Experimental Study on Combustion Noise of Common rail Diesel Engine using different blend biodiesel

Chien Hsing Li, Yong-Yuan Ku, Ko-Wei Lin Automobile Research and Testing Center

Copyright © 2013 SAE Japan and Copyright © 2013 SAE International

ABSTRACT

Due to the energy safety and environment protection, increase the percentage of biodiesel blend has become one of world wide strategies. In the past research, using biodiesel would affect the engine performance and increase the exhaust emission. Fortunately, these problems can be solved through the rapidly development of engine control technologies and lightweight structure design. However, the consideration of light/downsizing engine design with the same power has brought out much combustion noise. According to the higher and higher proportion has been widely used over the world. There was less researches focus on the different blending biodiesel impact on combustion noise. The combustion noise correspond to different blending biodiesel (D100,B5,B8,B20,B40,B100) which made form waste cooking oil has been discussion in this study. The experimental by using engine which meet EURO-4 was designed to caught spectrum of the combustion noise via transient window which under the constant engine speed of 1500rpm, 2000rpm, 2500rpm, with different torque at 30%, 50% and 70% of each speed, respectively. The overall result shows that different proportion blending of biodiesel related to combustion noise directly and the combustion noise raises 0.5 dB (ref. 1 bar) by using 20% biodiesel blend. In this study the influence trend curve map of engine speed and torque values correspond to combustion noise have been gotten. It would help us to get a win-win-win situation and to optimize the objective function of engine control model in the future.

INTRODUCTION

Biodiesel has been utilized widely all over the world to reduce dependency on fossil fuel and to slow down the climate change of environment. In Taiwan, biodiesel has been prompted from 1992 until now. In 2008, Taiwan government has mandated the 1% biodiesel policy and increased the proportion to the 2% in 2010. At the meanwhile, the new policy of increasing percentage up to 5% in 2016 has been announced and the high percentage biodiesel such as 8% has promoted to the fleets. The advantages of biodiesel have been widely discussed and compared with fossil diesel, biodiesel produces lower engine emissions of carbon monoxide (CO), hydrocarbons (HC), particular matter (PM), polycyclic aromatic hydrocarbons (PAHs), and carbon dioxide (CO2), which significantly reduce the impact of greenhouse effect on the earth [1~3].However there were fewer researches on the influence of combustion noise which utilized biodiesel in the past. As the noise and emissions legislations are stricter in this century. The major purpose of this study was to

investigate the impact of combustion noise of engine which equipped with common rail injection system which utilized the different blend biodiesel. In the past research, in the characteristic analysis of combustion noise and composition, Norbert W. Alt et al. [4] has presented the clear defined the combustion noises comprehensively. In the same time to introduce the procedures to improve the combustion noise and compare the combustion noise compositions and characteristics of frequencies of gasoline and diesel engine which at the different operation conditions and power output. That study gave the excellent definition of combustion noise and objective function. In general, engine noise can be divided into two main components including combustion noise and mechanical noise. Combustion noise mainly depends on the design of combustion chamber and fuel injection system and can be controlled by modulating the rate of fuel injection and improving the combustion process. Mendez et al. [6] conducted the study which using multiple injection strategies in diesel combustion to discus the optimization of emission and noise. In this study, engine emission, fuel consumption and the relative of combustion noise has been presented. The results showed, during the stroke process of the engine, to apply the fuel twice on the right time can reduce 50% of the NOx emission and raise the EGR rate. Ignition delay can reduce the heat release to decrease the combustion noise. Nevertheless, multi injection strategies bring also some drawbacks. Indeed, most of the injection strategies presented a very accurate control of the fuel delivery. High performance injection systems such as piezoelectric or direct acting injector are necessary to well control fuel quantities on each injection despite pressure waves in the common rail system. Zaw Win et al. [7] used the Taguchi methods to investigate the combustion noise, emission and fuel economy of diesel engine which equipped with direct injection. In this study, the results showed engine speed was significant influence on engine noise, moderately affected to smoke and CO2 emission as well as slight effect on fuel consumption. But HC, NOx and CO emissions have insignificant effect on engine speed. There has the optimization of combustion, emission and fuel economy which under the operating condition was engine speed at 2000 rpm, 40% throttle and ignition at 28 degree before TDC.

EXPERIMENTAL METHODS

The diesel engine DAYEH 2.5TD was utilized in this study as well as the engine equipped with common rail fuel injection system. The basic specifications of the test engine are shown in Table 1. In this study, the injection timing and injection pulse width were not been changed, the different

percentage of biodiesel (D100, B5, B8, B20, B40, B100) blend were used that in order to investigate the characteristics of combustion noise. The basic testing conditions are shown in Table2.

Table 1. Test engine specifications

Item Specification Displacement 2500 c.c. Rated Power 105 kW/3200 rpm~3600 rpm Rated Torque 340 N.m/1800 rpm~2500 rpm Cylinders In-line 4 Cylinder/4 Stroke Compression Ratio 17.5：1 Fuel System Common Rail Specific Power 42 kW/litre Min BSFC 215 g/kWh Air Intake System Turbocharged with intercooler Emission Devices Catalytic converter, EGR

Table 2. Test condition

fuel Engine speed(rpm) Torque(Nm) Load (%)

D100,B5,B8 ,B20,B40,B100

1500 84 30%

140 50% 196 70%

2000 102 30% 170 50% 238 70%

2500 102 30% 170 50% 238 70%

The properties of biodiesel are different from fossil diesel fuel, such as higher cetane index and viscosity, lower sulfur content, and no aromatic hydrocarbon. The heating value is lower than fossil diesel, with the increasing of biodiesel blended, heating value decreased. Because of lower heating value, the fuel composition would be increase in the past research. The biodiesel (B100) was made from waste cooking oils by local manufacturers in Taiwan and complied with the biodiesel standard CNS 15072 which worked out based on EN 14214. Table 3 shows the properties of the test fuels. The equipment of engine combustion pressure sensor and encoder has been setup of this study was shown as Figure 1~ Figure3.The Spectrum analysis instrument SC305-UTP, KISTLER combustion pressure sensor and encoder have been employed in the experiment. The sampling rate of combustion pressure was 51.2Hz and the variation of pressure wave could be acquired via signal processing which was combustion noise has been defined in this study. In additional, triaxial acceleration transducers have been pasted on the cylinder head to measure the vibration after ignition in the cylinder. Through the angle domain transfer, the related of crank angle, combustion pressure and engine vibration could be acquired. Therefore the different fuels and different testing conditions related to the combustion characteristics also been evaluated. The combustion transducer has installed on the hole of engine glow plug and the encoder has installed on the end of Dynamometer. The axial and crank connect by flange that could keep the rotating synchronously.

Table 3. Testing Fuel properties

Test item D100 B5 B8 B20 B40 B100 Cetane index 53.0 53.3 54 53.1 52.3 53.7 Density,15℃

(g/ml) 0.834 0.837 0.839 0.844 0.853 0.882

Copper strip corrosion (3hr

at 50℃) 1a 1a 1a 1b 1a 1a

Flash point℃ 76 75.5 76.5 78.5 85 174 Kinematic

viscosity, 40℃(mm2/s)

2.5 2.591 3.287 2.798 3.143 4.286

Carbon residue (on

10% distillation

residue) wt%

<0.1 <0.1 <0.1 <0.1 <0.1 0.1

Distillation, IBP(℃) 181 186.3 197 189.7 194.5 -

Distillation, T95(℃) 345 352.7 353.1 350 346.5 -

Distillation, FBP(℃) 355 352.8 360.5 350.3 349.4 -

Ash (wt%) <0.01 <0.001 <0.001 <0.001 0.003 0.004

Sulfur content (ppm) <2 <1 7.0 1.4 2.6 3.1

Lubricity, corrected wear scar diameter (1.4 wsd, 60℃) (μm)

- 246 225 299 416 259

Fatty acid methyl ester

content %(v/v) 0.10 5.0 8.44 20.1 40.4 99.1

Water and sediment (vol%)

- 0.005 0.01 0.005 0.005 -

Water content (ppm)

0.004% 107 183 140 226 354

Total contaminant

(ppm) - 0.5 6.2 0.4 1 2.9

Pour point (℃) - -21 -9 -15 -9 -

Cold filtration plug point

(CFPP) (℃) -24 -21 -6 -16 -12 -1

Oxidation stability, 110℃ (g/m3)

10 0.1 0.26 0.1 0.5 10.9*

Heating value (kcal/kg) - 10888 10890 10680 10441 -

Figure 1. combustion pressure senor on Engine in the Test Lab.

Figure 2. vibration senor on Engine head

Figure 3. Encoder of Engine

HORIBA CVS 9400T was the sampling equipment and HORIBA MEXA 7500D was exhaust emission analysis instrument has been used in this study was shown as Figure 4. Through the emission measurement system, HC, NOx, CO has been measurement under steady state testing. Furthermore, the data of measurements has been compared to the combustion characteristics which to evaluate the influence of using different fuels on common rail engine. The engine control parameters such as injection timing and injection pulse width have been maintained at the same over the testing procedure, in order to investigate the influence of combustion and emission of engine which using different fuels.

Figure 4. HORIBA MEXA 7500D was exhaust emission analysis instrument in the Test Lab.

Figure 5 showed the transformation of schematic of angle domain . While measure the combustion pressure signal, the angle signal has been acquired via encoder. To combine the combustion pressure and pulse of encoder signal that the crank angle of TDC and BDC have been defined. Therefore, the stroke of engine combustion and other signal have been obtained. At the meantime, at each testing condition 20 sampling data has been used to average, in order to investigate the effect on NVH of engine using different blend biodiesel.

Figure 5. Schematic of angle domain transfer

Figure 6 showed the transformation of schematic of combustion pressure signal frequency domain. The frequency information has been obtained by transferring the combustion pressure signal using Fast Fourier Transform (FFT) method. The effective bandwidth was 25.6kHz, analytic number was 8192 and the transient window has been utilized in this experiment. The cut off percentage has been set depending on duration of pressure wave related to transient window at each cycle. As the same before, at each testing condition 20 sampling data has been used to average.

Figure 6. Schematic of frequency domain transfer

Figure 7 showed the influence of transient window related to combustion pressure frequency. During the FFT process, when the time domain signal of combustion pressure has not weighted with any window, the frequency domain will cause

the phenomenon of harmonic. The harmonic probably produce from variation of engine speed, compress signal, ignition signal as well as combustion signal. That was a barrier to judgment the characteristic of noise because of the combination of all kind of harmonic frequencies. Therefore to cut off the transient window of the combustion pressure wave would get the better advantage to discuss the frequency area which produces by combustion noise.

Frequency (Hz)

NO time windowtransient window

Frequency (Hz)

NO time windowtransient window

Figure 7. Influence of time widow related to signal of combustion pressure frequency

SUMMARY/CONCLUSIONS

By using experimental design and analysis method, the transformation of angle domain, frequency domain and emission have been discussed in this study. In term of transformation of angle domain, Figure 8 showed the combustion pressure and angle domain transformation. Figure 8 showed the curve of angle domain which was under engine operation conditions including 1500 rpm/ 84 Nm, 2000 rpm/ 170 Nm and 2500 rpm/ 238 Nm, respectively.

The crank angle from -10° to 15° has been discuss in this figure. With the different testing condition, the variation of combustion pressure wave has been observed obviously. In theory, the curve of combustion pressure wave could be smooth. In fact, the combustion pressure wave of diesel engine showed slight shaking which showed in this figure.

Speaking of phenomenon of combustion pressure wave, engine speed from 1500rpm to 2000rpm, engine torque output at 50% load, the shaking at those operation conditions was almost the same. However, Shaking of combustion pressure wave, while engine speed up to 2000rpm; especially increase the percentage of biodiesel blend was observed obviously.

1500 rpm 30% Nm

3541475359657177838995

-10 -5 0 5 10 15crank angle

bar

D100 B5B8 B20B40 B100

(a) 1500 rpm/ 30%/ 84 Nm

2000 rpm 50% Nm

45515763697581879399

105

-10 -5 0 5 10 15crank angle

bar

D100 B5B8 B20B40 B100

(b) 2000 rpm/ 50%/ 170 Nm

2500 rpm 70% Nm

657177838995

101107113119125

-10 -5 0 5 10 15crank angle

bar

D100 B5B8 B20B40 B100

(c) 2500 rpm/ 70%/ 238 Nm

Figure 8. Transformation of combustion pressure and angle domain at different operating conditions

Figure 9 presented the engine operating condition at 1500 rpm/ 84 Nm, 2000 rpm/ 170 Nm and 2500 rpm/ 238 Nm, respectively, crank angle was from -180° to 180°. In the different testing condition, near-field noise and angle domain of engine vibration has not been observed clearly, that was different from combustion pressure signals. It need to be discussed by frequency transformation.

1500 rpm 30% Nm

0100200300400500600700800900

1000

-180 -108 -36 36 108 180crank angle

m/s

2D100 B5B8 B20B40 B100

(a) 1500 rpm/ 30%/ 84 Nm

2000 rpm 50% Nm

080

160240320400480560640720800

-180 -108 -36 36 108 180crank angle

m/s

2

D100 B5B8 B20B40 B100

(b) 2000 rpm/ 50%/ 170 Nm

2500 rpm 70% Nm

060

120180240300360420480540600

-180 -108 -36 36 108 180crank angle

m/s

2

D100 B5B8 B20B40 B100

(c) 2500 rpm/ 70%/ 238 Nm

Figure 9. Transformation of angle and engine vibration at different operating conditions

Speaking of transformation of frequency domain, Figure 10 showed the combustion pressure frequency transformation. Figure 10 showed the curve of angle domain which was under engine operation conditions including 1500 rpm/ 84 Nm, 2000 rpm/ 170 Nm and 2500 rpm/ 238 Nm, respectively.

There were obvious frequency peaks have been observed transferring the time domain pressure wave by using FFT. When utilized different fuels, the frequency of combustion characteristic has variation in the small range. The frequencies have been distributed at 2100 Hz, 6200 Hz, 11000 Hz and 15000 Hz, respectively. In the figure, under the different operating conditions and using different fuels, the response of frequency domain almost showed the same trends. The pressure wave signals have been judged by crank angle at TDC. The transformation data of combustion pressure frequency domain, in other words, the pressure value related to the limitation of frequency from 5612.7 Hz to 7071.5 Hz has been calculated by integration, and the influence of combustion noise which engine using different fuels have been discussed.

1500 rpm 30% Nm

-100

-80

-60

-40

-20

0

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

r. 1

bar

D100 B5B8 B20B40 B100

(a) 1500 rpm/ 30%/ 84 Nm

2000 rpm 50% Nm

-100

-80

-60

-40

-20

0

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

r. 1

bar

D100 B5B8 B20B40 B100

(b) 2000 rpm/ 50%/ 170 Nm

2500 rpm 70% Nm

-100

-80

-60

-40

-20

0

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

r. 1

bar

D100 B5B8 B20B40 B100

(c) 2500 rpm/ 70%/ 238 Nm

Figure 10. Transformation of combustion pressure frequency domain at different operating conditions

Figure 11 showed the transformation of engine vibration frequency domain. In this figure, the high correlation of characteristic frequency has been observed, the vibration response of 6000 Hz.

It showed the response has the high sensitivity influence of engine vibration during the process of diesel engine including injection, explosion and combustion. That means if we want to improve the diesel engine vibration, which could do effort on research of strategies of injection control and combustion modification as well as strengthen the stiffened of engine block. The quality of vibration could be improved.

1500 rpm 30% Nm

-30

-18

-6

6

18

30

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

f. 1m

/s2

D100 B5B8 B20B40 B100

(a) 1500 rpm/ 30%/ 84 Nm

2000 rpm 50% Nm

-30

-24

-18

-12

-6

0

6

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

f. 1m

/s2

D100 B5B8 B20B40 B100

(b) 2000 rpm/ 50%/ 170 Nm

2500 rpm 70% Nm

-30

-24

-18

-12

-6

0

0 5120 10240 15360 20480 25600frequency (Hz)

dB re

f. 1m

/s2

D100 B5B8 B20B40 B100

(c) 2500 rpm/ 70%/ 238 Nm

Figure 11. Transformation of engine vibration frequency domain at different operating conditions

The influence of combustion noise which using different fuels have analyzed in this paper was shown in Figure 12. To analyze the single fuel under the different operation condition, with the increasing of percentage of biodiesel, combustion noise has the trend of increasing. Each 20% increased of biodiesel, the combustion noise increased 0.5dB. (Reference 1 bar)

Figure 13 showed the trend of engine vibration which using different fuels. As the above-mentioned, to average the engine vibration values which using single fuel under the different operation condition. With the increasing of biodiesel blended, engine vibration has the trend decreased. Each 20% increased of biodiesel, the engine vibration decreased 0.25dB (Reference value 1m/s2). It contrasted with combustion noise.

To evaluate the emission of NOx by using the same calculation method, Figure 14 showed the trend of NOx emitted which using different fuels. With the increasing consecrations 20% of biodiesel, NOx decreased 10.6 ppm averagely.

Figure 15 showed the peak combustion pressure when using different testing fuels. With the increasing of biodiesel blended, the peak of combustion pressure decrees obviously. Because of the heating value of biodiesel is lower than pure

fossil fuel, the combustion temperature become lower when increasing the percentage of biodiesel. In contrast, it has the advantages of diesel engine emission.

combustion noise

-39

-38.5

-38

-37.5

-37

-36.5

-36

D100

B5

B8

B20

B40

B100

diesel fuel

dB re

f. 1

bar

combustion noise

Figure 12. Trend of combustion noise using different testing fuels

E/G head cover vibration

14

14.5

15

15.5

16

16.5

17

17.5

18

D100 B5 B8 B20 B40 B100diesel fuel

dB re

f. 1

m/s

2

vibration

Figure 13. Trend of engine vibration using different testing fuels

NOx

920

940

960

980

1000

1020

D100

B5

B8

B20

B40

B100

diesel fuel

ppm

NOx

Figure 14. Trend of NOx emision using different testing fuels

combustion pressure

90919293949596979899

100

D100

B5

B8

B20

B40

B100

diesel fuel

bar

combustion noise

Figure 15. Trend of peak combustion pressure using different testing fuels

CONCLUSIONS

The results obtained in this work indicate the influence of different blend of biodiesel. The conclusions are detailed as follows:

The different blend of biodiesel D100,B5, B8,B20,B40,B100 have been utilized to common rail engine to evaluate the impact on the NVH in this study. The method of this study was using exchange of angle domain and frequency domain to analyze influence of the combustion characteristics and to evaluate the concentrations of emission.

To summarize, with the increasing of biodiesel blend, combustion noise has trend of increasing. Each of 20% biodiesel increased the combustion noise increase 0.5dB, in contrary, the vibration of engine decrease 0.25dB. In terms of emission, increasing 20% biodiesel, the NOx emission has decreased 10.6ppm approximately.

The engine equipped with common rail which using different percentages of biodiesel has employed in this study. The injection taming and injection pulse width during the experiment has not been changed; result showed the unique engine control strategy was not suitable. Therefore, that could become the variation of noise and combustion. In order to evaluate the influence of engine noise which using the unique control strategy, there were important issues to modify the engine control parameters to fit the different fuels or special fuel need to be researched in the future.

REFERENCES

1. C.A. Sharp, S.A. Howell, J. Jobe, 2000, “The Effect of Biodiesel Fuels on Transient Emissions from Modern Diesel Engines, Part I Regulated Emissions and Performance,” SAE paper 2000-01-1967.

2. V. Camobreco, J. Sheehan, J. Duffield, M. Graboski, 2000, “Understanding the Life-Cycle Costs and Environmental Profile of Biodiesel and Petroleum Diesel Fuel,” SAE paper 2000-01-1487.

3. M.A. Kalam, H.H. Masjuki, 2002, “Biodiesel from palmoil - An analysis of its properties and potential,” Biomass Bioenergy, Vol. 23, pp. 471-479.

4. Norbert W. Alt, Norbert Wiehagen and Christoph Steffens, and Stefan Heuer, 2001, “Comprehensive Combustion Noise Optimization,” SAE Paper 2001-01-1510.

5. Sylvain Mendez and Benoist Thirouard, 2008, “Using Multiple Injection Strategies in Diesel Combustion: Potential to Improve Emissions, Noise and Fuel Economy Trade-Off in Low CR Engines,” SAE Paper 2008-01-1329.

6. Zaw Win, R. P. Gakkhar, S. C. Jain and M. Bhattachharya, 2002, “24 Noise, Emissions and Fuel Economy Investigation on a Small DI Diesel Using Taguchi Methods,” SAE Paper 2002-32-1793.

CONTACT INFORMATION

Chien Hsing Li Project Engineer Automotive Research and Testing Center, Taiwan leo_li@artc.org.tw

Yong-Yuan Ku Project Engineer Automotive Research and Testing Center, Taiwan tom@artc.org.tw

Ko-Wei Lin Project Engineer Automotive Research and Testing Center, Taiwan linkewei@artc.org.tw

ACKNOWLEDGMENTS

The authors would like to thank the financial support by Bureau of Energy, Ministry of Economic Affairs (R.O.C.) under contract 102-D0107 which has made this report successful.