International Journal of Computer Communication and Information System ( IJCCIS)

– Vol2. No1. ISSN: 0976–1349 July – Dec 2010

Design and Development of Torque Testing Rig for a Gearbox

Amruta Lomate1, Suhas Mohite2, and Rahul Shinde3
1
Mechanical Department, Bharati Vidhyapeeth, College of Engineering, Pune, Maharashtra
Amu2384@[Link]
2
Mechanical department, Govt. College of Engineering, Karad, Maharashtra
3
Executive director, Cyclo Transmission Ltd. Satara, Maharashtra
mohitess@[Link], cyclobox@[Link]

Abstract - Gearbox is an indispensae element of Different types of torque testing rig available in market
power transmission drives of most mechanical are studied. From above literature it is found out that each
systems. Therefore, it is very essential to assure the method has some advantages and disadvantages. For torque
performance of gear drives before they are put to testing back to back, digi-torque and multi-plate brake
use to ensure the trouble free functioning of systems are studied “Ref. [1-3]” For back to back
systems. While testing performance of the gearbox, arrangement two similar gear boxes are required, which
it is importance to check the torque carrying may not be available in some of the cases. In digi-torque
capacity at rated speeds. This work presents the method the flywheel was used to apply the torque, which is
design and development of a torque testing rig for not suitable for high torque measurement. Both these
0.5-10 kN-m capacity carried out for a gearbox methods are not suitable for this application.
having multi-plate brake system. While calibration
Multi-plate brake torque testing system (torque
of torque testing rig experimentally measured data
applying) is suitable because it can provide wide range of
is compared with the theoretical calculations and a
torque (from 500 N-m to 10 KN-m) at different speeds. The
good agreement between experimental and
brake plates are partially wetted to avoid wear and control
theoretical calculations is observed.
the temperature of brake plates. For actuating multi-plate
Index terms –Gear box, design, torque, calibration brake system, actuators are needed. While studying
different brake actuators [4]., it was found that hydraulic
I. INTRODUCTION
actuator is suitable for applying the pressure for multi plate
Gearbox is an indispensable element of power brake system, because it provides fast response along with
transmission drives of most mechanical systems such smooth engagement by controlling the rate of pressure
as vehicles, machinery and cranes, etc. The main built-up with a pressure control valve. By referring to the
function of the industrial gear drive is to reliably literature in the point of torque measurement, it was found
transmit torque and rotary motion between prime out that canister type load cell can provide high torque
mover and driven piece of equipment, at acceptable measurement because this load cell is suitable for high
level of noise, vibration and temperature. When one or torque or load range and also eliminates the need of prior
more of the preceding operating characteristics exceeds checking [5]. Finally it is decided to design and develop a
allowable limits, it can lead to failure of the gearbox. hydraulically actuated multiplate brake torque testing rig
Manufacturing companies or customers have to bear for different gearbox, maximum torque of 10 kNm at rated
big cost and time for the failure of gear drive system speed.
during operation. Therefore, it is very essential for the
manufacturer to assure the quality of gear drives before II. METHODOLOGY
they are put to use to ensure trouble free functioning of The proposed system has multi plate brake assembly as
systems. a main and core component. The assembly is mounted in
between the bracket and fixed on and around the shaft.
While testing the performance of gearbox, it is
importance to check the torque carrying capacity at
different speeds. At present, Cyclo transmission ltd.
does not have any torque testing facility. So to measure
the torque; company has to rely on theoretical
calculations according to ISO (International Organization
for Standardization) and AGMA (American Gear
Manufacturers Association) standards to specify torque
ratings. Therefore, it is very necessary for them to have
a test rig for gearboxes to assure quality of the gearbox.
Torque testing rig is an essential facility that every gear
manufacturing company must have.
To fulfil above needs of the torque testing rig, a
literature survey was undertaken with the view to Figure 1 Outline of the proposed gearbox test rig
finalize its features and specification.
12
 International Journal of Computer Communication and Information System ( IJCCIS)
– Vol2. No1. ISSN: 0976–1349 July – Dec 2010

Multi-plate brake system is used to apply the Where, Mt is maximum torque, F is working force, Z is
torque on the gearbox. For actuation of the brake, number of matching surfaces, µ is coefficient of friction,
hydraulic actuator was used. Brake system is directly D10 is inner diameter of outer brake plate and d2i is outer
coupled with the gearbox with the help of coupling or diameter of inner brake plate. Total number of brake plate
gear pair. As the shaft is coupled with gearbox shaft, is 27. Shaft is basically designed for crushing, shear,
brake force is transmitted to the gearbox and at the torsional and bending type of loading. Maximum shear
same time torque arm (torque arm is fixed with brake stress theory is used for determining diameter of shaft
assembly). Torque arm in turn exert the same amount because material of the shaft is ductile
of force on load cell.
16
To maintain normal temperature of the brake plate, τ= 2 2
(k b M b ) + ( k t M t ) (2)
proper cooling system is designed. The total system is πd 3
divided into following sub systems, viz., Where, τ is maximum shear stress, d is diameter of shaft, k b
A. Brake assembly and kt are shocks and fatigue factors respectively and Mb is
B. Actuation system bending moment. The minimum diameter of the shaft is
120mm.
C. Cooling system
D. Torque measurement and display unit Actuation System
III. DESIGN AND MANUFACTURING The hand operated hydraulic actuator has capacity of
20 bars is used.
To design a torque testing rig for a gearbox is an
innovative idea. On the basis of the conceptual design, C. Thermal Calculations and Design of Cooling
the main system is divided in to a number of System
subsystems carrying out specific functions as follows.
In case of multi plate brake, heat dissipation is the
A. Brake assembly main problem. As the temperature of oil increased above
B. Actuation system 900C, oil does not obeyed its properties, because that wear
of the brake plates will be takes place so to avoid such
C. Cooling system
problem water cooling jacket is provided.
D. Torque measurement and display unit
The overview of the test rig depicting different sub Torque Measurement and Display Unit
systems is as shown in “Fig. 2”. For torque transfer from brake assembly to the load
cell, torque arm of 1000 mm was used. For torque
measurement canister type load cell of capacity 2000 kg is
used with digital display. Resolution of ± 0.1 kg-m in
100 kg-m is used.
After design manufacturing and assembly is done as
per drawings. The actual view of test rig is as shown in
“Fig. 3”

Figure 2 Overview of the designed torque testing rig

A. Brake Assembly
Brake assembly is the main part of test rig. During
design of brake assembly, following procedure was
followed. Selection of materials “Ref. [6, 7]”, design
and analysis of various components like shaft, brake
plate, piston, cylinder, torque arm, selection of
bearings, oil seals, and piston rings etc were done. Figure 3 Torque testing rig
Some of the important calculations are shown below. IV. EXPERIMENTAL WORK
For finding number of brake plates required we used In the experimental work the set-up the test rig as
uniform wear criteria for brake plate design. “Ref. [8,
shown in “Fig.4” and by using the few gearboxes
9]” calibration of test rig have been done in following way. The
µFZ × ( D1O + d 2 i ) analytically determined torques values were compared with
Mt = (1) the measured torque values for different models of the gear
4
13
International Journal of Computer Communication and Information System ( IJCCIS)
– Vol2. No1. ISSN: 0976–1349 July – Dec 2010

box. The analytical and the experimental readings are Figure 5 Comparison of analytical and experimental
compared. torque values

14
Table I.
Comparison of analytical and experimental torque
values
Sr. Experimenta Analytical Absolute %
No. Pressure l Torque Torque error Error
1 2 1044 1002 42 4.12
2 3 1545 1503 42 2.79
3 5 2550 2505 45 1.79
Figure 4 Schematic arrangement of the test rig for 4 8 4052 4008 44 1.09
calibration 5 10 5050 5010 40 0.79
These results are given for different pressure 6 12 6055 6012 43 0.71
values ranging from 2 bars to 20 bars. The calibrated 7 15 7559 7515 44 0.59
measurement units (load cell, digital display and 8 16 8061 8016 45 0.56
pressure gauge) are used. For finding theoretical value 9 18 9067 9018 49 0.54
following equation is referred.
10 19 9568 9519 49 0.51

µ × π × Z × ( D − d ) × ( D1O + d 2 i ) 2 2 11 20 10069 10020 49 0.49

Mt = × Ph
16 It is observed that the experimental readings for torque
(3) are higher than analytical values by an absolute margin of
Where, Ph is hydraulic pressure, D is outer diameter of 40-50 Nm for the entire torque range. In terms of the
piston and d is inner diameter of piston. The test rig can percentage this accounts for approximately 5 to 0.5%.
also be tested for duty cycle test of some of the gear Further, higher error (5%) is observed at lower torque (and
boxes. lower pressure) whereas, lower error (0.5%) is observed at
V. RESULTS AND DISCUSSION higher torque values. The reasons for this error could be
attributed to various causes that are not considered in
The testing has been done on different types of theoretical calculations. The possible causes are mentioned
gearboxes, different types of testing is done the results below.
are discussed below. Calibration of the test rig carried
out by comparing experimental result with theoretical a) In analytical calculation, friction in the bearings,
values. The test results of a few models are done, one is oil seals and other moving parts is not considered.
discussed below. b) The tested gear boxes have a lower operating
1. Gear box Model number 4010-4,[11]. speed (5-10 rpm) at such speeds the stick-slip
phenomenon in between the brake plates is more
Specifications: Input power- 10H.P., significant, this could be one of the sources of the
Output speed- 5.75 r.p.m, error.
Output torque- 10.5 Nm. c) Inertia of the system.
Testing was carried out for 10 minutes at different d) When the hydraulic pressure increases the friction
constant pressure values and readings were observed in the various parts increases.
after every 2 minutes to allow for the system to get e) Errors in load cell readings.
stabilized.
f) Errors in display unit.
12000 Experimental Torque, Nm

10000
Analytical Torque, Nm g) The error in the pressure gauge also the pressure
indicator is analogue display so the manual
8000
mistake may be occurred.
Torque, Nm

6000

4000 The obtained results were compared to the

2000
theoretically calculated and a good agreement between the
0
experimentally obtained and theoretically calculated results
2 3 5 8 10 12 15
Pressure, Bar
16 18 19 20
were seen. The analytical and experimental values are
closely matched.
A. Comparison of errors in experimental values with
respect to torque
 942

941

940
International Journal of Computer Communication and Information System ( IJCCIS)
939
– Vol2. No1. ISSN: 0976–1349 July – Dec 2010

Torque, Nm
938

937

After calibration of the test rig, it was found that in output torque 940 936

the system there are three types of error, error in load Nm. 935

cell, error in brake assembly and error in digital 0 10 20 30 40

Ti me ,mi nu te s
50 60 70

display. After calibration of test rig, total error was

Figure 7
determined for each torque value, which is tabulated in
Comparison of
the “Fig. 6”
analytical and
5 experimental
% Error in load cell
% Experimental error torque readings
4
with time
3
From above graph it was observed that the torque
% Error

2 value is fluctuating in-between 936 to 941Nm with respect

to time. That variation is negligible (approx. 0.5%).
1

C. Relation between power and outlet temperature

0
0 200 400 600 800 1000 Temperature is an important parameter while
Load, Kg
calibrating the test rig, because if the oil temperature is
increased beyond 900 C, the lubricating oil will lose its
properties and rig will not work efficiently. Cooling jacket
is used for heat dissipation purpose. Water is used as a
Figure 6 Error analysis of torque testing rig coolant. The mass flow rate of the water was taken as 9.5
lit/minutes.
While actual testing of the gearboxes,
corresponding error factor should be added in the
experimental readings. Gear box was tested for 10
minutes. Reading was observed for every two minutes 45
Inlet Temperature
and average torque reading was taken, which is shown
Outlet Temperature
in below “Table 2”. 40
C
0
Temprature,

Table 2 Tests conducted on the different models 35

(output torque)
30
Sr. Model Actual Experimental
No No torque values Output speed 25
1 4005-2 556 590 88.5 0 1 2 3 4 5 6 7 8
Inpu t pow er (H P )
2 4005-2 754 794 65.3

3 4005-2 1049 1090 47

4 4006-3 2276 2321 20.8 Figure 8 Comparison of temperature and input power
at different models
5 4006-3 3142 3184 15
It is observed in “Fig. 8” that as the input power
6 4007-3 3847 3889 12.3 increases the temperature values also increase steadily.
7 4007-3 4290 4331 11 XI. CONCLUSION
8 4007-3 5210 5250 9.1 Design and development of hydraulically actuated
multi-plate brake system is carried out for measuring
9 4008-3 6500 6542 7.27 torque on the gearbox. System is designed for measuring
torque from 0.5 kNm to 10 kNm at rated speed. The test
10 4008-3 7280 7323 6.52 rig is calibrated with theoretical values. During calibrating
11 4007-4 8477 8517 5.35 of the systems, it is found out that the maximum error in
the measured torque using the test rig is about 50 Nm
(carried out on few gear boxes using the test rig and the
From the above table, it was observed that all the results are in good agreement with the theoretical values.
gear boxes are successfully running for output torque at The test rig can also be used for peak load test and duty
given speed. cycle test of some of the gear boxes. Temperature of
B. Duty cycle test cooling water is measured and it is found to be well within
the accepted limits. The overall performance of the test rig
For the gearbox, it is necessary to carry out the is found to be trouble free.
duty cycle test. One such result of the duty cycle tests
conducted on gearbox is presented in “Fig.7”. The gear XII. ACKNOWLEDGMENTS
box model 4005-2 with an output speed 52.4 rpm and

15
International Journal of Computer Communication and Information System ( IJCCIS)
– Vol2. No1. ISSN: 0976–1349 July – Dec 2010

The authors wish to thank Cyclo transmission Ltd.,

Satara for providing not only an opportunity to work on
this project but also providing all the technical know-
how and practical experience in the factory as well as
in the design office.
XIII. REFERENCES
[1] Walter Musial, “Dynamometer test bed”, National
renewable
energy laboratory wind technology center, 2001, pp. 1-3
[2] Automation Technology -Motor Testing System brochure
(MTS 3800), “Theory of Digitorque Opration”,Dayton,Ohio
2002, pp. 1-2
[3] Paul G. Marting And Nick J. Colarelli, “Low Speed Plate
Brake Tester”, SAE technical paper series 901701, 2007, pp. 5-
8
[4] “Power transmission products fundamentals” Power
transmission design magazine 1997, pp. A136-A141
[Link]/eptoot/public/schools/[Link]
[5] “Transaction in measurement and control”, a technical
reference series brought by Omega vol-3 © Putman Publishing
Company and OMEGA Press LLC.1998,pp57-61
[Link]/literature/transactions/volume3/[Link]
[6] Bauccio M. “ASM Metals Reference Book” 3rd edition,
ASM International, Materials Park, OH, 1993
[Link]/materials/materials_home
[7] Harvey P. D. “Engineer Properties of Steel”, American
Society for Metals, Metal Park, OH, 1982 [Link]
[8] V. B. Bhandari, “Design of machine elements” Tata
Mcgraw Hill, Second edition-2008, pp. 228-290,323-385,444-
495
[9] Harvey P. D. “Engineer Properties of Steel”, American
Society for Metals, Metal Park, OH, 1982 [Link]
[10] Joseph E. Shigley, “Mechanical engineering design”, Tata
Mcgraw Hill, Sixth edition, 2006, pp. 985-1034, 445-496
[11] Cyclo planetary and cycloidal drives –
Manuals, Cyclo Transmission Ltd. edition 2006 pp. 47-49

16