Analysis and Weight Reduction of a Tractor’s Front Axle

Dilip K Mahanty, Vikas Manohar, Bhushan S Khomane, Swarnendu Nayak

Tata Consultancy Services, India
Swarup Udgata
International Auto Limited,India
Abstract
Engineering components with optimum use of material and easy manufacturability is a direction where
prior simulation through finite element method is found to be very useful. Front Axle of Tractor is one of
the major and very important component and needs very good design as this part experiences the worst load
condition of the whole tractor. The objective of this paper was to analyse the new design of the front axle
of tractor for Thirteen (13) different Certification Test load conditions. The existing design has no field
failure reports; so the results of the existing design were taken as basis for comparison with results of the
proposed models. Based on the finite element analysis results, redesign was carried out for the front axle
for weight optimisation and easy manufacturability. This led to five proposed designs of the front axle
which were evolved based on the above objectives. The proposed designs were evaluated for selected worst
load cases of the existing design. The finite element analysis of new models yielded displacements and
stresses close to the existing design. The increase in stresses were close to 15 % for all five models. The
increase in displacement was not significant but all the new designs conceived had met the structural
requirement. It was also observed that for the proposed designs there was a significant reduction in weight
(approximately 40 %) and the proposed models did not involve a lot of welding, thereby significant savings
of manufacturing was observed. The components used in the assembly were also found to be cost effective
like smaller diameters bearing, smaller knuckle size etc. The reduction in cost of production and weight
significantly reduced the cost of the new design of Front Axle. This analysis work showcases the use of
finite element analysis as a method for reduction of cost in terms of materials and manufacturing.

Introduction
Front Axle of Tractor is one of the major and very important component and needs very good design as this
part experiences the worst load condition of the whole tractor. The objective of this work was to analyse the
current design of the tractor front axle and evaluate the proposed designs for reduction in weight and for
better manufacturability.
The current design was analysed for 13 different Certification Test load cases. Five different models were
proposed based on ease of manufacture and weight reduction. The welding and forming operations required
in the proposed models were less than the current model. Also some of the connected components like
bearings, bushes etc. have been redesigned for improved performance and decrease in cost. These were
done based on field feedback.
The Certification load cases as defined for the project is specified below:

a) Drop Test
In this case a pit of 2.5 feet deep and 2 feet wide and 5 feet long is dug on a very hard ground. The tractor
comes on to it at maximum speed of 35 Kmph and one of the front wheel is allowed to fall into it. The
tractor engine pushes the tractor further till the end of the pit and the engine keeps on humming in this case
even after tractor has reached the end of the pit for some time and finally the engine stops. This is for
tractors with 35/55 HP capacity.
b) Torture Test
The tractor is run on a test track which is having various types of humps/road conditions.
The conditions are described below:
Wheel 1 Condition Wheel 2 Condition
Ok Pot Hole ok Pot Hole
Ok Pot Hole ok Hump
Ok On a small radius Hump ok On a big radius Hump
(i.e. the radius of hump vary across the width of road )
Ok On a Plane Road ok On a Slope
(Like agriculture plot boundary)

Ok On a V road with Humps ok On a V road with

Humps
(The road height is less at the centre and on an inverted V road with humps)

c) ‘8’ Shaped Track Test

The Tractor runs at 35 Kmph speed on a ‘8’ shaped track with three medium sized humps positioned at
120o to each other in each circle of ‘8’. The Steering has to be turned till its locking position is reached
while negotiating a curve.

d) The Impact Test

One side of the tractor collides head on against a rigid wall at a speed of 35 Kmph.

e) Extender wide open Test

The front axle extenders are fully extended and the tractor runs on either type of the V road at a speed of 15
Kmph and also on one side slope condition.

f) Pit Test
At speed of 30 Kmph the tractor goes in side a pit of 10 feet deep with 20 degree slope on either side and
comes out.

g) Worst load test

In this load case all the worst load conditions, except the impact load would be applied on the axle and the
analysis would be carried out. Also an extender wide open test with worst load case would be carried out.
Procedure
The geometric model for the current configuration was created based on the drawings provided. Small
fillets and blends were ignored while creating the model. The geometric model was made using
Unigraphics V16.0 (See Figure 1).

Figure 1 - Solid Model of Current design

The proposed models were based on the sketches, drawings and discussions with the manufacturing
engineers. Five proposed models were created (See Figure 2).

Figure 2 - Solid Model of Proposed design

The geometric models created in Unigraphics were imported into ANSYS V5.6 and FE meshing were
carried out with appropriate type of finite elements.
The current model was meshed using SOLID 45 and SOLID 92 Tetrahedral elements available in ANSYS.
The inner box, outer box and the torch portions were meshed using the SOLID45 elements and the knuckle,
pin, hub, bearings and bushes were meshed using the SOLID92 elements. The bearings, knuckle, hubs and
the bushes were restrained by coupling nodes (See Figure 3).

Figure 3 - Finite Element Model of Current Design

The proposed model was meshed using shell elements and solid tetrahedral elements. The boxes were
meshed using the SHELL63 elements. SHELL63 was considered for the analysis as it saved a lot of time
and effort for meshing of the proposed models with varying thickness. The knuckle, pin, hub, bearings,
torch and bushes were meshed using SOLID92 elements (See Figure 4).

Figure 4 - Finite Element Model of Proposed Design

Major components of the Front axle where FE meshing have been carried out were,
• Outer box
• Inner box with torch
• Hub
• Knuckle and pin
• Bushes and bearings.
The details of the loads applied on the model were calculated and given in the appendix-A. A typical load
and boundary condition is given in Figure 5.

Figure 5 - FE Model of Current Design with Boundary Conditions

The individual components have been coupled together so that there is no free motion between
components. The bearings and bushes have been coupled to the pin and the knuckle on the inner side and
on the outer side they have been coupled with the sleeve / hub. The extender is coupled with the outer box
at the end of the extender and at the bolts. In the proposed model the torch is coupled to the extender. The
vertical restraints are applied on the bolts of the hub. The axial and transverse restraints are applied at the
pin in case of the current model and on the sleeve in the proposed model.
The loads as applied on the current and proposed model are given in Table – 1
Table 1: Directional loads applied for different Certification Test Conditions

Applied Loads

Sr No. Load Case (N)

FX FY FZ

Drop Test
1. 0 12000 7584
(One Wheel in Pit)
Torture Test
2. 0 24000 20120
(Both Wheels in Pit)
Torture Test
3. (One Wheel in Pit; 0 24000 20120
Another Wheel in Hump)
Torture Test
4. (One Wheel on Plane ; 11427 12000 0
Another on Slope)
Torture Test
5. 4368 24000 20120
(‘V’ Road With Hump)
Torture Test
6. ( Small Radius Hump; 0 24000 7388.4
Big Radius Hump)
Extender Wide Open Test
7. ( One Wheel on Slope; 11427 12000 0
One on Plane )
Extender Wide Open Test
8. 4368 24000 20120
( ‘V’ Road with Hump)

9. Pit Test 1640 28510 0

10. ‘ 8’ Shaped Track Test 25000 24000 20120

One Side Impact Test

11. 0 12000 23751.3
(Stops in 0.1 Second)
One Side Impact Test
12. 0 12000 2375.13
( Stops in 1 Second )

13. Worst Loading Condition 25000 28150 20120

The loads applied on the proposed model were:
Worst loading condition
One side impact (stops in 0.1 second) and
Extender wide with worst loading for the models which could be extended.

Analysis
The static analysis of the Front axle was carried out for the 13 load cases mentioned in Table 1 for the
current model. Based on the analysis of the current model of the front axle, five different configurations for
the proposed models of the front axle were created. The challenge was to reduce the weight of the axle in
such a manner that manufacturability would be easier and design criteria were also met.
All the proposed designs were analysed for the worst load case simulation as obtained for the current
model.

Analysis Results & Discussion

Displacement and Stress Results for Current Model

The analysis of the Current model yielded results as specified in Table 2.

Table2: Maximum Stress and displacements in different components

for 13 load cases

Sr Max. Max. Maximum Stress

Load Case Stress Disp.
No. ( N/mm2)
2
(N/mm ) (mm)
Hub Knuckle Box
Drop Test 131
1. 0.7061 27.5 127.5 131
(One Wheel in Pit) (Box )
Torture Test
2. 332.173 1.14 249 197 231
(Both Wheels in Pothole)
Torture Test
58
3. (One Wheel in Pothole; 333 1.123 182 245
(Bolt)
Another Wheel in Hump)
Torture Test
4. (One Wheel on Plane; 84.652 1.769 30.87 31.22 56.66
Another on Slope)

Torture Test 331.7 1.142 243 197 230.44

5.
(‘V’ Road with Hump)
 Max. Maximum Stress
Max.
Sr
Load Cases Stress Disp. (N/mm2)
No.
(N/mm2) (mm) Hub Knuckle Box

Torture Test
6. (Small Radius Hump; 143.3 0.44 134.5 133 119
Big Radius Hump)
Extender Wide Open Test
7. (One Wheel on Slope; 176.88 0.7181 39.6 77.58 158.74
One on Plane)
Extender Wide Open Test
8. 453 2.372 259 270 385
(‘V’ Road with Hump)

9. Pit Test 283 0.64 283 109 131

(Bolt)

241
10. ‘ 8’ Shaped Track Test 353.5 1.286 341 334.6
(Bolt) (Bearing)

One Side Impact Test

11. 320 .55 119 46 320
(Stops in 0.1 Second)
(Bolt) (Bearing)
One Side Impact Test 119
12. 0.2725 119 45.03 56
(Stops in 1 Second ) (Bolt)
384 333
13. Worst Loading Condition 384 1.278 241
(Bolt)

As can be seen the worst load case causes the maximum stress in the hub and the box. . The “8” shaped
track test causes worst load in the knuckle. The extender wide test causes the maximum stress in the box.
But this is a very rare loading event. The maximum displacement for all the 13 load cases is due to the load
case 8 (extender wide open test – on a V road). This is a very rare loading event. These are the results of the
current model and as the current model has reported no failures in the field, these results have been taken as
baseline to compare against the results obtained from the analysis of the proposed models.

Displacement and Stress Results for Modified Design

The displacement and stress values obtained for the modified designs are presented in the Table 3
 Table-3: Maximum Stress and displacements in different omponents
for proposed models.
Maximum Stress

Sr Max. Max. ( N/mm2)

Load Case Disp.
No. Stress
( mm ) Hub Knuckle Box
( N/mm2)

Design: Square Box with Extender

604.174 604 243.4

1. Worst Loading Case 3.017 387
(Hub Bolt) (Bolt) (Bearing)

345.155
159.5
2. Impact Test (Rib & Box 1.831 40.177 345.155
(Bolt)
Connection)

Extended Position 338.5

3. 864.4 8.06 864.4 524.4
(Worst Loading Case) (Bearing)

Design: Square Box without Extender

614.974 3.29 614.974 248.756

1. Worst Loading Case 407.413
(Hub Bolt) (Cap End) (Bolt) (Bearing)

318.223
1.72 174.5 43.5 318.223
2. Impact Test (Rib & Box
(Cap End) (Bolt) (Bearing)
Connection)

Design: ‘ U ‘ Shaped Box with Extender

3.117 619.417
1. Worst Loading Case 619.147 251.2 435.4
(Cap End) (Bolt)

413.58
1.587 166.1
(Bolt of 38.21 316.3
2. Impact Test (Cap End) (Bolt)
Extender)
 Max Stress Max Disp Max. Stress (N/mm2)
Sr.
Load Case
No. (N/mm2) (mm) Hub Knuckle Box

1073
Extended Position 8.835 890.9
3. (Bolt OF 351.042 1073
(Worst Loading Case) (Cap End) (Bolt)
Extender)

Design : ‘U’ Shaped Box without Extender

643.0 3.365 643.0 260.8

1. Worst Loading Case 257.86
(Hub Bolt) (Cap End) (Bolt) (Bearing)

240.422 1.864 178.638

2. Impact Test 44.394 240.422
(Box) (Cap End) (Bolt)
(Rib& Box
Connection)

Design : Square Box with Offset Knuckle (Option 4)

618.12
3.642 618.12 240.258
1. Worst Loading Case (Rib & Box 319.5
(Cap End) (Bolt) (Bearing)
Connection)

289.315 2.217 172.6 44.5

2. Impact Test 289.315
(Hub Bolt) (Cap End) (Bolt) (Bearing)
 Figure 6 - Total displacement plot for the current design

Figure 7 - Total displacement plot for the proposed design

 Figure 8 - Maximum equivalent stress plot for the proposed design

The weights of different models are given in Table 4.

Table 4. Weights of Different Models

Model Weight (KG)
Current Model 127.7
Square Shaped Model with Extender 67.982
Square Shaped Model without Extender 51.91
‘U’ Shaped Model with Extender 68.6
‘U’ Shaped Model without Extender 57.24

Conclusion
The results of the 13 load cases on current model indicate that the structure is safe overall except at some
localized locations where the stresses are high. It is very noticeable that the stresses die down within an
element or two. This shows that the stress is of a very localized nature. The existing design has no field
failure reports; so the results of the existing design were taken as basis for comparison with results of the
proposed models.
The proposed designs were evaluated for selected worst load cases of the existing design. The finite
element analysis of new models yielded displacements and stresses close to the existing design (refer table
3). The increase in stresses were close to 15% for all five models. The increase in displacement was not
significant but all the new designs conceived had met the structural requirement. It was also observed that
for the proposed designs there was a significant reduction in weight (approximately 40 %) and the proposed
models did not involve a lot of welding, thereby significant savings of manufacturing was observed. The
components used in the assembly were also found to be cost effective like smaller diameters bearing,
smaller knuckle size etc. The reduction in cost of production and weight significantly reduced the cost of
the new design of Front Axle.
Of the proposed models, the model U-box with extender was considered as the best based on the utilization
and application. This analysis work showcases the use of finite element analysis as a method for reduction
of cost in terms of materials and manufacturing.
.

References
1. ANSYS Theory Manual.
2. J. E. Shigley, C. R. Mischke, Mechanical Engineering Design, McGraw-Hill, 1989, Singapore.

Appendix-A

1. DROP TEST
For details of the loads (See Figure 9)
Velocity of the Tractor is 35 km/hr (9.722 m/sec).
Acceleration = - u2 /2S = - 31 m/sec2
Force = mass X acceleration = 7584 N.
Hence, vertical force on both wheels = 12000 N
Frontal force on Wheel, W1 = 7584 N

Figure 9 - Drop test details

2. TORTURE TEST WITH POT HOLE & HUMP

For details of the loads (See Figure 10)
 Assuming, velocity reduces from 35 km/hr ( 9.722 m/sec ) to 17.5 km/hr ( 4.861 m/sec) in 431
mm distance.
Acceleration = - (u2-v2)/2S = - 82.24 m/sec2.
Hence, frontal force on both wheel = 20120 N
And Vertical force due to impact load = 24000 N.

Figure 10 - Torture test with pothole & hump details

3. TORTURE TEST WITH TWO POT HOLES

For details of the loads (See Figure 11)
As size of potholes are same, from calculation of point no. 2,
Frontal force on both wheel = 20120 N
And Vertical force due to impact load = 24000 N.

Figure 11 - Torture test with 2 potholes details

4. TORTURE TEST WITH SMALL & BIG HUMPS
For details of the loads (See Figure 12)
Assuming velocity reduces from 30 km/hr ( 8.33 m/sec ) to 15 km/hr (4.17 m/sec ) at big hump.
Acceleration = - (u2-v2)/2S = - 30.16 m/sec2.
Hence, Frontal force on both wheel = mass X acceleration = 7388.4 N
And Vertical force due to impact load = 24000 N.

Figure 12 - Torture test with small & big humps details

5. IMPACT TEST (ONE SIDE COLLIDES AGAINST A RIGID WALL)

For details of the loads (See Figure 13)
Considering time required to stop = 0.1 sec.
Acceleration = - (u-v)/t = - 97.22 m/sec2. Hence, Frontal force on one wheel = 23751.28 N
And Vertical force = 12000 N.

Figure 13 - Impact test (one side collides against a rigid wall) details
6. TEST ON ‘V’ SHAPED ROAD
For details of the loads (See Figure 14)
By solving, R1 = 12770 N = R2
Hence, Transverse component = R1 sin200 =R2 sin200 = 4368 N
Due to humps on the road, Frontal component on both wheels = 20120 N ( from point no. 2).
Vertical force = 24000 N.

Figure 14 - Test on ‘ V ‘ shaped road details

7. TEST FOR ONE WHEEL ON PLANE & ANOTHER WHEEL ON SLOPE

For details of the loads (See Figure 15)
Considering, slope starts from the centre of the tractor ( the middle point of the wheels ),
The angle of the slope = tan –1 {914.4/(663 X cos43.60 )} = 62.30
So, R2 = 12906.4 N. Hence, Transverse component =R2sin62.30 = 11427 N
Vertical force on both wheels = 12000 N.

Figure 15 - Test for one wheel on plane & another on slope details

8. PIT TEST ON 200 SLOPE ROAD

For details of the loads (See Figure 16)
Total reaction force = R1 = 2mg = 4800 N.
So, R1 sin200 = 1640 N & R1 cos200 = 4510 N
Hence, Frontal force on both wheel = 1640 N
 Vertical force on both wheels = 24000 + 4510 = 28510 N. (For impact loading.)

Figure 16 - Pit test on 200 slope road details

9. TEST ON ‘8’SHAPED TRACK WITH HUMPS

For details of the loads (See Figure 17)
Assuming, radius of the track = 2.5 m.
And, velocity of the tractor = 35 km/hr ( 9.722 m/sec).
Cetrifugal force = ( m v2 / r ) = 9345 N .
Maximum bearing force = 25000 N ( given ).
Hence, Ttransverse component = 25000 N ( considering higher value).
Due to humps on the road, Frontal component on both wheels = 20120 N (From point no. 2).
Vertical force = 24000 N. ( for impact loading ).

Figure 17 - Test on ‘8’ shaped track with humps details

10. WORST LOADING CASE

For details of the loads (See Figure 18)
Considering all loading cases, maximum loads at all three directions are,
Transverse component = 25000 N, Frontal component on both wheels = 20120 N .
Vertical force = 28510 N.

Figure 18 - Worst loading case details