Erinç Uludamar, Erdi Tosun, Kadir Aydın: Articleinfo
Erinç Uludamar, Erdi Tosun, Kadir Aydın: Articleinfo
Erinç Uludamar, Erdi Tosun, Kadir Aydın: Articleinfo
Fuel
journal homepage: www.elsevier.com/locate/fuel
a r t i c l e i n f o a b s t r a c t
Article history: Noise and vibration of ignition compression engine is one of the most complicated field to cope with since
Received 8 December 2015 every mechanism that compose of the engine affect them separately. In this study, effect of various bio-
Received in revised form 8 January 2016 diesel on engine noise and vibration has been studied on an unmodified compression ignition engine.
Accepted 9 March 2016
Linear and non-linear based regression analysis based models were also developed to predict the rela-
Available online 16 March 2016
tionship between fuel properties and noise–vibration of the engine with excellent agreement. The results
revealed that noise and vibration of the engine decreased with increasing biodiesel ratio until pure bio-
Keywords:
diesel usage and in linear regression analysis, correlation coefficient (R) values were 0.784 and 0.741 for
Biodiesel
Compression ignition engine
vibration and sound pressure level, correspondingly. In non-linear regression analysis, correlation coeffi-
Engine vibration cient (R) values were 0.814 and 0.777 for vibration and sound pressure level, correspondingly. In linear
Regression analysis regression analysis, mean absolute percentage error (MAPE) values were 0.747% and 0.126% for vibration
Sound pressure level and sound pressure level, correspondingly. In non-linear regression analysis, mean absolute percentage
error (MAPE) values were 0. 682% and 0.342% for vibration and sound pressure level, respectively.
Ó 2016 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fuel.2016.03.028
0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
E. Uludamar et al. / Fuel 177 (2016) 326–333 327
In literature, although it has been many studies about perfor- Tanaka APM-7 Automated Pensky-Martens Closed Cup Flash Point
mance and exhaust emission characteristic of internal combustion Tester for flash point determination.
engines operated with alternative fuels, there is deficiency about Tests fuels were prepared as pure low sulphur diesel (D100),
their noise and vibration characteristics which are crucial issue 80% diesel, 20% sunflower biodiesel (SB20); 60% diesel, 40% sun-
for passenger comfort. With the usage of equations obtained from flower biodiesel (SB40); 40% diesel, 60% sunflower biodiesel
regression analysis, estimation of engine characteristic fuelled with (SB60); 20% diesel, 80% sunflower biodiesel (SB80); 80% diesel,
various biodiesels can be possible without time consuming and 20% canola biodiesel (CaB20); 60% diesel, 40% canola biodiesel
costly experimental works. Therefore, in this study, noise and (CaB40); 40% diesel, 60% canola biodiesel (CaB60); 20% diesel,
vibration characteristic of a direct ignition engine which was 80% canola biodiesel (CaB80); 80% diesel, 20% corn biodiesel
fuelled with low sulphur diesel, sunflower, canola, corn biodiesels (CoB20); 60% diesel, 40% corn biodiesel (CoB40); 40% diesel, 60%
and their blends were investigated at different engine speeds. Then corn biodiesel (CoB60); 20% diesel, 80% corn biodiesel (CoB80),
linear and non-linear based regression analysis models were devel- by volume and pure sunflower (SB100), canola (CaB100), and corn
oped to predict the relationship between fuel properties and (CoB100) biodiesels were also used for engine tests.
noise–vibration of the engine with excellent agreement.
2.2. Experimental engine
2. Material and method
In this work, experiments were performed on a four stroke, four
cylinder, and direct injection diesel engine. Experimental layout is
Through this study, experiments were conducted in Petroleum
shown in Fig. 2 and engine specifications are given in Table 1.
Research and Automotive Engineering Laboratories of the Depart-
The engine was warmed up to constant operating temperature
ment of Automotive Engineering at Çukurova University. The
and engine fuel line was cleaned before introduce with other fuel
methodology involves fuel properties, engine properties, data
and waited until the engine became stable and then experimental
acquisition equipment, and calculations for post processing.
data were gathered. Fuels were tested at 1200, 1500, 1800, 2100,
and 2400 rpm engine speeds under no load condition.
2.1. Test fuels
2.3. Measuring system, accelerometer and microphone
In this study, sunflower, canola and corn oil were used as raw
materials for biodiesel production since they are easily available Acoustic and vibration data were recorded via SoundbookTM uni-
in local oil companies in Turkey. versal portable measuring system with SAMURAI v2.6 software
Biodiesel samples were produced via the transesterification from SINUS Messtechnik GmbH. It is capable to double integration
reaction. In this reaction, methanol and sodium hydroxide were of the time signal as filtering according to ISO 10816, ISO 7919 and
used as reactant and catalyst. The mixture was heated up to 60 °C ISO 2954 standards and sound level meter (SLM) of the software
and kept at this temperature for 90 min by stirring. After the reac- meets the Class 1 SLM according to IEC 60651, IEC 60804, DIN
tion period, the crude methyl ester was waited at separating funnel EN 61672-1:2003 standards allowing simultaneous measurements
for 8 h. And then, crude glycerine was separated from methyl ester. with the frequency weightings A, C, Z.
Finally, the crude methyl ester was washed by warm water until the Triaxial ICPÒ accelerometer with 1.02 mV/(m/s2) sensitivity and
washed water became clear and then it dried at 110 °C for 1 h. ±4905 m/s2 measurement range was adhered on engine block with
Finally washed and dried methyl ester was passed through a filter. quick bonding gels to measure the vibration even in high frequency
Steps of biodiesel production are also briefly illustrated in Fig. 1. range. Data were collected in three orthogonal of the test engine.
Fuel properties were analysed by; Zeltex ZX 440 NIR petroleum Sound pressure level of the engine was measured by using a GRAS
analyzer with an accuracy of ±0.5 for determining cetane number; half-inch free-field standard microphone set which has 50 mV/Pa
Tanaka AKV 202 auto kinematic viscosity test for determining the sensitivity, 3.15 Hz–20 kHz frequency range and 17 dB(A)–149 dB
viscosity; Kyoto electronics DA-130 for density measurement and dynamic range. The microphone was placed at 1 m distance from
IKA-Werke C2000 Bomb Calorimeter for gross heating value the block surface.
determination. Tanaka AFP-102 for cold filter plugging point; Time and frequency domains were analysed in x (longitudinal),
y (lateral) and z (vertical) axes. Since RMS (root mean square) value
is the most relevant measure of vibration level (it gives an ampli-
tude value by considering time history of the wave), the results
were described as RMS value. Formula that was used to calculate
RMS value was presented in Eq. (1). aw (m/s2) represents the
weighted acceleration and T represents measurement time.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z
1 T 2
aw ¼ a ðtÞdt ð1Þ
T 0 w
Table 1
Technical specifications of the test engine. compression ignition engine, these parameters were selected as
Brand Mitsubishi canter input for estimation.
Model 4D34-2A One of the most widely used method to relate dependent vari-
Configuration In line 4 able(s) with independent variable(s) is regression analysis. It is a
Firing order 1–3–4–2
statistical process to find functional relationship between vari-
Type Direct injection diesel with glow plug
Displacement 3907 cc ables. Regression analysis is mostly used for prediction.
Bore 104 mm In linear regression, the function is linear equation and depen-
Stroke 115 mm dent variable can be expressed as a function of independent vari-
Power 89 kW @ 3200 rpm
able(s) in form of [23]:
Torque 295 N m @ 1800 rpm
Oil cooler Water cooled
Weight 325 kg
Y ¼ b0 þ b1 X 1 þ b2 X 2 þ . . . þ bn X n ð3Þ
In non-linear regression, the function is non-linear equation and
dependent variable can be expressed as a function of independent
MATLAB. Parameters which were selected as input in order to variable(s) in form of [23]:
estimate outputs should be determined according to correlation
coefficients between them. Therefore, all proper predictors which
Y ¼ a0 X a11 X a22 . . . X ann ð4Þ
have higher correlation coefficients were selected to define each
of predicted values. Since density, cetane number, kinematic vis-
cosity, and gross heating value have strongly affected combustion where Y is dependent variable, b0 to bn and a0 to an are equation
characteristics which directly related to noise and vibration of a parameters and X1 to Xn are independent variables.
Table 2
Fuel properties of the test fuels.
Test fuels Density (kg/l) Cetane number Kinematic viscosity at 40 °C (mm2/s) Gross heating value (cal/kg) Flash point (°C)
D100 0.837 59.3 2.7 45,857 74.5
SB20 0.844 56.5 3.1 44,246 101.5
SB40 0.854 53.6 3.4 43,430 106.5
SB60 0.869 51.1 3.7 42,998
SB80 0.876 49.0 4.1 41,874
SB100 0.886 44.5 4.5 39,149 >140
CaB20 0.846 56.9 3.2 43,413 76.7
CaB40 0.857 54 3.5 42,986 83.9
CaB60 0.861 51.8 3,8 41,568
CaB80 0.869 50 4.0 39,974
CaB100 0.883 46 4.7 38,363 120.5
CoB20 0.847 55.4 3 42,266 85.5
CoB40 0.856 52.1 3.2 41,881 91.5
CoB60 0.862 48.9 3.5 39,368
CoB80 0.870 45 3.8 38,158
CoB100 0.887 42.2 4.2 37,476 >140
EN590 820–845 Min 51 2.0–4.5 – Min 55
ASTM D 6751 – Min 47 1.9–6.0 – Min 93
EN 14214 860–900 Min 51 3.5–5.0 – Min 120
E. Uludamar et al. / Fuel 177 (2016) 326–333 329
Fig. 3. Time and frequency domain signal of the engine that fuelled with SB20 at 1500 rpm.
Fig. 4. Time and frequency domain signal of the engine that fuelled with CoB100 at 2400 rpm.
330 E. Uludamar et al. / Fuel 177 (2016) 326–333
80 80
Diesel Diesel
SB20 CoB20
70 70
SB40 CoB40
SB60 CoB60
SB80 CoB80
atotal (m/s2)
60 SB100 60 CoB100
atotal (m/s2)
50 50
40 40
30 30
1200 1500 1800 2100 2400 1200 1500 1800 2100 2400
Engine Speed (rpm) Engine Speed (rpm)
Fig. 5. atotal versus engine speed when the engine fuelled with sunflower biodiesel. Fig. 7. atotal versus engine speed when the engine fuelled with corn biodiesel.
100
80
Diesel
98
Sound Pressure Level (dB[A])
Diesel SB20
CaB20 SB40
70
CaB40 96 SB60
CaB60 SB80
CaB80 SB100
a total (m/s2)
60 CaB100 94
92
50
90
40
88
86
30
CaB20
Quality measurements of test fuels were performed according CaB40
to TS EN 14214 biodiesel standard and EN 590 diesel standard. Fuel 96 CaB60
properties are given in Table 2. CaB80
CaB100
94
CoB20
CoB40 diesel blend fuels are shown.
96 CoB60 Sound pressure level of the engine was also increased by
CoB80
CoB100 increasing engine speed. Average sound pressure level found
94 slightly lower than that of low sulphur diesel fuel with 0.6 dB[A],
0.3 dB[A], 0.2 dB[A], for SB20, CaB20, CoB20; 0.5 dB[A], 0.4 dB[A],
92 0.4 dB[A], for SB40, CaB40, CoB40; 0.7 dB[A], 0.7 dB[A], 0.7 dB[A],
for SB60, CaB60, CoB60; 0.8 dB[A], 0.8 dB[A], 0.7 dB[A], for SB80,
90 CaB80, CoB80; and 0.6 dB[A], 0.6 dB[A], 0.7 dB[A], for SB100,
CaB100, CoB100, respectively. Decrement of sound pressure level
88 may be related with the decrement of engine vibration. Therefore,
parallel to decrement trend of engine block vibration, sound pres-
86 sure level was also decreased by biodiesel addition.
1200 1500 1800 2100 2400
3.4. Results of linear and non-linear regression analysis
Engine Speed (rpm)
Fig. 10. Sound pressure level versus engine speed when the engine fuelled with In this study, both linear and non-linear regression analyses
corn biodiesel. were applied to experimental data.
Forecasted parameters are vibration and sound pressure level of
the internal combustion engine. Density, cetane number, viscosity
It is noticed that the values of atotal increased with the increase and lower heating value are selected as input parameters to predict
of engine speed due to the increment of inertia forces such as crank vibration and sound pressure level of engine.
rotation speed. The average of engine vibration (atotal) decreased by SPSS which is a software package used for statistical analysis
1.72%, 2.37%, 0.5% for SB20, CaB20, CoB20; 2.38%, 2.79%, 0.98% for was used to perform regression analysis. Results of analysis can
SB40, CaB40, CoB40; 2.28%, 2.44%, 2.11% for SB60, CaB60, CoB60; be seen from the following Eqs. (5)–(8):
4.60%, 5.76%, 3.19% for SB80, CaB80, CoB80 and 2.80%, 3.87%,
3.72% for SB100, CaB100, CoB100, respectively when compared Vibration ¼ 71:126 ð19:43 densityÞ ð0:123 CNÞ
with D100. Variations in fuel properties may explain the differ- ð1:042 viscosityÞ þ ð0:000121 LHVÞ ð5Þ
ences. Since biodiesel contains extra oxygen, combustion quality
of biodiesel is better than combustion of conventional diesel fuel. Sound level ¼ 85:097 þ ð10:596 densityÞ þ ð0:041 CNÞ
That’s why; many researchers have concluded on that, oxides of
nitrogen exhaust emission generally increases with biodiesel addi- ð0:366 viscosityÞ ð3:266 105 LHVÞ ð6Þ
tion by the same reason [12–14]. Enhancement of combustion with
Þ CN0:101
0:003
the usage of biodiesel means lowering vibration of the diesel Vibration ¼ 35:318 ðdensity
engine block.
0:109
viscosity ðLHV0:082 Þ ð7Þ
3.3. Sound pressure level of the engine Sound level ¼ 103:276 ðdensity
0:127
Þ ðCN0:017 Þ
0:02
Experiments were conducted in a sound insulated room. Fur- ðviscosity Þ ðLHV0:011 Þ ð8Þ
thermore, periphery of dynamometer was covered by a wooden
box with sound absorber panels.
52.0
51.5
Experimental LR NLR
51.0
50.5
50.0
49.5
Vibration
49.0
48.5
48.0
47.5
47.0
46.5
46.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Data
95.0
Experimental LR NLR
94.5
93.5
93.0
92.5
92.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Data
Fig. 12. Comparison of actual and predicted sound pressure level data.
[1] Atalay M, Anafarta N, Sarvan F. The relationship between innovation and firm
performance: an empirical evidence from Turkish automotive supplier
industry. Proc – Soc Behav Sci 2013;75:226–35.
[2] Holm M, Garcia AC, Adamson G, Wang L. Adaptive decision support for shop-
Eqs. (5) and (6) linearly calculates vibration and sound pressure floor operators in automotive industry. Proc CIRP 2014;17:440–5.
level of engine when we put the input parameters into equation. [3] Ganapathy T, Gakkhar RP, Murugesan K. Influence of injection timing on
performance, combustion and emission characteristics of Jatropha biodiesel
On the other hand, Eqs. (7) and (8) non-linearly calculates vibra- engine. Appl Energy 2011;88:4376–86.
tion and sound pressure level of engine when we put the input [4] Ong HC, Mahlia TMI, Masjuki HH, Honnery D. Life cycle cost and sensitivity
parameters into equation. analysis of palm biodiesel production. Fuel 2012;98:131–9.
[5] Alptekin E, Canakci M, Ozsezen AN, Turkcan A, Sanli H. Using waste animal fat
The results of analysis can be seen from the following Figs. 11 based biodiesels–bioethanol–diesel fuel blends in a DI diesel engine. Fuel
and 12. As seen from figure there is good agreement between 2015;157:245–54.
actual and predicted data. Correlation coefficient (R) and mean [6] Arbab MI, Masjuki HH, Varman M, Kalam MA, Imtenan S, Sajjad H. Fuel
properties, engine performance and emission characteristic of common
absolute percentage error (MAPE) were selected as performance
biodiesels as a renewable and sustainable source of fuel. Renew Sustain
parameter and results are given in the following Table 3. Energy Rev 2013;22:133–47.
It can be concluded that, vibration and sound pressure level can [7] Zhang H, Zhou Q, Chang F, Pan H, Liu X-F, Li H, et al. Production and fuel
be predicted with an acceptable accuracy with the usage of linear properties of biodiesel from Firmiana platanifolia L.f. as a potential non-food oil
source. Ind Crops Prod 2015;76:768–71.
and non-linear regression. [8] Gopal K Nantha, Karupparaj R Thundil. Effect of pongamia biodiesel on
emission and combustion characteristics of di compression ignition engine.
Ain Shams Eng J 2015;6:297–305.
4. Conclusion [9] Mohsin R, Majid ZA, Shihnan AH, Nasri NS, Sharer Z. Effect of biodiesel blends
on engine performance and exhaust emission for diesel dual fuel engine.
The objective of this study was to investigate the effect of bio- Energy Convers Manage 2014;88:821–8.
[10] Cheung CS, Man XJ, Fong KW, Tsang OK. Effect of waste cooking oil biodiesel
diesel fuels on an unmodified compression ignition engine with
on the emissions of a diesel engine. Energy Proc 2015;66:93–6.
regard to noise and vibration. Throughout the experiments the [11] Habibullah M, Masjuki HH, Kalam MA, Rizwanul Fattah IM, Ashraful AM,
engine fuelled with low sulphur diesel, sunflower biodiesel, canola Mobarak HM. Biodiesel production and performance evaluation of coconut,
palm and their combined blend with diesel in a single-cylinder diesel engine.
biodiesel, and corn biodiesel blends. Furthermore, both linear and
Energy Convers Manage 2014;87:250–7.
non-linear regression analysis were applied to experimental data [12] Tüccar G, Tosun E, Özgür T, Aydın K. Diesel engine emissions and performance
in order to predict vibration and sound pressure level of engine. from blends of citrus sinensis biodiesel and diesel fuel. Fuel 2014;132:7–11.
According to experiments and results, following summarises [13] Özener O, Yüksek L, Ergenç AT, Özkan M. Effects of soybean biodiesel on a DI
diesel engine performance, emission and combustion characteristics. Fuel
were brought out: 2014;115:875–83.
[14] Sayin C, Gumus M, Canakci M. Influence of injector hole number on the
Sound pressure level and vibration acceleration of the test performance and emissions of a DI diesel engine fueled with biodiesel–diesel
fuel blends. Appl Therm Eng 2013;61:121–8.
engine increased with engine speed. [15] Taghizadeh-Alisaraei A, Ghobadian B, Tavakoli-Hashjin T, Mohtasebi SS.
Averagely, compared to low sulphur diesel fuel, vibration and Vibration analysis of a diesel engine using biodiesel and petrodiesel fuel
sound pressure level decrement was observed with biodiesel blends. Fuel 2012;102:414–22.
[16] How HG, Masjuki HH, Kalam MA, Teoh YH. An investigation of the engine
usage. performance, emissions and combustion characteristics of coconut biodiesel in
The dominant frequency was regardless of fuel type. It signifi- a high-pressure common-rail diesel engine. Energy 2014;69:749–59.
cantly affected by engine speed due to upward and downward [17] Redel-Macías MD, Pinzi S, Leiva D, Cubero-Atienza AJ, Dorado MP. Air and
noise pollution of a diesel engine fueled with olive pomace oil methyl ester
piston movement.
and petrodiesel blends. Fuel 2012;95:615–21.
E. Uludamar et al. / Fuel 177 (2016) 326–333 333
[18] Elshaib AA, Kamal MM, Elahwany AA. Performance of a diesel engine fueled by [23] Bilgili M, Sahin B. Comparative analysis of regression and artificial neural
waste cooking oil biodiesel. J Energy Inst 2014;87:11–7. network models for wind speed prediction. Meteorol Atmos Phys
[19] Siavash NK, Najafi G, Hasanbeigi R, Ghobadian B. Acoustic analysis of a single 2010;109:61–72.
cylinder diesel engine using biodiesel fuel blends. Energy Proc 2015;75:893–9. [24] Dabrowski Z, Komorska I, Identification of engine vibroacoustic model for
[20] Lewis P, Fitriani H, Arocho I. Engine variable impact analysis of fuel use and diagnostics of mechanical defects. In: Proceedings of PACAM XII; 2012. p. 2–6.
emissions for heavy-duty diesel maintenance equipment. Transp Res Rec: J [25] Czech P, Lazarz B, Madej H, Wojnar G. Vibration diagnosis of car motor
Transp Res Board 2015;2482:8–15. engines. Acta Technica Corviniensis 2010;3:37–42.
[21] Wongchai B, Visuwan P, Chuepeng S. The vibration analysis of diesel engine [26] Tomaszewski F, Szymański G. Frequency analysis of vibrations of the internal
with hydrogen-diesel dual fuel. Am J Appl Sci 2013;10:8–14. combustion engine components in the diagnosis of engine processes. Arch
[22] Hirano I, Kondo M, Uraki Y, Asahara Y. Using multiple regression analysis to Transp 2012;24:117–25.
estimate the contributions of engine-radiated noise components. JSAE Rev
1999;20:363–8.