Flow measurements using combustion image velocimetry in diesel engines

(1)

Flow measurements using combustion image velocimetry in diesel engines

HENRIK W. R. DEMBINSKI

Licentiate thesis

Department of Machine Design Royal Institute of Technology SE-100 44 Stockholm

TRITA – MMK 2012:03 ISSN 1400-1179 ISRN/KTH/MMK/R-12/03-SE

(2)

TRITA – MMK 2012:03 ISSN 1400-1179

ISRN/KTH/MMK/R-12/03-SE

Flow measurements using combustion image velocimetry in diesel engines Henrik W. R. Dembinski

Licentiate thesis

Academic thesis, which with the approval of Kungliga Tekniska Högskolan, will be presented for public review in fulfilment of the requirements for a Licentiate of Engineering in Machine Design. The public review is held at Kungliga Tekniska Högskolan, Brinellvägen in 83, room B319 Gladan, 1^th of March at 10:00.

(3)

Abstract

This work shows the in-cylinder airflow, and its effects on combustion and emissions, in a modern, heavy-duty diesel engine. The in-cylinder airflow is examined experimentally in an optical engine and the flow field inside the cylinder is quantified and shown during

combustion, crank angle resolved. Cross-correlation on combustion pictures, with its natural light from black body radiation, has been done to calculate the vector field during the

injection and after-oxidation period. In this work, this technique is called combustion image velocimetry (CIV). The quantified in-cylinder flow is compared with simulated data, calculated using the GT-POWER 1-D simulation tool, and combined with single-cylinder emission measurements at various in-cylinder airflows. The airflow in the single cylinder, characterised by swirl, tumble and turbulent intensity, was varied by using an active valve train (AVT), which allowed change in airflow during the engine’s operation. The same operation points were examined in the single-cylinder engine, optical engine and simulated in GT-POWER.

This work has shown that the in-cylinder airflow has a great impact on emissions and combustion in diesel engines, even at injection pressures up to 2,500 bar, with or without EGR and load up to 20-bar IMEP. Swirl is the strongest player to reduce soot emissions.

Tumble has been shown to affect soot emissions negatively in combination with swirl.

Tumble seems to offset the swirl centre and the offset is observed also after combustion in the optical engine tests. Injection pressure affects the swirl at late crank angle degrees during the after-oxidation part of the combustion. Higher injection pressure gives a higher measured swirl. This increase is thought to be created by the fuel spray flow interaction. The angular velocity in the centre of the piston bowl is significantly higher compared with the velocity in the outer region of the bowl. Higher injection pressure gives larger difference of the angular velocity.

Calculated swirl number from the CIV technique has also been compared with other calculation methods, GT-POWER and CFD-based method. The result from the CIV

technique are in line with the other methods. CFD-based calculations, according to [62], has the best fit to the CIV method. The GT-POWER calculations shows the same trend at low swirl number, but at high swirl number the two methods differs significantly.

(4)

Acknowledgements

This thesis presents the results of the project “Transient Air Management”, commissioned by Scania CV AB, the Royal Institute of Technology (KTH) and the Swedish energy agency. I want to thank my supervisor, Prof. Hans-Erik Ångström, my colleague at KTH and Scania.

My family, for the support during this period of hard work.

(5)

List of publications

I. An Experimental Study of the Influence of Variable In-Cylinder Flow, Caused by Active Valve Train, on Combustion and Emissions in a Diesel Engine at Low λ

Operation. SAE paper: 2011-01-1830. Henrik W. R. Dembinski, Hans-Erik Ångström.

II. Optical study of swirl during combustion in a CI engine with different injection pressures and swirl ratios compared with calculations. SAE paper: 2012-01-0682.

Henrik W. R. Dembinski, Hans-Erik Ångström.

III. The effects of injection pressure on swirl and flow pattern in diesel combustion.

Submitted to: International Journal of Engine Research IJER-12-0006. Henrik W. R.

Dembinski, Hans-Erik Ångström.

(6)

Introduction

The compression-ignited engine, named the diesel engine, after its inventor Rudolf Diesel (born in Paris 18 March 1858 and died 29 September 1913), has been the main power source used in heavy-duty vehicles for a long period of time. The reason is mainly

economical, as the diesel engine is more efficient, and historically cheaper fuel together with higher durability compared with its competitor, the otto engine (also named after its inventor, Nicolaus Otto, born in Holzhausen 14 June 1832 and died 26 January 1891).

Over the decades, since the diesel engine was patented in 1894, many improvements to the engine have been carried out. However, there are still many challenges left to deal with.

Emission legislation is increasingly strict, together with higher demands on engine efficiency.

This equation is not very easy to solve. Higher engine efficiency does not automatically mean decreased emissions; rather the opposite is true with today’s engines. A better

understanding of the processes in the engine is one key to solving the emission/efficiency equation. This work is just one small piece of a giant puzzle that many scientists around the world are trying to piece together.

The basic principle of the diesel engine

The basic principle of the compression ignition (CI) engine is to increase the temperature of the air in the cylinder, typically in the order of 1,000 K, well over the self-ignition temperature of the fuel, and then inject the fuel. In this way, the combustion can be controlled in a better way compared with the premixed spark ignited (SI) engine. The temperature increase is caused by the increase in the cylinder pressure during compression. In a pressure volume (PV) diagram, see Figure 1, the operation of a CI engine can be observed. The PV diagram comes from the original patent by Rudolf Diesel and shows, in principle, the operation of the engine. When the pressure and the temperature increases, due to the change in cylinder volume (from 1 to 2 in diagram), the fuel is added to the compressed air in the cylinder [1].

The fuel injection maintain until 3 in the original patent. The maximum pressure starts to drop under the expansion and the cylinder pressure does not increase more than the compression pressure. In a modern engine, the PV diagram looks like Figure 2. The cylinder pressure greatly increases during combustion. In the original patent, the aim was to also use water injection. According to the patent, the exhaust temperature can be reduced to below the inlet temperature(!) of the engine by carrying out mechanical work on the crankshaft. The engine should not need any external cooling either. Neither of these two statements has been proven to work in reality, but the diesel process does work and powers many different applications.

Figure 1. Pressure and volume diagram from Rudolf Diesel’s patent [1]. Pressure is shown on the y-axis and volume is shown on the x-axis.

(10)

However, when the fuel is injected into the hot air in the cylinder, the relatively cold fuel starts to increase in temperature and starts to evaporate. Energy is taken from the hot air in the cylinder and the fuel starts to decompose into other chemical substances, for example Hydroxide (OH). OH plays an important role during the ignition delay period (and also in the after-oxidation period). Ignition delay period is the time between when the fuel has started to be injected until when exothermal reaction is observed (typically on the cylinder pressure trace). OH is a reactive substance that, at a certain critical mass, ignites the fuel. The ignition delay in CI engines is often explained by the Arrhenius correlation, which takes the in-

cylinder average temperature and pressure, together with the activation energy of the fuel, into account to calculate the ignition delay [2]. However, as shown later in this work, other things, such as airflow in the cylinder, affect the ignition delay. After the first premixed

combustion phase the diffusion combustion part takes over. The fuel is continuously injected, mixed, transformed and combusted in a non-premixed diffusion flame. Temperature and pressure rapidly increase in the cylinder. After some crank angle degree (CAD) the fuel injection stops and, shortly afterwards, so does the diffusion flame period. The after-oxidation combustion period takes then over, where the mix of air/fuel residuals that are in the cylinder, and proceeding the combustion of unburned products. This process is maintained under the expansion until the in-cylinder temperature decreases to a level where the exothermal

reactions stop, due to the chemical timescale being too long. In other words, the temperature is too low for the after-oxidation to happen due to a decrease in cylinder pressure caused by volumetric expansion (piston movement). When the volume reaches the highest value, the exhaust valve opens and the piston lowers the volume again, resulting in the exhaust being evacuated out from the cylinder. When the piston reaches the top dead centre (TDC), with the lowest cylinder volume, the exhaust valve closes and the inlet valve opens. Fresh air is sucked into the cylinder, typically at a slightly lower pressure compared with the exhaust stroke. The cycle starts over again when the piston reaches the bottom dead centre (BDC) and the inlet valve closes. In Figure 2, two load cases are plotted in the PV diagram, medium and high load. As seen, the overall pressure is higher in the high-load case compared with the medium-load case. In the high-load case, the inlet and exhaust pressure are increased due to turbocharging. In this way, the air density is increased in the engine, more fuel can be added and the load on the engine can be increased significantly. The drawback with

turbocharging is that the exhaust back pressure increases, which is caused by the exhaust turbine that drives the turbo compressor. This means that the residual gases left in the cylinder increase with the back pressure and reduce the volumetric efficiency. But the increased inlet pressure more than compensates with increased trapped air mass in the cylinder.

(11)

Figure 2. PV diagram showing a four-stroke diesel engine operating on medium and full load.

Historical view

In the original patent from Rudolf Diesel, he describes that solid, liquid and gaseous fuels can be used in his invention. The cooling of the engine was not required due to the controlled combustion. He compared his patent with other internal engine patents (i.e. Otto’s patent).

According to Diesel, the fast explosion that appears in premixed engines increases the heat and pressure to extremely high levels, which leads to the problem of lubricating and cooling the engine. Thanks to his more controlled way of introducing the fuel after the compression and by introducing it over a controlled duration, he claimed he could have greater control over the combustion velocity. With this control, the fuel is combusted with the same velocity as it provides mechanical work on the flywheel. In this way, no extra heat is produced that needs to be cooled away. In the patent, an insulating jacket may be provided outside the cylinder liner. Rudolf Diesel’s original engine, which did not have a cooling function, was not a good idea. He missed the fact that the heat transfer from the in-cylinder air to the engine also acts during compression. The combustion also produces a lot of heat that dissipates into the engine components and cannot be isolated. But still today some research has been done on isolating components in the cylinder [3]. In Diesel’s patent, he also showed that water injection could be used in the compression stroke to further increase the efficiency. He even believed that the exhaust gases should, by introducing water during the compression stroke, be cooler than the surrounding air. However, it has not been possible to show this in the reality.

The first fuel injection system that Diesel showed in the first patent was based on coal powder. In Figure 3, the mechanism is shown at four different positions, where the

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

Volume

PCYL1 [bar]

0 20 40 60 80 100 120 140 160 180 200

Medium load, IMEP 11 bar High Load, IMEP 20 bar

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

Volume

PCYL1 [bar]

0 1 2 3 4 5 6 7 8 9 10

(12)

combustion chamber is placed under the device and the coal is fed from above, at points a, b, c and d. In the first position, the coal powder is fed into the notch. The notch is then rotated down to position b, where the powder is introduced into the compressed air in the cylinder (TDC combustion). The powder is pressurised and starts to combust when it comes into contact with the compressed air. By controlling the amount of powder that is in contact with air (the small throat area), the combustion velocity is thereby also controlled.

Figure 3 Rudolf Diesel’s patented injection system for coal powder.

(13)

In the patent, a fuel system for liquid fuels was also described, where the fuel is introduced into the cylinder by a higher pressure than the highest possible cylinder pressure. For this, he aimed to use some kind of feed pump (not shown in this patent) that could pressurise the fuel up to those levels. In his next patent [4], he developed the injection system further, especially for liquid and gaseous fuels. He understood that mixing in a non-premixed combustion process was important. Different layouts of the combustion chamber with different injection systems are shown in this patent. The pre-chamber combustion geometry is introduced as well as the first version of today’s combustion bowl geometry for direct injected diesel engines, Figure 4.

Figure 4 The first version of a piston bowl (K in Fig. 3) and the first pre-chamber diesel (Fig. 5) [4].

Today, liquid fuel is typically used in the diesel process, but natural gas can also be used in a diesel engine. For example, the Westport dual fuel system [5], where a small pilot diesel injection is injected (and combusted) and, shortly after, the main gas injection starts to be injected.

Diesel engine combustion

Diesel combustion is a sort of non-premixed combustion where the fuel is injected, typically in liquid phase directly in the preheated combustion air. The combustion process is divided into three different stages, the premixed- (ignition delay), diffusion- and after-oxidation period. The classical paper from Dec [6] shows how the process proceeds and where the different emissions are created. This is a simplified conceptual model, used to simplify the discussion about diesel combustion.

(14)

Premixed combustion period

When the fuel is injected with high velocity into the hot air (which has a lower velocity), droplet break-up occurs. The velocity difference increases the evaporation rate, and thereby also the heat and air exchange around the droplet [7]. When enough fuel has evaporated and mixed with oxygen at the right temperature, and the reaction time is long enough, ignition of the fuel can occur.

The ignition delay in a diesel engine is commonly explained by the Arrhenius correlation [2].

In the correlation, the ignition delay is based on the chemical reaction time needed for the fuel to evaporate, decompose and excite the activation energy (E_A). The environmental conditions that effect the reaction times is temperature (T) and pressure (p). A and n are constants depending on fuel and, to some extent, the injection and airflow characteristics [2].



 





 



 ^

T R p E

A ⁿ ^A

id exp



₍₁₎

where R is the gas constant.

The ignition of fuel is a complex chemical reaction with many different reaction steps.

Detailed reaction models for diesel-like fuels can be found in [8]. A model that takes the turbulent mixing and history during self-ignition into account can be found in [9].

When ignition occurs, the premixed combustion phase starts. When the premixed fuel is consumed, the combustion passes over to a diffusion flame period where the rate of heat release (HR) is governed by the amount of fuel that is injected per unit of time. The combustion flame starts to look something like the image in Figure 5.

Diffusion combustion period

The injected fuel core is expanding in the area when it propagates from the injector. Hot air is mixed with the fuel and the fuel starts to evaporate and decompose. First, the fuel is

combusted at low λ with soot production as a result, the grey area in Figure 5. Air mixes into the flame and the combustion propagates. The yellow area is typically what is easy to observe when lots of black body radiation creates a bright yellow light. Most of the thermal nitrogen oxide (NOx) is created on the surface of the flame, marked in green, where the temperature is high and oxygen is located.

(15)

Figure 5 Dec’s classical combustion plume [6].

In Figure 6, particle image velocimetry (PIV) measurements on the air entrainment in non- burning spray, the instantaneous velocity field are showed. The spray area increases downstream from the spray when air is mixed up with the fuel spray. Local vortices are created around the spray and increase the air-mixing rate.

Figure 6 Fuel spray in a spray chamber that shows air entrainment into the spray visualised by PIV measurements [10].

When the Dec model is compared with pictures from an optical engine, Figure 7, it can clearly be seen that the geometry and airflow in the engine strongly affect the combustion.

When only one flame is injected it is easy to see how it splits into two halves when it reaches the edge of the bowl. The swirl direction is marked in the picture and shows that the flame

(16)

that is on the leeward side of the spray moves in the direction of the swirl. The flame on the opposite side travels some distance before it is moved back again in the direction of the swirl. When eight sprays are introduced in the combustion chamber, Figure 7, the reflected flames contribute and are reflected back to the bowl centre. If this geometry is done correctly, the reflected flames form a recirculation zone in the bowl and improve the mixing of residual gases, heat and unburned oxygen. This is of paramount importance for diesel combustion.

Figure 7 One combustion flame compared with eight combustion flames at 1,500-bar injection pressure.

The arrows show the swirl direction and the reflected combustion flames.

After-oxidation period

The after-oxidation period is of great importance to reduce smoke emissions and much (in the order of 25–45%) of the total HR is released during the after-oxidation part of the combustion. At the end of injection (EOI), when the injection needle starts to close, the fuel flow through the injection nozzle decreases with the fuel velocity. The fuel penetration is thereby limited and the fuel spray creates less turbulence. The mixing with air is restricted with higher soot production as a result, this is shown in Figure 8. When the spray velocity decreases, the bright soot illumination is also seen in the centre of the bowl. When the injection has ended, the bright illumination increases in the middle of the bowl. This bright section in the middle of the bowl has a high soot content that needs to be oxidised. This means that a fast needle closure is beneficial to reduce the late soot production.

(17)

Figure 8. Combustion at 500 bar injection pressure, swirl number 6.3, end of injection period.

The soot and unburned residuals have time to be reduced during the after-oxidation period.

Increased airflow is beneficial during this period. The after-oxidation period is maintained as long as the in-cylinder temperature is high enough. During the expansion stroke, the volume increases, and thereby decreases the pressure and temperature in the cylinder. The time window required to reduce the emissions can be affected by the injection timing. A later injection gives a shorter time window for after-oxidation, with higher soot emissions as a result.

This description of diesel combustion is valid for “normal diesel combustion”. Other types of combustion modes that can be applied in a diesel engine are, for example, “low-temperature combustion” modes. With high amounts of exhaust gas recirculation (EGR) and long ignition delay, all of the fuel is injected into the cylinder typically before combustion. Therefore, the fuel has a longer mixing time [11], and the rich fuel zones are thereby avoided with

decreased soot and NOx formations. Examples of these types of combustion modes can be seen in [12], [13], [14], [15], [16], [17], [18] and [19]. One drawback with this type of

combustion is that it is best suited for low-load operation and is very difficult to control in a fast engine transient. This is basically the reason why this is not used in this work, which deals with fast engine transients and fairly high load.

(18)

Basic turbulent flows

Most of the flow in nature and engineering applications is turbulent. Turbulent flow is

especially the dominant flow in combustion engines. Turbulence can be observed all around us. When we stir our drink or tap on a glass of water we can easily observe turbulence. The cumulus clouds are turbulent, the boundary layer around a sailboat is turbulent and so on.

Without turbulence many things would not work. For example, the combustion in an SI premixed engine would be too slow, with the result that the maximum possible engine speed would be limited to a couple of revolutions per minute. Turbulence is characterised by its irregular random nature with circular motions (eddies) in all three dimensions, see Figure 9. It behaves randomly in time and space and always occurs at high Reynolds numbers (Re). Re is defined by,







ud  ud



Re (2)

where µ is the dynamic viscosity, ρ is the density, d is the pipe-flow diameter or a characteristic length for the problem, is the free-stream velocity and ʋ is the kinematic viscosity.

Figure 9. This picture is a side view of the large eddies in a turbulent boundary layer. Laser-induced fluorescence is used to capture the quasi-periodic coherent structures. The flow is from left to right. Note the reflection from the polished surface [40].

Turbulence cannot maintain itself; it depends on its environment to obtain energy. Turbulent flows are generally shear flows. If the energy supply is shut off, the turbulence quickly dissipates and is transformed to heat. In a diesel engine, where lots of both large eddy and small eddy turbulence is created during inlet stroke, only the large flows can survive for a longer time in the cylinder, and thereby has a chance to effect combustion, when the inlet valves are closed (energy supply). The mean small eddy turbulence lifetime is much smaller than the time for induction and compression [41].

The transition from laminar to turbulent flow does not have an exact critical Re number, where the flow is either one or the other and the transition is poorly understood. The scientist Osborne Reynolds (born 1842 and died 1912), who first experimented on high Re numbers in pipe flow, showed a critical Re (where the flow becomes turbulent) at 13,000. His

experiments were repeated in the 1970s, in Manchester, where Reynolds’ original

experimental apparatus still exists. The critical Re number was found to be much less than

(19)

13,000. The reason can be found in the transition region, where the flow is very sensitive to external disturbances, as the flow is going from laminar to turbulent.

The circular motions in the turbulent layer increase the mixing of the reactants and products significantly, and thereby also combustion velocity. The oxidation, fuel pieces, radicals and heat from the combustion are mixed and exposed to each other more rapidly compared with laminar combustion. A wide range of length scales exists in turbulent flow, from the biggest dimensions of the flow field to diffusive actions of molecular viscosity. The smallest scales, called Kolmogorov’s microscale, have relatively small timescales, which makes them quite statistically independent of the big and relatively slow vortices. The large eddies lose most of their kinetic energy with one turnover time to the smaller-length scales (called dissipation).

This means that turbulence is a strongly damped non-linear stochastic system [42].

A schematic sketch of a turbulent boundary layer near a wall is seen in Figure 10. From the stochastic boundary layer profile, a time-average thickness profile can be plotted that describes the mean thickness of the boundary layer,

 (x )

. The instantaneous velocity, u , can be divided into two velocity components, mean (U or u) and a fluctuating part (u),

U u

u  (also called Reynolds’ decomposition). In Figure 10, u is the fluctuating velocity in x-direction, vin y-direction and w in z-direction (not shown in this picture).

At the wall, all the velocity components are zero, and outside the boundary layer the velocity is the same as the free-stream velocity U_. A time-average velocity profile for the turbulent layer can be plotted that describes the behaviour of the mean velocity. Compared with the laminar case, the turbulent-velocity profile has a higher flow velocity near to the wall. The difference is explained by the fact that the transverse transport (transport in the y-direction) of momentum and vorticity in laminar flow is driven by the viscous shear stress in the fluid. In the turbulent case, the transverse transport is driven by convection, by the fluctuating

turbulent eddies vortices (or, in other words, by the turbulence itself of the circular motions).

This gives a higher velocity, momentum and friction near the wall in the turbulent case. The drag coefficient, Cd, is thereby also higher compared with the laminar case.

Figure 10. Turbulent boundary layer with time-average thickness, turbulent- and laminar-velocity profile sketches.

The Reynolds’ decomposition:

i i

i U u

u    (3)

u’

v’

y x

U



(20)

The flow in engines in not stationary. But to simplify calculations in engine flow, the velocity components can be defined as stationary for a small period of time (or for a small CAD window). The mean velocity for stationary flows is defined as [42]:



^



  ^t ^T

t i

i T udt

U T ⁰

0

lim 1 (4)

and the mean value of the fluctuating part (velocity) is zero by definition; and can be written as:

 



^

  

 ^t ^T

t i i

i T u U dt

u T ⁰

0

1 0

lim (5)

In turbulent flows, the convection dominates over the molecular diffusion. Turbulence has fast shifts in pressure and velocity. To describe the motion of the gas, two basic equations are required: the mass conservation and momentum equation. First, we have the mass conservation equation. If we think about a control volume, the mass in the control volume (

Mcv) is the mass transported into the volume minus the mass transported out of the volume:

 

^

 



in out

cv m m

t

M   (6)

The mass can also be described as velocity and density, which gives the mass conservation or the continuity equation [43]:

 

0





 

 U

t



(7)

where



 

 ³

1

i xi (8)

The momentum equation (Navier–Stokes equation) is based on Newton’s second law and relates the fluid particle acceleration to the surface forces and body forces. The Navier–

Stokes equation is defined as [43]:

U Dt p

DU 1 2

 (9)

The Navier–Stokes equation, together with the continuity equation, describes the conservation of mass, momentum and energy in a flow field. With the restriction of an incompressible flow field, there is no need for the energy equation to describe the flow. In compressible flow, like supersonic flow, or when heat transfer is involved, the energy equation cannot be ignored.

When a velocity difference exists between the fluid particles, initiated by fluid motion, shear stress is generated. When a particle with a mass is transported in a flow with different velocity, compared with the particle, the shear stress occurs and wants to accelerate or slow down the particle. The total mean shear stress for two dimensional flow, as showed in Figure 10, can be written as:

v y u

U   



 

 (10)

(21)

Where  uv is the turbulent stress (or Reynolds’ stress) and the other component is the viscous stress. Reynolds’ stress is an internal stress that acts on mean turbulence flow. The viscous stress is acting on the particle by the fluid viscosity.

When the turbulence dissipates, the large eddies lose their kinetic energy first to smaller- length scales and then to heat. The rate of conversion of turbulence into heat by molecular viscosity is called the dissipation rate, ε. At the smallest turbulent vortices, it is still the fluid viscosity that dissipates kinetic energy to heat.

Vorticity is a fluid parcel with a tendency to rotate around itself. If the vorticity is zero, the fluid parcel can still move in a curve, but it does not rotate around its own axis. Vorticity has the dimension of frequency (sec^-1).

z y

x e e

e

u 

 











 



 











 



 











 





 y

u x v x

w z u z

v y

 w (11)

Swirl and tumble flow

The airflows inside the cylinder are commonly characterised by swirl, tumble and turbulent intensity. Swirl and tumble are large-scale vortices that can exist inside the cylinder, one at a time or in combination. The vortices are created during the inlet stroke, when the piston moves down and inlet air passes over the inlet ports. Depending on the inlet port design, swirl and/or tumble vortices are created and conserved (or dissipate slowly) in the cylinder when the inlet ports are closed. Unlike small-scale vortices, these big-scale vortices do not dissipate so fast and survive a sufficient time and can thereby effect the combustion and after-oxidation. Friction against cylinder walls and in the airflow by itself makes the swirl flow to slowly dissipate, it is just a matter of time. Swirl is the angular velocity around the cylinder centre axis and tumble is the perpendicular velocity to the cylinder axis. Both variables can be normalised towards the engine crankshaft velocity and the dimensionless numbers, swirl number (SN) and tumble number (TN), can be defined as:

Engine Swirl

SN 

  (12)

Engine Tumble

TN 

 (13)

where

Swirl

 Air angular velocity around cylinder centre axis.

Tumble

 Air angular velocity perpendicular to the cylinder axis.

Engine

 Engine angular crankshaft velocity.

When both vortices exist, they are combined to create one big resulting vortex. Typically, swirl is used in DI diesel engines and tumble in SI engines. In diesel engines, the after- oxidation part of the combustion is of great importance. To reduce soot emissions, swirl has the opportunity to survive the compression and the combustion, and thereby affect the combustion in the latter part of the engine cycle. With a variable valve train it is possible to control SN and TN. Variable valve trains are in production in light-duty engines [20], [21], but

(22)

this is not yet common in heavy-duty diesels, although research and development is in progress [22]. Port designs on diesel engines have historically been very important [23], [24].

SN has been an important factor in good combustion and low smoke at moderate injection pressures. Recently, research on the injection system using higher injection pressure and EGR has shown significantly decreased emissions. Although SN has a demonstrable effect on emissions and combustion [25], it has not changed appreciably until the introduction of today’s high-injection pressures. Today, some manufacturers have nearly no swirl

(quiescent) in their combustion systems.

During the engine cycle, from that when swirl is created and the inlet valves are closed, the swirl changes in velocity during compression and combustion. During compression, the swirl velocity increase at late CAD when air flow is forced into the piston bowl. This “spin-up” is created when the conserved momentum in the airflow is forced into smaller radius. When the piston moves down again, the opposite happens. The flow is also slowing down due to the friction against the cylinder walls. But if the swirl flow is assumed to be constant for a small time window, the swirl vortex conservation can be observed if the continuity and momentum equation, mentioned earlier, is rewritten to polar-cylindrical coordinates for the mean flow.

The details can be seen in [44]. The tangential equation from the momentum equation describes the conservation of mean flow angular momentum in the flow. If a fluid is moved outside its normal track around the cylinder axis (increased radius), where it maintains its angular velocity in the swirl, it will have an angular momentum loss compared with the surrounding fluid particles. The centripetal forces acting on the fluid element are lower than the opposite net pressure force caused by the mean radial pressure gradient, with the result that the fluid is forced into its normal equilibrium orbit. This demonstrates why swirling flow in a cylinder can be stable and does not dissipate into smaller-scale turbulence as fast, as with other types of turbulent flows, when the energy source is closed. With this above-mentioned statement, it is easy to understand why swirling flows in an engine are often modelled as a solid-body rotation. Even if PIV measurements show that a perfect solid-body rotation does not exist [45], [46], this can be a decent assumption in modelling.

Tumble is of paramount importance for an SI engine to increase combustion velocity. The tumble vortex is transformed to small-scale turbulence around TDC, due to the geometric change of the combustion chamber during compression. Fuel and air are premixed before combustion, and a spark plug ignites the mixture. The flame front propagates through the premixed air/fuel. First, the flame propagation is laminar (around 0.3 m/s), but later it becomes turbulent (in the order of 10–80 m/s depending on for example engine speed), as turbulence exists in the cylinder [2]. At TDC, the tumble vortex is transformed to small-scale turbulence, which greatly influences the flame velocity. Without the turbulence, the flame velocity should remain around laminar combustion velocities with a low flame speed, and thereby unable to combust all of the mixture in the cylinder before the expansion stroke has been finished. Turbulence has many characterising elements, in this work turbulent intensity is used to characterise the small scale vortices. The turbulent intensity is normalized with the mean piston speed, and defined as:

Vp

NTI u´ (14)

(23)

 

^the^piston

 

^m/s

of ity Mean veloc

m/s intesity Turbulent

´



 Vp

u

Turbulence is also of great importance for diesel combustion. Compared with an SI engine, which has a long time to mix air and fuel, the diesel process is a non-premixed combustion.

The mixing is done while the combustion takes place (in the diffusion phase of the

combustion). The spray creates lots of turbulent flows, which increase the mixing process.

Higher injection pressure means increased turbulent kinetic energy.

Squish flow

A squish region is the flat area on the piston that is located on the outer radius. Typically, this area is designed to have a small clearance to the cylinder head, and in a diesel engine there are several reasons for this [47]. First of all, as much of the air trapped in the cylinder as possible should be located in the piston bowl cavity at TDC to contribute in the combustion when the fuel is injected. For a given compression ratio (higher compression ratio gives higher efficiency to a certain level), the heat transfer at TDC is high, and by decreasing the volume outside the bowl, the heat transfer is also lowered. During the late stage of the

compression, right before TDC, the volume that is above the squish area (the squish volume) rapidly decreases. The air trapped in this volume is forced towards the centre of the cylinder and creates a strong flow. This squish flow can thereby strongly affect the combustion, depending on when the fuel injection starts. The flow affects the swirl in the bowl and, depending on how strong the swirl flow is, the squish flow contributes in different ways.

According to [44] and [47], in a light-duty engine at moderate SN, the squish flow flows into the bowl horizontally and creates a rotating vortex in the bowl, as illustrated in Figure 11. At high SN, the squish flow is deflected from the horizontal track to follow the bowl geometry down into the piston cavity. This results in a change in the direction of the created vortex in the bowl.

(24)

Figure 11. Squish flow interaction with swirl flow in the piston bowl [44], numerical simulation of flow velocity at different SN. Flow velocity (Sp) is expressed in mean piston speed.

Cycle-to-cycle variations

All engines have cycle-to-cycle variations. It is normally coefficient of variation (CoV_IMEP) of indicated mean effective pressure (IMEP) that is the measuring variable. Definition:

 

avg N

i

avg i

IMEP IMEP N



IMEP





 ¹

2

IMEP

1

CoV (15)

where IMEPavg is the average of all measured cycles (N).

A rule of thumb is to keep CoV_IMEP bellow 3% on engines in vehicles to maintain a good drivability. When CoV_IMEP rises above this value, problems with efficiency and emissions typically arise as well. The main reason to cycle variations is based on the fluctuating nature of turbulent airflows inside the cylinder. Typically, CoV_IMEP is higher for SI engines compared with CI engines. The main reason is that the combustion velocity in an SI engine is

dependent on the in-cylinder turbulence. Small changes in the turbulence give a larger change in the combustion propagation velocity in a SI engine, because of the premixed combustion. Specially the ignition event and the first laminar flame propagation is effected due to λ stratification in the cylinder. In the CI engine, the fuelling is induced more or less during the combustion and the HR rate is thereby mainly controlled by how fast the fuel enters the combustion chamber. Still, the cycle-to-cycle variations of the airflow also influence the combustion for CI engines. In [26], an investigation of the cycle-to-cycle variations on SN was done using PIV measurements for two different SN settings. The engine tested was a 4-stroke SI engine with a 2-valve cross-flow cylinder head and the swirl was controlled either with a shrouded poppet valve with SN 6.0 and TN 2.0 (as determined in a steady-state flow bench) and a standard poppet valve with SN 0.7 and TN 2.5. At TDC, the

(25)

measured (with PIV) SN for the shrouded valve decreased to a mean of SN 5.2, and for the std valve it increased to a mean of SN 0.9. It was found that the low swirl case has more cycle-to-cycle variations in SN compared with the high-SN case. A stronger swirl vortex seems to be beneficial to have small-cycle variations in SN.

Emissions from diesel engines

The emission legislation for heavy-duty diesel engines has been strongly restricted since 1992, when the Euro I emission standards were set until the coming levels, Euro VI that is seen in Table 1. The most challenging emissions from a diesel engine are particulate matter (PM) and NOx. These emissions have decreased greatly to today’s “hard-to-measure” levels.

The reason for lowering the emissions is justified by the fact of how they affect humans and environment. Both soot and NOx irritate the respiratory organs [30]. The harmfulness of NOx was found by tyre manufacturers when ground-level ozone caused rubber tyres to fracture.

The NOx in combination with HC from combustion was found to be the reason for both smog and ozon. PM consist of Polycyclic aromatic hydrocarbons (PAH), solid coal and condensed hydrocarbons. If the particles are small enough, they pass the cilia and go into the lungs.

Exposure to ultra-fine particles may cause cardiovascular diseases [28]. PAH are considered to be highly carcinogenic. Carbon monoxide (CO) interferes with haemoglobin, which impairs the ability to pick up and transport oxygen. CO has 300 times higher affinity compared with the affinity of oxygen. This means that the haemoglobin takes up the CO instead of oxygen.

Table 1. Emission standards for HD diesel engines [27].

Emissions formation in a diesel engine

NOx emissions can be formed by four different mechanisms during combustion with air: the Zeldovich mechanism, the Fenimore mechanism, the nitrous oxide- (N2O) intermediate mechanism and the NNH mechanism [7]. The Zeldovich (or thermal) mechanism is the dominating mechanism at high-temperature combustion over a wide range of λ and is the mechanism that is most used to explain NOx formation from diesel engine combustion. The Fenimore mechanism is particularly important in rich combustion. Fenimore (also called prompt NO) discovered that NO was rapidly produced in a laminar premixed flame zone a long time before thermal NOx had time to form. The hydrocarbon radicals react with nitrogen and form amines or cyano compounds. These compounds are then converted to

(26)

intermediate compounds that, after a while, form NO. The N2O-intermediate mechanism has an important role at very lean, low-temperature combustion processes (typically gas turbines in lean operating conditions). Those three mechanisms can contribute in premixed and diffusion flames. The NNH mechanism is a relatively new discovered reaction pathway, and it seems to be particularly important in hydrogen combustion and methane combustion. NOx can also be created if the fuel by itself contains nitrogen, which typical diesel fuel have only in very low concentrations.

But as earlier mentioned, the Zeldovich mechanism is the dominating mechanism in diesel combustion. As it is called today, the extended Zeldovich [7] version has three reactions:

O+N2 ↔NO+N N+O₂↔NO+O N+OH↔NO+H

The first reaction has a very high activation energy due to strong triple bond in the N2 molecule. The Zeldovich mechanism needs four things to form NOx: heat, time, oxygen and nitrogen. If some of these components are reduced or removed, the NOx formation

decreases. The Fenimore mechanism is linked to the combustion chemistry of hydrocarbons.

HC are usually low from conventional diesel combustion. Some of the fuel that is injected is formed into different hydrocarbons that are not present in the fuel. For example, it is found small amounts of methane, formaldehyde and aromatics in the exhaust. The influence on the human body has been discussed earlier, so the understanding of how these emissions are created is of paramount importance. PAHs are formed under fuel-rich conditions, and they are important precursors in soot formation. Another source of HC comes from injection needle sack volume [2]. When the injection has ended, some fuel is left in the needle sack volume that is ventilated to the combustion chamber. During the expansion stroke, when the cylinder pressure decreases, the evaporating fuel that is left in the sack is evacuated into the cylinder. When the pressure and temperature decrease during the expansion, the HC is not combusted before the exhaust valve opens and evacuates the HC emissions. In Figure 12, a bright soot cloud is created during cylinder pressure drop at some CAD after EOI. At later CAD, when pressure and temperature decreases even more, the HC in the sack leaks out without forming soot around the injector tip. By reducing the sack volume, this emission can be reduced.

(27)

Figure 12. Optical engine picture of the combustion at the after-oxidation phase of combustion. The blue circle indicates an injection sack leak after the injection has ended.

PM occurs in incomplete, cold and/or rich combustion. PM is kinetically controlled. PM starts to grow when PAH molecules are conglomerated and start to form a particle. The particle then starts to grow by addition of acetylene and coagulation. In Figure 13, the different steps of PM formation are shown. The medicines used to reduce the soot emissions are the three

“t’s” of combustion: time, temperature and turbulence [32]. This means, just increase the temperature in the combustion and wait for a longer time with a good mixing, and the soot is no longer a problem. This can be done with earlier start of injection (SOI). But, as always,

“there is no free lunch”, and NOx starts to form. The reason for this is the longer residual time and higher cylinder temperature, which increases thermal NOx formation. This is the

classical NOx-soot trade-off problem that diesel engines struggle with. By reducing one of the emissions, the other emission starts to increase.

(28)

Figure 13. Schematic soot formation in combustion [32].

CO comes from incomplete and cold combustion in diesel engines. Increased EGR, late combustion facing, lack of oxygen and, as shown later in this work, too high swirl can increase the CO emissions. HC and CO can easily be treated in an oxidation catalyst

mounted after the engine. The after-treatment system for NOx and soot is more complex and expensive. If the engine-out emissions can be reduced, the after-treatment system can be made simpler or ignored. This is not an easy goal.

Engine transients

Fast engine load build up, called transient, is a challenge to keep the exhaust emissions at legal levels. In Figure 14, a transient load build up is plotted in forms of IMEP and inlet plenum pressure at 1,000 rpm for a turbo diesel engine. The requested load, the blue line, is the desired value and the red line shows the actual load on the engine – and these values differ. The problem is that the air supply from the turbocharger unit, the green line, needs some time to build up the boost pressure. The amount of fuel needs to be restricted during the boost pressure build up before it is at a normal operating level. If the fuel mass is not restricted, the in-cylinder λ will go under a critical level and the engine will start to produce high smoke levels. In the engine, electronic control unit (ECU) functionality to detect and control low λ operation is constantly developed. Different models are implemented in the ECU to predict the oxygen trapped in the cylinder, allowing the right amount of fuel to be added for each cycle during the transient. As shown in examples of control system models [33], [34], [35], [36], the λ cannot pass below a critical value, typically  1.25–1.30 for an engine without a diesel particle filter, before smoke emissions increase rapidly. Besides a disappointed driver, who wants engine power as fast as possible, there is also a longer time period where the engine operates at unfavourable conditions. As long as the air is restricted, an EGR engine cannot use EGR (as the turbo pressure is too low). The in-cylinder mean

(29)

temperature increases due to less gas mass being kept in the cylinder, with higher NOx emissions as a result. The engine efficiency is also negatively affected when the SOI needs to be retarded to maintain NOx levels at acceptable levels.

If the combustion system can maintain low emissions of PM and NOx at a lower λ, the

available exhaust energy and engine torque will be increased. This results in a faster build up of boost pressure. Also, it can be traded against lower emissions during the transient.

Figure 14. Turbo diesel engine transient from 3-bar IMEP to full load, in this case, 23-bar IMEP.

Requested load, engine-out load and inlet pressure plotted versus time.

In the following, the tests that have been performed in this work and the attached

papers/journal are based on this engine transient. Load points are chosen from this line and repeated in a single-cylinder engine and an optical engine with the same boundary

conditions. The only difference is that the respective load point is repeated at steady-state conditions in the laboratory engines where the transient phenomena can be observed in detail. The advantage is that relatively fast load build up can be examined and changes in;

for example, in-cylinder flow and injection parameters can be applied and studied in a controlled way.

1 1.3 1.6 1.9 2.2 2.5

0 5 10 15 20 25

-1 0 1 2 3 4 5

Inlet pressure [bar]

IMEP [bar]

Time [s]

Req. Load IMEP [bar]

IMEP engine out [bar]

Inlet pressure [bar]

Load 1

Load 3

Load 2

Load 4

(30)

Project motivation

The main project focus is on how airflow in the cylinder affects combustion and the emissions from a diffusion flame diesel engine at transient operation. As explained above, emissions from an engine transient become increasingly important. During a transient, the emissions increase. By creating a better understanding of the processes, the combustion can be controlled in a better way during the transient. Ignition delay, diffusion combustion and after- oxidation period are the main stages in diesel combustion. The after-oxidation period has been shown by other work [37], [38] to be of paramount importance for engine-out emissions.

During after-oxidation, the mixing process is of paramount importance. Optical engine measurements on flow structure during the after-oxidation part of combustion can give important knowledge on the flow structure that affects the emissions from the engine.

The main objectives of this work are to:

1. Demonstrate the potential emission reduction when variable swirl and tumble is applied to the combustion system at high-injection pressure.

2. Quantify the airflows in the cylinder before and during combustion. This will be done with 1-D simulations and experimentally.

3. Look into the after-oxidation part of combustion and understand the effects on this when injection pressure and swirl are changed.

Method

Fast load increase means low λ operation with less remaining oxygen for the after-oxidation of the fuel. The demands on the mixing process in the cylinder are therefore higher. The method for this work is to first measure what is happening during an engine transient, examine where the critical points are and repeat them in a controlled way with different in- cylinder airflow. The repetition is done in a single-cylinder engine with AVT, which is opened up for variable in-cylinder airflow. Simulation tool GT-POWER, together with constant flow-rig measurements, are used to quantify the airflow before combustion. To create an

understanding on how the flow field influences the combustion, optical engine measurements are taken using a high-speed camera. The flow inside the cylinder during combustion is quantified with optical calculations in a cross-correlation program. Simulated, measured and processed data on flow quantities, emissions and combustion data is combined to examine the airflow effect on diesel combustion during transient.

(31)

Methodology

The experimental equipment and simulation tools used in this work can be seen in Figure 15.

Data from three different engines was used: a six-cylinder turbocharged production engine 1), a single-cylinder engine 6) and an optical engine 9). The six-cylinder engine was a Scania DC13 Euro 5, equipped with EGR. The engine was used to measure the transient behaviour and the data provided 2) was used to set up a single-cylinder and optical-engine test series.

The single-cylinder engine used in the tests was equipped with a Lotus AVT system 7). With this system, the valves were hydraulically controlled, enabling the valve profiles 4) to be shifted during operation. With different valve profiles, the airflow in the cylinder can be affected over a wide range. The airflow was quantified as SN, TN and normalised turbulent intensity (NTI). The 1-D simulation program, GT-POWER, 5) was used to calculate these quantities. The two tested cylinder heads were measured in a constant flow rig 3). Swirl and tumble at valve lifts from 1 to 15 mm, with increments of 1 mm, were measured for each valve individually and with the two valves together. Valve profiles 4) were created in MATLAB. These were then used in the Lotus AVT system and in GT-POWER. In this way, the SN, TN and NTI were calculated and then used to plot, for example, emissions measured in the single-cylinder engine, as a function of airflow in the cylinder 8). The tested load points in the single-cylinder engine were then repeated in the optical engine 9). The combustion pictures 10) was captured with a high-speed Phantom camera. The pictures were then evaluated using LaVision DaVis 7.2 PIV software 11). Velocity vector fields were calculated 12) and CAD resolved during the injection and after-oxidation part of the combustion. The results from single-cylinder tests were put together with the velocity vector figures calculated from the high-speed combustion film. Data evaluation was then done from this set of data.

(32)

Figure 15. Overview of used experimental equipment and calculation tools.

Test equipment

The three different test engines used in this work have the same, or nearly the same,

combustion system layout, presented in Table 2 and Table 3. Some deviation was, of course, inevitable, due to the different engine constructions, as seen in Table 3. The engine layout was based on a Scania DL Euro 5 combustion system equipped with a common rail XPI injection system capable of injection pressures up to 2,500 bar. This is a swirl-supported combustion system with a re-entry combustion bowl design that does not need any extra after-treatment system for Euro V emission legislation when EGR is used. It has a 4-valve cylinder head with a centrally placed 8-hole injector.

80°70°60°50°40°30°20°10°0°-10°-20°-30°-40°-50°-60°-70°

-4 -2 0 2 4 6 8

1471013 Angle on cylinder head

Tumble value

Valve lift [mm]

One valve tumble 6-8

4-6 2-4 0-2 -2-0 -4--2

300 350 400 450 500 550

0 2 4 6 8 10 12 14

CAD

Valve lift [mm]

15 mm std 10 mm std 5 mm std 15 mm step 10 mm step 5 mm step

1 1,3 1,6 1,9 2,2 2,5

0 5 10 15 20 25

-1 0 1 2 3 4 5

Inlet pressure [bar]

IMEP [bar]

Time [s]

Req. Load IMEP [bar]

IMEP engine out [bar]

Inlet pressure [bar]

Load 1

Load 3

Load 2

Swirl

Tumble

Injection pressure 1000 bar, load 1b

0 1 2 3 4 5 6 7

0.5 1 1.5 2 2.5 3 3.5

4 Smoke [FSN]

one valve two valves

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Flow - bench

GT-POWER

Single cyl

Lotus AVT

PIV DaVis

Emissions data combined with flow data

Combine data

300 350 400 450 500 550

0 2 4 6 8 10 12 14

CAD

Valve lift [mm]

1

2 3 4

5 6 7

8

9 10 ¹¹

12

(33)

Table 2. Engine specifications.

Compression ratio 17.3:1

No. of valves 4

Injection system Scania common rail XPI

Injector holes 8

Spray angle [deg]

(° between cyl.head and spray)

Injector hole diameter (inner/outer) [mm]

Max Injection pressure [bar] 2500 Common test engine data

16

0.187 / 0.163

The difference between the tested engines can be seen in Table 3. The fist obvious difference is that bore and stroke differs slightly between the engines. The reason is that 127/154 mm is the old engine configuration and 130/160 mm is the new configuration. The optical engine was updated with the new bore but not the new stroke. The combustion system (injection system, bowl geometry, cylinder head design and so on) is the same for all test engines.

Table 3. Test engine individual specifications.

Engine type Optical engine Scania single cylinder

Scania DL Euro 5 engine

Bore/stroke [mm] 130/154 127/154 130/160

Connecting rod [mm] 255 255 255

Valve system Camshaft Active valve train Camshaft

Single-cylinder engine with an AVT system

The single-cylinder engine was equipped with a Lotus AVT system that is a fully hydraulic system. One hydraulic cylinder was coupled to each valve in the cylinder head, as seen in Figure 16. Hydraulic oil, with a pressure of approximately 200 bar, was supplied to each side of the piston inside the hydraulic cylinder, making the piston move. The oil flow was

controlled by a servo valve that directs the oil to one side of the piston at a time. In this way, the engine valves are controlled with a good level of accuracy.

(34)

Figure 16. The Lotus AVT valve actuator with inlet valve (a) and the actuators on the cylinder head (b) Different valve profiles can be implemented in the AVT system. They are created in MATLAB and then uploaded to the system. During engine operation it is then possible to change the valve profiles. Some examples of valve profiles used in this work are seen in Figure 17. The system enables SN variations between 0.4 to 6.7 and TN from 0.5 to 4.0 with the used cylinder head configuration.

Figure 17. The tested valve profiles. The dark-blue profile is the standard valve profile used in a standard engine.

Optical engine

The optical engine layout can be seen in Figure 18, with the two different tested piston bowls.

On the original piston, a piston extension is mounted that leads to the optical piston and the liner that it is fitted into. The camera, Phantom v7.3, is installed next to the engine and the combustion light is transferred to the camera by a mirror mounted inside the piston

extension. The engine is capable of running with cylinder pressures up to 160 bar. A titanium clamping ring was mounted above the piston glass to fix it. This restricted the field of view to a diameter of 80 mm, compared with the total cylinder bore of 130 mm. Two different shapes

300 350 400 450 500 550

0 2 4 6 8 10 12 14

CAD

Valve lift [mm]

b

a

(35)

on the piston bowl glass were tested, bowl-shaped piston and flat piston bowl. A schematic spray path is plotted, Figure 18, in the two piston bowls. The optical engine had a normal camshaft valve mechanism. To change the in-cylinder airflow, two different cylinder heads were used. To further extend the possible airflow in the cylinder of the optical engine, one of the inlet ports was also blocked (or not).

Figure 18. Principal layout of the optical engine on the left. On the right, the two tested piston bowl shapes with plotted schematic spray.

Due to the long piston extension, the effective compression ratio is lower than the geometrical compression ratio and decreases with cylinder pressure. At 160 bar, the distance between cylinder head and squish area on the piston increased by 1.5 mm compared with atmospheric pressure. To compensate for the lower compression ratio, the boost pressure and inlet temperature was increased so the motoring cylinder pressure at TDC in the optical engine was equal to the single-cylinder engine. The λ was slightly higher in the optical engine compared with the single-cylinder engine. The increase in inlet

temperature was also done to compensate the increased ignition delay in the optical engine, since only one combustion event was performed during the measurement.

Steady-state flow rig

A honeycomb type steady-state flow rig was used to generate swirl and tumble data for the GT-POWER simulations. In Figure 19, a sketch of the flow rig can be seen with the most important parts marked with a number. 1) the tested cylinder head is mounted on a cylinder liner 2) with an inner diameter equal to the engine bore. A honeycomb torque meter 3) is mounted in the stagnation chamber 4), which measures the torque that is created by the swirling flow from the cylinder head. The pressure difference between the stagnation

chamber and the atmosphere is created by the fan 8) and monitored by 5) and kept constant for all measurements at 25 mbar. To measure the airflow that is passing the cylinder head, a roots blower is mounted in 6), which works as an airflow meter. By monitoring the pressure

Flow measurements using combustion image velocimetry in diesel engines