Inlet manifold design and theory in regards to maximising volumetric efficancy

Symbols and Abbreviations

F/I – forced induction

n/a – naturally aspirated

di – diesel injection

si – sequential injection in this case but usually refers to sports injection

map- manifold absolute pressure

fwd- front wheel drive

rwd- rear wheel drive

3d- 3 dimensional

cad- computer aided design

cc- cubic capacity

ecm- engine control module

ecu- engine control unit

rpm- revolutions per minute

k- unit of heat transfer Kelvin

cc- cubic centimetres

id- internal diameter

od – outside diameter

mps- meters per second

mph- miles per hour

bhp- brake horse power

ft-lbs- foot pounds of torque

bdc- bottom dead center

tdc- top dead center

l- litre

x,y,z- axis in relation to cad

mm- millimetre unit of measurement

int – intake

exh- exhaust

v/e- volumetric efficency

deg- degree

bar- unit of pressure

Chapter 1 – inlet manifold parameters design feasibility

When talking about the design of an inlet manifold there are many parameters that need to be discussed such as;

Will the inlet be fitted to a Forced induction (f/a) or naturally aspirated (N/A) engine?
What are the gains required? Maximum Torque, power, or fuel economy?
At which point will the design and manufacture costs outweigh the target market?
How will the fuel be delivered to the combustion chamber? DI (direct injection), SI (sequential injection) or Carburettor.
What sensors will be fitted to the inlet such as air mass, MAP (manifold absolute pressure), air temp, as the placement of these may affect the true reading of the sensors?
What drive configuration will the vehicle be as this may affect the location of the intercooler and the inlet may need redesigning for FWD (Front Wheel Drive) or RWD (Rear Wheel Drive vehicles) depending on how much space available in the engine bay?
How will the manifold be designed, 3D CAD (Computer Aided Design) or hand drawn?
How will the manifold be produced, what fabrication techniques are going to be employed?
Testing and fault correction. No matter how accurate the design and science behind why the manifold is designed how it is, some slight changes may be required to squeeze maximum flow.

Possible Problems

The inlet manifold in designed will be fitted to the Ford Escort Rs Turbo, retrofitted with a 2000cc Zetec engine, This engine configuration is very common in the performance ford scene, minimal changes will have to be made to the design for RWD vehicles just the location of brake servo vacuumed take off.

Due to the lack of space in escort engine bay the shape of the inlet manifold will be compromised. The ideal configuration would have the throttle body situated central between cylinders 2/3 this will provided even air flow to all cylinders. Unfortunately this configuration is not feasible due to space limitation as the throttle body will be mounted at the R/H side of the plenum closest to cylinder 4, although this is not ideal as the cylinders closest to the throttle body will receive more air then the cylinders that are further away. There is a solution to this, if the inlet plenum is manufactured into a wedge shape with the largest end at the throttle body and the smallest end at cylinder 1, this will increase the Velocity of the air as it enters the plenum this is caused by the air being forced toward cylinder 4 due to the air being compressed due to the reduction of area as it travels towards the narrow end of the plenum.

The aim of the inlet manifold when manufactured is to increase Engine Power whist keeping the vehicle driveable with smooth power delivery and allowing for maximum air flow throughout the entire rev range, fuel economy will not be taken into account in the design of the inlet as this is not required by the end user.

The Zetec engine uses SI to deliver the correct fuelling to the combustion chamber by the use of 4 injectors, these will be replaced with higher impendence injectors with a different flow rate to suit the desired power output, the standard injector configuration atomises the fuel direct into the intake ports of the cylinder head, although this may or may not be the best place to inject fuel so research is needed into the ideal placement of the injectors.

The Zetec engine will be using an aftermarket electronic Control Module (ECM) to keep all the fuelling parameters in check and to allow for the injectors to deliver the fuel required to match the increased flow of the new manifold, the ECM unit will need to gather various information from the inlet manifold by the use of a number a sensors. These sensors may operate differently depending on their position in the intake track for example if the map sensor was to be fitted to a high pressure area where airflow is reaching maximum speed, would the reading for the Manifold pressure be same if fitted to a different location on the manifold? Also where is the ideal place to read air temperature as this may be effected by the heat radiated from the hot engine?

Inlet manifold Design

For the inlet manifold to increase engine performance it needs to provide the following:

“1. To provide as direct a flow as possible to each cylinder

2. To provide equal quantities of charge to each cylinder

3. To provide a uniformly distributed charge of equal mixture strength to each cylinder

4. To provide equal aspiration intervals between branch pipes

5. To provide the smallest possible induction tract diameter that will maintain adequate air velocity at low speed without impeding volumetric efficiency in the upper speed range

6. To create as little internal surface frictional resistance in each branch pipe as possible

7. To provide sufficient pre-heating to the induction manifold for cold starting and warm-up periods

8. To provide a means for drainage of the heavier liquid fraction of fuel

9. To provide a means to prevent charge flow interference between cylinders as far as possible

10. To provide a measure of ram pressure charging” [1]

Once all the different variables and science behind how differently sized and shaped intake runners affect the torque delivery at various different RPM sites, have been taken into account the physical design on the manifold can be started, so a prototype model can be produced using various different manufacturing techniques. Allowing testing using the flow bench and alteration of the inlet manifold to allow for testing of increased flow over the standard cast inlet to prove the inlet designed will increase the volumetric efficiency of the engine.

Inlet manifold surface finish

“Smooth internal walls, compared with those with a rough surface finish, produce the least viscous drag on the movement of charge-whether it is pure air or a mixture of air and fuel particles-consequently marginally higher volumetric efficiency can be obtained with smooth induction tracts” [2]

“Slow moving air-fuel mixtures tend to precipitate fuel particles, in particular the heavier ones, onto the walls of the tract, they then tend to merge into surface films. This effect becomes more pronounced with increased throttle opening.

With rough surface finishes the tract offers considerable resistance to the flow of liquid particles suspended in the air stream near the walls, these particles then cling to the walls where they then spread and merge with each other to form films. As these films thicken they become unstable so that they are dragged back into the main air stream and immediately new fuel particles will be thrown onto the tract walls where they again accumulate before repeating the breakaway cycle. This process of build-up and breakaway of liquid fuel from the tract walls produces a continuous and effective mixing mechanism for the charge as it moves towards the inlet port.

Conversely, a smooth surface finish reduces the surface flow resistance keeping the film thickness to a minimum, consequently there will be very little mechanical break-up or mixing of the fuel particles with the air stream until the erratically suspended fuel particles impinge on the underside of the inlet valve head.” [3]

Conclusion

A rough surface will cause resistance to the flow of liquid suspended in the air causing it to cling to the walls. This will create a film of fuel coating the tract walls that will then be swept back into the air as the film thickens. The film will constantly replenish and be swept into the airstream. This provides effective mixing of the charge as it enters the engine.

A smooth surface reduces the flow resistance reducing the thickness of the film of fuel clinging to the tract wall. There will be minimal mechanical break up as a result of this causing an undistributed mixture.

Inlet runner Section shapes

Circular

“Circular-sectioned tracts provide the minimum surface area for any cross-sectional shape. They therefore offer the least resistance to charge flow and this gives the highest volumetric efficiency and, correspondingly, the maximum mass of charge per unit time. However, the charge column moving through a circular section tends to generate a longitudinal twist or whirl which causes the globules of fuel in the air stream to be thrown onto the tract walls. As a result, the distribution of mixture strength across the section may be very uneven. A disadvantage of the circular section is that the semi-circular floor of the tract provides only a relatively small surface area for evaporation of that fuel, existing in a liquid state, which has made its way to the bottom of the tract”[4]

Segment sections

“This is a compromise which retains the high volumetric efficiency of the circular-sectioned tract and the large flat floor area of the rectangular-sectioned tract. It maximises the passage vaporization ability and simultaneously minimises charge column swirl in the segment-shaped section. This semi-circular section provides, firstly, the least resistance to flow due to the circular portion of the tract, secondly it offers a large, flat floor area for rapid evaporation of the liquid content of the fuel, and thirdly the flat floor minimizes charge column swirl and therefore the charge retains its initial inlet density as it flows through the tract.”[5]

Conclusion

The shapes of intake runners are a contributing factor to the volumetric efficiency of the engine, the effects and pros and cons of the circular section and the semicircle section runners have been researched, and have found that the circular tract provides the least resistance to charge flow and gives the highest volumetric efficiency, although the circulate shape of the tract causes the fuel/air mixture to swirl, which causes the fuel particles in the air to stick to the tract walls, and as a result of this the distribution of the mixture may be uneven, and with the tract being circular there is very little surface area preventing minimal evaporation of the fuel which has made its way to the bottom of the tract.

A compromise to the shape of the tract needs to be made to maintain maximum volumetric efficiency whilst minimising the swirling motion of the air fuel mixture caused by the cylindrical shape of the runners. The semi-circle shaped tract solves this problem. The high flow characteristic of the circular tract is maintained whist minimising swirl due to the flat floor of the tract, the floor also mineralizes the charge swirl.

Chapter 2 – material specification

Material Specification

A decision had to be made to choose what material the inlet manifold is to be made of for the final model to be tested on a road car, it was decided that the following materials would be analysed and compared to decide the material which will be used.

Plastic
Steel
Aluminium

When comparing the materials a number of different variable were taken into account;

The application, in this case an inlet manifold fitted to a turbocharged sports car.

Strength, the material needs to be able to withstand high boost pressures on a regular basis, without risk of leaks, cracks or destruction.

Plastics are a new age material and can be produced to suit any use, plastics molecular structure can be adjusted to change its strength property’s by adding different ingredients to the mixture.

Steel has been used since the first engine used in an automobile, steel has a high tensile strength, and the ability to withstand high pressures before failing.

Aluminium is used a lot in the aeronautical industry mainly because it is lightweight, but also is very strong as new compositions of aluminium are being produced every day. Aluminium has the tendency to tear before shattering.

Machining

Plastic can be manufactured or moulded using an techniques known as injection moulding, this is known as Thermosetting (this is when the properties of a material will only allow the material to become a molten liquid state once, this is necessary as the manifold would melt if thermosetting plastics weren’t used). Plastic pellets are heated until liquid and then injected into a mould and rotated until the inside of the mould has even coverage. This cuts down any manufacturing time and costs.

Steel is a dated material and is costly and time consuming to machine due to its tensile strength, steel can also be cast using a sand cast but this will leave a rough surface texture, and would require further hand machining to make the surface smooth.

Aluminium is inexpensive to machine due to the fact it tears and not shatters like steel this allows for the machining work to be completed quicker with less risk of the tooling becoming damaged.

Life span

Plastic is not bio-degradable (meaning the material will never decompose) and will never rust or oxidise like steel or aluminium, so in theory a plastic manifold would last forever.

Steel is susceptible to rust due to its high iron content this over time could affect the performance of the manifold as the surface texture will change affecting the flow of air to the engine.

Aluminium can be anodised (a chemical process where aluminium is submerged into a chemical solution and a positive charge is applied to the work piece being anodised turning the piece into an anode. The material surface then attracts the chemical solution). Anodising increases the surface strength of aluminium and reduces the chances of the metal oxidising.

Weight

Plastic has the least mass of the 3 chosen materials so would be ideal in a motorsport application

Steel has the highest mass due to the high iron ore content.

Aluminium is at a half-way point between steel and plastic.

Surface Texture

Plastic is very smooth when moulded due to its self-levelling properties. The plastic will obtain the same texture as the mould in which it was cast in.

Steel can be molested to achieve smooth surface but this will be time consuming due to the complexity of the inlet manifold

Aluminium has a smooth surface texture reducing corrections having to be made to the surface.

Heat dissipation

The ability to disperse heat from the inlet manifold is a very important factor. This is due to the high air charge temperatures created by the compressed air being delivered from the turbo charger.

Material	Temperature 25 ∙c
Aluminium	250k
Steel	43k
Plastic	15k

Table 1Material heat properties

Above can be found the thermal conductivity of the 3 materials. Energy is transferred in the area of decreasing temperature since higher temperatures are associated with higher molecular energy. Materials with high conductivity will radiate more thermal energy to the surrounding area thus allowing more heat to dissipate from the inlet manifold.

Expansion

Plastics tend to expand very little if at all, this is down to the thermosetting properties of the material as the plastics molecular property will only change the first time it is heated.

Steel has a rapid expansion rate when heated. This has the potential to cause cracks over time when the inlet has been heated and cooled numerous times making the material brittle.

Aluminium also has a fast expansion rate but unlike steel the chances of the material becoming fractured are slim due to the material being malleable

Costs

Plastic parts are cheap to produce on a mass scale but very expensive to make just one as the mould would need to be made and jigs to accommodate the mould in the injection moulding machine would also need to be manufactured.

Steel is very cheap as a raw material and also fairly inexpensive to manufacture.

Aluminium is more expensive than steel but the tooling costs requires to machine aluminium will be a lot less, this is mainly because of aluminium’s shearing/tearing property’s . The time required to machine aluminium is also a lot less then steel.

Material Specification matrix analysis chart

	Plastic	Steel	Aluminium
Strength
Machining
Life span
Weight
Surface Texture
Heat dissipation
Expansion
Costs

Total	3	3	8

Table 2materials matrix analysis chart

Table 3 material comparison bar chart

Chapter 3 – inlet manifold science

A properly designed induction system for a NA engine must be both well thought out and well designed, it simply not just a case of having the biggest or the best. In order to create such an inlet literature research on the subject must be carried out, some of the theories are going to put to the test using 1D simulations and flow testing are;

Pulse tuning, rarefaction waves and effect on volumetric efficiency

The easiest way to explain the theory behind pulse tuning is that the engine is basically an air pump, but it is wrong to consider that air flows through the engine, instead it pulses, each cylinder will pulse every 2 revolutions of the crank. These pulses can be used to gain a positive effect, being able to move the point at which the engine makes most power. This can be seen when you blow into a glass bottle full of water, the air molecules inside vibrate creating a high frequency sound, as the quantity of water decreases the frequency at which the molecules vibrate also slows. This is due to the fact that the distance the air has to travel effects the frequency at which the air is vibrating.

“The natural supercharging effect occurs when the engine is running, a column of air moves through the induction tract passageway from the point of entry to the inlet port and the valve and then into the cylinder every time the inlet valve opens the reduction in cylinder pressure produces a negative pressure wave which travels at the speed of sound (343mps for inlet)” [figure? Page 275]

In regards to the pulse tuning of an internal combustion engine When the inlet valve opens all the charge down he inlet runners and ports is moving very quickly, when the valve closes its compressed against the back of the valve head. Think of it like a spring, it then pulses back up the port and through the inlet runners until it reaches atmospheric pressure. The pressure here is relatively high compared to the vacuum of the port, so the pressure wave bounces back down the inlet runners and port. If you get it right, just as the inlet valve opens, the air is pushed in, thus increasing volumetric efficiency, which increases torque.

“Through the column of air from the back of the inlet valve to the open atmospheric end of the tract. When the pressure wave reaches the atmosphere rarefaction occurs; that is, the air at the tract entrance suddenly becomes less dense and therefore creates a depression. Instantly, the surrounding air rushes in to fill this depression. As a result, a reflected positive pressure wave is produced due to the inertia of the air, and this causes the pressure pulse to travel back to the inlet valve port. It is this first reflected pressure wave that if correctly timed, is responsible for ramming the air into the cylinder towards the end of the induction period. When the pressure-wave again reaches the back of the inlet valve it reverses its direction and is reflected outwards. Thus, these negative and positive pressure waves are continuously reflected backwards and forwards between the open intake and the inlet valve port but with decaying amplitude until the inlet valve close. [Figure? Page 275]

Cars that need a low power bands have long runners as the wave has more time, increasing torque in the low power band 1500-3500 as the torque is needed when cruising and lower engine speeds.

“To utilise fully the pressure-wave pulse it must be times so that its first positive pressure-wave arrives at BDC towards the end of the induction period at its peak amplitude. Therefore, it is important to know the time it takes for a pressure wave to travel through the column of air from the open inlet valve to the intake entrance, to be reflected and then to travel back again to the inlet valve.”[Page 275]

Cars the need high power bands like bike’s and F1 engines have almost no runners, this cause’s a pulse which is most efficient at an higher rpm, creating more torque at higher engine speeds, this would not be practical on a road car as it is impractical to let the engine rev that high due to road speed limits and fuel efficiency.

How is inlet length calculated?

Inlet length is calculated using, crankshaft angular displacement, engine speed, speed of sound, the formula to calculate inlet length can be seen bellow.

Speed of sound =

Therefore, time taken to travel the tract length and back again

t= time for the pulse to travel the tract length and back again

L= length of tract from open end to inlet valve head (mm)

C=speed of sound through air (approximately 330m/s)

The crankshaft angular displacement during the same interval of time would be as follows.

Crankshaft displacement = time to travel tract length and back again X angular speed

Therefore

Where = crankshaft angular displacement (deg)

N = engine crankshaft speed (RPM)

For this example we will use an rpm of 7000.

=333mm

Chapter 4 – 1D Engine Simulation GT-POWER

The standard Zetec engine was then created in Gt-power; this involved using measurements obtained from various components from the engine such as, inlet length and runner diameter, inlet port length and diameter, valve sizes, bore/stroke, conrod length, compression ratio, injector delivery rate, camshaft data [figure?].

The accuracy of this data was very important in order for a realistic simulation to be created for testing and analysing of data.

Intake manifold

The inlet manifold needed to be completed next using various measurements and pictures to relate to the configuration of runners and flow splits. Intake runners and flow splits are needed to transfer end environment to inlet port. The pipe round template is selected from the template library this was then renamed int runner, the attributes in the main folder where then filled with the data obtained from measuring of the standard Zetec inlet. The diagram below left, show an illustration with diameters of intake runners, lengths and the dimensions and areas of the inlet plenum.

Figure 2 above. Ford Zetec EFI inlet manifold

Figure 1 above, illustration of a Ford Zetec inlet manifold and its dimensions

Figure 3 pipe round template renamed intrunner and filled with attributes and measurements from a standard Zetec inlet.

A flow split is now needed to link the 4 intake runners to create a plenum (a chamber with equalized atmospheric pressure)

The flow split is drawn on paper first to find the measurements needed to create the flow split, there is 1 input from the throttle butterfly and 4 outputs for the intrunner to feed each cylinder with charge.

The Zetec features a 1-4 flow split this is where the plenum delivers air flow to 4 intake runners using only a single collector.

Figure 4 drawing of standard zetec 4-1 flow split

The drawing now needs to be measured using a compass to find the angle each port is in relation to the x y and z axis.

Axis	Angle In Degree’s
X	0	180	90	45	135
Y	90	90	90	90	90
z	90	90	0	135	135

The flow split template is selected from the template library. The boundary data tab is then selected and the form is filled with the following data which was previously obtained from the drawing.

The part is also renames man-fs

Figure 5 Gt-Suite flow split data from zetec inlet

It is wise to then view the flow split in 3d to check the data entered is correct and that all the ports are in the correct location in relation to the x,y,z axis.

Figure 6 inlet 4-1 flow split viewed in 3d

The man-fs is then placed on the workspace and linked to the in runner’s when doing this, it is crucial that the parts and linked in the correct order using the number forms, for example our 3d model of the flow split the largest port number 3 is connected to the throttle butterfly and must be the third component to be linked. It is also important that the components are linked in the correct direction as GT-Power engine simulations runs from left to right.

Intake ports and valves

The next component created were the inlet ports, these consist of a pipe round and a flow split due to the design utilising Siamese ports, the pipe round template was then renamed [intport] with a diameter of 66mm and a length of 80mm this single port then splits into 2 separate ports known as Siamese ports, a flow split was created and renamed siemese-sfs and filled with the following criteria.

Figure 7 a screen print captured in GT Power Zetec inlet port

Inlet valves were now needed in order to complete the induction side of the cylinder head,

The valveCamCon template was the selected from the component library and dragged onto the workspace this was then renamed intvalve and filled with the following criteria obtained from prior measurements of the Zetec cylinder head

Int valve diameter	32mm
Exh valve diameter	28mm
Valve Lift Int	9.4mm
Valve Lift Exh	9.55mm
Cam Duration Int	258
Cam Duration Exh	250
Valve lash	0.1mm

Table 4 zetec valve train data

The camshaft profile was then created using measurements previously obtained from measuring camshaft lift over 360 degrees.

this data was then used to create a lift array on GT-Suite, the data was entered into the table, because we have seat theta=0 at max lift any measurement before max lift is in – degree and after max lift +degree and max lift as 0degree’s, the reason the measurements had to be entered into the table in this way is to allow GT to used max lift as a reference point for altering and defining camshaft timing.

The flow array tab was then selected as reference array had to be created using this simple mathematical equation, Array of valve lift or valve lift divided by Valve Reference Diameter, as specified in Flow Coefficient Lift Unit: must be monotonic and start at zero.

The following data was then entered.

Table 5 zetec lift array data

This step was repeated using different data to complete the exhaust camshaft.

0mm	0
1mm	0.0312
2mm	0.0625
3mm	0.09375
4mm	0.125
5mm	0.15625
6mm	0.1875
7mm	0.21875
8mm	0.25
9mm	0.28125
9.4mm	0.3125

Cylinders and crank train

The engine cylinders and pistons were then created, the EngCylinder template was then selected from the template library and dragged into the workspace, and The EngCylinder is somewhat unique, because the majority of the input to the object comes in the form of reference objects. The engine performance Gt-suite tutorial was then followed to complete the cylinder.

Exhaust System

And exhaust system in now require by the engine in order to create a realistic simulation, similar to the inlet manifold the exhaust system consists of pipe rounds and flow splits.

A drawing was made from measurements obtained measuring a Zetec exhaust system.

The exhaust ports measured 56mm diameter, the exhaust runners measure 150mm from cylinder head mating surface to the exhaust down pipe port measuring 60mm diameter.

The exhaust down pipe has a length of 1000mm. this gives a total length of 1190mm

For the purpose of this simulation a full exhaust system is not necessary in order to obtain sufficient results,

The drawing was then split up into pipe rounds and flow split to allow for the exhaust manifold to be created in GT-Power, the following boundary data was entered to create the 4-1 exhaust flow split.

Link ID Number	1	2	3	4	5
Part Name	exhport-1	exhport-2	exhpipe-1	exhport-3	exhport-4
Adjacent Part Diameter	56.0	56.0	60.0	56.0	56.0
Angle wrt X-axis	0	180	90	45	135
Angle wrt Y-axis	90	90	90	90	90
Angle wrt Z-axis	90	90	0	135	135
Characteristic Length mm	40	40	40	40	40
Expansion Diameter mm	56	56	60	56	56

Table 6 zetec 4-1 exhaust flow split data

Fuel injector

Fuel injector’s are required in order to deliver fuel to the air mixture, the fuel injector is controlled by the ECM unit, the ECM unit knows the required amount of fuel needed by the engine by the use of a AMF (air mass flow meter), and the AMF measure the density and quantity of the air. In GT-Power fuel delivery quantity is distinguished by selecting a component of the inlet tract in which an Air Mass reading will be taken.

The InjAfSecConn template is selected from the template library and copied into the work space,

The values entered for the Zetec fuel injector can be found below.

[Type a quote from the document or the summary of an interesting point. You can position the text box anywhere in the document. Use the Drawing Tools tab to change the formatting of the pull quote text box.]

Connecting components

Now all components have been created they can now be placed on the workspace and connected together, this is done from left to right,

Figure 12 the complete zetec engine built in GT-Power

All components are now linked the next step is to run the engine simulation,

Case Setup

In order to obtain sufficient result from the engine being run a case setup needs to be assigned ,Case setup allows different parameters previously set by the user (any value in closed square brackets e.g. [RPM]) to be replaced by different values set in the case setup dialogue box.

When the engine crank train was created a value of [rpm] was entered for engine speed, this has created a RPM tab in the case setup, this allows us to create different cases for different rpm sites, 7 cases were assigned in 1000rpm intervals from 1000 idle speed to a max engine speed of 7000rpm,

With the case setup now complete the simulations can be run, once the simulation is finished the results can be analysed in GT-Post.

Figure 14 Gt-Post loading screen

Chapter 4 – GT-POST

When running engine simulations GT-Power will bring up a black dialogue box containing all the information being processed to obtain engine performance results, it is wise to watch for any errors or restricted time-steps as this will show you which components are hindering engine performance, or any errors causing power loss.

Figure 15 Gt-Post processing dialogue

Figure 16 Gt-Post processing dialogue complete simulation

Once the simulation has completed text similar to the above should be displayed, press any key to continue, the engine can now be viewed in GT-Power.

Chapter 5 Analysing Gt-post data

Once the simulation is complete the data obtained can be viewed, due to the fact that it is the inlet manifold in which we are most concerned the types of data which are relevant to our testing are,

Power/torque
Ve (volumetric efficiency)
Static pressure present in the intake (allows us to analyse pulse tuning)

Figure 17 standard zetec power graph

The above power graph [figure17] is used as a base reference for the standard zetec engine. This allows us to make comparisons between the standard inlet system and the individual throttle body system.

The 1D simulation produced

124.47ft-lbs max torque at 4000rpm

136.84bhp at 7000rpm

Zetec cam timing diagram

Figure 18 above cam timing diagram of the zetec engine

The inlet camshaft begins opening the intake valve at 350 degrees 10 prior to TDC the exhaust valve is still closing at this point and is not fully closed until 370 degrees this produces a overlap of 20 crank angle degrees promoting the scavenging effects of the induced mixture expelling the burnt charge. This can be seen in [figure 18]

Rarefaction waves

Figure 19 above is a graph showing the static pressure inside the inlet port of cylinder 1 with the standard zetec manifold this can be analysed to see the effects of pulse tuning

The graph in [figure?] shows the harmonics occurring in the inlet system the high spots occur when the pulse reaches the valve and the low spots happening when the pulse reaches atmosphere and reverses back to the valve, these pulse can be tuned to arrive at the inlet valve just as its closing thus increasing volumetric efficiency [figure?] and forcing the maximum amount of charge in without the help of a F/I system.

It is not known whether or not the zetec inlet manifold was designed to maximise pulses, or if the theory was even put into place, but by analysing the graph a pulse of 1.22 bar 0.22 bar above atmospheric pressure arrives at the inlet valve at the correct time this can be seen occurring at -100degrees. This pulse happens at 4000rpm with the inlet being a length of 610mm when calculated in the inlet length formula this optimises the pulse at 3850rpm, the calculations for this can be seen in [figure 19a].

“The inlet system featured in [figure?] Is the standard zetec with a total intake track length of 66cm from atmosphere to the back to the valve, this was transposed from the following formula

Speed of sound =

Therefore, time taken to travel the tract length and back again

t= time for the pulse to travel the tract length and back again

L= length of tract from open end to inlet valve head (mm)

C=speed of sound through air (approximately 330m/s)

The crankshaft angular displacement during the same interval of time would be as follows.

Crankshaft displacement = time to travel tract length and back again X angular speed

Therefore

Where = crankshaft angular displacement (deg)

N = engine crankshaft speed (RPM)

The standard zetec inlet is to deliver the strongest pulse to the intake valve at 3850rpm to increase maximum power and overall torque.

Inlet length to optimise pulse tuning at 3850rpm =607mm (66cm) from valve to atmosphere

The length of the standard zetec inlet is 610mm from valve to atmosphere this optimises the rpm site of 3850rpm.”[Figure 19a]

Volumetric efficiency

Figure 20 Volumetric efficiency graph for standard zetec inlet

The positive effects of pulse tuning can be seen in [figure?] volumetric efficiency is at its highest at 4000rpm this is where the pulse is tuned to arrive at the inlet valve at the correct time, this concludes that pulse tuning does create a positive effect on performance.

Zetec engine with optimised inlet manifold

The zetec engine previously programmed with the ITB inlet manifold was opened in Gt-Post to analyse the results.

Figure 21 above power graph of zetec engine fitted with individual throttle bodies in GT-Post

The above power graph is very similar to the previous graph in looks [figure], peak power still occurs at 7000rpm with 157.3bhp a increase of 20.46bhp this is a large increase considering only changes to the inlet system had been made. Torque was also improved 143.41ft-lbs occurring at 4000rpm an increase of (18.94ft-lbs).

Pulse tuning

Figure 22 above static pressure graph for zetec engine with ITB manifold

It can be seen in [figure?] the effects the shortened individual throttle body manifold made to the rpm in which the strongest pulse arrives, a pulse with a strength of 1.3 bar (0.3 above atmospheric pressure) arrived at the intake valve just before closure at 7000rpm, where as in previous tests of the standard inlet the pulses arrived much earlier at 4000rpm with less amplitude, for maximum performance and volumetric efficiency at high RPM the shorted individual throttle body manifold is much better at achieving this.

The volumetric efficiency in which this manifold produced can be seen in [figure?]

The graph in [figure22] shows the harmonics occurring in the inlet system the high spots occur when the pulse reaches the valve and the low spots happening when the pulse reaches atmosphere and reverses back to the valve, these pulse can be tuned to arrive at the inlet valve just as its closing thus increasing volumetric efficiency [figure23] and forcing the maximum amount of charge in without the help of a F/I system.

Volumetric efficiency

Figure 23 zetec fitted with individual throttle bodies, volumetric efficiency graph

It is apparent that the volumetric efficacy has increased through the entire RPM range with over 100% occurring from 3500rpm onwards, this shows that the new inlet system has improved the efficiency of the engine from ideal to redline.

Volumetric efficiency comparison

[Figure 23] above left volumetric efficiency graph zetec with individual throttle bodies.

[Figure 22] above right volumetric efficiency graph zetec with standard inlet manifold.

On comparison of the volumetric efficiency results it appears that the individual throttle bodie inlet manifold not only creates a higher maximum volumetric efficacy, but also created higher volumetric efficiency over the zetec engine entire RPM range.

The standard zetec engine produces a volumetric efficacy of 0.777% at 1000rpm whereas the zetec with ITB’S produces 0.94% this is a 16.3% increase in V/E.

The standard zetec then has a steady rise to its max V/E of 92% at 4000rpm which is also the calculated optimised rpm range for pulse tuning to occurs see figure 19a.

Whereas the ITB zetec has a steady rise in V/E from 1000rpm 0.94% to 7000rpm with a max V/E of 105.25% this is proof of the pulse tuning creating positive results due to the filling capacity of the engine being over 100% proving the air is being physically pushed into the engine due to pulse arriving at the correct timing.

Chapter 6 3D Cad

The first step in designing a prototype component is to construct a 3D Model of the component in question, Autodesk Inventor Professional was the chosen software for this task.

A 3 Dimensional model enables the component to be rotated and mated to other components thus allowing the programmer to find any manufacturing constraints and possible problems with the design. The inlet manifold flange was the first component to be created; measurements and hand drawing were drafted allowing for an accurate CAD model to be created.

Figure 24 above 3d cad drawing of the inlet flange

The throttle bodies would then connect to the inlet flange, the original design utilised silicone piping to connect the itb’s to the manifold, this method of joining would cause a reduction of flow due to the steps created in the intake track. The above design was then re-designed adding flanges to the intake runners to allow the throttle bodies to be bolted to the manifold this would reduce restriction and remove any steps.

Figure 25 above is the revised drawing of the intake flange.

3D Cad Throttle bodies

The individual throttle bodies were now constructed using inventor pro.

Dimensions previously obtained using a venire calliper was used to create an accurate replication of the throttle bodies. Flanges were also added to each end of the throttle bodies allowing them to bolt onto the manifold flange and plenum, allowing for an air tight seam.

Figure 26 above, 3d cad model of throttle bodies.

The throttle spindle required allowing all 4 throttle butterflies to open and close simultaneously was then constructed.

This consisted of a brass bar 10mm in diameter and 400mm in length with 40mm cut outs 6mm deep at 92mm centres allowing fitment of the throttle butterflies.

Figure 27 above 3d cad rendering of throttle spindle

3D Cad Throttle spindle assembly

Throttle butterflies were then created with a diameter of 40mm and a depth of 2mm, with 2 holes offset from the centre by 10mm, these allow for secure fixture to the throttle spindle.

Figure 28 above 3d rendering of the throttle butterfly

A top section was then created with holes offset 10mm from the centre with countersunk wholes to reduce flow restrictions .

Figure 29 above exploded throttle spindle assembly

Chapter 7Fabrication of individual throttle bodies

Due to the high cost of aftermarket throttle bodies a set of O.E.M throttle bodies were sourced from a Suzuki GSXR-600 motorcycle, these we ideal due to the I/D (Internal Diameter) being the same as the inlet runners of 40mm. The throttle bodies had to be extensively modified in order to line up with the Zetec’s 92mm port spacing. This included re-manufacturing the brass throttle spindle to the correct length and machining new slots to accommodate the throttle butterflies.

In order to keep the inlet runners as straight as possible the throttle bodies had to be cut apart to create 4 separate throttle bodies as originally they were in pairs.

The image to the left shows the throttle bodies with their original spacing, throttle bodie 1 and 2 are joined as are 3 and 4. The blue line shows where cuts had to be made to make the throttle bodies individual.

The throttle bodies were stripped down completely leaving only the cast aluminium throttle bodies, cuts were then made between bodies number 1, 2. This allowed for the throttle bodies to be re-spaced to their respective location. In order to find the cylinder spacing an inlet manifold gasket was used as a template for inlet port spacing. The throttle bodies were then placed above each port allowing for templates for steel plate to be made to affix the throttle bodies in their new location.

2 sets of plates were made from 1mm steel sheet with holes to allow threading fixing’s for drilled in the throttle bodies and then threaded with an M0.5mm engineers tap for the M0.5mm bolts.

The plates were the attached allowing for measurements to be obtained. This step was crucial in order to complete a CAD (computer aided drawing) of the throttle spindle, which was to be made from brass on an engineer’s lath.

Construction of throttle spindle

Due to changes to the cylinder spacing a new throttle spindle was manufactured from 10mm brass bar.3D CAD drawing previously created [see figure 27] of the revised spindle design was used as a engineers drawing to aid the manufacturing process.

Figure 33 above is a schematic of the throttle spindle.

The brass spindle had to be turned down to the correct size this involved the work piece being placed in a lath to be turned to the correct dimensions, there was no room for error and the tolerances were very high due to the fact the inlet manifold will be under positive pressure and air would be able to escape.

Figure 34 above throttle spindle after turning

In order for the throttle cable and linkage assembly to be connected the 6mm end would require a M6 thread to allow the linkage to be secured.

Figure 35 M6 thread added.

The throttle spindle was then machined to accept the throttle butterflies, the meant milling on a ‘’T milling machine’’. To allow for the cut outs seen in [figure 3d cad rendering of throttle spindle] a jig had to be made to support the spindle and to hold it central whilst the cut outs were made.

Firstly a cad drawing of the jig was made, this allows for simple machining and to find and problems before they would be encountered during manufacturing.

The jig consisted of a 50mm billet of steel with a length of 150mm and width of 100mm. This was then processed on the milling machine to insure the top surface was true, this involved removing 1/1000 of a inch from the surface. A 20mm wide 8mm deep channel was then cut allowing the throttle spindle to be secured in place by 2 clamps which are tightened by M10 bolts. A 50mm wide channel 8mm deep was then cut central to the work piece, this cut was made to allow the cutting tool to cut the brass spindle without removing material from the steel jig.

Figure 37 above various images during construction of the spindle jig.

Construction of Inlet manifold

An inlet flange was needed in order to affix the throttle bodies to the cylinder head. The 3d cad illustration [figure] of the inlet flange was converted into a 2d cad drawing [figure], this was then sent off to be water cut using precision equipment

The flange is made from 6mm plate steel a minimal depth of 5mm was needed to reduce warp age and increase tensile strength to prevent the manifold form distorting over time.

Fabrication of inlet runners

The inlet runners were made from bar steel on the lath. This involved turning a 50mm bar down to 42mm and boring the centre out to 38mm.

Figure 39 42mm bar hollowed out to 38mm.

The now hollow bar was now cut into 4 50mm lengths. The runners could now be welder onto the inlet flange. This involved clamping the flange to the welding bench to try and limit distortion caused by heating the metal flange up, the four runners were then tacked onto the flange using a TIG set (tungsten inert gas welder). Once all four runners were tacked onto the flange they were fully seam welded.

Welding of inlet flange

Figure 41 image of zetec flange with runner seem welded

The 50mm inlet runner was then welded onto the flange this involved bolting the flange to a aluminium block to act as a heat sink to minimise warp age.

The flange was then tack welded on each side then fully seam welded, the insides of the ports were then cleaned up using carbide burs and air die grinder operating at 28,000rpm.

Figure 42 carbide bur

Completed ITB manifold

Figure 43 completed inlet manifold

The individual throttle body manifold was not complete and assembled.

This allowed flow testing of the cylinder head and inlets to be carried out.

Chapter 8 Porting of cylinder head

Figure 44 image of the standard zetec inlet port

a small degree of rough porting was done to the head to increase flow results this included removing factory casting marks which increase turbulence and slow down airflow, and knife edging the dividing edge between the two ports to make the transition between single and Siamese more streamlined.

Figure 46 silicone casting of port

Silicone casts were then made of the inlet port this involved coating the inlet ports with a mould release agent (Vaseline in this case due to shoe string budget) and then filled with silicone this was given 70 hours to dry the moulds were then pushed out the ports to give me a rough idea of where material needed to be removed to increase flow. Unfortunately no pictures were taken at this stage due to loss of data.

Knife edging inlet ports

Figure 47 knife edged ports

The inlet ports were then knife edged using the carbide burrs see [figure 42] this involved rough removal of excess material and casting, also making the transition from one port to two as sharp as possible.

Figure 48 Complete inlet port after finishing with 260grit drum wheel

The inlet ports were then finished by removing any rough edges and smoothing all the marks and scratched left by the carbide burrs.

Chapter 9 flow testing

Flow testing

Figure 49 flow testing bench setup, the zetec cylinder head can be seen to the left of caption

The zetec cylinder head was then attached to the flow test to allow testing of the flow rating of the inlet ports and inlet manifold. This involved fitting a bar in place of the cams with 2 bolts which can be tightened to simulate the cams lifting the valves.

Two DTI gauges are used for each inlet valve to allow the amount of valve lift in mm to be monitored. Data about the zetec engine and port sizing was entered into the super flow program.

Testing was going to be done on the standard zetec inlet manifold but due to the design it would not physically fit on the flow bench without large adapters being manufactured. This meant that a comparison of flow abilities of the standard inlet and ITB inlet not being possible, instead a test without any inlet was carried out to see if the ITB inlet caused any resistance over the cylinder head without a inlet manifold fitted.

Figure 50 screen shot captured in port flow analyzer of zetec head with no inlet

Figure 51 screen shot captured in port flow analyzer of zetec head with ITB inlet

Figure 52 ITB and tests without manifold overlaid

It is apparent by looking at the graph that when the zetec head was fitted with individual throttle bodies the air flow increased by as much as 15liters per second , with 8-10mm of lift following the same path as the head without manifold, this leads me to believe that the ITB manifold not only causes less restriction than no manifold but also increases the engines breathing ability.

Chapter 10 Conclusion

All the testing performed on the zetec engine both theoretical and practical have lead me to believe that inlet design, and pulse tuning in particular play a massive part in how engine performance in produced and tuning the power band of the engine.

From the results obtained in get suite and the various different length and shaped inlets helped me decided the optimum length and design of inlet manifold for the 2.0 zetec. It was concluded though vast research that the individual throttle bodie was the best available induction system due to its high throttle response speed and allowing placement of the throttle butterfly much closer to the valve whereas a conventional single throttle inlet this is not possible.

If this project was to be taken further, more research and development testing of bell mouths would be carried out as they reduce drag created by flowing into a straight pipe they also increase velocity similar to the venture effect.

It would also have been highly beneficial if the standard zetec inlet manifold could have been flow tested on the flow bench as the comparison between the ITB manifold and no manifold is very vague although an improvement can be seen. But due to time constraints and costing of the establishment buying in all the jigs to test various different manufactures cylinder heads this is just not possible in the time scale available.

Results obtained from analysing the static pressures inside the inlet ports on GT-Suite lead me to believe that the length on inlet created will create maximum power at 7000rpm as this can be seen on both power graphs and volumetric efficacy diagrams, further testing involving fitting the inlet manifold to a road going vehicle and testing the back to back on a dynameters against the standard inlet would further fortify this.

Research for further developments

Bell Mouth

The “bell mouth” is the opening, at which the intake runner meets atmospheric pressure. Bell mouths are available in many different shapes and sizes, which creates a problem, which bell mouth is best for our application,”

The perfect bell mouth will allow air to pass whist maximising CD “The effectiveness of the flow at the end of a pipe in an internal combustion engine is expressed numerically as a ‘discharge coefficient’, i.e., a Coefficient of Discharge or CD”

The CD of the different bell mouths can be measure and compared using a steady flow rig, similar to the Super flow rig found in the test cell at Blackpool and the Fylde collage.

The pipe ends under testing, in this case the bell mouth, is placed before a settling tank/plenum and a steady flow of air is sucked through it into the plenum by the use of a vacuum pump, similar to one found in a domestic vacuum cleaner. Some commercial flow testing rigs will situate the pressure tank under water to increase atmospheric pressure to allow testing for applications using forced induction. Due to the fact the inlet system being designed for a Naturally Aspirated engine, the bell mouth is exposed to pressure ratios of 1.1 or less, so a commercial flow bench is not required for testing.

Figure 53 a simple drawing of a flow bench

Bell mouth design

Figure 54 bell mouth spillway

The bell mouth pipe end has been tried and testing in many different applications, above is an image of a bell mouth spillway. The bell mouth spillway is used to control the release of flow from a dam into a downstream area, the spillway releases flood so that the water does not overflow.

The pipe end of the spillway has been designed to replicate the bell mouth found in the internal combustion engine, this is to mineralize flow loss.

Other shapes and design of bell mouths have also been used over the years for spill ways, but the above design was found to be most effective due to its elliptical profile.

Stepped bell mouth spillway

Figure 56 above a stepped bell mouth spillway

The principle around the stepped bell mouth spillway is to make the flow of water turbulent to help break up possible ice deposits. The desired effect would not be positive on an internal combustion engine, but is a good example of how air would react to a stepped bell mouth.

Figure 57 above is a stepped bell mouth spillway

Elliptical bell mouth spillway

Figure 58 above is an elliptical bell mouth spillway

The elliptical bell mouth allows a smooth and even flow of water (or air in an automotive application) to enter the pipe end this can be seen below in figure ??.

Figure 59 above is a smooth elliptical bell mouth spillway

This design reduced turbulence whist maintaining maximum flow, notice how the water moves in the same manor around the perimeter of the bell mouth.

Straight pipe bell mouth spillway

Figure 60 above is a straight pipe bell mouth spillway

The straight pipe bell mouth causes the greatest flow loss, this is due to the fact the pipe end does not provide a smooth transition for the water/air to enter the pipe, this causes the flow path to become unpredictable and unstable. The image of the straight pipe bell mouth spillway in figure ?, is an example of the path and nature in which air would enter a straight pipe, notice the water is not clinging to the walls unlike the elliptical bell mouth, this reduced the velocity of the water and creates turbulence.

Literature Review

In this review Heinz Heisler, Advance engine technology book will be discussed.

A quote from Heinz Heisler, Advance engine technology says

1. To provide as direct a flow as possible to each cylinder

2. To provide equal quantities of charge to each cylinder

3. To provide a uniformly distributed charge of equal mixture strength to each cylinder

4. To provide equal aspiration intervals between branch pipes

5. To provide the smallest possible induction tract diameter that will maintain adequate air velocity at low speed without impeding volumetric efficiency in the upper speed range

6. To create as little internal surface frictional resistance in each branch pipe as possible

7. To provide sufficient pre-heating to the induction manifold for cold starting and warm-up periods

8. To provide a means for drainage of the heavier liquid fraction of fuel

9. To provide a means to prevent charge flow interference between cylinders as far as possible

10. to provide a measure of ram pressure charging” [1]

This is going to be the criteria in which the inlet manifold will have to conform to

When designing an inlet manifold for a four cylinder engine the design needs to be taken into account, Heinz Heisler Advance engine technology quotes

There is a compromise to smooth surface walls as stated in, Heinz Heisler Advance engine technology

Heinz Heisler has contradicted his argument by stating the pros of a smooth surface finish and downsides of a smooth finish.

Figures

Figure 1 above, illustration of a Ford Zetec inlet manifold and its dimensions. 23

Figure 2 above. Ford Zetec EFI inlet manifold. 23

Figure 3 pipe round template renamed intrunner and filled with attributes and measurements from a standard Zetec inlet. 24

Figure 4 drawing of standard zetec 4-1 flow split 25

Figure 5 Gt-Suite flow split data from zetec inlet 26

Figure 6 inlet 4-1 flow split viewed in 3d. 26

Figure 7 a screen print captured in GT Power Zetec inlet port 27

Figure 8 standard zetec cam data table. 28

Figure 9 diagram of standard zetec exhaust manifold and dimensions. 30

Figure 10 fuel injector timing tab. 32

Figure 11 fuel injection data. 32

Figure 12 the complete zetec engine built in GT-Power. 33

Figure 14 Gt-Post loading screen. 34

Figure 13 case setup Gt-Suite. 34

Figure 15 Gt-Post processing dialogue. 35

Figure 16 Gt-Post processing dialogue complete simulation. 35

Figure 17 standard zetec power graph. 36

Figure 18 above cam timing diagram of the zetec engine. 37

Figure 19 above is a graph showing the static pressure inside the inlet port of cylinder 1 with the standard zetec manifold this can be analysed to see the effects of pulse tuning. 38

Figure 20 Volumetric efficiency graph for standard zetec inlet 42

Figure 21 above power graph of zetec engine fitted with individual throttle bodies in GT-Post 43

Figure 22 above static pressure graph for zetec engine with ITB manifold. 44

Figure 23 zetec fitted with individual throttle bodies, volumetric efficiency graph. 45

Figure 24 above 3d cad drawing of the inlet flange. 47

Figure 25 above is the revised drawing of the intake flange. 48

Figure 26 above, 3d cad model of throttle bodies. 49

Figure 27 above 3d cad rendering of throttle spindle. 49

Figure 28 above 3d rendering of the throttle butterfly. 50

Figure 29 above exploded throttle spindle assembly. 50

Figure 30 Gsxr-600 throttle bodies). 51

Figure 32 Gsxr-600 throttle bodies). 52

Figure 31 zetec inlet gasket 52

Figure 33 above is a schematic of the throttle spindle. 53

Figure 34 above throttle spindle after turning. 53

Figure 35 M6 thread added. 54

Figure 36 3d cad rendering of the jig required to hold the throttle spindle central during machining. 54

Figure 37 above various images during construction of the spindle jig. 55

Figure 38 Water cut inlet manifold flange. 55

Figure 39 42mm bar hollowed out to 38mm. 56

Figure 40 inlet runner. 56

Figure 41 image of zetec flange with runner seem welded. 57

Figure 42 carbide bur. 57

Figure 43 completed inlet manifold. 58

Figure 44 image of the standard zetec inlet port 59

Figure 46 silicone casting of port 59

Figure 45 silicone casting of inlet port 59

Figure 47 knife edged ports. 60

Figure 48 Complete inlet port after finishing with 260grit drum wheel 61

Figure 49 flow testing bench setup, the zetec cylinder head can be seen to the left of caption. 62

Figure 50 screen shot captured in port flow analyzer of zetec head with no inlet 63

Figure 51 screen shot captured in port flow analyzer of zetec head with ITB inlet 63

Figure 52 ITB and tests without manifold overlaid. 64

Figure 53 a simple drawing of a flow bench. 66

Figure 54 bell mouth spillway. 67

Figure 55 Bell mouths fitted to a triumph spitfire. 67

Figure 56 above a stepped bell mouth spillway. 68

Figure 57 above is a stepped bell mouth spillway. 68

Figure 58 above is an elliptical bell mouth spillway. 69

Figure 59 above is a smooth elliptical bell mouth spillway. 69

Figure 60 above is a straight pipe bell mouth spillway. 70

Tables

Table 1Material heat properties. 16

Table 2materials matrix analysis chart 18

Table 3 material comparison bar chart 18

Table 4 zetec valve train data. 28

Table 5 zetec lift array data. 29

Table 6 zetec 4-1 exhaust flow split data. 31

Mobile Mechanic & Auto electriction in Blackpool

Inlet manifold design and theory in regards to maximising volumetric efficancy

Symbols and Abbreviations

Chapter 1 – inlet manifold parameters design feasibility

Possible Problems

Inlet manifold Design

Inlet manifold surface finish

Conclusion

Inlet runner Section shapes

Segment sections

Conclusion

Chapter 2 – material specification

Chapter 3 – inlet manifold science

Chapter 4 – 1D Engine Simulation GT-POWER

Intake manifold

Intake ports and valves

Cylinders and crank train

Exhaust System

Fuel injector

Case Setup

Chapter 4 – GT-POST

Chapter 5 Analysing Gt-post data

Zetec cam timing diagram

Rarefaction waves

Volumetric efficiency

Zetec engine with optimised inlet manifold

Pulse tuning

Volumetric efficiency

Volumetric efficiency comparison

Chapter 6 3D Cad

3D Cad Throttle bodies

Chapter 7Fabrication of individual throttle bodies

Construction of throttle spindle

Construction of Inlet manifold

Fabrication of inlet runners

Welding of inlet flange

Chapter 8 Porting of cylinder head

Knife edging inlet ports

Flow testing

Chapter 10 Conclusion

Research for further developments

Bell Mouth

Bell mouth design

Stepped bell mouth spillway

Elliptical bell mouth spillway

Straight pipe bell mouth spillway

Literature Review

Figures

Share this:

Leave a comment Cancel reply

North west mobile mechanics

Archives