RAIL WHEEL

INTERACTION
- Nilmani, Prof. Track
RAIL WHEEL INTERACTION

• RUNNING OF A RAILWAY VEHICLE OVER A

LENGTH OF TRACK PRODUCES DYNAMIC
FORCES BOTH ON THE VEHICLE AND ON
THE TRACK

• THE INTERACTION AFFECTS BOTH TRACK

AND RAILWAY VEHICLE – RAIL WHEEL
INTERACTION
EFFECT OF VEHICLE ON TRACK
LARGE DYNAMIC FORCES

– DETERIORATION OF TRACK GEOMETRY

– TRACK COMPONENT WEAR & DAMAGE

– NOISE
 EFFECT OF TRACK ON
VEHICLE
• SAFETY

• RIDING COMFORT

• COMPONENT WEAR & DAMAGE

UNDERSTANDING RAIL -WHEEL
INTERACTION

• DERAILMENT BY FLANGE MOUNTING

• WHEEL CONICITY AND GAUGE PLAY
• WHEEL OFF-LOADING
• CYCLIC TRACK IRREGULARITIES-
RESONANCE & DAMPING
• CRITICAL SPEED
• TRACK / VEHICLE TWIST
SELF CENTRALIZING CONED WHEELS
Sinusoidal motion of wheelset
SINUSOIDAL MOTION OF VEHICLE
 PLAY BETWEEN WHEEL SET
AND RAILS

• G = Gw + 2 tf + s
• G is track gauge 1676 mm (BG)
• Gw is wheel gauge 1600 mm (BG)
• tf is flange thickness –
28.5 mm new; 16mm worn out
• s is standard play
= 19mm for new wheel
= 44mm for worn out wheel
• Mean
Position

• Typical
(Asymm
etrical)
Position

• Extreme
Position
EFFECT OF PLAY

Lateral Displacement Y = a sin wt

a  amplitude = /2 = Play/2
Lateral velocity = aw cos wt
Lateral Acceleration = -aw2 sin wt
Max acc = -aw2

Angular Velocity w = 2 
2f 

KLINGEL’S FORMULA (1883)

Wave Length 0 of a Single wheel

0 rG
= 2
2
G = Dynamic Gauge
r = Dynamic Wheel Radius
 = Conicity
1
0  ;Frequency  

CONCLUSIONS
•With increase , 0 reduces, f increases – oscillations
increase – instability
•For high speed  – low 1 in 40 on high speed routes
•Worn out wheel  increases – increasing instability

For wheel set (MULTIPLE RIGID WHEELS)

2
l 
   1   
G

l= Rigid wheel base

EFFECT OF PLAY
4 v2 2
acc  a.
2

1
acc 
2

•As conicity increases Lateral Acceleration

Increases
•acc  a  /2 play
•As play increases Lateral Acceleration
Increases
CONCLUSIONS
• EXCESSIVE OSCILLATIONS DUE
TO
– Slack Gauge
– Thin Flange
– Increased Play in bearing & Journal
– Excessive Lateral and Longitudinal
Clearances
• Increased Derailment Proneness
Wheel-set on Curve
THE PROCESS OF FLANGE CLIMBING
DERAILMENT
SECTIONAL PLAN OF WHEEL
FLANGE AT LEVEL OF FLANGE
TO RAIL CONTACT
ZERO ANGULARITY (PLAN)
POSITIVE ANGULARITY (PLAN)
NEGATIVE ANGULARITY (PLAN)
EXAMPLES OF WHEEL SET COFIGURATION
WITH POSITIVE ANGULARITY
ZERO ANGULARITY
(ELEVATION)
POSITIVE ANGULARITY
(ELEVATION)
NEGATIVE ANGULARITY
(ELEVATION)
FORCES AT RAIL-WHEEL CONTACT AT
MOMENT OF INCIPIENT DERAILMENT
 FORCES AT RAIL-WHEEL CONTACT AT
MOMENT OF INCIPIENT DERAILMENT
• Resolving Along Flange Slope
R= Q cos  + Y sin …. 1.
For safety against derailment
• Derailing forces > stabling forces
• Y cos  + R > Q sin 
• Substituting R from equation 1
 Y cos  +  (Q cos  + Y sin ) > Q sin 
 Y (cos  +  sin ) > Q (sin  -  cos )

Y (sin    cos  )
 
Q (cos    sin  )
Nadal’s Equation (1908)
Y tan   

Q 1   tan 
For Safety: LHS has to be small. RHS has to be
large
Y  Low
Q  High
  Low
tan   Large
 FACTORS AFFECTING
SAFETY
Flange Slope

•  = 90º would indicate higher safety.

• However, with slight angularity, flange contact

shifts to near tip.

• Safety depth for flange reduces resulting into

increase in derailment proneness
 FACTORS AFFECTING
SAFETY
Flange Slope

• ANGULARITY is inherent feature of

vehicle movement. If the vehicle has
greater angularity,  should be less for
greater safety depth of flange tip.
• However, there is a limit to it, as this
criterion runs opposite to that indicated
by Nadal’s formula.
 FACTORS AFFECTING
SAFETY
Flange slope

• On I.R., for most of rolling stock  = 68º 12’

(flange slope 2.5:1)
• For diesel and electric locos, the outer wheels
encounter greater angularity for negotiation of
curves and turnouts. For uniformity, same 
adopted for all wheels.
•  kept as 700 on locos upto 110 kmph
•  kept as 600 on locos beyond 110 kmph
 FACTORS AFFECTING
SAFETY
Flange slope

• With wear  increases, but results in greater

biting action, hence increase in .
 FACTORS AFFECTING
SAFETY
OTHER FACTORS INFLUENCING NADAL’S
FORMULA

–  INCREASES WITH INCREASED ANGULARITY, 

(PROF. HEUMANN)

 
(acting upwards for positive
0.0 0.0
angularity)
0.02 0.27
 FACTORS AFFECTING
SAFETY
OTHER FACTORS INFLUENCING NADAL’S
FORMULA

•Greater eccentricity (positive angularity)

increases derailment proneness as flange safety
depth reduces.
•Persistent Angular Running
•As positive angularity increases derailment
proneness, persistent angularity leads to greater
chances of derailment.
 FACTORS AFFECTING
SAFETY
• Not possible to know values of Q, Y, ,  and
eccentricity at instant of derailment.
• Calculations by NADAL’s formula not to be
attempted.
• Qualitative analysis by studying magnitude of
defects in track/vehicle and relative extent to
which they contribute to derailment proneness,
should be done.
 DEFECTS/FEATURES
AFFECTING 
1. Rusted rail lying on cess, emergency x-
over
2. Newly turned wheel – tool marks
3. Sanding of rails (on steep gradient,
curves)
4. Sharp flange (radius of flange tip < 5mm)
increases biting action
 DEFECTS/FEATURES CAUSING
INCREASED ANGLE OF ATTACK

• Excessive slack gauge

• Thin flange (<16mm at 13mm from flange tip for
BG or MG)
• Excessive clearance between horn cheek and
axle box groove
• Sharp curves and turnouts
• Outer axles of multi axle rigid wheel base subject
to greater angularity, compared to inner wheel
 DESIGNED ANGULARITY WHILE
NEGOTIATING CURVE

PLAY HELPS THE WHEEL NEGOTIATE

CURVE
 DEFECTS/FEATURES FOR
INCREASED ANGLE OF ATTACK

• DEFECTS / FEATURE FOR INCREASED

POSITIVE ECCENTRICITY

• WHEEL FLANGE SLOPE BECOMING STEEPER

(THIS DEFECT REDUCES SAFETY DEPTH AS
ECCENTRICITY INCREASES)
 DEFECT/FEATURES CAUSING
PERSISTENT ANGULAR RUNNING

• DIFFERENCE IN WHEEL DIA MEASURED ON

SAME AXLE
• INCORRECT CENTRALISATION & ADJUSTMENT
OF BRAKE RIGGING AND BRAKE BLOCKS
• WEAR IN BRAKE GEARS
• HOT AXLE
• HIGHER COEFFICIENT OF FRICTION
• DIFFERENT BEARING PRESSURES
 STABILITY ANALYSIS
• Q & Y – Instantaneous values, measurement by
MEASURING WHEEL

• Hy = Horizontal force measured at axle box level

• Q = (Vertical) spring deflection x spring

constant
 STABILITY ANALYSIS
Nadal’s Formula
Y tan    Dry Rail 0.33
 Wet Rail 0.25
Q 1   tan  Lubricated Rail 0.13
Rusted Rail 0.6
•for =68º, =0.25
RHS works out to 1.4, rounded off to 1.0 for a
factor of safety
•On I.R. Hy/Q measurement over 0.05 sec.
•It is one of the criteria for assessing stability of
Rolling Stock
• LIMITING VALUES OF Y/Q RATIO VS TIME
DURATION (JAPANESE RAILWAY)
• DIRECTION OF SLIDING FRICTION AT TREAD OF
NON-DERAILING WHEEL
CHARTET’S FORMULA
Y Qo
 K1  K 2
Q Q
tan   
K 1   '
1   tan 
K2 = 2(’+  )

  2
'

 = Angle of coning of wheel  = 1/20 = 0.05

 = 0.25 K1 = 2 K2  0.7
FOR SAFETY

y Qo
 2  0.7
Q Q

 Y >2Q – 0.7Qo
 2Q <Y + 0.7Qo
As Y  0 2Q <0.7Qo  Q < 0.35Qo
Instantaneous Wheel Load Q should not
drop below 35% of nominal wheel load Qo.
For safety, the Q limited to 60 % of Qo
EFFECT OF TWIST ON VEHICLE
 REFERRING TO FIG.
• a = Distance between centres of the
spring A&B bearing on the wheel set
• PA = Load reaction in spring A
• PB = Load reaction in spring B
• G = Dynamic gauge
• R1 = rail reaction under wheel –1
• R2 = rail reaction under wheel-2
e = amount of overhang of spring centre
beyond the wheel rail contact point.
Contd..

• Let the rail under wheel-2 be

depressed suddenly by an extent
½ Zo, so that the instantaneous
value of R2 becomes zero, i.e. the
wheel load of wheel-2 drops to
zero.
Contd..
• Under effect of lowering of rail, spring B
would elongate and load reaction PB
would drop
Taking moments about spring B
• PA a+ m (G/2+e)
= R1 (G +e) + R2e
= R1 (G+e) (since R2 = 0) (i)
Taking moments about spring A
• PB a+ m (G/2+e)
= R1e + R2 (G+e)
= R1e (since R2 = 0) (ii)
 Contd..

Subtracting (i) from (ii)

(PA-PB) a = R1G
or PA – PB = R1 G/a
Now, R1 + R2 = T/2
 R1 = T/2, since R2 = 0
 PA – PB = T/2 *G/a
 Contd..

• Difference in deflections of the springs A &

B (owing to difference in the load reactions
viz PA – PB = f (PA – PB), where f is specific
deflection (assuming f to be the same for all
the springs)
• That is, the difference in deflections of the
two springs. = f *T/2* G/a
• By geometry, the above difference implies
a difference in the levels of the two rails
under the wheel set under consideration,
to be

2
 T G G T G 
f   f  
 2 a a 2 a
• Obviously, this is the extent by which
the rail under wheel 2 would required
to be depressed to reduce R2
instantaneously to zero.
• i.e. 2
T  G 
2 Z0  f  
1
2a
 EFFECT OF TRACK AND
VEHICLE TWIST

• Track Defect that will completely off

load the wheel
Zo = fT (G/a)2
• This equation is given by Kereszty
 EFFECT OF STIFFNESS OF
SPRINGS

• Larger ‘f’ i.e. deflection per unit,

better from off loading point of
view
 LOADED / EMPTY
CONDITION OF VEHICLE

• Larger the ‘T’, Better it is

• An empty wagon is more prone for
derailment
• G/a RATIO - should be Large

• Overhang should be less

• G/a ratio less for MG than BG

VARIATIONS IN SPRING
STIFFNESSES
One spring
Zs = x/2 (G/a)
x  Defect in one spring

Diagonally opposite springs

Zs = x (G/a) = 2Zs
 TORSIONAL STIFFNESS OF
VEHICLE UNDER FRAME
Converted to Equivalent Track Twist

• Zu =  T/4 (G/a)2
•   Specific deflection of a corner of
under frame
• Torsionally flexible under frame is
desirable
• Riveted under frame desirable as against
welded one
TRANSITION OF A CURVE
• Track
– Cant Gradient should be as flat as possible

• Effect on vehicle
– Zb = i.L,
– i  Cant Gradient
– L = wheel base

• Longer wheel base is not desirable

PERMISSIBLE TRACK TWIST

Zperm = 0.65 Zo – Zs + Zu – Zb
(If one spring is defective)

Zperm = 0.65 Zo – 2Zs + Zu – Zb

(If two diagonal springs are defective)
 Motions of Vehicle
• (a) Linear oscillation; (b) rotational
oscillation
Motions of Vehicle

HUNTING: COMBINED ROLLING + NOSING (VIOLENT

MOTION)
 TRACK & VEHICLE DEFECTSS
CAUSING VARIOUS PARASITIC
MOTIONS
A. TRACK DEFECTS PARASITIC MOTION
• X-Level • Rolling
• Loose Packing • Bouncing, Rolling
• Low Joint • Pitching
• Alignment • Nosing, Lurching
• Slack Gauge • Nosing, Lurching
• Versine Variation • Nosing, Hunting
VEHICLE DEFECTS CAUSING
PARASITC MOTION
A. VEHICLE DEFECT PARASITIC MOTION
• Coupling • Shuttling, Nosing
• Worn wheel • Hunting, Nosing,
Lurching
• Ineffective spring • Bouncing, Pitching,
Rolling
• Side Bearer • Rolling, Nosing
Clearance • Nosing
• In-effective Pivot
Contd…..

TRACK DEFECT MODE OF AFFECTS’

OSCILLATIONS VALUE
Low joint Bouncing & Pitching Q
Unevenness
Loose Packing
Alignment Lurching, Nosing & Y
Gauge Fault Rolling Q
Twist Rolling Q
The above track defects when occurring in cyclic form would
cause external excitation
Hence, Adequate Damping Necessary
EFFECT OF CYCLIC TRACK
IRREGULARITY ON VEHICLE
Contd…
• Oscillation mode of vehicle will be
bouncing and pitching
• For speed v = 13 m/s, excitation freq.
t=13m
• For speed v = 26 m/s’ excitation freq. =
2 cps, t=13m
• This is “forcing” frequency, f = v/t
 Contd…
• “Natural” frequency of a vehicle in a particular
mode of oscillation : Frequency of osc. in that
mode, when system oscillates freely, after
removal of external forcing frequency.
• For simple spring of stiffness “k” & mass “m”
natural freq.,

1 k
fn= 2 m

• For 2 stage suspension system, there would be

2 natural frequencies.
 RESONANCE
Natural Frequency =
1 k
2 m
k = Spring Stiffness
m = Mass
If frequency caused by external excitation
is equal to natural frequency, resonance
occurs under no damping condition
• FOUR TYPES OF FREE OSCILLATIONS FOR SAKE OF
COMPARISON
• MAGNIFICATION FACTOR VERSUS FREQUENCY RATIO
FOR VARIOUS AMOUNTS OF DAMPING FOR SIMPLE
SPRING MASS DAMPER SYSTEM SHOWN.
 PRIMARY HUNTING
• WHEN THE BODY OSCILLATIONS
ARE HIGH WHILE THE BOGIE IS
RELATIVELY STABLE
• EXPERIENCED AT LOW SPEEDS
• MAINLY AFFECTS RIDING COMFORT
 SECONDARY HUNTING
• WHEN THE BODY
OSCILLATIONS ARE RELATIVELY
LESS. WHILE THE BOGIE
OSCILATIONS ARE HIGH
• EXPERINCED AT HIGH SPEEDS
• AFFECTS VEHICLES STABILITY
CRICITAL SPEED
• THE SPEED AT THE BOUNDARY
CONDTION BETWEEN THE STABLE &
UNSTABLE CONDITON IS CALLED AS
CRITICAL SPEED OF THE VEHICLE
• THE SPEED FOR WHICH THE ROLLING
STOCK IS CLEARED FOR THE SERVICE
IS NORMALLY ABOUT 10 TO 15% LESS
THAN THEC RITICAL SPEED AT WHICH
THE VEHICLE HAS BEEN TESTED.
 FACTORS AFFECTING CRITICAL
SPEED
1. VEHICLE WHEEL PROFILE
2. RAIL HEAD PROFILE, INCLINATION &
GAUGE
3. RAIL WHEEL COEFF. OF FRICTION
4. AXLE LOAD AND DISTRIBUTION OF
VEHICLE MASS
5. DESIGN AND CONDITION OF VEHICLE
SUSPENSION
LATERAL STABILITY OF
TRACK
LATERAL TRACK DISTROTION
DUE TO
• LESS LATERAL STRENGTH
• EXCESSIVE LATERAL FORCES
BY VEHICLE
LATERAL STABILITY OF
TRACK
THIS STUDY IS IMPORTANT
FOR
• ASSESSMENT OF STABILITY OF
ROLLING STOCK
• INVESTIGATION OF
DERAILMENTS
PRUD’HOME’S FORMULA
–Hy > 0.85 (1+P/3)
–Hy> LATERAL FORCE
–P = AXLE LOAD (t)
 Allowable Twist in Track
RDSO letter no CRA 501 dtd 29.04.83
Speed Peak value of Peak value of Twist
(KMPH) UN on 3.6m TW on 3.6m (mm/M)
chord (mm) chord (mm)
75 14 13 1 in 276
60 16 15 1 in 240
45 22 22 1 in 163
30 24 25 1 in 144
15 33 30 1 in 120
 Allowable change in gauge
• Maximum gauge variations permitted are not
laid down. In any case, the safety tolerances
are not laid down in the manual
• Para 237(8)(a): It is a good practice to maintain
uniform gauge over turnouts
• If there is a derailment over P & Cs, the gauge
variation is often a point of controversy.
• We shall use the allowable variation in versines
to check if the gauge variation is within limits