You are on page 1of 10
Alexandria Engineering Journal (2016) xxx , xxx – xxx HOSTED BY Alexandria University Alexandria Engineering

HOSTED BY

Engineering Journal (2016) xxx , xxx – xxx HOSTED BY Alexandria University Alexandria Engineering Journal

Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej

www.elsevier.com/locate/aej www.sciencedirect.com ORIGINAL ARTICLE Application of zero-dimensional
www.elsevier.com/locate/aej www.sciencedirect.com ORIGINAL ARTICLE Application of zero-dimensional

ORIGINAL ARTICLE

Application of zero-dimensional thermodynamic model for predicting combustion parameters of CI engine fuelled with biodiesel-diesel blends

Hariram V. a , * , Bharathwaaj R. b

a Department of Mechanical Engineering, Hindustan Institute of Technology & Science, Hindustan University, Chennai, Tamil Nadu, India b Department of Mechanical Engineering, S.A. Engineering college, Chennai, India

Received 30 March 2016; revised 10 August 2016; accepted 25 August 2016

KEYWORDS

Zero-dimensional model; Combustion; Heat release; Rate of pressure rise

Abstract Biodiesel from non-edible vegetable oil seems to be a promising alternate for petro-diesel in the present energy scenario. This study analyses the experimental and theoretical effects on the blends of Bee Wax biodiesel with straight diesel on combustion parameters. A zero-dimensional mathematical model is developed to analyse the rise in in-cylinder pressure along with Wiebie’s heat release correlations, ignition delay, gas dynamics model, heat transfer model and frictional model. The combustion parameters include in-cylinder pressure rise, net heat release and rate of pressure rise are investigated and found to be higher for straight diesel and deteriorated with the increase in blends of BWB. The theoretical simulation also supports the experimental data with constant injec- tion timing, speed and compression ratio.

2016 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Honey and beekeeping have a long history in India with oldest records in the form of paintings by prehistoric man in the rock shelters. With the development of civilization, honey acquired a unique status in the lives of human beings. The recent past has witnessed a revival of the industry in the rich forest regions along the sub-Himalayan mountain ranges and the Western Ghats, where it has been practiced in its simplest form. Several natural plant species provide nectar and pollen to honey bees.

Thus, the raw material for production of honey is available free from nature. Beehives neither demand additional land space nor do compete with agriculture or animal husbandry for any input. Beekeeping constitutes a resource of sustainable income generation to the rural and tribal farmers providing valuable nutrition in the form of honey, protein rich pollen, brood and beehive wax. During the 1980s, an estimated one million beehives had been functioning under various schemes of the Khadi and Vil- lage Industries Commission with the development of apicul- ture using the indigenous bee, Apis cerana, Apis mellifera, which gained popularity in Jammu & Kashmir, Punjab, Hima- chal Pradesh, Haryana, Uttar Pradesh, Bihar and West Ben- gal. Wild honey bee colonies of the giant honey bee and the oriental beehive have also been exploited for collection of

* Corresponding author. E-mail address: connect2hariram@gmail.com (V. Hariram). Peer review under responsibility of Faculty of Engineering, Alexandria University.

1110-0168 2016 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: V. Hariram, R. Bharathwaaj, Application of zero-dimensional thermodynamic model for predicting combustion parameters of CI engine fuelled with biodiesel-diesel blends, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.08.021

2 V. Hariram, R. Bharathwaaj Nomenclature h crank angle (in ) Cr compression ratio Crl

2

V. Hariram, R. Bharathwaaj

Nomenclature

h

crank angle (in )

Cr

compression ratio

Crl

connecting rod length

V swept

swept volume (in mm 3 )

BDC

bottom dead centre

TDC

top dead centre

D

bore diameter (in mm)

l

stroke length (in mm)

DI

direct injection

C–H–O

carbon hydrogen oxygen equilibrium system

FTT

finite temperature thermodynamics model

bTDC

before top dead centre

EGT

exhaust gas temperature in degree celsius

U

internal energy (in joules)

W

work done (in kW)

C v

x

t

i

t

g

specific heat at constant volume (in kJ/kg K)

mass fraction starting time of injection (in s) starting of combustion (in s) stagnation pressure (in bar)

specific heat ratios for gasses coefficient of heat transfer (in W/m 2 )

velocity of gasses (in m/s) mean piston speed (in mm/s)

p

k

h

t

V p

BWB 10 beeswax biodiesel 10% straight diesel 90%

BWB 20 beeswax biodiesel 20% straight diesel 80%

o

c

CA

crank angle (in )

HRR

heat release rate (in J/CA)

honey. Both natural and cultivated vegetations in India consti- tute an immense potential for development of beekeeping. About 500 flowering plant species have been identified as major and minor sources of nectar and pollen. In the recent years the exotic honey bee has been introduced representing a wide variety of bee fauna that can be utilized for the devel- opment of honey industry in the country. There are several types of indigenous and traditional hives including logs, clay pots, wall niches, baskets and boxes of different sizes and shapes. India, potentially produces about 120 million bee colo- nies discharging over 1.2 million tons of honey and 15,000 tons of beeswax. Organized collection of forest honey and beeswax using improved methods results in an additional production of at least 120,000 tons of honey and 10,000 tons of beeswax [22]. Internal combustion engines, mainly compression ignition engines are used in transportation industries. They are also used in power production and agricultural applications as well. CI engines operate at relatively higher efficiency when com- pared with spark ignition engines, due to their higher compres- sion ratio. Currently, investigations are carried out to reduce the diesel emission and provide cleaner and more efficient engine systems. Biodiesels will be a key in achieving the above goal and thermodynamic simulation of the internal combus- tion engines providing enhanced knowledge to understand and configure new engines technologies. Simulation is the process of creating a virtual system to understand the properties of real time system without physical interventions. The development in the field of computers has opened the doors of computational environment. Diesel engine simulation models are classified into three categories, namely zero-dimensional single zone models, quasi-dimensional multi zone models and multi-dimensional models in which zero- dimensional models provide the essence of thermodynamic property of a system modelled. On the other hand, the multi-dimensional models resolve the space of modelled system on a fine grid thus providing special information. However, computational time and coding constraints keep this model away from consistently using them in-design optimization pro- cess. Combustion process of compression ignition engine is a complicated phenomenon based on thermodynamic models that incorporate the energy equation for closed system and

species conservation equation that predicts the heat release pattern and pressure of the system. Working fluid undergoes a continuous change in its thermodynamic state through the series of the engine operational cycles which mainly consists of constant pressure suction, adiabatic compression, constant pressure combustion, isentropic expansion and constant pres- sure exhaust events where the thermodynamic states are to be computed with the help of programs for diesel operating cycle. Many researchers conducted experiments based on mathe- matically modelling of engine fuelled with bio-diesel for pre- dicting the performance and combustion parameters. A few of them are discussed below. Ramadhas et al. studied the the- oretical model for rubber seed oil methyl ester fuelled engine and analysed effect of relative air fuel ratio and compression ratio on the engine performance for different fuels. They used single zone thermodynamic model for predicting the perfor- mance of compression ignition engine by considering funda- mental assumption about cylinder charge, specific heats to solve energy equation and also Hohenberg equation in heat transfer [1] . Raut theoretically modelled a compression ignition engine depending on the characteristics of fuel using thermodynamic model for predicting the properties of working substance in suction, compression, power and exhaust strokes during the engine operation. The model predicted the performance of compression ignition engine in terms of brake power and brake thermal efficiency for all fuels and also calculated the volume as a function of crank angle by taking cylinder bore, stroke length of the engine and length of the connecting rod. The author used Wiebie heat release model to analyse the heat release pattern of the engine and Pflaum equation to predict the heat transfer coefficient and heat lost to the walls of the engine [14] . Rakopoulos et al. elaborated the two zone models for DI diesel engine with the use of mass, energy and state equations for determining the local temperature and pressure. Also, the author applied chemical equilibrium scheme for C– H–O system to predict the exhaust gas composition and used simulation model for various loads and compression ratios along with Annand’s equation for determining the heat trans- fer coefficient. The authors have also elaborated the use of fuel

Application of zero-dimensional thermodynamic model 3 spray model, air entertainment model for two zone application

Application of zero-dimensional thermodynamic model

3

spray model, air entertainment model for two zone application and sensitivity analysis discussing their procedure [5]. Gonica et al. reported the effect of engine load and biodie- sel concentration on the performance of diesel engine running with diesel-biodiesel blends using theoretical model based finite temperature thermodynamic (FTT) approach which was based on temperature dependent variable specific heats. The applica- tion of FTT was intensely discussed for compression ignition engine and exhibited a linear increase in power for varying loads at constant engine operating parameters [11] . Gogoi and Baruah simulated a single zone combustion model for pre- dicting the performance of CI engine fuelled with biodiesel blends. They analysed the effect of compression ratio and engine speed on BP and BTE with Karanja biodiesel blends. The model prediction resulted in exhibiting similar trend for biodiesel concentration up to 50% and showed better perfor- mance with 60% blend ratio [21] . Canakci, experimentally investigated the influence of using biodiesel blends on performance, combustion and emission characteristics in compression ignition engines in which, the brake specific fuel consumption increased with the addition of biodiesel on comparison with straight diesel [15] . Patil devel- oped a thermodynamic model for predicting the performance of CI engine fuelled with diesel and POME biodiesel blends using MATLAB and studied its empirical simulation [19] . Sivalingam et al. validated the combustion and emission parameters of direct injection CI engine fuelled with blends of bio-ethanol emulsion. In-cylinder pressure, ignition delay, NO x and smoke were determined and compared with mathe- matically simulated results at all loads. The results exhibited a marginal 5% variation between the experimental and theo- retical data with diesel fuel and 3% variation with biodiesel operations.

In this present work, zero dimensional thermodynamic model was developed to analyse the in-cylinder pressure, rate of heat release and rate of pressure rise to understand the com- bustion behaviour of the DI CI engine and the same was val- idated using biodiesel obtained from wax esters of Apis melifera (honey bee).

2. Materials and methods

In the present experimental investigation, bio-diesel was obtained from wax esters of Apis melifera (honey bee). Since honey bees are grown commercially for the production of honey, the source of raw material is sustainable. In the process of extracting wax from the beehive, honey bees were not killed leaving the ecological balance undisturbed. Commercially, honey bees are domesticated for honey and wax is obtained as the by-product of domestication process which makes it more reliable and renewable. The test engine used for this experimental investigation is direct injection compression igni- tion Kirloskar 240PE test engine equipped with eddy current dynamometer having 87.5 mm and 110 mm as bore and stroke length respectively with capacity of 661 cc that delivers 3.5 kW at 1500 rpm. The schematic diagram of the test rig is shown in Fig. 1 with pressure, temperature, crank angle, airflow and fuel flow measuring instruments and the test engine specification is given in Table 1. A high speed data acquisition system (DAS) is used in the experimental setup to transmit and analyse the data. Many ways have been devised to extract, refine and lighten the colour of beeswax. A widely preferred method is pressur- ized extraction method in which the beeswax is maintained at 90 C under water at pressurized condition to melt the

at 90 C under water at pressurized condition to melt the Figure 1 Schematic diagram of

Figure 1 Schematic diagram of experimental setup.

4 V. Hariram, R. Bharathwaaj Table 1 Engine specifications. Table 3 Composition of beeswax. Description

4

V. Hariram, R. Bharathwaaj

Table 1 Engine specifications.

Table 3 Composition of beeswax.

Description

Specifications

Composition of beeswax

In %

Make Model Bore diameter Stroke length Rated power Injection timing Connecting rod length Capacity Speed Compression ratio Loading type Load cell Pressure sensor Temperature sensor

Kirloskar 240PE/four stroke 87.5 mm 110 mm 3.5 kW@1500 rpm 23 bTDC 230 mm 661 cc 1500 rpm

Acid esters

4

Di-esters

6

Tri-esters

3

Free alcohols

1

Cerotic acid

4.4

Lacceryl palmitate

2

Lignoric acid

1

Melissic acid

2

Montanic acid

2.6

17.5:1

Myricyl palmitate

23

Eddy current dynamometer 0–50 kg, stain gauge type 0–350 bar K Type, PT 100, RTD thermocouple

Myricyl cerotate

12

Myricyl hypogaeate

12

Psyllic acid

1.5

Hydrocarbons

13.5

wax into liquid form. Acidic medium was prepared by adding 5% sulphuric acid to water containing processed beeswax to lighten the colour and increase the yield of liquid beeswax called slumgum. 5% of hydrogen peroxide was mixed to the slumgum to avoid bleaching action and later recovered. The biodiesel from beeswax was obtained by two stage trans-esterification process. The primary stage comprises of acid catalysed esterification with the molar ratio of 1:2:0.04 of wax, methanol and concentrated hydrochloric acid for reducing the free fatty acid content from 14.5% to >2%. Base catalysed esterification with methanol and sodium hydroxide was followed with treated wax ester. The yield of biodiesel was 85% at molar ratio of 1:0.25, reaction temperature of 60 C, for NaOH concentration of 0.02% by weight. The prop- erties of biodiesel were studied with respect to diesel and tab- ulated in Table 2 and the composition of Beeswax biodiesel is given in Table 3. BWB showed higher kinematic viscosity at 40 C, which was determined to be 3.5 mm 2 /s whereas for straight diesel, the kinematic viscosity was noticed to be 2.6 mm 2 /s. However, BWB-Diesel blends showed an increase in kinematic viscosity (i.e.) BWB10 and BWB20 exhibited 2.69 mm 2 /s and 2.78 mm 2 /s respectively. The calorific value of BWB, BWB10 and BWB20 also showed a similar trend with variations between 44.8 MJ/kg and 44.1 MJ/kg for blends of BWB. Density at 15 C of the BWB was found to be 880 kg/m 3 which is higher by 2.3% than straight diesel and the increase in blend ratio resulted in grad- ual increase in density of the fuel. Cetane number of the bio- diesel was determined using molecular weight from Gas

Chromatography Mass Spectrometry analysis and by employ- ing the Iodine Value and Saponification value relations

respectively, where D is the number

of double bond, A i is mass composition from GC/MS result and MW i is the molecular weight. Cetane number of the BWB was found to be 48 whereas the value for the straight die- sel lies between 46 and 53. When the BWB was blended with diesel, the Cetane number was found to be lying between 50 and 49 for BWB10 and BWB20 respectively. The cloud and pour point of beeswax biodiesel was found to be 12 C and 15 C which was lower on comparison with straight diesel. At lower temperature, especially in northern parts of India, clogging of fuel filter and fuel injector takes place which can be overcome by installing fuel line heaters as well as by adding pour point additive such as ethylene glycol up to 1.5% of the biodiesel by volume.

P 254 DXA i and P 560 A i

MW i

MW i

2.1. Mathematical modelling

2.1.1. Engine kinematics model

The fundamental geometry of reciprocating piston engine is shown in Fig. 2. Using the engine kinematics model volume every crank angle can be calculated by solving [20]

Vð hÞ ¼ V swept

V swept ¼ p

4 D 2 l

2

4

Cr Cr 1

1 cos 2h

2

1

þ 2

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3

s

2

2 Crl

l

sin 2 h 5 ð 1Þ

ð 2Þ

Table 2 Physio-chemical properties of straight diesel and biodiesel blends.

Property

ASTM standards

Diesel

BWB

BWB 10

BWB20

Kinematic viscosity at 40 C (mm 2 /s) Cetane number Calorific value (MJ/kg) Density at 15 C (kg/m 3 ) Flash point ( C) Carbon residue (%) Acid value, mg KOH Iodine value Cloud point ( C) Pour point ( C)

3.5–5

2.6

3.5

2.69

2.78

51

48

50.7

50.4

45.5

38.5

44.8

44.1

860–900

860

880

862

864

45

110

57

63

0.17

0.04

<0.8

0.35

0.4

121

34

12

31

29

40

15

37

35

Application of zero-dimensional thermodynamic model 5 Figure 2 Engine kinematics. where V s w e

Application of zero-dimensional thermodynamic model

5

Application of zero-dimensional thermodynamic model 5 Figure 2 Engine kinematics. where V s w e p

Figure 2 Engine kinematics.

where V swept is volume displaced by engine when piston moves from BDC to TDC in mm 3 , D represents cylinder bore diam- eter in mm, while l is stroke length of the engine in mm, Crl is length of connecting rod in mm. Cr represents the compression ratio, Crl represents the length of connecting rod in mm and h is crank angle in radians. The first derivative of Eq. (1) gives the change in cylinder volume for change in crank angle.

dV ¼ V swept

dh

2

2

6

4

3

7

q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sin h 5

1

sin 2h

2

2 Crl

l

2 sin 2 h

ð 3Þ

2.1.2. Zero dimensional combustion model

The thermodynamic or fluid dynamic model was developed based on energy conservation or complete analysis of fluid motion in the cylinder predicting the performance and com- bustion characteristics of the engine. Zero dimensional model predicts and analyses the characteristics of thermodynamic properties of the engine solving energy equations whereas multi-dimensional momentum conservation equations allow to visualize the gas flow and combustion products. Therefore, Zero dimensional model is suitable for observing the effects of variation in heat release rate and in-cylinder pressure parame- ters of engine operation. The zero-dimensional model [2,19] used in the study was based on the first law of thermodynamics applied to a closed cycle for duration when both the inlet and exhaust valves are closed (i.e. for the period of combustion). The energy equation for above process is written as

dQ c dQ h ¼ dU þ dW

dh

dh

dh

dh

ð 4Þ

where dQ c denotes the heat generated by the fuel due to com-

d

h

bustion, dQ h represents heat transferred to walls by the means of in cylinder gasses. dU describes the change in internal energy of the system while dW gives the rate of work transferred from the system. The above equation signifies the net heat release rate which is defined as difference between heat generated by the fuel due to combustion and heat that is convicted to the cylinder walls due to in cylinder gasses. Eq. (4) is rearranged as

dQ c dQ h

5Þ

d

h

dh

d

h

dT

dh

þ p dV dh

dh

dh

¼ MC v

ð

The rate of change of internal energy of the system is a function of temperature. Here, M , C v , T, p and V are in- cylinder instantaneous mass, specific heat of gases at constant volume, instantaneous temperature, in-cylinder pressure at the given instant of crank angle and instantaneous volume respectively.

2.1.3. Heat release model

In this present study, the combustion model was developed based on equally distributed heat release phenomena (i.e. heat is liberated evenly from all the areas of the combustion cham- ber). Since single zone combustion model predominantly give the accurate results in case of constant pressure combustion, Wiebie’s function was employed to determine the mass frac- tion of the fuel burnt and which in turn applied for finding the heat release rate due to combustion of fuel. Wiebie co-relation for predicting heat release rate is expressed as [13]

x b ð hÞ ¼ 1 exp

"

a

h h o Dh

mþ1

#

ð 6Þ

where x b denotes mass fraction of fuel burnt at given crank angle h, Dh represents the duration of combustion, ‘‘a is a parameter that symbolizes the completion of combustion. Wie- bie assumed x max as 0.99; hence, value of ‘‘ ais taken as 6.908. ‘‘m is the parameter that delineates the rate of combustion.

When Eq. (6) is differentiated once with respect to x and mul- tiplied with Q av , gives the relation for determining the heat release rate as [16]

Dh

dQ c

dh

¼ að m þ 1Þ

Q

av

h h o

Dh

D

h

m

"

exp a

h h o

D

h

mþ1

#

ð 7Þ

Here Q av is average heat release rate.

2.1.4. Ignition delay model

Ignition delay is termed as the time difference between start of injection of fuel and start of combustion. Physical delay and chemical delay are two broad classifications of delay period that occurs in CI combustion. Ignition delay was calculated as a function of air fuel mixture temperature, pressure that accounts for physical delay and air-fuel equivalence ratio, fuel properties that characterize the chemical delay. In actual com- bustion, ignition delay plays a major role in heat release rates. The empirical correlation developed by Wolfer [3,17] given in Eq. (8) is integrated by Runge-Kutta method to determine ignition delay period.

Z t g

dt dð p; TÞ ¼

1

K t i Z t g

dt

t

i

pð tÞ

a

exp

A

RTð tÞ

¼ 1

ð 8Þ

t

i

The constant values relevant for direct injection diesel engine are K = 2272, a = 1.19, A/ R = 4650. Where t i is time when injection starts, t g time when combustion starts. K is thermal conductivity, A is activation energy of fuel and R is universal gas constant.

2.1.5. Gas exchange modelling

It is equally necessary to model the intake and exhaust process to push the results towards the actual values. So, intake and exhaust processes are considered as a function of cylinder pres- sure and mass flow rate in and out of the engine. By consider- ing the gas as ideal and applying the energy equation for exhaust process,

dp

dt ¼ k p p N dt

1 dN

1

dV

e

V dt

ð

9Þ

6 V. Hariram, R. Bharathwaaj Similarly for intake process, dp 10 Þ dt Here d

6

V. Hariram, R. Bharathwaaj

Similarly for intake process,

dp

10Þ

dt

Here dN is mass flow rate through the valves which is selected by knowing the critical pressure ratio which is the function of specific heat ratios ( k ) of the gasses alone. It is given by Eq. (11) .

¼ k r

1 dN

1

dV

i

N dt

dt

V dt

ð

p c ¼

k þ 1

2

ð

k

k

1

Þ

ð 11Þ

Case (i): when pressure ratio p/p o is less than p c , the flow is subsonic and the mass flow rate is given by

dN

dt

dN

dt

¼

A v p

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

v

u

u

t

k

2

RT ð k 1Þ k þ 1

k 1

k

"

p

p

o

k

1

k

1

#

ð

12Þ

Case (ii): when pressure ratio p /p o is greater than p c , the flow is supersonic for this criteria mass flow rate is given by

¼

A v p

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

s

k

2

RT k þ 1

k

k

þ 1

1

ð

13Þ

Here ‘‘ k is specific heat ratios and subscripts p and r denote products and reactants respectively when applied for either cases of inlet and exhaust process. A v is effective valve flow area which depends on the valve lift profile of an engine. p o is stagnation pressure for that particular temperature and p is static pressure in the manifold [7–9] .

2.1.6. Heat transfer model

Since considerable amount of heat is transferred to the atmo- sphere by exhaust gases and cooling water by means of convec- tion, net work done by the working fluid for entire cycle is

W n ¼ I

p þ Dp DV

2

ð 14Þ

where W n indicates the net work output of the engine, D V is change in volume and Dp is change in pressure due to heat transfer process and it is given by [12]

Dp ¼ h c A s ð T s T f Þ

ð 15Þ

p

f C v T f

m

DT

where A s is surface area of engine, T s is surface temperature, T f , m f , and C v are temperature, mass and specific heat ratio of working fluid respectively, and h c is heat transfer coefficient and is determined by any of the following methods.

Ashley-Campbell equation : h c ¼ 0: 3D 0: 2 p 0 : 8 # 0 :8

T 0 : 5

ð 16Þ

where D is bore diameter, p is pressure, # is velocity of gasses and T is temperature.

Sitkei and Ramanaiah equation : h c ¼ 0: 014ð 1 þ bÞ p 0: 7 V 0: 8

p

T 0: 5 V

A

ð 17Þ

where b = 0.05–0.10, V p is mean piston speed, and A is area of bore.

Hohenberg equation : h c ¼ 130p 0: 8 ð V p þ 1: 4Þ 0: 8

T 0 :5 V 0 : 06

ð

18Þ

Pflaum equation : h c ¼

p

ffiffiffi

p T f ð v p Þ

ð 19Þ

fð v p Þ ¼ 3 2:57 ½1 e ð 1: 5 0 : 416V p Þ

20Þ

The above equation is considered as positive when V p is greater than 3.6 m/s. If V p is less than 3.6 m/s, the equation becomes negative.

ð

2.1.7. FMEP model

Due to frictional losses, there is a considerable decrease in the indicated power of the engine due to frictional losses so brake power is dependent on frictional power. In general nine fric- tional mean effective pressure (FMEP) losses are considered

[20,10].

Mean effective pressure lost to overcome friction due to gas

pressure.

FMEP1 ¼ 0: 42ð p a p inf Þ

l

D 2

0:088 Cr 1 :33 0 : 39Vp =100

where p inf is intake manifold vacuum. MEP absorbed by piston rings.

FMEP2 ¼ 10 0: 377 l npr

D 2

where npr is number of piston rings present. MEP lost due to piston.

FMEP3 ¼ 12: 85 Psl

100 V p

100 V p

Dl

1000

10

ð 21Þ

ð 22Þ

ð 23Þ

where Psl is piston skirt length in mm. Blow by losses.

FMEP4

¼ p

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

p a p inf

"

0: 121Cr

0:4

ð 0: 0345 þ 0: 001055CrÞ

where Cr is compression ratio and N is speed. MEP lost by throttling.

FMEP5 ¼ 2: p 75 þ p inf

e

MEP lost in valve gear.

FMEP6 ¼ 0: 226 30

4N

GH 1: 75

1000

D

2

l

N

1000

1:185

#

ð24Þ

ð

ð

25Þ

26Þ

where G is number of valves and H is intake valve diameter. Pumping loss.

FMEP7 ¼ 0: 0275

N

1000

1: 5

MEP lost in bearing friction.

FMEP8 ¼ 0: 0564

D

N

l

1000

ð

27Þ

ð 28Þ

MEP lost in combustion chamber.

FMEP9 ¼

r

ffiffiffiffiffiffiffiffiffiffiffiffi

pimep

11: 45

0: 0915

N

1000

1 :7

ð

29Þ

where pimep is intake manifold pressure.

Application of zero-dimensional thermodynamic model 7 2.2. Combustion chemistry When a chemical reaction occurs, the

Application of zero-dimensional thermodynamic model

7

2.2. Combustion chemistry

When a chemical reaction occurs, the bonds within the mole- cules of reactants are broken, and atoms and electrons rear- range to form products. In combustion reactions, rapid

oxidation of the fuel results in energy release with the forma- tion of combustion products. Considering fuel as C a H b O c which is oxidized to form products, the combustion equation

is

given by

C

a H b O c þ xð O 2 þ 3 :77N 2 Þ !dCO 2 þ eH 2 O þ fN 2 ð 30 Þ

The atomic balance of each species in the above C–H–O is given as C: a = d, H: b = 2d, O: 2x = 2 d + e. Stoichiometric air fuel ratio is the amount of air required to completely combust one mole of fuel. The molecular formula of the fuel is determined as C 18 H 35 O 2 and therefore, equation for the combustion of the fuel is written as

C 18 H 35 O 2 þ að O 2 þ 3: 77N 2 Þ !18CO 2 þ 17: 5H 2 O þ a 3: 77N 2

ð 31 Þ

Hereby considering the oxygen balance 2 + 2 * a = 18 * 2 + 17.5. Solving the above equation gives ‘‘aas 12.875 (i.e.) 12.875 mol of air is required to completely burn 1 mol of fuel. The above equation is rewritten as

C 18 H 35 O 2 þ 12 :875 ð O 2 þ 3: 77N 2 Þ ! 18CO 2 þ 17 :5H 2 O þ 48 :53N 2

ð 32 Þ

Specific heat is the factor that is dependent upon tempera- ture of the combustion chamber. The following relations are be used to determine the specific heat of the in cylinder gases

For

400 6 T 6 1600 ;

Cp ð TÞ ¼ BL þ CL =T

ð

33

Þ

For

1600 6 T 6 6000 ;

Cp ð TÞ ¼ BH þ CH=T

ð

34

Þ

n

Cp ¼ X Cp k N k

 

ð

35

Þ

k ¼1

where BL , BH , CL , and CH are constants depending upon the species ‘‘ k is number of product species present.

2.3. Fuel spray model (fuel atomization and spray penetration

model)

The fuel jet that emerges out of the nozzle generally forms a cone-shaped spray at the tip of the nozzle known as atomiza- tion breakup regime, where the fuel is fragmented into smaller

droplets size less then nozzle diameter. For jets in the atomiza- tion regime, the spray angle h is given by

tan

h

2 ¼

1

A 4p

q

g

q

l

1=2

p

ffiffiffi

3

6

ð

36 Þ

where q g and q l are gas and liquid density of fuel, and A is con- stant for given geometry which is normally taken as 4.9. The speed at which the fuel penetrates into the combustion chamber also plays a major role in the mixing of fuel droplets with air. The spray penetration speed is given as

S ¼ 3: 07

Dp

q g

1 =4 ð td n Þ 1 =2

294

T g

1= 4

ð 37 Þ

where Dp is pressure drop across the nozzle, t is time after the start of injection and d n is nozzle diameter (see Table 4). The simulation was carried out using of MATLAB which divided into four major subroutines such as Ideal cycle simula- tion, Fuel cycle simulation, Progressive combustion simulation and Actual cycle simulation. Initially, in Ideal cycle simulation (ICS), air is considered to be working medium to analyse the pressure at different state points by using basic thermodynamic relations. ICS was followed by Fuel Cycle Simulation (FCS) in which fuel properties and adiabatic flame temperature are determined to identify the amount of heat generated by the fuel during the combustion process. The FCS is further modi- fied by considering actual combustion where zero dimensional combustion model incorporates the ignition delay equation and Wiebie’s heat release correlation to study the pressure and heat released in the Progressive cycle simulation (PCS). Finally Actual cycle simulation is carried out to eliminate the assumptions made by considering the pressure loss due to heat transfer with the surroundings, pressure lost due to gas

exchange process and frictional losses to obtain pressure- crank angle curves. The pressure data thus obtained are simu- lated for heat release rate with the equation obtained by apply- ing first law of thermodynamics to closed system [21] . The details of molecular weight, carbon presence, oxygen and hydrogen content were obtained from the molecular formula derived from the Gas Chromatography Mass Spectrometry result of beeswax biodiesel along with other thermodynamic

constants [4] .

3. Results and discussion

3.1. In-cylinder pressure and rate of heat release

The simulation was carried out using programs coded in latest version MATLAB software where the thermodynamic equa- tions are solved to predict the in-cylinder pressure, net heat release rate and rate of pressure rise for given fuel properties and various engine loads. The experiments were conducted at 0%, 50% and 100% load conditions. Fig. 3 shows the vari- ation between in-cylinder pressure and net heat release for actual and mathematical model running with straight diesel, BWB10 and BWB20 as fuel at 0% load condition. The simu- lated results showed higher value than the actual output from the engine. This may be due to static error in the instrument, ambiguity in the calibration of the equipment and due to

Table 4 Simulation data for fuel and air.

Simulation data

Values

Molecular weight No of carbon present in fuel No of hydrogen present in fuel No of oxygen present in fuel Universal gas constant (kJ/kg K) Ratio of specific heats (for air) Specific heat at constant pressure (for air) (kJ/kg K) Specific heat at constant volume (for air) (kJ/kg K) Ambient temperature (in K)

283

18

35

2

0.287

1.4

1.005

0.718

300

Ambient pressure (in bar)

1.03

8 V. Hariram, R. Bharathwaaj human errors while obtaining the results from the equipment which

8

V. Hariram, R. Bharathwaaj

human errors while obtaining the results from the equipment which are un-avoidable. The variations in simulated and exper- imental may be also due to assumptions such as air as ideal gas with constant C p and C v values but in real-time scenario, speci- fic heat is the function of temperature. Under simulation study, the rate of combustion was considered to be uniform through- out the combustion duration, but actually rate of combustion depends on concentration of reactants [18] . It is clearly seen from Fig. 3, that the peak pressure for straight diesel fuel operating at 0% load was around 48 bar whereas the simulated results show the peak pressure of 52.9 bar which accounts to 8% deviation between simulated and experimental results. Similarly, for BWB 10 and BWB 20, experimental peak pressure was found to be 45.9 bar and 44.8 bar respectively whereas the simulated peak pressure was 47.2 bar and 49.6 bar showing the error of 7% and 11% respectively. It was also noticed that peak pressure during sim- ulation is shifted away from TDC by 1–2 CAD whereas exper- imentally, it comes closer to TDC. In compression ignition engine, pressure is mainly depen- dent upon the amount of fuel burnt during pre-mixed combus- tion phase. If more fuel is burnt in this phase, then the peak cylinder pressure would be high. In normal case, amount of fuel burning in premixed phase basically depends upon the ignition delay period. If ignition delay is more, then premixed combustion phase is reduced and hence only less amount of fuel is burned in this phase. This in turn reduces the peak pres- sure inside the cylinder. Ignition delay is characteristics of combustion that can vary with respect to Cetane number of the fuel. Here in this case, the Cetane number of straight diesel was more than that of BWB10 and BWB20 and hence the curve shows an increased pressure for diesel as 48 bar whereas BWB10 and BWB20 exhibited 45.9 bar and 44.5 bar respectively. Fig. 3 also shows the net heat release rate for all the three fuel blends used. It can be seen that the heat release for diesel is higher than other two blends which may be due to marginal increase in density influencing fuel droplet atomization and higher latent heat of vaporization of bio fuels and hence it

higher latent heat of vaporization of bio fuels and hence it Figure 3 In-cylinder pressure and

Figure 3 In-cylinder pressure and net heat release comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 0% load.

vaporizes quickly forming combustible air fuel mixture and instantly burns by absorbing the heat from the premixed com- bustion phase. Due to better fuel atomization, enhanced com- bustion of fuel leads to maximum heat release of 20.32 J/CA for diesel, 19.78 J/CA for BWB10 and 18.97 J/CA for BWB20 blends. Fig. 4 depicts the comparison of combustion parameters for experimental and predicted results of straight diesel, BWB10 and BWB20 at 50% load operation which exhibits similar trend to that of 0% load operation. It can be noticed from the Fig. 4 which shows the increase in pressure from 48.08 bar to 60.14 bar for straight diesel, from 45 bar to 57.29 bar for BWB10 and 45 to 66.15 bars for BWB20 which may be due to Ignition delay depending in-cylinder pressure, temperature and spray atomization. The increase in in- cylinder pressure results in reduction in physical period of igni- tion resulting in maximum fuel burning in the pre-mixed com- bustion phase. This effect raises the overall temperature of the combustion chamber and leads to steady state heat release dur- ing the diffused combustion phase. Another reason being higher calorific value of straight diesel compared to BWB10 and BWB20 resulting in better rate of heat release. In case of 50% load and 0% load operation of the engine, the heat release rate of BWB10 shows a slight increase than that of die- sel which also may be due to additional oxygen content in the BWB10 blend and increase in ignition delay for lower loads. Fig. 5 shows the variations in the experimental and simu- lated results of in-cylinder pressure and net heat release rates for straight diesel, BWB10 and BWB20 operating at 100% load. At this operating condition, the pressure was found to be increased up to 69 bar for diesel and 67 bar for BWB10 and 65 bar for BWB20 blends which may be due to increase in the engine load to the maximum extent which draws higher power accordingly. During this period, the in-cylinder temper- ature and exhaust gas temperature rise to a considerable extent reducing the chemical delay phase of the ignition delay period and hence early pre-mixed combustion occurs which shows the drastic rise in pressure and heat release of the engine. At 100%

rise in pressure and heat release of the engine. At 100% Figure 4 In-cylinder pressure and

Figure 4 In-cylinder pressure and net heat release comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 50% load.

Application of zero-dimensional thermodynamic model 9 load operation, the effect of oxygen on combustion showed

Application of zero-dimensional thermodynamic model

9

load operation, the effect of oxygen on combustion showed negligible effect which may be due to increase in in-cylinder temperature leading to formation of mono-atomic oxygen hav- ing tendency to oxidize with carbon and nitrogen molecules present in the air fuel mixture leading to the formation carbon-monoxide and oxides of nitrogen in the exhaust across all biodiesel blends. Exhaust gas temperature (EGT) is a phe- nomenon which describes the extent of combustion inside the combustion chamber. High EGT signifies complete oxidization of air–fuel mixture along with emission such as oxides of car- bon and nitrogen. In the present scenario, EGT of diesel is higher when compared with the other blends of BWB10 and BWB20 which may be due better combustible and high flammability properties of straight diesel on comparison with BWB10 and BWB20 blend ratios.

3.2. Rate of pressure rise

Fig. 6 shows the rate of pressure rise for experimental and sim- ulated values of engine fuelled with straight diesel, BWB10 and BWB20 at 0% load condition, where the rate of pressure rise was found to be 1.1 bar for diesel and 0.8 bar for BWB10 and 1.1 bar for BWB20 when compared with theoretical data which show 1.21 bar, 1 bar, and 1.16 bar for straight diesel, BWB10 and BWB20 respectively with an average of 7–9% variation in experimental and theoretical results. Fig. 7 depicts the changes in rate of pressure rise for modelled and experi- mental results for straight diesel, and BWB10 and BWB20 at 50% load operation which shows a significant increase on BWB20 for both simulated and experimental results on com- parison with 0% load condition which may be due higher latent heat of vaporization and lower combustion duration. Fig. 8 shows the comparison between rate of pressure rise for straight diesel, BWB10 and BWB20 for experimental and simulated model at 100% load operations which exhibits a higher value of 6.4 bar, 5.68 bar and 5.53 bar for straight die- sel, BWB10 and BWB20 blend ratio. It can be seen that the rate of pressure rise lies very close between BWB10 and BWB20 due to their marginal variation in density which leads

BWB20 due to their marginal variation in density which leads Figure 5 In-cylinder pressure and net

Figure 5 In-cylinder pressure and net heat release comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 100% load.

straight diesel, BWB10 and BWB20 blend ratios at 100% load. Figure 6 Rate of pressure rise

Figure 6 Rate of pressure rise comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 0% load.

of straight diesel, BWB10 and BWB20 blend ratios at 0% load. Figure 7 Rate of pressure

Figure 7 Rate of pressure rise comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 50% load.

straight diesel, BWB10 and BWB20 blend ratios at 50% load. Figure 8 Rate of pressure rise

Figure 8 Rate of pressure rise comparison for experimental and simulated models of straight diesel, BWB10 and BWB20 blend ratios at 100% load.

to better atomization of the fuel particles in the combustion chamber [6] . It can be also noted that the rate of pressure rise is more when the fuel is complete atomized. Since the increase

10 V. Hariram, R. Bharathwaaj in latent heat of vaporization and relatively lower density of

10

V. Hariram, R. Bharathwaaj

in

latent heat of vaporization and relatively lower density of

straight diesel than BWB10 and BWB20 blends, diesel shows a better atomization and hence the in-cylinder pressure rises

instantly during the diffused combustion phase.

4. Conclusion

A comprehensive zero-dimensional model was developed to

estimate the in-cylinder pressure, rate of heat release and rate

of pressure rise in a cylinder compression ignition direct injec-

tion engine using straight diesel and blends of BWB. In this study, thermodynamic approach was used to predict the in- cylinder pressure, and Wiebie’s and Wolfer’s relation are

applied to analyse heat release correlations and ignition delay respectively along with gas dynamics, heat transfer and FMEP equations. At 0% load, the cylinder pressure for straight diesel was found to be 48.09 bar and 52.9 bar for experimental and simulated studies. With BWB10 and BWB20, the in-cylinder pressure was found to be 45.97 bar and 44.8 bar for experimen-

tal and simulated results. At 50% load and 100% load condi-

tions, a similar trend is noticed in the P-h diagram in which straight diesel exhibited higher peak. The rate of heat release for straight diesel, BWB10 and BWB20 at 0%, 50% and 100% load shows an increasing trend between 20.38 J/CA and 57.96 J/CA which may be due to enhanced pre-mixed combustion phase which is also supported

by the simulated results with a deviation of 7.2%. The rate of

pressure rise for simulated straight diesel, BWB10 and BWB20

was found to be 8.3% higher than the experimental data across

all loading conditions.

References