moment is called the stabilizing or aligning moment.
this angle. This moment is called the stabilizing or aligning moment.
The sensor used in the tests directly measures the value of the aligning moment, it is the moment
Themarked
sensoras usedMZ. in the tests
Figure directly
13 shows the Mmeasures the value
Z course recorded of the
during onealigning moment,
of the turning it is the
maneuvers. Themoment
as MZ . Figure
marked importance of the M13 showsisthe
Z moment MZ course
revealed recorded
during taxiing, during
especially onespeed.
at high of the Theturning
higher themaneuvers.
MZ
value, theof
The importance easier
the itMisZtomoment
maintain isa straight
revealed direction,
duringwhile too high
taxiing, a stabilizing
especially moment
at high can cause
speed. The higher
significant steering forces when cornering, especially when taxiing at low speed.
the MZ value, the easier it is to maintain a straight direction, while too high a stabilizing moment can
Sensors 2019, 19, x FOR PEER REVIEW
cause significant steering forces when cornering, especially when taxiing at low speed. 12 of 15

Figure
Figure 13. Sample
13. Sample courseof
course ofthe
the aligning
aligning moment,
moment,MZ.MZ .

3.5. Effect of Weather Condidtions

As already mentioned, the study was carried out over four day—in October, November, December,
and March. The purpose of such a research program was to check to what extent the worsening
deterioration of weather conditions, in terms of lowering the external temperature and increasing
 Figure 13. Sample course of the aligning moment, MZ.

3.5. Effect of Weather Condidtions

As already mentioned, the study was carried out over four day—in October, November, December,
and March.
Sensors 2020, 20, The
227 purpose of such a research program was to check to what extent the worsening 12 of 15
deterioration of weather conditions, in terms of lowering the external temperature and increasing
humidity of air and soil, affects the measured values of forces and moments, as well as the resulting
3.5. Effect of Weather Condidtions
coefficients. Based on the measurements carried out, mean values of repetitions (minimum of five
repetitions for each
As already day) of individual
mentioned, the study wasmeasurements werefour
carried out over determined. A collection
day—in October, of final data
November, in the
December,
formMarch.
and of bar charts
The is presented
purpose in Figures
of such 14–16, where
a research program average
was to values
checkof rolling
to what resistance
extent thecoefficient,
worseningkRR,
friction coefficient, μ, and stabilizing moment, M , are shown for four days, representative
deterioration of weather conditions, in terms of lowering the external temperature and increasing
Z for particular
weather periods.
humidity of air and soil, affects the measured values of forces and moments, as well as the resulting
The rolling
coefficients. resistance
Based on thecoefficient
measurementsincreases, andout,
carried this is
meana significant
values ofincrease. The explanation
repetitions (minimum for that
of five
is quite obvious:
repetitions In autumn,
for each late autumn,
day) of individual winter, and early
measurements werespring, there isAa collection
determined. gradual increase
of finalin humidity
data in the
of both
form of air
barand soil,
charts is which causes
presented increasing
in Figures deformation
14–16, of the values
where average surfaceofand, as aresistance
rolling result, an coefficient,
increase in
krolling resistance.
RR , friction coefficient, µ, and
Although we did not construct
stabilizing moment, M any relationship
Z , are shown forbetween
four days,therepresentative
rolling resistance
for
coefficient weather
particular and soil moisture
periods. (see Table 1), these parameters correspond well to each other.

Sensors 2019, 19, x FOR PEER REVIEW 13 of 15

covered surfaces. A slight increase in the braking friction coefficient was observed for the
measurement carried out in March. Perhaps the reason was a stronger solar operation, which dried
the surface, especially grass.
Figure 14. Average peak values of rolling resistance coefficient, kRR , for the four measurements.
Figure 14. Average peak values of rolling resistance coefficient, kRR, for the four measurements.

The second term, braking friction coefficient shows an inverse relationship, i.e., it decreases
during the period considered. In this case, the cause is similar, although the increasing humidity
works slightly differently, namely, lower grip (=more slipperiness), especially in the case of grass-

Figure 15. Average peak values of braking friction coefficient, µ, for the four measurements.
Figure 15. Average peak values of braking friction coefficient, μ, for the four measurements.

The last of the parameters analyzed, the aligning moment, shows an interesting tendency (Figure
16). No significant impact of the weather period is observed during the months October–November,
while a significant decrease in the MZ moment value in December was recorded. In March, the MZ
value increased significantly compared to the December measurements.
 Figure 15. Average peak values of braking friction coefficient, μ, for the four measurements.

The last of the parameters analyzed, the aligning moment, shows an interesting tendency (Figure
16). No significant impact of the weather period is observed during the months October–November,
while a significant decrease in the MZ moment value in December was recorded. In March, the MZ
Sensors increased
value 2020, 20, 227 significantly compared to the December measurements. 13 of 15

Figure 16. Average peak values of aligning moment, MZ , for the four measurements.
Figure 16. Average peak values of aligning moment, MZ, for the four measurements.
The rolling resistance coefficient increases, and this is a significant increase. The explanation for
4. Conclusions
that is quite obvious: In autumn, late autumn, winter, and early spring, there is a gradual increase
in humidity of both
An airfield air and soil,
experiment waswhich causesin
conducted increasing
order to deformation of the gear
measure landing surface and,forces
wheel as a result,
and
an increaseAinfive-element
moments. rolling resistance. Although
wheel force sensorwewasdidused
not construct
to measureanyFXrelationship
and FZ forces between
and Mthe rolling
X, MY, and
resistance
M coefficient and soil moisture (see Table 1), these parameters correspond well to each other.
Z moments acting on an aircraft wheel during a low-speed taxiing. The measurement was

conducted usingterm,
The second braking
the PZL friction
104 Wilga coefficient
35A as a testshows anThe
aircraft. inverse
wheelrelationship,
force sensori.e.,was
it decreases
installed during
on the
the period considered. In this case, the cause is similar, although the increasing humidity
left landing gear in place of the wheel. Forces and moments were measured on a grassy surface of works slightly
differently,
the Rzeszównamely,
Jasionkalower grip (=more
aerodrome, slipperiness),
four times on days of especially in the
autumn and case of
spring grass-covered
months. surfaces.
The results show
A slight increase in the braking friction coefficient was observed for the measurement carried out in
March. Perhaps the reason was a stronger solar operation, which dried the surface, especially grass.
The last of the parameters analyzed, the aligning moment, shows an interesting tendency
(Figure 16). No significant impact of the weather period is observed during the months
October–November, while a significant decrease in the MZ moment value in December was recorded.
In March, the MZ value increased significantly compared to the December measurements.

4. Conclusions
An airfield experiment was conducted in order to measure landing gear wheel forces and moments.
A five-element wheel force sensor was used to measure FX and FZ forces and MX , MY , and MZ moments
acting on an aircraft wheel during a low-speed taxiing. The measurement was conducted using the
PZL 104 Wilga 35A as a test aircraft. The wheel force sensor was installed on the left landing gear in
place of the wheel. Forces and moments were measured on a grassy surface of the Rzeszów Jasionka
aerodrome, four times on days of autumn and spring months. The results show changes in wheel
force and moment values during various maneuvers: Acceleration, turning left–right, and deceleration
(braking).
A significant influence of weather conditions on the values of determined coefficients kRR and µ,
as well as on the aligning moment MZ , was noted. The rolling resistance coefficient increased during
the considered weather period. However, the friction coefficient decreased, as did the stabilizing
moment. The explanation for these trends is the increase in air and soil humidity, which translates into
increased soil deformability and reduced adhesion.
Further research is planned in the area of applied machine learning to analyze the impact of
unpaved airfield surface conditions on aircraft ground performance. First of all, we are planning
to certify the wheel force system for flying in order to conduct flight tests. It is interesting how the
forces and moments act during takeoff, as well as during landing, especially at touchdown. Also,
we are planning to perform measurements on winter surfaces, different types of snow, ices, and at
thawing conditions. Also, measuring wheel forces on higher grasses is also planned in the future.
Sensors 2020, 20, 227 14 of 15

Another planned research is dynamic modeling of the airplane motion during takeoff and landing
ground roll.
Regarding the wheel force sensor design, we are planning to develop and build a smaller version,
possible to install into a wheel of a very light aircraft.

Author Contributions: Conceptualization, J.P., P.B.; Methodology, J.P., P.B.; Software, P.B., T.Ł., Validation,
J.M., Formal Analysis, A.T. and J.M.; Investigation, J.P., T.Ł. and J.L.; Resources, J.J.; Data Curation, J.P., T.Ł.
and J.L.; Writing-Original Draft Preparation, J.P.; Writing-Review and Editing, P.B. and T.Ł., Visualization, J.L.;
Supervision, A.T. and D.B.; Funding Acquisition, P.B. All authors have read and agreed to the published version
of the manuscript.
Funding: The project/research was financed in the framework of the project Lublin University of Technology—Regional
Excellence Initiative, funded by the Polish Ministry of Science and Higher Education (contract No. 030/RID/2018/19).
Acknowledgments: The authors are highly indebt to Grzegorz Piwowarski from the Aeroclub of Lublin and
Zdzisław Nowak from the Aeroclub of Rzeszów, for their help in organizing the ground tests with the PZL 104
Wilga 35A airplane.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Pytka, J.; Łyszczyk, T.; Gnapowski, E. Monitoring Grass Airfield Conditions for the GARFIELD System.
In Proceedings of the 5th International Workshop on Metrology for Aerospace (MetroAeroSpace), Rome,
Italy, 20–22 June 2018.
2. Pytka, J.; Tarkowski, P.; Budzyński, P.; Józwik, J. Method for Testing and Evaluating Grassy Runway Surface.
J. Aircr. 2017, 54, 229–234. [CrossRef]
3. Pytka, J.; Józwik, J.; Budzyński, P.; Łyszczyk, T.; Tofil, A.; Gnapowski, E.; Laskowski, J. Wheel dynamometer
system for aircraft landing gear testing. Measurement 2019, 148, 106918. [CrossRef]
4. Hinüber, E.L.V.; Reimer, C.; Schneider, T.; Stock, M. INS/GNSS integration for aerobatic flight applications
and aircraft motion surveying. Sensors 2017, 17, 941. [CrossRef] [PubMed]
5. Baiocchi, V.; Napoleoni, Q.; Tesei, M.; Costantino, D.; Andria, G.; Adamo, F. First tests of the altimetric and
thermal accuracy of an UAV landfill survey. In Proceedings of the 5th International Workshop on Metrology
for Aerospace (MetroAeroSpace), Rome, Italy, 20–22 June 2018; pp. 403–406.
6. Petritoli, E.R.; Leccese, F.; Cagnetti, M. Takagi-Sugeno Discrete Fuzzy Modeling: An IoT Controlled ABS for
UAV. In Proceedings of the IEEE International Workshop on Metrology for Industry 4.0 and IoT, MetroInd
4.0 and IoT, Naples, Italy, 4–6 June 2019; pp. 191–195.
7. Pecora, A.; Maiolo, L.; Minotti, A.; de Francesco, R.; de Francesco, E.; Leccese, F.; Cagnetti, M.; Ferrone, A. Strain
gauge sensors based on thermoplastic nanocomposite for monitoring inflatable structures. In Proceedings of
the 4th International Workshop on Metrology for Aerospace (MetroAeroSpace), Benevento, Italy, 29–30 May
2014; pp. 84–88.
8. Petritoli, E.; Leccese, F.; Leccisi, M. Inertial navigation systems for UAV: Uncertainty and error measurements.
In Proceedings of the 6th International Workshop on Metrology for Aerospace (MetroAeroSpace), Torino,
Italy, 19–21 June 2019; pp. 218–222.
9. Petritoli, E.; Leccese, F. High accuracy attitude and navigation system for an autonomous underwater vehicle
(AUV). Acta IMEKO 2018, 7, 3–9. [CrossRef]
10. Petritoli, E.; Leccese, F.; Cagnetti, M. A high accuracy buoyancy system control for an underwater glider.
In Proceedings of the IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea
Health Parameters MetroSea, Bari, Italy, 8–10 October 2018; pp. 257–261.
11. Eling, C.; Klingbeil, L.; Kuhlmann, H. Real-time single-frequency GPS/MEMS-IMU attitude determination of
lightweight UAVs. Sensors 2015, 15, 26212–26235. [CrossRef] [PubMed]
12. Burkard, H.; Calame, C. Rotating Wheel Dynamometer with High Frequency Response. Tire Technol. Int.
1998, 1998, 154–158.
13. Loh, R.; Nohl, F.W. Mehrkomponenten-Radmessnabe. Einsatzmoeglichkeiten und Ergebnisse. ATZ 1992, 94,
44–53, (In German with English Summary).
14. Kuchler, M.; Schrupp, R. Mehrkomponenten-Motorradmessnabe/Multiaxial Motorcycle Wheel Load
Transducer. VDI-FVT Jahrbuch VDI Verlag Dusseldorf 2002, 10, 91–119, (In German with English Summary).
Sensors 2020, 20, 227 15 of 15

15. Pytka, J. Wheel Dynamometer for Off-Road Vehicle Testing; SAE Technical Paper Series; Paper No. 2008-01-0482;
Society of Automotive Engineers: Warreendale, PA, USA, 2008.
16. Bigoš, M.P.P.; Kelemen, M.; Tonhajzer, R.; Šima, M. Measuring method for feedback provision during
development of fuel map in hexadecimal format for high-speed racing engines. Measurement 2014, 50, 203–212.
17. Feng, L.; Chen, W.; Wu, T.; Wang, H.; Dai, D.; Wang, D.; Zhang, W. An improved sensor system for wheel
force detection with motion-force decoupling technique. Measurement 2018, 119, 205–217. [CrossRef]
18. Meymand, S.Z.; Ahmadian, M. Design, development, and calibration of a force-moment measurement
system for wheel–rail contact mechanics in roller rigs. Measurement 2016, 81, 113–122. [CrossRef]
19. Yuan, C.; Luo, L.P.; Yuan, Q.; Wu, J.; Yan, R.J.; Kim, H.; Shin, K.S.; Han, C.S. Development and evaluation of a
compact 6-axis force/moment sensor with a serial structure for the humanoid robot foot. Measurement 2015,
70, 110–122. [CrossRef]
20. Lyle, K.H.; Jackson, K.E.; Fasanella, E.L. Simulation of Aircraft Landing Gears with a Nonlinear dynamic
Finite Element Code. J. Aircr. 2002, 38, 142–147. [CrossRef]
21. Chester, D.H. Aircraft Landing Impact Parametric Study with Emphasis on Nose Gear Landing Conditions.
J. Aircr. 2002, 39, 394–403. [CrossRef]
22. Wei, X.; Nie, H. Dynamic Analysis of Aircraft Landing Impact Using Landing-Region-Based Model. J. Aircr.
2005, 42, 1631–1637. [CrossRef]
23. Pytka, J.; Józwik, J.; Łyszczyk, T.; Budzyński, P.; laskowski, J.; Gnapowski, E. Measurement of Forces and
Moments Acting on Aircraft Landing Gear Wheel. In Proceedings of the 6th International Workshop on
Metrology for Aerospace (MetroAeroSpace), Torino, Italy, 19–21 June 2019.
24. Pytka, J.; Budzyński, P.; Kamiński, M.; Łyszczyk, T.; Józwik, J. Application of the TDR Moisture Sensor for
Terramechanical Research. Sensors 2019, 19, 2116. [CrossRef] [PubMed]
25. Mitschke, M.; Wallentowitz, H. Dynamik der Kraftfahrzeuge. 4. Auflage; Springer: Berlin/Heidelberg,
Germany, 2004.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).