- Email: [email protected]
The Influence of LPG Fuelling on Diesel Engine Cycle Variability
Available online at www.sciencedirect.com
ScienceDirect Procedia Technology 22 (2016) 746 – 753
9th International Conference Interdisciplinarity in Engineering, INTER-ENG 2015, 8-9 October 2015, Tirgu-Mures, Romania
The Influence of LPG Fuelling on Diesel Engine Cycle Variability Alexandru Cernata,*, Constantin Panaa, Niculae Negurescua and Cristian Nutua a
University Politehnica of Bucharest, Blvd. Splaiul Independentei no. 313, Bucharest 70000, Romania
Abstract Liquid Petroleum Gas physic-chemical properties lead to specific aspects of the in-cylinder combustion and mechanical running of diesel engines. Comparative to the classic diesel engine, the cycle variability coefficients (COV) are presented at dual fuelling for maximum pressure, maximum pressure rise rate, maximum pressure angle, indicated mean effective pressure, heat release rate and cycle angle points of mass fraction burned. The calculated values of these parameters are bigger comparative to the normal values of diesel engines, but don’t exceed the maximum admitted values that provide engine reliability. From COV values analysis, the maximum admitted LPG cycle dose is also established. © 2016 2016The TheAuthors. Authors. Published Elsevier © Published by by Elsevier Ltd.Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the “Petru Maior” University of Tirgu-Mures, Faculty of Engineering. Peer-review under responsibility of the “Petru Maior” University of Tirgu Mures, Faculty of Engineering Keywords: LPG dose; cycle variability; indicated mean effective pressure; lean mixture; homogeneity.
1. Introduction In the actual context of pollutant emission reduction, the Liquid Petroleum Gas (LPG), a worldwide alternative fuel used on a large scale for internal combustion engines, can be a viable alternative fuel also for diesel engines. Even its properties make is suitable for spark ignition fuelling it can be use also for diesel engine fuelling if specific fuelling methods are applied [1, 2, 3]. The heat release, the energetically and pollution performance of the engine are influenced by different burning properties of the LPG comparative to diesel fuel. Higher combustion rate of air-LPG mixtures leads to the increase of maximum pressure and maximum pressure rise rate, with consequences on cycle variability. From this point of view, the maximum LPG cycle quantity admitted inside the cylinder can be limited for efficiency, mechanical reliability, noise, smooth running reasons and also because of the cycle variability
* Corresponding author. Tel.: +400723470021. E-mail address: [email protected]
2212-0173 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the “Petru Maior” University of Tirgu Mures, Faculty of Engineering doi:10.1016/j.protcy.2016.01.034
747
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
limitation. LPG raised vapour pressure assure forming of air-LPG mixtures of higher homogeneity. A burning laminar velocity comparable with the one of gasoline (0.46 m/s) [1, 4] and starting the combustion into a relative lean high homogeneity air-LPG mixture influences the combustion process and cycle variability [5, 6]. Wide inflamabillity limits of the liquid petroleum gas 2,1…10,4% versus 0,6 … 5,5% for diesel fuel will influence the combustion cycle variability and may leads to an improvement of the combustion process and of the pollutant emissions [1, 4, 5, 7, 8, 9]. The cycle variability of air-LPG mixtures combustion is analyzed comparative to diesel fuelling for a 1,5 litre direct injection supercharged diesel engine, type K9K from Dacia Logan. The engine is fuelled with diesel fuel and liquid petroleum gas by diesel gas method. The LPG dose is injected inside the inlet manifold and electronic controlled by a secondary electronic control unit (ECU) connected back to back with the main engine ECU. 2. Study of Diesel Engine Cycle Variability for LPG Fuelling The cycle variability can be characterized by coefficients of in-cylinder pressure variation. The intensity of the cycle variety phenomena is defined by the coefficient of cycle variability (COV), as relation (1) shows. The coefficient of cycle variability is defined as a relative average deviation of maximum values of the studied parameters, as pressure [10]. For n consecutive cycles, if is considered a normal distribution of the deviation probabilities, the squared average deviation can be calculated and the cycle variability coefficient is defined as: n
¦ai
n
¦ (a i − i =1n
)2
i =1
(COV )a i =
n −1 n
¦ai
⋅100%
(1)
i =1
n where n is the number of cycles (50 consecutive cycles), a is the parameter of which variability is study and is defined by indicated mean effective pressure (IMEP), maximum pressure pmax, maximum pressure rise rate (dp/dα)max, maximum of heat release rate (dξ/dα)max and the angle where maximum pressure occurs, αpmax in the cycle number “i” [10]. Generally, for the cycle variability evaluation for regimes with injection timing closer to the value of injection timing for maximum torque brake (MTB), the coefficient of variation for maximum pressure is suitable, (COV)pmax. The COV of maximum pressure angle, (COV)αpmax , the moment per cycle when the maximum pressure occurs, is used for characterization of the combustion cycle variability during the initial phase of combustion. The variation of the indicated mean effective pressure (IMEP), appreciated by (COV)IMEP, is the most suitable instrument to define the engine response to the combustion process variability [10]. From this point of view the limit value of (COV)IMEP define practically the limit of mixture leaning. This cycle coefficient can also indicate the variability of flame development during the rapid phase of combustion. The physical-chemical properties of LPG show the great perspectives for spark ignition engine use (a high octane number, a laminar flame velocity closer to gasoline laminar flame velocity, wide inflamability limits) and difficulties for compression ignition engine use (a lower cetane number, high autoignition temperature) fact that implies special solutions for mixture forming and combustion control and also influences the cycle variability of the combustion process [1, 2, 3, 6, 7]. The laminar flame velocity of air-LPG mixtures which is comparable with the laminar flame velocity of air-gasoline mixtures will leads to the acceleration of the combustion process with effects of maximum pressure rise rate and on heat release rate [1, 2]. The air-LPG mixture in a higher state of homogeneity, due to propane higher vapour pressure will leads to a high velocity combustion process, after the flame nucleus is forming inside the cylinder next to the diesel fuel jet boundary [1, 4, 5, 7]. The initial very lean mixtures of air-LPG achieved inside the cylinder before combustion starts will influence the combustion cycle variability and engine mechanical running. The
748
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
quantity of LPG injected into the inlet manifold is established by an energetically substitute ratio xc which take into consideration the lower heating values of both fuels:
xc =
Ch LPG ⋅ Hi LPG ⋅100 Ch DF ⋅ Hi DF + Ch LPG ⋅ Hi LPG
(2)
where: ChDF, ChLPG diesel fuel and LPG fuel consumptions, HiDF, HiLPG lower heating values of diesel fuel and LPG. a
minimum 170
maximum
b
mean
minimum
mean
150 p max [bar]
160 p m ax [bar]
maximum
155
150 140 130
145 140 135 130 125
120
120
0
10
20
xc [%]
30
40
50
0
2
4
xc [%]
6
8
10
Fig. 1. (a) Maximum pressure versus xc at 2000 min-1, 100% load; (b) Maximum pressure versus xc at 4000 min-1, 100% load.
For different substitute ratios of diesel fuel by LPG, were analyzed individual pressure and heat release diagrams, in terms of maximum pressure, maximum pressure rise rate and heat release rate. For full load, 2000 and 4000 min-1 regime, the increase of LPG cycle dose leads to the increase of cycle maximum pressure and maximum pressure cycle dispersion, thus the dispersion of successive running cycle’s increases, as Fig. 1 shows. a
maximum
b
mean
minimum
maximum
mean
17
23 22 21 20 19 18 17 16 15 14
16 α p max [CAD]
α p max [CAD]
minimum
15 14 13 12 11
0
2
4
xc [%]
6
8
10
0
10
20
xc [%]
30
40
50
Fig. 2. (a) Maximum pressure angle versus xc at 2000 min-1, 100% load; (b) Maximum pressure angle versus xc at 4000 min-1, 100% load.
Also, the angle value of maximum pressure, αpmax, is registered closer to the Top Dead Centre (TDC), with is generally related to a rapid, brutal combustion, Fig. 2. At 2000 min-1, the dispersions between the values registered in consecutive cycles for maximum angle pressure tends to decrease for average substitute ratios comparative to diesel fuelling. For 4000 min-1, the significant quantities of LPG will lead to the increasing of dispersion between values of maximum pressure angle comparative to diesel fuelling especially for xc=30…40. The cycle variability is accentuated because of the initial lean air-LPG mixtures formed inside the cylinder before combustion start. Thus,
749
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
the combustion will starts in a relative lean mixture that amplifies the cycle variability. This aspect is related with the increase of maximum pressure values and of maximum pressure cycle variability when the LPG dose increases. a
mean
minimum
mean
b
maximum
minimum
maximum
10
7
(dp/d α ) max [bar/CAD]
(dp/d α)max [bar/CAD]
7.5 6.5 6 5.5 5 4.5 4
9 8 7 6 5 4 3
3.5 0
2
4
xc [%]
6
8
10
0
10
20
xc [%]
30
40
50
Fig. 3. (a) Maximum pressure rise rate versus xc at 2000 min-1,100% load; (b) Maximum pressure rise rate versus xc at 4000 min-1,100% load.
The forming of higher homogeneity air-LPG mixture leads to higher combustion velocity, during the performing mixtures combustion phase, by kinetics of chemical reactions at low temperatures. This fact implies the increase of in-cylinder maximum pressure with 10% at maximum diesel-LPG substitute ratio, and also the increase of maximum pressure rise rate value, Fig. 1 and Fig. 3. For this operating regime the values registered for averaged cycles, shown as “mean”, are completed with the minimum and maximum values registered in the individual cycles, shown as “minimum” and “maximum”, respectively. The increase of maximum pressure rise rate is maximum for xc=9,25 and is with 68% bigger comparative to the reference value. This maximum pressure rise rate increasing will be limited thru the value of substitute ratio at xc=10, at maximum torque regime, leading to values that won’t exceeded 7 bar/CAD. For the other xc substitute ratios, the increasing percents in maximum pressure rise rate are much lower, 23% for xc=2,59 and 33,8% for xc=4,48, respectively. The lower increasing comparative to the reference regime is for xc=5,98 with 18%, with an absolute value of 4,5 bar/CAD [7]. At 4000 min-1 speed regime, the LPG cycle quantity is much bigger and the in-cylinder pressure reaches values with 23% higher comparative to the diesel fuelling, for xc=40. Values of 160 bar registered in some cycles, leads to necessity of substitute ratio limitation in order to maintain engine reliability. From xc=20, the cycle dispersion of maximum pressure angle increases comparative to maximum pressure values, Fig. 1. Also the maximum pressure rise rate registered raised values comparative to the only diesel fuelling, Fig. 3. The most important increase of 130% is for xc=40 substitute value, for which the maximum pressure rise rate is in the area of 8…10 bar/CAD. For the other substitute ratios the increasing percents for maximum pressure rise rate remains significant: 74% for xc=18,3 and 82% for xc=29,5 versus diesel fuel. For a substitute ratio of xc=20…35, the maximum pressure rise rate values are in the area of xc=9,25 from maximum torque speed, 2000 min-1. For xc=40, the maximum pressure rise rate climbs to higher values around 10 bar/CAD. This fact becomes a new limitation criterion for the substitute ratio value. The limitation of xc is also required by variability coefficients in terms of maximum pressure rise rate (COV)(dp/dα)max, 5.129% at xc=0, 2.761% at xc=2.58, 3,624% at xc=5,98%, 4,124 at xc =9,25 for 2000 min-1 and 8,496% at xc=0, 6,641% at xc=18,3, 9,207% at xc=29,5, 7,835% at xc=40% for 4000 min-1. Following the aspects that lead to the establishing of a maximum substitution ratio, can be also taken into consideration the values of cycle variability coefficients for incylinder maximum pressure and indicated mean effective pressure, (COV)pmax and (COV)IMEP, respectively. In order to evaluate the engine response to cycle variability of the combustion process for different LPG cycle quantities, the coefficient of variation for indicated mean effective pressure and maximum pressure were calculated and presented versus xc values. The cycle variability can be very well appreciated by the values of these two coefficients and they are related with the engine adaptability for traction [10]. The normal automotive manoeuvrability is assured if (COV)pmax, respectively (COV)IMEP, don’t exceeded 10%. [10]. The coefficient (COV)pmax don’t exceed values of 4% at 2000 min-1 speed and full load regime, Fig. 4(a), being registered a slightly increases of the coefficient value with the LPG dose.
750
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
a
b
4
3.5 (COV)pmax [%]
(COV)pmax [%]
3.5 3 2.5 2 1.5 1
3 2.5 2 1.5 1 0.5
0.5
0
0 0
2
4
6
8
10
0
10
20
xc [%]
xc [%]
30
40
50
Fig. 4. (a) Coefficient (COV)pmax versus xc at 2000 min-1, 100 % load; (b) Coefficient (COV)pmax versus xc at 4000 min-1, 100 % load.
At 4000 min-1 speed regime, the higher dose of LPG leads to a much significant continuously increase of (COV)pmax. Even the maximum value doesn’t exceed 3%, with no reason for xc limitation, the further increases tendency of (COV)pmax with xc, especially for xc>30%, can be take into consideration, Fig. 4(b). For MTB speed regime, the IMEP cycle variability, (COV)IMEP, is 3 times higher for maximum LPG cycle dose comparative to diesel fuelling, Fig. 5(a). From this point of view the substitute ratio limitation at xc = 9.25% becomes obvious. If is admitted as limit value [(COV)IMEP ]max = 10%, a limit value in substitute ratio of xc = 8% results [7]. For xc=40% the value of coefficient (COV)IMEP is 14 times bigger comparative to reference value, as Fig. 5(b) shows. a
b
14
14 (C O V ) IM E P [% ]
(COV) IMEP [%]
12 10 8 6 4
12 10 8 6 4 2
2
0
0 0
2
4
xc [%]
6
8
10
0
10
20
30
40
50
xc [%]
Fig. 5. (a) Coefficient (COV)IMEP versus xc at 2000 min-1, 100% load; (b) Coefficient (COV)IMEP versus xc at 4000 min-1, 100% load.
Thus, become necessary the limitation of the LPG dose at 40% in order to maintain the normal engine running for dual fuelling. If the admitted limit is [(COV)IMEP]max=10%, then the substitute ratio will be limited to xc=25%. Thus, the maximum pressure and maximum pressure rise rate are limited at pmax=150 bar and (dp/dα)max=6 bar/CAD, respectively. At 4000 min-1, the significant increases in LPG cycle dose leads to the increasing of incylinder maximum pressure rise rate due to the increase of cycle heat release, seeing that is no variation for the medium values of maximum cycle heat release (dξ/dα)max, in the investigated rage of xc, Fig. 6(b). The maximum values of the heat release rate, from the individual cycles, continuously increase. The maximum heat release obtained values are according with the variation of the maximum pressure and with the variation of the maximum pressure rise rate at the same operating regime. The maximum heat release rate diagrams present beside value from the averaged cycles, “mean”, also the minimum and maximum values from the individual cycles, “minimum” and “maximum”, respectively. The most important increase of 160% is registered for the maximum LPG dose, for the other quantities the increasing percent being 39% at xc=18.3 and 28% at xc=29.5. This higher values lead to the limitation of the LPG dose at 40%.
751
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
a
minimum
b
maximum
mean 420
350
370
(d ξ /d α ) max [J/C AD ]
(dξ/dα)max [J/CAD]
mean 400
300 250 200 150 100
minimum
maximum
320 270 220 170 120 70
50
20
0 0
2
4
xc [%]
6
8
0
10
10
20
xc [%]
30
40
50
Fig. 6. (a) Maximum heat release rate versus xc at 2000 min-1, 100% load; (b) Maximum heat release rate versus xc at 4000 min-1, 100% load.
The calculated values for maximum heat release rate cycle variability coefficients, (COV)(dξ/dα)max, are 34,1% at xc=0%, 86,22% at xc =2,58, 56,54% at xc =5,98, 52,01% at xc =9,25 for 2000 min-1 and 35,35% at xc =0, 22,85% at xc =18,3, 62,73% at xc =29,5 and 32,66% at xc =40 for 4000 min-1. a
b
80
100 CO V α 10 %
COV α 5 %
100
60 40 20 0
80 60 40 20 0
0
2
4
6
8
10
0
2
4
c
d COV α 90 %
COV α 50 %
35 30 25 20 15 10 5 0 0
2
4
6 xc [%]
6
8
10
6
8
10
xc [%]
xc [%]
8
10
40 35 30 25 20 15 10 5 0 0
2
4 xc [%]
Fig. 7. (a) COV for α5% of mass fraction burned versus xc; (b) COV for α10% of mass fraction burned versus xc; (c) COV for α50% of mass fraction burned versus xc; (d) COV for α90% of mass fraction burned versus xc at 2000 min-1, 100% load.
The position of the moments of 5%, 10%, 50% and 90% from the heat release versus TDC is directly influenced by these phenomena’s. For 2000 min-1, the conventional 5% and 10% of heat release appears much earlier comparative to diesel fuel, for all substitute ratios, in a direct proportional dependence with the increase of LPG dose. The 90% of mass fraction burned is archived later on cycle comparative to diesel fuel, because of a lower heat release quantity during the main phase, 50%. The increase tendency of combustion process variability, reflected in the variability of the angles α5, α10 and α50 with the increasing of the LPG cycle dose is also remarked. Thus, only
752
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
for diesel fuelling the values of the cyclic variability coefficients increases from (COV)α5=28.3%, (COV)α10=33%, (COV)α50=30%, (COV) α90=29.9%, to values of (COV)α5=49.4%, (COV)α10=80%, (COV)α50=31.3%, (COV) α90 =34.7% for the maximum LPG dose, xc=9.25, Fig. 7. For 4000 min-1 appears the same phenomena of an earlier release of 5% and 10% of burned mass for dual fuelling, Fig. 8. The increase of LPG dose leads to the acceleration of the heat release during combustion, the acceleration of the combustion process and of the sooner achieving on cycle of the 50% and 90% mass fraction burned, comparative to diesel fuelling. Also the amplification of the combustion cycle variability, explained by α5, α10, α50, α90 angles variability, is present when the LPG cycle dose increases. For diesel fuelling the values of coefficients (COV)α5=45.5%, (COV)α10=29.5%, (COV)α50=4.5%, (COV) α90=6.1% are much lower comparative to the values registered for diesel-LPG fuelling: (COV)α5=89.5%, (COV)α10=64.1%, (COV)α50=84.6%, (COV)α90=46.8%, values calculated for maximum LPG dose, xc=40 and presented in Fig. 8. Also this phenomenon is related with a combustion process closer to engine top dead centre and leads to the increasing of the maximum pressure rise rate during the combustion process. a
b
100
COV α 10 %
COVα 5 %
80 60 40 20 0 0
10
20
30
40
80 70 60 50 40 30 20 10 0
50
0
10
xc [%]
30
40
30
40
50
xc [%]
d
90 80 70 60 50 40 30 20 10 0
50 COV α 90 %
COV α 50 %
c
20
40 30 20 10 0
0
10
20
xc [%]
30
40
50
0
10
20
50
xc [%]
Fig. 8. COV for α5% of mass fraction burned versus xc; (b) COV for α10% of mass fraction burned versus xc; (c) COV for α50% of mass fraction burned versus xc; (d) COV for α90% of mass fraction burned versus xc at 4000 min-1, 100% load.
3. Conclusions The experimental study shows the main cycle variability characteristics of the engine running at dual fuelling comparative to classic fuelling solution. Thus, the main conclusions can be formulated: 1. The diesel fuel and LPG engine fuelling leads, comparative to classic diesel fuelling, to an increasing tendency for maximum pressure rise rate (dp/dα)max and for cycle variability (COV)IMEP. 2. The xc substitute ratio value is limited by the maximum values or by the maximum variation interval of same running characteristic parameters of the engine, as is follows: if the maximum admissible level of the cycle variability is (COV)IMEP =10%, for IMEP, then the substitute ratio will be limited to xc=8% for 2000 min-1 and to xc=25% at 4000 min-1, respectively; if the maximum pressure rise rate is limited to the value (dp/dα)max=8
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
bar/CAD, then the LPG substitute ratio will be limited to xc=10% at 2000 min-1 and to xc=40% at 4000 min-1, values that will lead to a maximum acceptable in-cylinder pressure level of 160 bar. 3. The partial substitution of diesel fuel by liquid petroleum gas leads to the increase of the heat release rate during the rapid combustion phase, showed by the raised level of the maximum heat release rate (dξ/dα)max and earlier achievement per engine cycle of 5% and 10% of mass fractions burned, from the heat release (decreasing of α5 and α10 angles values). This influence is correlated with the increasing tendency for maximum pressure rise rate values and for in-cylinder maximum pressure values. 4. The partial substitution of the diesel fuel by LPG leads to the increase of cycle variability at the beginning of the combustion process, shown by the variability of the angles Į5 and Į10 with a direct influence on cycle variability for indicated main effective pressure IMEP, (COV)IMEP and maximum pressure, (COV)pmax. The 90% mass fraction burned is archived later on cycle comparative to diesel fuel, because of a lower heat release quantity during the main phase, 50%. Also, the increased cycle variability registered at the beginning of the combustion process leads to the increase of the cycle variability at the end of combustion, explained by the variability of the Į90 angle.
Acknowledgements The work has been funded by the Sectoral Operational Programme Human Resources Development 2007-2013 of the Ministry of European Funds through the Financial Agreement POSDRU/159/1.5/S/134398.
References [1] Popa M G, Negurescu N, Pana C. Diesel Engines. Processes. Bucharest: MatrixRom; 2003. [2] Goto S, Lee D, Wakao Y, Honma H, Moiri M, Akasaka Y, Hashimoto K, Motohashi M, Konno M. Development of an LPG DI Diesel Engine Using Cetane Number Enhancing Additives. Alternative Fuels; SAE Paper 1999-01-3602;1999. SP-1482. [3] Goto S, Shakal J, Daigo H. LPG engine research and development in Japan. Int’l Congress on Transportation Electronics; 1996. [4] Pana C, Negurescu N, Popa M G, Cernat A. Experimental Aspects of the Use of LPG at Diesel Engine, Journal of Scientific Bulletin University Politehnica of Bucharest, Series D; vol.72; Iss. 1, Bucharest; 2010. pp. 25-34 [5] Goto S, Lee D, Shakal J, Harayama H, Honjyo F, Ueno H. Performance and emissions of an LPG lean-burn engine for heavy duty vehicles. SAE Paper 1999-01-1513; 1999. [6] Goto S, Lee D, Harayama H, Honjyo F, Ueno H, Honma H, Wakan Y, Mori M. Development of LPG SI and CI Engines for Heavy Duty Vehicles. F2000A171; Korea: Seoul 2000 FISITA World Automotive Congress; 2000. [7] Cernat A. Contributions to the Study of a Diesel Engine Running on Diesel Fuel and Liquid Petroleum Gas. Doctoral Thesis; University Politehnica of Bucharest; 2010. [8] Pana C, Negurescu N, Popa M G, Cernat A. Ecologic Automotive Diesel Engine Achieved by LPG Fueling. Journal of Environmental Protection and Ecology; Vol. 13; No. 2; 2012. 6pp. 88–699 [9] Pană C, Negurescu N, Popa M G, Boboc G, Cernat A. Reduction of NOx, Smoke and BSFC in a Diesel Engine Fueled with LPG. Czech Republic, Praha; MECCA Journal of Midle European Construction and Design of Cars; Vol III; No. 4; 2005. pp. 37- 43 [10] Heywood J B. Internal Combustion Engines Fundamentals. New York: McGraw-Hill Book Company; 1988.
753
ScienceDirect Procedia Technology 22 (2016) 746 – 753
9th International Conference Interdisciplinarity in Engineering, INTER-ENG 2015, 8-9 October 2015, Tirgu-Mures, Romania
The Influence of LPG Fuelling on Diesel Engine Cycle Variability Alexandru Cernata,*, Constantin Panaa, Niculae Negurescua and Cristian Nutua a
University Politehnica of Bucharest, Blvd. Splaiul Independentei no. 313, Bucharest 70000, Romania
Abstract Liquid Petroleum Gas physic-chemical properties lead to specific aspects of the in-cylinder combustion and mechanical running of diesel engines. Comparative to the classic diesel engine, the cycle variability coefficients (COV) are presented at dual fuelling for maximum pressure, maximum pressure rise rate, maximum pressure angle, indicated mean effective pressure, heat release rate and cycle angle points of mass fraction burned. The calculated values of these parameters are bigger comparative to the normal values of diesel engines, but don’t exceed the maximum admitted values that provide engine reliability. From COV values analysis, the maximum admitted LPG cycle dose is also established. © 2016 2016The TheAuthors. Authors. Published Elsevier © Published by by Elsevier Ltd.Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the “Petru Maior” University of Tirgu-Mures, Faculty of Engineering. Peer-review under responsibility of the “Petru Maior” University of Tirgu Mures, Faculty of Engineering Keywords: LPG dose; cycle variability; indicated mean effective pressure; lean mixture; homogeneity.
1. Introduction In the actual context of pollutant emission reduction, the Liquid Petroleum Gas (LPG), a worldwide alternative fuel used on a large scale for internal combustion engines, can be a viable alternative fuel also for diesel engines. Even its properties make is suitable for spark ignition fuelling it can be use also for diesel engine fuelling if specific fuelling methods are applied [1, 2, 3]. The heat release, the energetically and pollution performance of the engine are influenced by different burning properties of the LPG comparative to diesel fuel. Higher combustion rate of air-LPG mixtures leads to the increase of maximum pressure and maximum pressure rise rate, with consequences on cycle variability. From this point of view, the maximum LPG cycle quantity admitted inside the cylinder can be limited for efficiency, mechanical reliability, noise, smooth running reasons and also because of the cycle variability
* Corresponding author. Tel.: +400723470021. E-mail address: [email protected]
2212-0173 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the “Petru Maior” University of Tirgu Mures, Faculty of Engineering doi:10.1016/j.protcy.2016.01.034
747
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
limitation. LPG raised vapour pressure assure forming of air-LPG mixtures of higher homogeneity. A burning laminar velocity comparable with the one of gasoline (0.46 m/s) [1, 4] and starting the combustion into a relative lean high homogeneity air-LPG mixture influences the combustion process and cycle variability [5, 6]. Wide inflamabillity limits of the liquid petroleum gas 2,1…10,4% versus 0,6 … 5,5% for diesel fuel will influence the combustion cycle variability and may leads to an improvement of the combustion process and of the pollutant emissions [1, 4, 5, 7, 8, 9]. The cycle variability of air-LPG mixtures combustion is analyzed comparative to diesel fuelling for a 1,5 litre direct injection supercharged diesel engine, type K9K from Dacia Logan. The engine is fuelled with diesel fuel and liquid petroleum gas by diesel gas method. The LPG dose is injected inside the inlet manifold and electronic controlled by a secondary electronic control unit (ECU) connected back to back with the main engine ECU. 2. Study of Diesel Engine Cycle Variability for LPG Fuelling The cycle variability can be characterized by coefficients of in-cylinder pressure variation. The intensity of the cycle variety phenomena is defined by the coefficient of cycle variability (COV), as relation (1) shows. The coefficient of cycle variability is defined as a relative average deviation of maximum values of the studied parameters, as pressure [10]. For n consecutive cycles, if is considered a normal distribution of the deviation probabilities, the squared average deviation can be calculated and the cycle variability coefficient is defined as: n
¦ai
n
¦ (a i − i =1n
)2
i =1
(COV )a i =
n −1 n
¦ai
⋅100%
(1)
i =1
n where n is the number of cycles (50 consecutive cycles), a is the parameter of which variability is study and is defined by indicated mean effective pressure (IMEP), maximum pressure pmax, maximum pressure rise rate (dp/dα)max, maximum of heat release rate (dξ/dα)max and the angle where maximum pressure occurs, αpmax in the cycle number “i” [10]. Generally, for the cycle variability evaluation for regimes with injection timing closer to the value of injection timing for maximum torque brake (MTB), the coefficient of variation for maximum pressure is suitable, (COV)pmax. The COV of maximum pressure angle, (COV)αpmax , the moment per cycle when the maximum pressure occurs, is used for characterization of the combustion cycle variability during the initial phase of combustion. The variation of the indicated mean effective pressure (IMEP), appreciated by (COV)IMEP, is the most suitable instrument to define the engine response to the combustion process variability [10]. From this point of view the limit value of (COV)IMEP define practically the limit of mixture leaning. This cycle coefficient can also indicate the variability of flame development during the rapid phase of combustion. The physical-chemical properties of LPG show the great perspectives for spark ignition engine use (a high octane number, a laminar flame velocity closer to gasoline laminar flame velocity, wide inflamability limits) and difficulties for compression ignition engine use (a lower cetane number, high autoignition temperature) fact that implies special solutions for mixture forming and combustion control and also influences the cycle variability of the combustion process [1, 2, 3, 6, 7]. The laminar flame velocity of air-LPG mixtures which is comparable with the laminar flame velocity of air-gasoline mixtures will leads to the acceleration of the combustion process with effects of maximum pressure rise rate and on heat release rate [1, 2]. The air-LPG mixture in a higher state of homogeneity, due to propane higher vapour pressure will leads to a high velocity combustion process, after the flame nucleus is forming inside the cylinder next to the diesel fuel jet boundary [1, 4, 5, 7]. The initial very lean mixtures of air-LPG achieved inside the cylinder before combustion starts will influence the combustion cycle variability and engine mechanical running. The
748
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
quantity of LPG injected into the inlet manifold is established by an energetically substitute ratio xc which take into consideration the lower heating values of both fuels:
xc =
Ch LPG ⋅ Hi LPG ⋅100 Ch DF ⋅ Hi DF + Ch LPG ⋅ Hi LPG
(2)
where: ChDF, ChLPG diesel fuel and LPG fuel consumptions, HiDF, HiLPG lower heating values of diesel fuel and LPG. a
minimum 170
maximum
b
mean
minimum
mean
150 p max [bar]
160 p m ax [bar]
maximum
155
150 140 130
145 140 135 130 125
120
120
0
10
20
xc [%]
30
40
50
0
2
4
xc [%]
6
8
10
Fig. 1. (a) Maximum pressure versus xc at 2000 min-1, 100% load; (b) Maximum pressure versus xc at 4000 min-1, 100% load.
For different substitute ratios of diesel fuel by LPG, were analyzed individual pressure and heat release diagrams, in terms of maximum pressure, maximum pressure rise rate and heat release rate. For full load, 2000 and 4000 min-1 regime, the increase of LPG cycle dose leads to the increase of cycle maximum pressure and maximum pressure cycle dispersion, thus the dispersion of successive running cycle’s increases, as Fig. 1 shows. a
maximum
b
mean
minimum
maximum
mean
17
23 22 21 20 19 18 17 16 15 14
16 α p max [CAD]
α p max [CAD]
minimum
15 14 13 12 11
0
2
4
xc [%]
6
8
10
0
10
20
xc [%]
30
40
50
Fig. 2. (a) Maximum pressure angle versus xc at 2000 min-1, 100% load; (b) Maximum pressure angle versus xc at 4000 min-1, 100% load.
Also, the angle value of maximum pressure, αpmax, is registered closer to the Top Dead Centre (TDC), with is generally related to a rapid, brutal combustion, Fig. 2. At 2000 min-1, the dispersions between the values registered in consecutive cycles for maximum angle pressure tends to decrease for average substitute ratios comparative to diesel fuelling. For 4000 min-1, the significant quantities of LPG will lead to the increasing of dispersion between values of maximum pressure angle comparative to diesel fuelling especially for xc=30…40. The cycle variability is accentuated because of the initial lean air-LPG mixtures formed inside the cylinder before combustion start. Thus,
749
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
the combustion will starts in a relative lean mixture that amplifies the cycle variability. This aspect is related with the increase of maximum pressure values and of maximum pressure cycle variability when the LPG dose increases. a
mean
minimum
mean
b
maximum
minimum
maximum
10
7
(dp/d α ) max [bar/CAD]
(dp/d α)max [bar/CAD]
7.5 6.5 6 5.5 5 4.5 4
9 8 7 6 5 4 3
3.5 0
2
4
xc [%]
6
8
10
0
10
20
xc [%]
30
40
50
Fig. 3. (a) Maximum pressure rise rate versus xc at 2000 min-1,100% load; (b) Maximum pressure rise rate versus xc at 4000 min-1,100% load.
The forming of higher homogeneity air-LPG mixture leads to higher combustion velocity, during the performing mixtures combustion phase, by kinetics of chemical reactions at low temperatures. This fact implies the increase of in-cylinder maximum pressure with 10% at maximum diesel-LPG substitute ratio, and also the increase of maximum pressure rise rate value, Fig. 1 and Fig. 3. For this operating regime the values registered for averaged cycles, shown as “mean”, are completed with the minimum and maximum values registered in the individual cycles, shown as “minimum” and “maximum”, respectively. The increase of maximum pressure rise rate is maximum for xc=9,25 and is with 68% bigger comparative to the reference value. This maximum pressure rise rate increasing will be limited thru the value of substitute ratio at xc=10, at maximum torque regime, leading to values that won’t exceeded 7 bar/CAD. For the other xc substitute ratios, the increasing percents in maximum pressure rise rate are much lower, 23% for xc=2,59 and 33,8% for xc=4,48, respectively. The lower increasing comparative to the reference regime is for xc=5,98 with 18%, with an absolute value of 4,5 bar/CAD [7]. At 4000 min-1 speed regime, the LPG cycle quantity is much bigger and the in-cylinder pressure reaches values with 23% higher comparative to the diesel fuelling, for xc=40. Values of 160 bar registered in some cycles, leads to necessity of substitute ratio limitation in order to maintain engine reliability. From xc=20, the cycle dispersion of maximum pressure angle increases comparative to maximum pressure values, Fig. 1. Also the maximum pressure rise rate registered raised values comparative to the only diesel fuelling, Fig. 3. The most important increase of 130% is for xc=40 substitute value, for which the maximum pressure rise rate is in the area of 8…10 bar/CAD. For the other substitute ratios the increasing percents for maximum pressure rise rate remains significant: 74% for xc=18,3 and 82% for xc=29,5 versus diesel fuel. For a substitute ratio of xc=20…35, the maximum pressure rise rate values are in the area of xc=9,25 from maximum torque speed, 2000 min-1. For xc=40, the maximum pressure rise rate climbs to higher values around 10 bar/CAD. This fact becomes a new limitation criterion for the substitute ratio value. The limitation of xc is also required by variability coefficients in terms of maximum pressure rise rate (COV)(dp/dα)max, 5.129% at xc=0, 2.761% at xc=2.58, 3,624% at xc=5,98%, 4,124 at xc =9,25 for 2000 min-1 and 8,496% at xc=0, 6,641% at xc=18,3, 9,207% at xc=29,5, 7,835% at xc=40% for 4000 min-1. Following the aspects that lead to the establishing of a maximum substitution ratio, can be also taken into consideration the values of cycle variability coefficients for incylinder maximum pressure and indicated mean effective pressure, (COV)pmax and (COV)IMEP, respectively. In order to evaluate the engine response to cycle variability of the combustion process for different LPG cycle quantities, the coefficient of variation for indicated mean effective pressure and maximum pressure were calculated and presented versus xc values. The cycle variability can be very well appreciated by the values of these two coefficients and they are related with the engine adaptability for traction [10]. The normal automotive manoeuvrability is assured if (COV)pmax, respectively (COV)IMEP, don’t exceeded 10%. [10]. The coefficient (COV)pmax don’t exceed values of 4% at 2000 min-1 speed and full load regime, Fig. 4(a), being registered a slightly increases of the coefficient value with the LPG dose.
750
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
a
b
4
3.5 (COV)pmax [%]
(COV)pmax [%]
3.5 3 2.5 2 1.5 1
3 2.5 2 1.5 1 0.5
0.5
0
0 0
2
4
6
8
10
0
10
20
xc [%]
xc [%]
30
40
50
Fig. 4. (a) Coefficient (COV)pmax versus xc at 2000 min-1, 100 % load; (b) Coefficient (COV)pmax versus xc at 4000 min-1, 100 % load.
At 4000 min-1 speed regime, the higher dose of LPG leads to a much significant continuously increase of (COV)pmax. Even the maximum value doesn’t exceed 3%, with no reason for xc limitation, the further increases tendency of (COV)pmax with xc, especially for xc>30%, can be take into consideration, Fig. 4(b). For MTB speed regime, the IMEP cycle variability, (COV)IMEP, is 3 times higher for maximum LPG cycle dose comparative to diesel fuelling, Fig. 5(a). From this point of view the substitute ratio limitation at xc = 9.25% becomes obvious. If is admitted as limit value [(COV)IMEP ]max = 10%, a limit value in substitute ratio of xc = 8% results [7]. For xc=40% the value of coefficient (COV)IMEP is 14 times bigger comparative to reference value, as Fig. 5(b) shows. a
b
14
14 (C O V ) IM E P [% ]
(COV) IMEP [%]
12 10 8 6 4
12 10 8 6 4 2
2
0
0 0
2
4
xc [%]
6
8
10
0
10
20
30
40
50
xc [%]
Fig. 5. (a) Coefficient (COV)IMEP versus xc at 2000 min-1, 100% load; (b) Coefficient (COV)IMEP versus xc at 4000 min-1, 100% load.
Thus, become necessary the limitation of the LPG dose at 40% in order to maintain the normal engine running for dual fuelling. If the admitted limit is [(COV)IMEP]max=10%, then the substitute ratio will be limited to xc=25%. Thus, the maximum pressure and maximum pressure rise rate are limited at pmax=150 bar and (dp/dα)max=6 bar/CAD, respectively. At 4000 min-1, the significant increases in LPG cycle dose leads to the increasing of incylinder maximum pressure rise rate due to the increase of cycle heat release, seeing that is no variation for the medium values of maximum cycle heat release (dξ/dα)max, in the investigated rage of xc, Fig. 6(b). The maximum values of the heat release rate, from the individual cycles, continuously increase. The maximum heat release obtained values are according with the variation of the maximum pressure and with the variation of the maximum pressure rise rate at the same operating regime. The maximum heat release rate diagrams present beside value from the averaged cycles, “mean”, also the minimum and maximum values from the individual cycles, “minimum” and “maximum”, respectively. The most important increase of 160% is registered for the maximum LPG dose, for the other quantities the increasing percent being 39% at xc=18.3 and 28% at xc=29.5. This higher values lead to the limitation of the LPG dose at 40%.
751
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
a
minimum
b
maximum
mean 420
350
370
(d ξ /d α ) max [J/C AD ]
(dξ/dα)max [J/CAD]
mean 400
300 250 200 150 100
minimum
maximum
320 270 220 170 120 70
50
20
0 0
2
4
xc [%]
6
8
0
10
10
20
xc [%]
30
40
50
Fig. 6. (a) Maximum heat release rate versus xc at 2000 min-1, 100% load; (b) Maximum heat release rate versus xc at 4000 min-1, 100% load.
The calculated values for maximum heat release rate cycle variability coefficients, (COV)(dξ/dα)max, are 34,1% at xc=0%, 86,22% at xc =2,58, 56,54% at xc =5,98, 52,01% at xc =9,25 for 2000 min-1 and 35,35% at xc =0, 22,85% at xc =18,3, 62,73% at xc =29,5 and 32,66% at xc =40 for 4000 min-1. a
b
80
100 CO V α 10 %
COV α 5 %
100
60 40 20 0
80 60 40 20 0
0
2
4
6
8
10
0
2
4
c
d COV α 90 %
COV α 50 %
35 30 25 20 15 10 5 0 0
2
4
6 xc [%]
6
8
10
6
8
10
xc [%]
xc [%]
8
10
40 35 30 25 20 15 10 5 0 0
2
4 xc [%]
Fig. 7. (a) COV for α5% of mass fraction burned versus xc; (b) COV for α10% of mass fraction burned versus xc; (c) COV for α50% of mass fraction burned versus xc; (d) COV for α90% of mass fraction burned versus xc at 2000 min-1, 100% load.
The position of the moments of 5%, 10%, 50% and 90% from the heat release versus TDC is directly influenced by these phenomena’s. For 2000 min-1, the conventional 5% and 10% of heat release appears much earlier comparative to diesel fuel, for all substitute ratios, in a direct proportional dependence with the increase of LPG dose. The 90% of mass fraction burned is archived later on cycle comparative to diesel fuel, because of a lower heat release quantity during the main phase, 50%. The increase tendency of combustion process variability, reflected in the variability of the angles α5, α10 and α50 with the increasing of the LPG cycle dose is also remarked. Thus, only
752
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
for diesel fuelling the values of the cyclic variability coefficients increases from (COV)α5=28.3%, (COV)α10=33%, (COV)α50=30%, (COV) α90=29.9%, to values of (COV)α5=49.4%, (COV)α10=80%, (COV)α50=31.3%, (COV) α90 =34.7% for the maximum LPG dose, xc=9.25, Fig. 7. For 4000 min-1 appears the same phenomena of an earlier release of 5% and 10% of burned mass for dual fuelling, Fig. 8. The increase of LPG dose leads to the acceleration of the heat release during combustion, the acceleration of the combustion process and of the sooner achieving on cycle of the 50% and 90% mass fraction burned, comparative to diesel fuelling. Also the amplification of the combustion cycle variability, explained by α5, α10, α50, α90 angles variability, is present when the LPG cycle dose increases. For diesel fuelling the values of coefficients (COV)α5=45.5%, (COV)α10=29.5%, (COV)α50=4.5%, (COV) α90=6.1% are much lower comparative to the values registered for diesel-LPG fuelling: (COV)α5=89.5%, (COV)α10=64.1%, (COV)α50=84.6%, (COV)α90=46.8%, values calculated for maximum LPG dose, xc=40 and presented in Fig. 8. Also this phenomenon is related with a combustion process closer to engine top dead centre and leads to the increasing of the maximum pressure rise rate during the combustion process. a
b
100
COV α 10 %
COVα 5 %
80 60 40 20 0 0
10
20
30
40
80 70 60 50 40 30 20 10 0
50
0
10
xc [%]
30
40
30
40
50
xc [%]
d
90 80 70 60 50 40 30 20 10 0
50 COV α 90 %
COV α 50 %
c
20
40 30 20 10 0
0
10
20
xc [%]
30
40
50
0
10
20
50
xc [%]
Fig. 8. COV for α5% of mass fraction burned versus xc; (b) COV for α10% of mass fraction burned versus xc; (c) COV for α50% of mass fraction burned versus xc; (d) COV for α90% of mass fraction burned versus xc at 4000 min-1, 100% load.
3. Conclusions The experimental study shows the main cycle variability characteristics of the engine running at dual fuelling comparative to classic fuelling solution. Thus, the main conclusions can be formulated: 1. The diesel fuel and LPG engine fuelling leads, comparative to classic diesel fuelling, to an increasing tendency for maximum pressure rise rate (dp/dα)max and for cycle variability (COV)IMEP. 2. The xc substitute ratio value is limited by the maximum values or by the maximum variation interval of same running characteristic parameters of the engine, as is follows: if the maximum admissible level of the cycle variability is (COV)IMEP =10%, for IMEP, then the substitute ratio will be limited to xc=8% for 2000 min-1 and to xc=25% at 4000 min-1, respectively; if the maximum pressure rise rate is limited to the value (dp/dα)max=8
Alexandru Cernat et al. / Procedia Technology 22 (2016) 746 – 753
bar/CAD, then the LPG substitute ratio will be limited to xc=10% at 2000 min-1 and to xc=40% at 4000 min-1, values that will lead to a maximum acceptable in-cylinder pressure level of 160 bar. 3. The partial substitution of diesel fuel by liquid petroleum gas leads to the increase of the heat release rate during the rapid combustion phase, showed by the raised level of the maximum heat release rate (dξ/dα)max and earlier achievement per engine cycle of 5% and 10% of mass fractions burned, from the heat release (decreasing of α5 and α10 angles values). This influence is correlated with the increasing tendency for maximum pressure rise rate values and for in-cylinder maximum pressure values. 4. The partial substitution of the diesel fuel by LPG leads to the increase of cycle variability at the beginning of the combustion process, shown by the variability of the angles Į5 and Į10 with a direct influence on cycle variability for indicated main effective pressure IMEP, (COV)IMEP and maximum pressure, (COV)pmax. The 90% mass fraction burned is archived later on cycle comparative to diesel fuel, because of a lower heat release quantity during the main phase, 50%. Also, the increased cycle variability registered at the beginning of the combustion process leads to the increase of the cycle variability at the end of combustion, explained by the variability of the Į90 angle.
Acknowledgements The work has been funded by the Sectoral Operational Programme Human Resources Development 2007-2013 of the Ministry of European Funds through the Financial Agreement POSDRU/159/1.5/S/134398.
References [1] Popa M G, Negurescu N, Pana C. Diesel Engines. Processes. Bucharest: MatrixRom; 2003. [2] Goto S, Lee D, Wakao Y, Honma H, Moiri M, Akasaka Y, Hashimoto K, Motohashi M, Konno M. Development of an LPG DI Diesel Engine Using Cetane Number Enhancing Additives. Alternative Fuels; SAE Paper 1999-01-3602;1999. SP-1482. [3] Goto S, Shakal J, Daigo H. LPG engine research and development in Japan. Int’l Congress on Transportation Electronics; 1996. [4] Pana C, Negurescu N, Popa M G, Cernat A. Experimental Aspects of the Use of LPG at Diesel Engine, Journal of Scientific Bulletin University Politehnica of Bucharest, Series D; vol.72; Iss. 1, Bucharest; 2010. pp. 25-34 [5] Goto S, Lee D, Shakal J, Harayama H, Honjyo F, Ueno H. Performance and emissions of an LPG lean-burn engine for heavy duty vehicles. SAE Paper 1999-01-1513; 1999. [6] Goto S, Lee D, Harayama H, Honjyo F, Ueno H, Honma H, Wakan Y, Mori M. Development of LPG SI and CI Engines for Heavy Duty Vehicles. F2000A171; Korea: Seoul 2000 FISITA World Automotive Congress; 2000. [7] Cernat A. Contributions to the Study of a Diesel Engine Running on Diesel Fuel and Liquid Petroleum Gas. Doctoral Thesis; University Politehnica of Bucharest; 2010. [8] Pana C, Negurescu N, Popa M G, Cernat A. Ecologic Automotive Diesel Engine Achieved by LPG Fueling. Journal of Environmental Protection and Ecology; Vol. 13; No. 2; 2012. 6pp. 88–699 [9] Pană C, Negurescu N, Popa M G, Boboc G, Cernat A. Reduction of NOx, Smoke and BSFC in a Diesel Engine Fueled with LPG. Czech Republic, Praha; MECCA Journal of Midle European Construction and Design of Cars; Vol III; No. 4; 2005. pp. 37- 43 [10] Heywood J B. Internal Combustion Engines Fundamentals. New York: McGraw-Hill Book Company; 1988.
753