IRJET- Analysis and Design of a Formula SAE Powertrain

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 03 | Mar 2019 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 448
Analysis and Design of a Formula SAE Powertrain
Prasantha Laxman Pujari
1Mechanical Engineering Department, G.V. Acharya Institute of Engineering and Technology, Shelu Road, Shelu,
Maharashtra – 410101, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - This project aims are to design and develop the
powertrain system for formula SAE race car and the design
that used for the car must fulfill the standard regulation that
made by SAE for Formula SAE competition. Besides that, the
load of the axle also been determined to get the maximum
torque applied using three different approached that are
engine performance, tire friction, and predicted acceleration
and the cornering force acting on the race car.
Key Words: Engine Unit, Cbr250R, Intake & Exhaust,
Cooling System, Fuel Pump
1. INTRODUCTION
The stiffness of mounting system determines the vibration
isolation ability of the transmitted path, which is the key
factor that affects the vibration and noise of vehicle.Inorder
to improve the vibration isolation ability of the powertrain
mounting system, considering the powertrainoffront wheel
drive car as the research object, the vibration decoupling
rate and its corresponding frequency of the powertrain
mounting system are analyzed by rigid body dynamics and
energy method
2. ENGINE UNIT
2.1 MODULE OBJECTIVES
To deliver a reliable and weight optimized engine system
with sufficient power and the possibility for low fuel
consumption. An engine has to be selected from either
Honda CBR250R or Honda CBR600RR.
2.2 ENGINE SPECIFICATIONS
Fig -1: Honda CBR600RR, Honda CBR250
Table -1: Engine Comparison
Engine Comparison
Specifications CBR 250R CBR 600RR
Bore (mm) 76 67
Compression Ratio 10.71 12.21
Cooling Liquid Liquid
Cylinders 1 4
Displacement (cc) 249.4 599
Engine Configuration Single -
Cylinder
Horizontal In -
Line
Engine Type 4 - stroke 4 - stroke
Fuel Injector Yes Yes
Stroke 55 42.5
Valves Per Cylinder 4 4
Peak Torque 23 Nm
@7000 rpm
66 Nm @ 11250
rpm
Peak Power 20 kW @8500
rpm
88 kW @ 13500
rpm
Power / Weight 0.67 kW / kg 1.25 kW / kg
2.3 REASONS FOR CHOOSING CBR250R
[1] The CBR250R engine stays in the power band for a
longer time than its 600cc counterpart over the
course of a lap.
[2] The CBR250R is around 40 kg lighter than the
commonly used CBR600RR. The lighter load of the
engine would enable the team to use a lighter
chassis, further reducing the weight.
[3] It would be easier for the driver to handle the
CBR250R on track than the 600RR, leading to
consistency in performance.
[4] Higher fuel efficiency can be achieved.
[5] Since Honda CBR600RR engines are not readily
available in India, importing spare parts becomesa
problem and leads to high costs.

2.4 SIMULATIONS TO SUPPORT THE CHOICE
Simulations have been performed for the Honda CBR250R
and the Honda CBR600RR engines on different track layouts
with varying weights of the car. The results indicate that at
lower weights, both engines perform similarly, while at
higher weights; the CBR250RR losesitsadvantage.However,
the difference in lap-times over a simulated circuit is less
than 1 second under ideal conditions. It is expected that
when the driver’s accuracyandreal conditionsaretakeninto
account, the performance difference would be much lesser.
Fig -1: Engine torque and power curves (black: torque,
blue: power)
Some of the results after the simulation are as follows:
Fig -2: Engine speed over the course of the lap (red: 250cc,
green: 600cc)
Variation of lap-times of the car with its net weight on a
custom-made autocross track
Weight of Car Lap Time
600 cc 250 cc 600 cc 250 cc
260 210 57.81 57.8
270 220 57.8 57.81
280 230 57.8 57.84
290 240 57.8 57.87
300 250 57.79 57.91
310 260 57.79 57.95
320 270 57.79 58.01
2.5 PERFORMANCE AT DIFFERENT TRACKS
Track Layout Lap Time (sec)
250 cc (Wt:
220 kg)
600 cc
(Wt: 270
kg)
Acceleration (75m) 5.48 5.01
Skid Pad test (Circle,
r=5m)
5.75 5.75
55.47 55.41
44.11 43.93
57.81 57.80
53.05 53.05
Consider Track 3, which has a combination of high-speed
straights and low-speed corners. On this track, the speed of
both the engines over the course of the lap is plotted and
compared. It is observed that the 250cc engine remains in
the power band for a longer duration than the 600cc engine.
This is possible because while the 600cc engine is
predominantly in the first and second gears only, the 250cc
engine makes full use of all the gears.

A smaller engine combined with its greater usage atfullopen
throttle improves the fuel efficiency too.
2.6 RESULTS
Parameter 250 cc 600 cc
Acceleration Poor acceleration
and top speed
Huge advantage.
Longer the
straights, better the
performance.
Skid Pad event Lighter car, hence
better cornering
and turning
Heavier car hence
may be slower
during cornering
Handling Lighter car and a
smaller engine
makes it very easy
to handle;
maximum
performance can be
extracted
Difficult to run the
engine at its
optimum
conditions
Fuel Efficiency Good Poor
Faster in tight and
twisty circuits
Faster in high-
speed circuits
Fig -3: Gear used over the course of the lap
3. INTAKE & EXHAUST
3.1OBJECTIVES
[1] To determine the appropriate length of the intake
runner to achieve high volumetric at high speeds
without compromising on volumetric efficiencyatlow
speeds.
[2] To determine the advantage of a plenum and to decide
on its maximum volume.
[3] To determine the gain in volumetric efficiency
achieved by using the new camshafts over the OEM
cams.[3]
3.2 PRINCIPLE
Concept: A wave which reflects from a closed end maintains
its phase whereas a wave reflecting from an open-end
changes its phase by 90 degrees. Whiletheengineisrunning,
the intake valve opens and closes once every two
revolutions. While the valve is open, air flows through the
valve from the intake manifold to the engine cylinder. As it
closes (and an area of flow decreases) the velocity of air
increases (through the equation of continuity). When the
valve closes, the flow is stopped abruptly and a compression
wave is created which travels upstream from the valve. This
is called the ramming effect. [6] This compression wave
reaches the open end, reflects back towards the valve as
rarefaction wave subsequently reflectsfromtheclosedvalve
as a rarefaction wave and reflectsagainfromtheopen endas
a compression wave flowing downstream. By the time this
compression wave occurs, we want the valve to open for the
next cycle. The length of the intake runner (location of the
open end) has to be accordingly tuned to ensure that the
compression wave reaches the valve just as the next cycle
begins.
When the intake valve opens, there is low pressure
downstream of the pipeand hencea rarefactionwavetravels
from the valve upstream the intake runner. This wave
reflects from the open end as a compression wave and
travels back towards the pipe. The rarefaction wave begins
just as the intake valve opens and the compression wave
must reach the valve just before it closes.
Both these effects help in pumping as much air as possible
into the cylinder when the intake valve is open and the
intake system is tuned for the same.
AVL Boost software has been used forintaketuning.Amodel
of the engine was simulated in this software and focus was
given only to volumetric efficiency.
3.3 RESTRICTOR DESIGN
A CD nozzle was used to house the 20 mm intake restrictor.
The converging section upstream of the restrictor is angled
at 140 and the diverging section downstream of the
restrictor is angled at 70 to the axis.[1] Ideally, the angle
should not be more than 70 to the axis because ofanyfurther
increase in the angle will result in flow separation in the
diverging part and also lead to vortices formation both of
which are undesirable because they cause pressureloss.For
the same reason, the diverging part is also longer than the
converging part. The open ends have a diameter of 38 mm
which corresponds to the diameter of the throttle butterfly
valve.

Fig -4: Intake restrictor- CD nozzle
3.4 DETERMINATION OF INTAKE RUNNER LENGTH
The entire intake runner has been modelled into three
sections, two of which are inside the engine block and the
other being the fabricated section. From upstream to
downstream, they are
[1] Fabricated intake section
[2] Small section of 20mm length and 38mm
diameter just inside the engine block.
[3] Split section which leads to two intake valves.
Fig -5: Intake restrictor- CD nozzle
Sections 2a and 3 were held constant and the length of
section 1 was changed. The diameter of section1 waskept at
38 mm. Using Helmholtz resonator principle, required
runner lengths were calculated so that the peak volumetric
efficiency occurs at 6000, 7000, 8000 and 9000 RPM. These
lengths were tested in the software.
The software was configured to run separate cases for all
speeds between 5000 RPM and 11000 RPM in intervals of
500 RPM and the variation of volumetric efficiency with
speed was plotted. This process was repeated for all the
runner lengths. The results are plotted below.
Fig -6: Simulation model in AVL Boost
Variation of volumetric efficiency with Intake and exhaust runner lengths
0.7
0.75
0.8
0.85
0.9
0.95
1
VolumetricEfficiencyAmb(-)
5000 6000 7000 8000 9000 10000 11000
speed (rpm)
in 250, ex 600 (-)
in 220, ex 600 (-)
in 220, ex 500 (-)
in 250, ex 500 (-)
Fig -7: Simulation model in AVL Boost
From the simulation it was observed that 250mm was the
optimum intake runnerlengthand600mmwastheoptimum
exhaust runner length that gavehighvolumetricefficiencyat
both low and high speeds.
3.5 EFFECT OF PLENUM
Constraint- Located downstream of throttle.
A plenum is used by all teams using 4-cylinder engines to
avoid choking at the restrictor and for the even distribution
of the air in all cylinders.[2] To determine whether it would
prove useful in a single cylinder engine, a plenum was
simulated in the software and the volumetric efficiency
curves were plotted for different plenum volumes.
3.6 SIMULATION RESULT
As plenum volume increased, volumetric efficiency increased
However, the softwaresimulatedonly steady-state response.
Since the throttle is upstream of the plenum, larger the
plenum, slower is the throttle response. Literature survey
showed that the plenum volume is typically 4 times that of
the engine displacement volume. Therefore a 1litreplenum
was chosen.

Fig -8: Effect of different plenum volumes based on steady
state simulation
Fig -9 Intake runner and plenum
3.7 OBSERVATION
[1] Very weak throttle response wasobserved whenthis
was mounted on the car (much weaker than
expected)
[2] Engine was not able to idle and died every time the
throttle was released. Engine breaking seemed to be
the only way to control it. We had ensure high RPM
even while slowing the car down by quickly shifting
to lower gears.
The idling problem could have been because:
While designing the runner, volumetric efficiency curve was
observed only between 5000 and 11000 RPM. It could have
been possible that this curve fell sharply at loweridlingRPM
(~ 1500 RPM) as a result of which the volumetric efficiency
could have been very low at idling RPM leading to shutting
down of the engine.[6]
When the engine was run at idle for quite some time, the
throttle response seemed to improve. However, the idling
issue could not be rectified with this intake system.
3.8 CHANGE OF INTAKE SYSTEM
The plenum was then removed from the design and a
different intake system was used, where the intake runner
was directly connected to the CD nozzle. This design:
[1] Gave good throttle response
[2] Engine remained very stable even at idling
Therefore, the new intake system (without the plenum)was
chosen.
3.9 CHANGE OF CAMSHAFTS
The OEM Honda camshafts were replaced by new Takeaway
camshafts that offered longer intake and exhaust durations
with the same lift. This configuration would improve
volumetric efficiency at high speeds and slightly lower that
at lower speeds, thereby shifting the power-speed curve to
the right.
As expected, idling became more difficult once these
camshafts were installed. To counter this, the butterflyvalve
was adjusted so that the idling RPM was increased from
1500 RPM to 2000 RPM, where the engine remained more
stable.
In the volumetric efficiency curve, it can be observed that
there are two intake peaks and one exhaust peak which
occurs between the two intake peaks. The exhaust runner
length in the simulation model was changed till the three
peaks gave a fairly flat volumetric efficiency curve. Through
iterations, a length of 600mm was chosen.
The muffler used for previous car was 30 cm in length and
had a diameter of 10 cm. This muffler barely managed to
clear the noise test recording 109.7dBattherequiredspeed,
which is too close for comfort.
Hence, a longer muffler (40 cm) was bought havingthesame
diameter (10cm) and packaging thickness. Since there is
packaging material for a longer distance, we expected the
output noise to be considerably lesser.
On performing noise test, a sound level of 110 dB was
recorded at 9000 RPM and 107 dB was recorded at 8500
RPM (designated RPM) with the new camshafts.
3.10 MUFFLER ROUTING
Front facing Rear facing
Advantages: Advantages:
Longer muffler length
possible. Easier to clear
Exhaust gases exit at the
rear.

noise test.
Low mounting of the
muffler, lowers the CG of the
car.
No need of firewall or
casing. Well shielded
from the driver.
Fewer bends in the exhaust
pipe.
Disadvantages: Disadvantages:
Unconventional layout. Packaging is difficult.
Pressure at end of tailpipe is
stagnant pressure, not static
pressure.
Higher mounting of the
muffler increases CG of
the car.
Extra bodywork needed to
cover the muffler, greater
weight.
Extra sharper bends in
the exhaust pipe, routing
is difficult.
3.11 MANUFACTURING OF INTAKE SYSTEM
3.11.1 CONSTRAINT
The intake system must remain within the envelope defined
by the top of the main roll hoop and the edge of rear tires.
Since the wheelbase has been reduced and the tires are
smaller owing to 10” rims, this envelope has become much
tighter than what it was for previous car
The entire intake system was manufactured out of FRP.
3.11.2 MOULD MAKING
A PVC pipe of OD 38 mm was taken to form the mould. In
order to bend it accordingly, it was first stuffed with special
sand and blocked on both sides. This is done to ensure that
the PVC does not fold while bending.
After both the sides are blocked and sealed, a hot air gun is
used to blow hot air onto the pipe. The air is uniformly
blown across the area where the pipe would be bent.
Adequate time was given in between for the heat to transfer
to the sand inside the pipe too.
When enough heat is provided, the PVC begins to bend. At
this point, the pipe is stretched and bent to the required
angle. Sharper the bend, greater the chance of folding in the
pipe. After the required bend is achieved, the pipe is cooled
and brought back to room temperature.
The bent pipe is cut into two longitudinally equal halves
which are used to lay FRP. Two halves of FRP are made
which are then joined together with another layer of FRP to
form a single pipe. The plenum is manufactured similarly
too.
Advantage: Easier to detach moulds from the FRP layer.
Disadvantage: Inner surface of the pipe is not smooth astwo
halves are joined together. This may impact air flow by
decreasing coefficient of discharge.
The throttle bodies are connected to the FRP mould using
small sections of a rubber tube of ID 45mm (corresponding
to OD of throttle body) and clamped. These must befastened
so that there is no air leakage from any of these joints.
3.12 MANUFACTURING OF EXHAUST SYSTEM
OEM exhaust pipes of the CBR250R are used to completethe
routing of the system. Three OEM pipes were used and
sections were cut such that when welded together, the
required configuration wasachieved.Duringmanufacturing,
the following constraints were kept in mind:
[1] The muffler must be parallel to the side impact
structure so that the mount on which the exhaust
rests may be as small as possible. Additionally, this
would keep the muffler as close to the chassis as
possible and ease driver egress.
[2] The muffler must be inclined in such a way that it
remains higher than thelowermostchassismember
at that point throughout the length.
The pipes were routed from behind the main roll hoop to
ensure that there was sufficient distance between the
exhaust runner and electricals to avoid heating.
The exhaust runner was welded to the flange of the muffler
which was connected to the muffler through six Allen bolts.
To avoid exhaust gas leakage, two exhaust gaskets were
placed between the muffler flange and the muffler.
3.13 FINAL ROUTING OF EXHAUST MANIFOLD
Length of Exhaust pipe (In CAD) = 60.729cm
Starting from the engine port, length description is given
below
[1] 53.374 mm straight-line
[2] 82.030 mm curve length (angle = 94-degree, bend
radius = 50 mm)
[4] 73.792 mm curve length (angle = 84.56-degree, bend
radius = 50 mm)
[6] Outer diameter of pipe in CAD = 45mm
[7] Inner diameter in CAD = 38 mm
3.14 MUFFLER DIMENSIONS
[1] Length = 600mm
[2] Outer diameter = 100 mm
[3] Inner diameter = 45mm

Fig -10 CAD showing the side view of the exhaust and its
connection to the engine
Fig -11 Overall view of system
4. COOLING SYSTEM
4.1 PRINCIPLE
To cool down the engine, a coolant is passed through
the engine block, where it absorbs heat from the engine. The
hot coolant is then fed into the inlet tank of the radiator
(located on the top of the radiator), from which it is
distributed across theradiatorcorethroughtubestoanother
tank on the opposite end of the radiator. As the coolant
passes through the radiator tubes on its way to the opposite
tank, it transfers much of its heat to the tubes which, in turn,
transfer the heat to the fins that are lodged between each
row of tubes. The fins then release the heat to the ambient
air. Fins are used to greatly increase the contact surface of
the tubes to the air, thus increasing the exchange efficiency.
The cooled coolant is fed back to the engine, and the cycle
repeats. Normally, the radiator does not reduce the
temperature of the coolant back to ambient airtemperature,
but it is still sufficiently cooled to keep the engine from
overheating.
The coolant we are using is normal tap water, because of its
boiling point being close to the standard engine operating
temperature - 110 C.
4.2 SET UP
Since side-pods were not present on the car, the positioning
of the radiator could be done as required. Possible locations
considered were
[1] Same place as PREVIOUS CAR, where side pods
would have been present
[2] Mounted to the main roll hoop at the lowest
position.
[3] Main roll hoop bracing
Locating it in the same place where side pods would have
come has the following issues
[1] The radiator has to be angled so that it fits within
the height of the side impact structure, meaning
lesser cooling area
[2] Additional firewall for the driver Very low position,
hence it will not be very effective for the cooling
pump to circulate water
[3] We thought this was the reason why wewerefacing
radiator issues in previous car
Locating it on the bottom most vertical structureofmain roll
hoop. Very low position with respect to engine.
We assumed that locating the radiator lowerthantheengine
would mean that there is no head available at the engine
inlet (cooling pump inlet). Hence, this would make water
circulation difficult, leading to cooling issues. Hence, we
decided to place the radiator on the roll hoop bracing higher
than engine inlet.
Fig -12 Rear view of the car showing radiator, catch can
and hoses

4.3 PROBLEMS AND SOLUTIONS
Problem 1 Bursting of radiator hoses while
running the car
Experiment 1
Hose clamps were
changed
Experiment 2
Checking water
circulation inside
the engine
Experiment 3
Changing hose
quality
Reason
This experiment
was carried out
because the
bursting occurred
mostly near the
end of the hoses,
where it was
clamped
Reason
This was because
the hoses were
getting very hot,
so suspicions
about the
circulation itself
and not the
radiator or hoses
developed
Reason
This was
because of
suspicions in
the quality of
hoses
Solution
Instead of using
the standard
“Worm Drive”
Clamps, a “Wire
Hose Clamp” were
used
Solution
The hoses were
removed and
water wasdirectly
filled into the
engine inlet
Solution
The standard
rubber hoses
were replaced
with metal
braided hoses
with a
maximum
pressure
capacity of
125bar
Observations &
Inference
The problem still
persisted.
Moreover, the
cracks appeared
even at the middle
Observations &
Inference
The cold water
poured inside
came out as steam
through the other
end. The outlet
water was so hot
that the bottle
could not be held
without
insulation. This
means that water
circulation and
heat transfer from
engine to water
was happening
properly
Observations &
Inference
No more
bursting of
hoses. The
metal braided
hoses are
capable of
sustaining the
pressure and
temperature of
the coolant
Problem 1 eliminated.
Therefore, the radiator has been placed on the main roll
hoop bracing and metal reinforced hoses have been chosen
to transport coolant water between the radiator and the
engine.
Problem 2 Steam released from the radiator after
twenty minutes or so of running the
engine
Experiment 1
Mounting the
catch can at the
same height of
radiator overflow
vent
Experiment 2
Increasing the
boiling point of
coolant
Experiment 3
To condense the
steam coming out
of catch can
Reason
We learnt that
radiators not only
vent water to the
catch cans but
also suck water
back from them
Reason
The pressure of
cap defines the
boiling point of
coolant. The
pressure was
expected to be
very less that
boiling starts
within very short
time.
Reason
The steam gets
condensed from
the water in the
catch can
Solution
The catch can was
mounted on the
bracing at the
same level of the
overflow vent
Solution
The radiator cap
was changed from
1.1 bar cap to 1.4
bar cap of KTM
Duke 390
Solution
Catch can was
filled with water
and a vent from
the cap was
immersed in this
water
Observations &
Inference
Steaming still
continued. The
amount of water
in the catch can
was 400ml and
the amount of
water lost by
radiator was
350ml. Hence,
50ml was
converted to
steam. This was a
huge loss, and it
will critically
affect the engine
performance
during endurance
Observations &
Inference
No considerable
positive results,as
steaming still
persisted and the
loss was about
50ml.We
understood
simulating race
track conditions
would not work,
and hence took it
to actual testingof
the whole car
Observations &
Inference
No steam was
observed during
the whole
endurance run
Running conditions were simulated by using a high speed
fan and running the car on jacks

Fig -13 Isometric view of the radiator
Fig -14 Isometric view of the car showing the coolant
hose connections to radiator and catch can
5. FUEL TANK
5.1 OBJECTIVES & CHALLENGES
The tank has to be securely attached to the vehicle all times
and if made of rigid material should not be used to carry
structural loads.
[1] Sloshing occurs at low volumes, which creates
imbalance to the car during braking, acceleration and
deceleration
[2] Sloshing also creates an interrupted supply of fuel to
the pump.
[3] The tank has to be easily mounted and accessible in
the car.
[4] The weight of the tank should be minimum to lower
down the CG of car.
5.2 MATERIAL OF THE TANK
The material of the tank should be checked for corrosion
with gasoline.
Aluminium
[1] Lesser weight to the tank
[2] Easily fabricated into any desired shape
[3] Good resistance for corrosion with gasoline
[4] Easy availability in the market
Plastic
[1] Reduces lot of weight but punctures in contact with
sharp things.
[2] Mounting of flexible material is quiet difficult.
[3] HDPE (High Density Poly Ethylene) is the best
material mostly used and heat sources should be
avoided in such case
Aluminium and Poly urethane were selected for
manufacturing the tank
[1] Aluminium for body
[2] Flexible PU hose for filler neck
5.3 MOUNTING
[1] Mounting should be below or close to the existing
CG, so the CG doesn't change much as the fuel tank
empties.
[2] The tank has to be away from heat sources like
exhaust.
[3] To avoid change in lateral weight distribution, tank
has to be mounted close to the horizontal CG.
[4] The tank is placed in an aluminum basket and for
secure mounting Velcro’s were used
5.4 Sloshing
The effect of sloshing can be reduced by many ways
Foam- The tank has to be designed such that there is
accessibility to stuff the foam. The fuel is added with foam
particles. These foam particles create stabilitytothefuel and
reduce the turbulence. The main disadvantage of foam is,
little bits of foam gets blocked in the pump when excess of

foam is used. This results in surging of fuel tank but by
careful baffling surge can be avoided.
Baffles- These baffles trap the fuel between them and result
in a constant supply of fuel to the intake manifold. Baffles
result in increasing weight and complexion of the tank.
Welding the baffles to the tank is also difficult.
Fig -15 Baffles
Filler neck- Advantage can be takeninassigningsomepartof
tank volume to neck. This ensures that significant amount of
fuel stores in the neck when the tank is full. The issue
regarding this is that the CG of the car goes a little bit up as
the volume of fuel is significantly distributed from top to
bottom.[7]
Low pressure lift pumps
It can be tricky to recover the fuel as it moves around in the
tank, particularly at low level -imaginedrinkingwitha straw
from water bumping around on a tray compared with
drinking the same water with a straw in a bottle.
The tank is therefore provided with a low pressure
electrically powered ‘lift pumps’ at the corners to suck the
fuel. These deliver the fuel to a small tank known as
collector. This collector is provided with an external pump.
This pump can constantly recover fuel to the intake
manifold. The issue regarding low pressure lift pumps is it
will increase in complexion and manufacturing of the tank.
5.5 Shape of fuel tank
One of the simplest design is taking a cuboid fuel tank and
placing internal baffles. Further we can modify the cuboid
shape by making the two opposite rectangular sides into
quadrilaterals by pulling down only the top edge to reduce
the effect of sloshing as shown in the below figure. Foam
can also added to further minimize sloshing. This design
has geometry, baffles, foam. This design is also easy to
manufacture.
Fig -15 Manufactured Fuel tank
Cylindrical tank
We can also design the tank in cylindrical shape. This
reduces the effect of sloshing to a greater extent but
mounting the cylindrical tank into the car is a major issue.
Conical tank
A conical tank can also be used with an internal suction
pump at the bottom. Mounting of a conical tank is also an
issue. But as the radius of the tank is high at thetopthe effect
of sloshing is more at the top even when the fuel is more
compared to other tanks. This is more useful when fuel is
low and baffles are used.[8]
Since the 250cc has a higher fuel efficiency, the fuel tank
capacity was kept at 4 liters. To reduce complexity, we split
the total volume as approximately 3 liters (3.072 lit) stored
in an aluminium cuboid and 1 liter (0.981 lit) in a flexible
fuel resistant hose made of polyurethane. The pump
occupied one fifth area of bottom part of the tank which left
approximately 200 sq.cm (205.76 sq.cm), forwhichsloshing
was very less and there wasn’t any space for baffles. Hence
baffles weren’t used. The transparent filler neck eliminated
the use of sight tube
5.6 Fuel Tank
The fuel tank is a trapezoidal box made from aluminium
plates of thickness 2mm. The main aim of making it a
trapezium and not a box is to make it have a clearance of
10mm from the flexing of seat. It was fitted with a bottom
mounted pump. The volume of the tank was designed to be
3L, as the mileage of the CBR250Rengineisaverage10kms/l
and we need to cover a distance of 22kms non stop. There
was a plate beneath the tank to lock the fuel pump to the
tank.

Fig -16 Seat Clearance
5.7 FUEL PUMP
Fuel must be recovered from the tank and supplied to fuel
injection system at pressures of 4 to 5 bar. For the recovery
of fuel from the cell there are two types of pumps
[1] External fuel pump
[2] Internal fuel pump
External fuel pump
In some cases the pump is located outside thetank andthese
doesn’t involve submersion. Thesepumpsarecalledexternal
fuel pumps. Fuel lines are drawn from the tank and
connected to these pumps. These pumps suck the fuel and
supply them to the fuel injection system.
[1] External fuel pumps are noisy
[2] External pump are easier to maintain and service
due to location but requires additional fittings and
space, adds to weight and cost report
[3] External pumps are mounted lower than the tank
cause they don't pump against gravity
5.8 INTERNAL FUEL PUMP
There are cases where pumps are inside or submerged in
the tank. These work under submersion. These are called
as internal fuel pumps.
[1] Internal fuel pumps are difficult to change.
[2] Internal the pump, it stays cooler because the fuel
keeps it cool.
[3] It also may be a safety factor in a crash because the
pump is not exposed.
[4] Internal fuel pumps are harder to maintain and
service
[5] An internal fuel pump does not require additional
plumbing and mounting brackets therefore it saves
weight and space (depending on design)
Property External pump Internal pump
Noise High Low
Maintenance Easy to maintain
and service
Difficult to
remove, maintain
and service
Weight Adds weight and
cost
Less comparedto
external pump
Additional fittings Requires
additional fittings
Doesn’t require
Space Adds space Consumes less
space
Removability Easily removable Difficult to
remove
Safety of pump Safety is a bit less Pump is more
safer as it always
stays cool
because of fuel
and lasts long
There are two types of internal fuel pumps
[1] Top mounted fuel pump
[2] Bottom mounted fuel pump
5.9 TOP MOUNTED FUEL PUMP
A hole is made on top of the tank and the pump is placed in
this hole carefully because it has to be submerged always in
the fuel for continuous supply to the engine. If the above
requirement is absent, in case of low fuel level it sucks only
air and burns itself.[9] Pump is highly accessible and can be
replaced immediately with ease if there is any failure at any
time independent of the fuel level.
Sloshing has a slight effect on top mounted fuel pump. The
pump must extend from top of the tank to bottom for
constant submerge where the sizeofthepumpincreases asa
result, size and weight of the tank increases comparedtothe
bottom mounted pump in order to compensate the volume.
Fuel can leak through the top-hole spilling over the driver
and the engine. Firewall and other safety precautions must
be designed to prevent the spill.
Fig -16 Top Mounted Fuel Pump

5.10 BOTTOM MOUNTED FUEL PUMP
[1] A hole is placed on the bottom of the fuel tank and
the pump is placed in this hole.
[2] As the pump and all its fuel lines are situated below
the tank accessibility to the pump is less.
[3] There is also a restriction that pump can be
removed only after all the fuel is consumed and
nothing is left over in the tank.
[4] There is no immediate replacement for the pump
during failure.
[5] As all the equipment is below, the tank has to be
mounted at a certain height above from the lowest
member of the chassis. This increases the CG of the
tank and the car which is the major issue regarding
the usage of this pump.
[6] The sloshing effect is minimal compared to top
mounted pump.
[7] The bottom of the tank must be carefully sealed
because there is a greatest possibility of leakage.
[8] A pan is generally placed below the tank to collect
the leaked fuel. This can be ignored in bottom-
mounted pump because all the leaked fuel will spill
over the ground where there is no problem.
[9] The size of the pump is small compared to top
mounted pump.
Fig -17 Bottom Mounted Fuel Pump
Table Comparison of top and bottom mounted fuel pumps
Property Top mounted Bottom mounted
Accessibility Easily accessible
as the pump is at
the top of the
tank
Accessibility is
less because the
pump is located
below the tank
Centre of gravity No effect on CG of
the car
The CG of the car
decreases
because of the
mounting of the
tank due to the
pump and all fuel
lines
Pump removal The pump can be
easily removed at
all fuel levels
The pump can be
removed only
after all the fuel is
consumed and
removing the
pump involves
pain over top
mounted
Sloshing effect Medium Low
Leakage Fuel can leak
from top and spill
over the driver
and engine
Fuel leakage has
no major
problems except
loss of fuel
Size The size of top
mounted pump is
high and hence
size of tank
increases
The pump
occupies less
volume
Submersion Care should be
taken such that
pump is always
submerged
Pump is always
submergedexcept
during low fuel
level and lateral
forces
The CAD model of the pump is made, to help intheassembly.
The pump has three nozzles at the bottom - two for filters
and one connecting to the engine intake.
[1] Seventh Plate: The seventh plateisanAluminiumplate of
6mm thickness, to hold the fuel pump with the fuel tank.
A innovative lockingmechanismhasbeencreated.Twoslots
for three of the holes are made. The pump will be inserted at
an angle and rotated to ensure locking.
Fig -18 Locking mechanism of the fuel pump to the
seventh plate

An extra length at the back has been provided for mounting
it to the chassis. One side mounting is done to ensurenoload
is carried by the tank. It has the same dimension as the base
of the fuel tank with M6 holes for holding itandM8holesfor
holding it to the chassis member. There is step provided for
3mm and diameter of 110mm for the outer diameter of the
pump.[10]
[2] Gasket: The gasket is made of Silicon. It is2mmthick and
its use is to fill the gap between the seventh plate and the
tank to ensure there is no fuel leakage.
[3] Box Channel: A 1inch aluminium box channel is used at
the bottom to simply support the assembly using M8 nuts
and bolts.
[4] Chassis Member: This 1inch solid MS memberwill bethe
one on which the fuel system is mounted. Two mounts are
welded onto it, with M6 holes that go into the seventh plate
mounts.
Fig -20 The L channel on which the fuel tank is resting is
in turn mounted to the chassis member.
REFERENCES
[1] Aho, Christopher, Scott Duncan, Dan Cullen, Dan
Swan, Adam Panzica, and Ryan Lehrmitt. 2009
Formula SAE Racecar.Tech. 2009. Print.
[2] Milliken, William. (1995). Race car vehicle dynamics
(1st ed.). Troy, MI: SAE International.
[3] Timmins, Steve. Formula SAE Powertrain Phase 4:
Performance Validation and Path Forward.
University of Delaware. 2010. Web.
[4] Young, D. (2011). A Brief Introduction to Fluid
Mechanics (5th ed.). Hoboken, NJ: Wiley.
[5] Alspaugh, David, Aquadro, Alessandro, Barnhill,
Dylan, Beasley, Nicholas, Bennett, Andrew, Francis,
John. Design and Analysis of a FSAE Racecar. Tech.
2012. Print.
[6] "Torsen 012000 print"
http://www.torsen.com/files/University%20Speci
al%20012000.pdf
[7] James, R. (2004) Design of an Aluminum Differential
Housing and Driveline Components for High
Performance Applications. Unpublished
undergraduate thesis, Massachusetts Institute of
Technology.
[8] Durand, K. (2005) Design of a Chain Driven Limited
Slip Differential and Rear Driveline Package for
Formula SAE Applications. Unpublished
undergraduate thesis, Massachusetts Institute of
Technology.
[9] "Assembly and Timing Notes"
http://www.torsen.com/files/012000%20Gear%2
0Assembly%20&%20Timing.pdf
[10] "NTN Radial Ball Bearings"
http://www.ntnamerica.com/Engineering/PDFs
1000/A 1000/A OOORadial.pdf

IRJET- Analysis and Design of a Formula SAE Powertrain

More Related Content

What's hot (20)

Similar to IRJET- Analysis and Design of a Formula SAE Powertrain (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Analysis and Design of a Formula SAE Powertrain