Analysis of Combustion Process in Industrial Gas Engine with Prechamber-Based Ignition System

Home , Combustion chamber, Ignition system

energies

Article Analysis of Combustion Process in Industrial Gas Engine with Prechamber-Based Ignition System

Rafał Slefarski´ 1,*, Michał Goł˛ebiewski 1, Paweł Czyzewski˙ 1, Przemysław Grzymisławski 1 and Jacek Wawrzyniak 2 1 Poznan University of Technology, Piotrowo 3 Str., Chair of Thermal Engineering, 60-965 Poznan, Poland; [email protected] (M.G.); [email protected] (P.C.); [email protected] (P.G.) 2 Polish Oil & Gas Company, Branch KRIO Odolanow, Krotoszynska 148, 63-430 Odolanow, Poland; [email protected] * Correspondence: [email protected]; Tel.: +48-61-665-22-18

Received: 29 December 2017; Accepted: 25 January 2018; Published: 2 February 2018

Abstract: Application of a pre-combustion chamber (PCC) ignition system is one of the methods to improve combustion stability and reduce toxic compounds emission, especially NOx. Using PCC allows the operation of the engine at lean combustion conditions or the utilization of low calorific gaseous fuels such as syngas or biogas. The paper presents the results of an experimental study of the combustion process in two stroke, large bore, stationary gas engine GMVH 12 equipped with two spark plugs (2-SP) and a PCC ignition system. The experimental research has been performed during the normal operation of the engine in an industrial compression station. It was observed that application of PCC provides less cycle-to-cycle combustion variation (more than 10%) and nitric oxide and carbon monoxide emissions decreased to 60% and 26% respectively. The total hydrocarbon (THC) emission rate is 25% higher for the engine equipped with PCC, which results in roughly two percent engine efficiency decrease. Another important criterion of engine retrofitting was the PCC location in the engine head. The experimental results show that improvement of engine operating parameters was recorded only for a configuration with one port offset by 45◦ from the axis of the main chamber. The study of the ignition delay angle and equivalence ratio in PCC did not demonstrate explicit influence on engine performance.

Keywords: prechamber; combustion stability; two-stroke engine; NO emission; THC emission

1. Introduction Despite the introduction of modern low-emission high-speed four-stroke engines into the natural gas compression network, the share of old two-stroke type engines is still high. This is mainly because of their reliability, large base of qualified staff and significant number of such engines remaining from their wide dissemination in the past. The introduction of new regulations on the emission of toxic substances into the atmosphere, indicates the necessity to upgrade two-stroke engines of the older type in order to meet the limits. The high investment cost of purchasing new units results in the interest of two-stroke engine operators to upgrade their systems to comply with the new emission standards. The main contaminants emitted during the combustion of natural gas in these engines are nitrogen oxides (NOx) and unburned hydrocarbons. One of the most common of the old type natural gas units are two-stroke GMVH engines. Depending on the type of configuration, operating conditions and required power it can consist of 8, 12 and 16 cylinders depending on different power, adjusted to the size of the powered gas compressors. They are used in the transmission system of natural gas transported from Russia across Poland to the countries of Western Europe as well as the national gas transmission network. About 100 units work in this system at present [1]. These engines

Energies 2018, 11, 336; doi:10.3390/en11020336 www.mdpi.com/journal/energies Energies 2018, 11, 336 2 of 15 are usually operated for more than 8000 h a year. They are used to compress natural gas after the treatment processes (cleaning raw natural gas which can contain inert gases like nitrogen or carbon dioxide) as well as to drive the public gas transport system distribution. Nitrogen oxides released into the atmosphere negatively affect humans and other living organisms, causing damages or diseases of the nervous system and contributing to the formation of smog. The formation of nitrogen oxides in the combustion process is closely related to the parameters of the combustion process and the type of fuel burned [2]. Hence, three major mechanisms of their formation in flames have been considered: the fuel mechanism, the prompt mechanism, and the thermal mechanism [3]. For natural gas burned in pressure systems with a high ratio of the amount of energy extracted from the combustion chamber unit, the main mechanism responsible for the formation of NOx is the thermal mechanism [4]. Its intensity is closely related to the temperature of the combustion and the amount of oxygen in the reaction zone [5]. Higher combustion temperatures in the reaction zone result in higher NOx emissions [6]. New emission standards set a more restrictive emission limits like theDirective for Medium Combustion Plants (MCP) [7], where for a gas engine supplied by gases fuels (natural gas or 3 non-standard gases) the NOx emission limit is 190 mg/m recalculated to oxygen content in exhaust gas equal to 15%. A vital role in fulfilling these environmental regulations is played by the application of the primary and secondary methods of toxic compounds reduction. In term of primary methods of NOx emission reduction, several approaches can be listed: creating internal or external exhaust gas recirculation, lean mixtures combustion (with high air excess ratio), modernization of ignition system burning of lean mixtures with high air excess ratio as well as modification of fuel and air supply system (high pressure of fuel injection). Internal gas recirculation can be achieved by applying different valve lift profiles for the intake valves. Research provided by Königsson et al. [8] indicated that a 20% hydrocarbon emissions reduction can be obtained by increasing the swirl number from 0.4 to 3.0. In case of external gas recirculation (EGR) Sorathia et al. [9] has proven that using this approach allows to decrease NOx emission by 45% for diesel engine when operating at full load for 15% of EGR. At the same level of EGR for GMVH-8 unit Rudkowski at al. [10] received 33% NOx reduction with decrease of overall engine efficiency by 2.75%. Moreover a 15% EGR rate allows effective reduction of NOx emission without significantly lowering the engine performance and emission [9]. The introduction of lean mixtures into the cylinder can reduce exhaust emission by lowering the combustion temperature, which prevents thermal nitrogen oxide formation and provides additional oxygen to oxidize unburned hydrocarbons and carbon monoxide. Ma et al. [11] investigated the spark ignition (SI) engine initially supplied with natural gas and its hydrogen blended mixtures influence on emissions and efficiency. For air equivalence ratio equal to 0.63 and with pure methane they achieved ten times the reduction of NOx (from 25 g/kWh to 2.5 g/kWh) compared to the case when equivalence ratio is equal to 1. The combustion of lean mixtures approach almost always involves the use of turbochargers. The use of lean air/fuel ratios is an important means of increasing engine efficiency and reducing exhaust emissions. The formation of nitrogen oxides (NOx) is primarily a function of temperature, so lower combustion temperatures resulting from lean combustion lead to reduced NOx emissions. Under lean operating conditions there is also an excess of oxygen available to oxidize carbon monoxide and unburned hydrocarbons. However, when the air-fuel ratio is directed toward the lean limit of combustion, misfire cycles occur and incomplete combustion results in rise of unburned hydrocarbons emission. A different method of increasing the lean limit is charge stratification [12], meaning that the mixture in the region of the spark plug electrodes is richer than the surrounding ultra-lean air/fuel mixture. Ma and Wang [13] investigated the extension of the lean combustion limit by introducing hydrogen to a natural gas spark-ignition engine. In the most favorable case with 50% hydrogen in supplied fuel they increased the lean combustion limit from φ = 0.59 (100% natural gas) to 0.42 [13]. Spark plug operation with ultra-lean air/fuel mixtures (φ > 0.57) and high pressures (>18 bars) is Energies 2018, 11, 336 3 of 15 problematic from the ignition point of view. In such conditions, the classic spark plug lifespan tends to be very short. In modern applications, electrode erosion is reduced by maintaining fuel concentration in the spark gap and keeping the electrode within safe operating range as well as spreading the discharge energy per electrode unit surface area over a wider space [14]. Due to difficulties in ignition of lean mixtures using spark plugs, to facilitate this process, the ignition system is often equipped with pre-combustion chambers (called also prechamber) Prechambers are usually mounted to the cylinder head, in which combustion is initiated before charge enters into the main combustion chamber. The standard pre-chamber represents about one to two percent of the clearance volume. The ignition energy in this case is about one million times higher than in standard spark plug [15]. Most prechambers work on air fuel mixtures with an equivalence ratio more than 1. It results in higher nitrogen oxides formation in prechambers and therefore has a significant effect on the overall emission of nitrogen oxides. The emission of nitrogen oxides for φ = 1.25 was more than 70% higher than in case with φ = 0.95 (the drop from 155 ppm to 40 ppm referred to 15% oxygen in exhaust gases) [16]. In very lean combustion, the vast majority of nitrogen oxides are formed in prechamber. Burning of leaner mixtures in pre-combustion chamber (PCC) results in lower NOx emission but affects the stable operation. Tozzi et al. [17] investigated novel PCC technology with unique vortex flow providing reduction of NOx emission without compromising engine stability. They confirmed 34% reduction of NOx with a one percent gain in engine efficiency compared to 30% reduction of NOx with 7.5% loss of engine efficiency in classical PCC approach. In connection with the introduction of new emission standards in Switzerland [18], Roethlisberger and Favrat [19] successfully modernized the Liebherr G 926 TI engine powered by natural gas by using pre-chambers in place of spark plugs to reduce hydrocarbons (THC), carbon monoxide and nitrogen oxide emissions, achieving lower emission values than the new regulations. Application of prechamber autoignition potential in a lean combustion regime demonstrated negative impact on NO, THC and CO emission (increase by 202%, 45% and 30% respectively) with decrease of fuel conversion efficiency by 17% comparing to SI ignition system [20] Therefore, the use of self-ignition in the pre-chamber was not considered in this article. The analysis of the GMVH-12 engine carried out by Rojewski [21] showed irregularity of pressures indicated in the cylinders. The modification of the fuel gas supply system in the form of swirling fuel valves was performed. After applying this solution, the pressure in the cylinders became more uniform, which positively influenced the engine efficiency. A different approach for decreasing THC, CO and NOx emission in gas engines are the secondary methods. In case of NOx reduction selective catalytic or non-catalytic reduction (SCR, SNCR) can be applied. However, these solutions require additional equipment investments, reagent injection, more complicated measuring and control system, which results in a significant increase of operating expenses. Considering the fact that the most favorable results were obtained with the use of precombustion chambers, it was decided to modernize the GMVH-12 engine using this technology. The main objective of investigations presented in this article was to analyze and quantify the benefits of implementing the prechamber ignition system in a two-stroke, large bore reciprocating gas engine fueled by natural gas, especially by NO emission study. In first part of the study the engine GMVH-12 was equipped with classical spark plug (SP) ignition system while in the second stage of investigation the ECO-JET prechambers were installed. The parameters used to quantify the effect of various ignition systems on combustion process in gas engine the variance of peak pressure (COV) as well as the emission of nitric oxide (NO), carbon monoxide (CO) and total hydrocarbons (THC) were examined. In addition, the specific fuel consumption index gi was calculated for all performed engine tests. Energies 2018, 11, 336 4 of 15 Energies 2018, 11, x 4 of 15

2.2. Experimental Experimental Setup Setup

2.1.2.1. Test Test Object TheThe investigated engine, engine, type type GMVH-12, GMVH-12, is is a a twelve-cylinder, two-stroke gas engine driving three-cylinderthree-cylinder gas compressors compressors by common crankshaft system (Figure 1 1).). TheThe compositioncomposition ofof thethe fuelfuel isis presented presented in in Table Table1 1..

FigureFigure 1. 1. SchemeScheme of of Cooper–Bessemer Cooper–Bessemer GMVH-12 GMVH-12 engine [21]. [21].

TableTable 1. Fuel gas composition. Component Fraction (%) Component Fraction (%) CH4 95.677 C2H6 CH4 95.677 1.183 C3H8 C2H6 1.183 0.167 C H 0.167 C4H10 3 8 0.018 C4H10 0.018 N2 2.955 N2 2.955 The GMVH-12 engine is turbocharged with ET-18 turbochargers powered by exhaust gas comingThe from GMVH-12 engine engineoperation. is turbocharged It consists of witha centrifugal ET-18 turbochargers blower and turbines powered mounted by exhaust on gasa common coming shaft.from engineThe compression operation. Itratio consists of the of turbocharger a centrifugal is blower 1:3. The and GMVH-12 turbines mounted engine uses on a a common transversely- shaft. flushedThe compression cylinder system. ratio of It the consist turbocharger of air injection is 1:3. Theinto GMVH-12the cylinder engine at an usesangle a to transversely-ﬂushed the cylinder head. Thencylinder the system.air stream It consist turns ofand air heads injection to the into exha the cylinderust windows. at an angleCurrently, to the the cylinder GMVH-12 head. engines Then the are air suppliedstream turns with and gas heads fuel by to thegas exhaustinjection windows. valves. They Currently, are located the GMVH-12 centrally enginesin the cylinder are supplied head withand theirgas fuel open–close by gas injection drive is valves. transmitted They aremechanically located centrally by cam in followers the cylinder from head the and cam their placed open–close on the connectingdrive is transmitted rod of the mechanically shaft. Pressure by camand the followers amount from of gas the supplied cam placed to the on thevalves connecting are controlled rod of theby theshaft. GOV Pressure 10/50 Gas and Engine the amount Governor of gas based supplied on engine to the load valves and are engine controlled rotational bythe speed. GOV The 10/50 ignition Gas isEngine provided Governor with two based spark on engine plugs loadlocated and on engine the cylinder rotational head. speed. The The main ignition parameters is provided of GMVH with twoare presentedspark plugs in locatedTable 2. on the cylinder head. The main parameters of GMVH are presented in Table2. ToTo study study the the influence inﬂuence of different of different ignition ignition systems systems on combustion on combustion process process in GMVH-12 in GMVH-12 engine, duringengine, duringthe second the secondstage stageof investigations of investigations the theengine engine was was equipped equipped with with scavenged scavenged ECO-JET ECO-JET prechamber. Accurate Accurate testing testing of of these these devices devices was was ca carriedrried out out at at Colorado Colorado State State University University [15,16]. [15,16]. The cross-sectional view of ECO-JET prechamber is displayed in Figure 22.. TheThe volumevolume ofof PCCPCC usedused 3 inin the the current work is 38.9 cmcm3,, which corresponds to 1.35% of thethe clearance volume. Initially the GMVH-12GMVH-12 engine was not designed designed for operation operation with with a a PCC PCC ignition ignition system. system. Therefore, Therefore, systems systems are are pre-assembled inin placeplace ofof oneone of of the the two two spark spark plugs plugs located located in in the the cylinder cylinder head. head. These These openings openings are are positioned at an angle (Figure 3) so the gas emerging from the prechamber through the single hole is able to spread in a considerable volume of the main chamber.

Energies 2018, 11, 336 5 of 15 EnergiesEnergies 2018 2018, 11, 11, x, x 5 5of of 15 15

Table 2. Basic parameters of gas engine. positioned at an angle (Figure3)Table so the 2. gasBasic emerging parameters from of gas the engine. prechamber through the single hole is able to spreadEngineEngine inSpecification Specification a considerable volume of the main chamber. Data Data EngineEngine type type Two-stroke Two-stroke RatedRated power power Table 2. Basic parameters of gas engine. 1790 1790 kW kW IntakeIntake TurbochargerTurbocharger with with intercooler intercooler EngineNumberNumber Speciﬁcation of of cylinders cylinders 12 12 in in V Data V configuration configuration Engine type355.6355.6 × ×374.12 374.12 mm—left mm—left Two-strokebank bank of of cylinders/355.6 cylinders/355.6 × ×372.14 372.14 mm— mm— BoreBore × ×stroke stroke Rated powerrightright 1790bank bank kW of of cylinders cylinders Intake Turbocharger with intercooler CompressionCompression ratio ratio 9.62:1—left 9.62:1—left bank bank of of cy cylinders/9.60:1—rightlinders/9.60:1—right bank bank of of cylinders cylinders Number of cylinders 12 in V conﬁguration BoreFuelFuel× stroke supply supply 355.6 × 374.12 mm—left bank Direct ofDirect cylinders/355.6 injection—poppet injection—poppet× 372.14 mm—rightvalve valve bank of cylinders RatedRatedCompression crankshaft crankshaft ratio speed speed 9.62:1—left bank of cylinders/9.60:1—right 330 330 rpm rpm bank of cylinders FuelFuelFuel system system supply pressure pressure Direct injection—poppet 0.21–0.35 0.21–0.35 MPa MPa valve Rated crankshaft speed 330 rpm ChargeFuelCharge system air air pressuretemperature temperature 0.21–0.35 633 633 MPa K K ChargeChargeCharge air temperature air air pressure pressure 6330.038 0.038 K MPa MPa ExhaustExhaustCharge gases airgases pressure temperature temperature 0.038 700 MPa 700 K K CompressionCompressionExhaust gases pressure pressure temperature in in cylinder cylinder 700 2.8 2.8 K MPa MPa Compression pressure in cylinder 2.8 MPa MaximumMaximum combustion combustion combustion pressure pressure pressure 6.1 6.1 MPa 6.1 MPa MPa ConstructionConstructionConstruction year year year 1973 1973 1973

FigureFigure 2. 2. Cross CrossCross section section section of of ECO-JET ECO-JET prechamber. prechamber.

(a()a ) (b(b) )

FigureFigure 3. 3. ( a()a( )aLocation )Location Location of of spark of spark spark plug plug plug (SP) (SP) (SP) or or pre-combustion orpre-combustion pre-combustion chamber chamber chamber (PCC) (PCC) (PCC) in in cylinder cylinder in cylinder head; head; head; ( b(b) ) ConfigurationsConfigurations(b) Conﬁgurations of of investigated investigated of investigated ignition ignition ignition system. system. system.

Energies 2018, 11, 336 6 of 15

The ECO-JET prechambers are equipped with a water jacket cooling system which is connected to the external water supply system. Thanks to this solution it was possible to use standard cylinder heads which initially do not allow cooling of the prechamber.

2.2. Experimental Procedure The study of the ignition system type influence on combustion process in large bore, stationary gas engines were divided into three parts. The purpose of the first stage was to analyze the operation of the GMVH-12 engine equipped with a classic (two spark plugs per cylinder) ignition system. Parameters characterizing engine operation, like peak pressure in cylinder, indicated mean effective pressure (IMEP) [22], specific fuel consumption and emission of toxic compounds (NO, CO, THC) were investigated. The tests were done for five rotational speeds from 290 up to 330 rpm with an accuracy of ±2 rpm. The indication process of each cylinder was prepared using measuring device Ultima EDI equipped with Kistler piezoelectric combustion sensors with the total accuracy of 1.5%. The flue gas composition was measured using a set of gas analyzers (Emerson Rosemount) with the following sensors: infrared (CO, CO2), chemiluminescence (NO), paramagnetic (O2) and the Flame Ionization Detector (THC). Accuracy of gas analyzers is presented in Table3.

Table 3. Basic parameters of gas engine. THC: Total hydrocarbon.

Analyzer Range Accuracy

O2 0–25% ±1 of full scale CO2 0–30% ±1 of full scale CO 0–5000 ppm ±1 of full scale NO 0–1000 ppm ±1 of full scale THC 0–5000 ppm ±1 of full scale

In the second step of study, a separate single cylinder was analyzed. Several different configurations of the ignition system were tested. In Case 1 two spark plugs were used, in Case 2 one SP and one PCC, while in Case 3 only one PCC. In Case 4 two PCC were installed instead of spark plugs. During the investigation of one PCC (Case 3) the prechamber was installed in two different places in the cylinder head located in the ports A and B (as presented in Figure3). In addition, in Case 3 and Case 4 the influence of ignition time and different stoichiometry in PCC were tested. The ignition time has been changed by decreasing the ignition angle to −2◦ before the top dead centre (TDC). For other tests, the ignition angle was set on −5.4◦ by engine automatic control system. Further, the stoichiometry in PCC was changed by supplying additional air to ECO-JET prechamber. The compressed air before the pre-chamber inlet was mixed with fuel and as a premix mixture injected into the engine cylinder. The amount of air supplied to the prechamber during normal operation is the result of the pressure exerted by the movement of the piston and diameter of the PCC outlet opening. Therefore, it is impossible to characterize the value of equivalence air ratio. The third part of the investigation has been prepared for gas engines equipped with one pre-combustion chamber per one cylinder mounted in port A (Figure3). During this stage of investigation the same parameters characterizing engine operation as in the first step were collected for five engine rotational speeds. Obtained data was used for analysis of the influence of prechamber installation on operation parameters of engine.

3. Results and Discussion

3.1. Preliminary Test with Two Spark Plugs Ignition In the ﬁrst stage of study each cylinder of GMVH-12 gas engine was equipped with two spark plugs, where the ignition delay was controlled automatically and set on an angle equal to −5.4◦ before the TDC. Main working parameters of engine are presented in Table4. The analysis of mean effective pressure distribution and overall emission have been collected for ﬁve rotational speeds (range from Energies 2018, 11, 336 7 of 15 Energies 2018, 11, x 7 of 15

Formula290 up to (1), 330 where rpm). Inthe addition, Qb is the the fuel specific gas consumption fuel consumption presented parameter in m3/h was and calculated Ni is average according power to 3 valueFormula of the (1), engine where calculated the Qb is the based fuel on gas the consumption indicated pressure presented of 12 in mcylinders/h and in N [kW].i is average Specific power fuel consumptionvalue of the engine for the calculated two spark based plugs onconfiguration the indicated ranged pressure from of 0.274 12 cylinders Nm3/kWh in [kW].to 0.269 Specific Nm3/kWh fuel forconsumption 290 and 330 for rpm the respectively. two spark plugs configuration ranged from 0.274 Nm3/kWh to 0.269 Nm3/kWh for 290 and 330 rpm respectively. Q Nm 3 g = Q Nm[ ] (1) Nb kWh gi = [ ] (1) Ni kWh The example of combustion pressure distribution trends for 12 cylinders at 300 rpm rotational The example of combustion pressure distribution trends for 12 cylinders at 300 rpm rotational speed is demonstrated in Figure 4. For all tests, the coefficient of variation for engine before speed is demonstrated in Figure4. For all tests, the coefficient of variation for engine before modifications COVE,I was calculated as the ratio of standard deviation of the main chamber peak modifications COVE,I was calculated as the ratio of standard deviation of the main chamber peak pressures and the average main chamber peak pressure from 12 cylinders [23]. Main chamber peak pressures and the average main chamber peak pressure from 12 cylinders [23]. Main chamber peak pressure was designated on 10 work cycles from one engine cylinder. For each cylinder, a series of pressure was designated on 10 work cycles from one engine cylinder. For each cylinder, a series IMEP measurements were performed. Figure 4 displays indicated mean effective pressure for 12 of IMEP measurements were performed. Figure4 displays indicated mean effective pressure for cylinder of GMVH-12 engine. 12 cylinder of GMVH-12 engine.

Table 4. Work parameters of GMVH-12 engine. Table 4. Work parameters of GMVH-12 engine. Manifold Air Manifold Air Ignition Fuel Power Fuel Consumption Manifold Manifold Power Fuel Consumption Pressure TemperatureIgnition Timing Timing FuelInjection Injection (kW) (Nm3/h) Air Pressure Air Temperature (kW) (Nm3/h) (◦BTDC) (◦BTDC) (kPa)(kPa) (◦C) (°C) (°BTDC) (°BTDC) 1382–1554 389–422 18.6–23.6 36–38 3.75–5.1 76 1382–1554 389–422 18.6–23.6 36–38 3.75–5.1 76

Figure 4. Indicated mean effective pressure for 12 cylinders of the GMVH-12 engine, rotational speed Figure 4. Indicated mean effective pressure for 12 cylinders of the GMVH-12 engine, rotational speed 300 rpm, spark plug ignition system. 300 rpm, spark plug ignition system.

The obtained COV E,I coverscovers the the range range from from 10.9 10.9 up up to to 13 13.4..4. The The lowest lowest value value was measured for 300 rpm and maximum value at 320 rpm. It shows hi highgh combustion instabilities caused mainly by misfiring.misﬁring. Second Second group group of of analyzed analyzed parameters parameters was was overall overall emission emission of of NO, NO, CO and THC toxic compounds from the whole engine. The results are presented in Figure5 5..

Energies 2018, 11, x 8 of 15 Energies 2018, 11, 336 8 of 15

Figure 5. Overall emission for gas engine GMVH-12 with spark plugs ignition system. Figure 5. Overall emission for gas engine GMVH-12 with spark plugs ignition system.

ItIt can can be be observed that that the the values values of of all measur measureded compounds decrease decrease with with an an increase increase in in the rotationalrotational speed. NitricNitric oxideoxide emissionemission is is about about 30% 30% lower lower than than presented presented by by Rudkowski Rudkowski et al.et [al.10 ][10] for fortwo two stroke stroke gas gas engine engine GMVH-8 GMVH-8 with with EGR, EGR, but but still still four four to fiveto five times times higher higher than than emission emission limits limits for fornew new engines engines [7]. [7]. The The gas gas analyzers analyzers for THCfor THC were were calibrated calibrated using using a mixture a mixture of CH of4 andCH4 Nand2 as N span2 as spangas, thusgas, thethus displayed the displayed results forresults high for hydrocarbons high hydrocarbons emission emission refer to molar refer fractionto molar of methanefraction of in methaneflue gases. in The flue maximum gases. The value maxi ofmum THC value (0.82%) of wasTHC obtained (0.82%) for was the obtained rotational for speed the rotational equal to 290 speed rpm equalwhich to is 290 about rpm 7.8% which of the is totalabout amount 7.8% of of the fuel total delivered amount to of the fuel engine. delivered For other to the speeds, engine. the For average other speeds,value was the also average on high value level was equal also toon 7.4%. high level equal to 7.4%. TheThe analysis analysis of of literature literature and and experience experience of of indu industrialstrial practice practice has has identified identified that that there there is isa possibilitya possibility to toachieve achieve improvement improvement in inemission emission of of toxic toxic compounds compounds emission emission reduction reduction by by modernizationmodernization ofof the the engine. engine. The The modification modification of ignition of ignition system system with use with of precombustion use of precombustion chambers chamberswas selected. was selected.

3.2.3.2. Study Study of of One One Cylinder Cylinder Work Work ToTo determine the best arrangement of the ignition system, one randomly selected cylinder was was investigatedinvestigated in in the the second second stage stage of of study. study. In In th thee preliminary tests, the cylinder was equipped with twotwo spark plugs,plugs, whilewhile in in the the further further attempts attempts the the prechambers prechambers were were used. used. The The investigation investigation has beenhas beendone done at one at rotational one rotational speed speed of 320 of rpm 320 for rpm three for configurations three configurations of ignition of ignition system withsystem PCC. with Initially PCC. InitiallyGMVH-12 GMVH-12 engine is engine equipped is equipped with two with spark two plugs. spark Therefore, plugs. Therefore, it was possible it was topossible study theto study influence the influenceof PCC location of PCC in location the cylinder in the head, cylinder in place head, of spark in place plugs of ports spark on plugs the combustion ports on the stability combustion as well stabilityas emission as well of THC, as emission NO and of CO. THC, For NO all and performed CO. For tests all performed coefficient oftests variation coefficient from of one variation cylinder from for different configurations of ignition system COV was calculated. The COV is the ratio of standard one cylinder for different configurations of ignitionC system COVC was calculated.C The COVC is the ratiodeviation of standard of main chamberdeviation peak of main pressures chamber and averagepeak pressures main chamber and average peak pressure main chamber from 10 workpeak cycles of one cylinder. The results of determined COV are presented in Figure6. pressure from 10 work cycles of one cylinder. The resultsC of determined COVC are presented in Figure 6. It can be noted that the most uniform combustion process inside the cylinder (the most uniform distributionIt can be of noted combustion that the pressure) most uniform was obtained combustion for caseprocess with inside only onethe cylinder PCC in port (the B. most The uniform value of distributionCOVC is 1.4%, of whichcombustion is 20% pressure) lower than was for obtained two spark for ignition case with configuration. only one PCC The in obtained port B. valuesThe value can be referred to results of Olsen and Lisowski investigations on single separated cylinder, where the COV of COVC is 1.4%, which is 20% lower than for two spark ignition configuration. The obtained values canhas be ranged referred from to 10% results to 25% of Olsen depending and Lisowski on equivalence investigations ratio in theon prechambersingle separated [16]. cylinder, For prechamber where

Energies 2018, 11, x 9 of 15 Energies 2018, 11, x 9 of 15 theEnergies COV2018 has, 11 ranged, 336 from 10% to 25% depending on equivalence ratio in the prechamber [16].9 For of 15 prechamberthe COV has installed ranged fromin port 10% A to the 25% combustion depending instabilities on equivalence increase. ratio inThis the phenomena prechamber could [16]. Forbe explainedprechamber by installedair-fuel mixturein port compositionA the combustion distributi instabilitieson in the increase.cylinder, Thisas near phenomena to the port could A the be installed in port A the combustion instabilities increase. This phenomena could be explained by air-fuel stoichiometricexplained by air-fuelcomposition mixture is present.composition This distributiconclusionon isin based the cylinder, on Computional as near to Fluid the portDynamics A the mixture composition distribution in the cylinder, as near to the port A the stoichiometric composition (CFD)stoichiometric calculation composition results of naturalis present. gas combustionThis conclusion process is based in GMVH-12 on Computional engine. It canFluid also Dynamics explain is present. This conclusion is based on Computional Fluid Dynamics (CFD) calculation results of that(CFD) more calculation stable combustion, results of natural in comparison gas combustion to 2-PCC process case, occurs in GMVH-12 in 2-SP configuration, engine. It can (COValso explainC = 2.3 natural gas combustion process in GMVH-12 engine. It can also explain that more stable combustion, andthat 1.7 more respectively stable combustion, for 2-PCC in and comparison 2-SP). Delivering to 2-PCC of case,additional occurs air in to2-SP the configuration, PCC-B, results (COV in increaseC = 2.3 in comparison to 2-PCC case, occurs in 2-SP configuration, (COVC = 2.3 and 1.7 respectively for intensityand 1.7 respectively of misfiring for (mixture 2-PCC isand out 2-SP). of high Delivering flammability of additional limits) and air tounburned the PCC-B, fuel results is injected in increase to the 2-PCC and 2-SP). Delivering of additional air to the PCC-B, results in increase intensity of misfiring mainintensity cylinder of misfiring which influences(mixture is the out dose of high composition. flammability The limits) ignition and occurs unburned only fuel for isone injected PCC-A to but the (mixture is out of high flammability limits) and unburned fuel is injected to the main cylinder which notmain all cylinder fuel is burned. which Thisinfluences hypothesis the dose can composition.be confirmed Theby analyzing ignition occurs the THC only emission, for one asPCC-A its value but influences the dose composition. The ignition occurs only for one PCC-A but not all fuel is burned. fornot PCC-A all fuel isis higherburned. than This for hypothesis test with can only be one confirmed PCC located by analyzing in port theB. The THC results emission, of influence as its value of This hypothesis can be confirmed by analyzing the THC emission, as its value for PCC-A is higher than ignitionfor PCC-A system is higher on emission than for of testtoxic with compound only ones for PCC one located cylinder in areport displayed B. The results in Figure of influence 7. of for test with only one PCC located in port B. The results of influence of ignition system on emission of ignition system on emission of toxic compounds for one cylinder are displayed in Figure 7. toxic compounds for one cylinder are displayed in Figure7.

Figure 6. Variance of peak pressure COVC for different ignition systems. Figure 6. Variance of peak pressure COVC for different ignition systems. Figure 6. Variance of peak pressure COVC for different ignition systems.

FigureFigure 7. 7. EmissionEmission of of NO NO and and THC THC for for different different igniti ignitionon systems systems from from one one cylinder cylinder of of GMVH-12. GMVH-12. Figure 7. Emission of NO and THC for different ignition systems from one cylinder of GMVH-12.

Energies 2018, 11, x 10 of 15

EnergiesFor2018 all, investigated11, 336 tests when using PCC (PCC-A, PCC-B, 2-PCC), NO emission was lower10 than of 15 in two spark plugs basic configuration. It is the result of spatial distribution of the combustible mixtureFor ignition all investigated by flame tests jet exiting when using from PCCthe PCC, (PCC-A, as its PCC-B, energy 2-PCC), is in the NO order emission of one was million lower times than thein two spark spark plug plugs discharges basic configuration. [15]. Therefore, It is the it allows result of to spatial ignite distribution leaner mixtures, of the combustiblereduce the peak mixture of combustionignition by flametemperature jet exiting and from consequently the PCC, as its re energyduce isthe in theNO orderformation of one millionthrough times the thethermal spark mechanism.plug discharges For [15PCC-B]. Therefore, and 2-PCC it allows configurations to ignite leaner the mixtures,reduction reduce of THC the peakwas also of combustion observed (comparedtemperature to and 2-SP consequently ignition). It reduceis likely the caused NO formation by the combustion through the stability. thermal In mechanism. the case with For PCC-A, PCC-B whereand 2-PCC the combustion configurations mixture the reduction is too rich of THCfor prop waser also ignition, observed the (compared increase of to THC 2-SP ignition).is significant It is comparedlikely caused to PCC-B by the test combustion results, and stability. even for In 2-SP the case ignition with configuration. PCC-A, where Based the combustion on the analysis mixture of the is resultstoo rich shown for proper in Figures ignition, 6 theand increase 7 it can of be THC conclude is significantd that the compared best ignition to PCC-B system test results,solution and for even the investigatedfor 2-SP ignition engine configuration. will be one Basedpre-combustion on the analysis chamber of the located results in shown port B. in For Figures this6 case and 7the it canlowest be emissionconcluded of that NO, the more best consistent ignition system combustion solution and for less the cycle-to-cycle investigated enginecombustion will be variation one pre-combustion is obtained, whichchamber leads located to the in lowest port B. THC For thisemission. case the lowest emission of NO, more consistent combustion and less cycle-to-cycleIn the next step combustion of single cylinder variation operation is obtained, stud whichy, the leads influence to the of lowest ignition THC time emission. and different stoichiometryIn the next in stepPCC of was single tested, cylinder but operationonly for study,two configurations the influence of ignitionignition timesystem: and the different pre- combustionstoichiometry chamber in PCC waslocated tested, in port but onlyA or forport two B. The configurations ignition delay of ignition was decreased system: the to − pre-combustion2° before TDC ofchamber the piston. located This in port solution A or port favorably B. The ignition affects delay the wasreduction decreased of tonitric−2◦ beforeoxides TDCformation of the piston.(NO decreasing—inThis solution favorably case of 2-SP affects ignition) the reduction but also of redu nitricce oxides the value formation of IMEP (NO and decreasing—in consequently case decreases of 2-SP theignition) engine but efficiency. also reduce The the results value ofof IMEPignition and dela consequentlyy influence decreases on emission the engine and combustion efficiency. The stability results are of presentedignition delay in Figure influence 8. on emission and combustion stability are presented in Figure8.

FigureFigure 8. 8. EmissionEmission of of toxic toxic compound compoundss versus versus ignition time.

ResultsResults presented inin FigureFigure8 show8 show that that nitric nitric oxide oxide reduction reduction is observed is observed only only for the for PCC-A the PCC-A case, case,and isand about is about 60% 60% when when ignition ignition delay delay is shifted is shifted to an to angle an angle of − of2◦ −before2° before TDC. TDC. For For PCC PCC located located in inport port B, B, the the NO NO increases increases by by 75%. 75%. Similar Similar trend tren isd observedis observed for for combustion combustion stability, stability, improvement improvement for forPCC-A PCC-A and and deterioration deterioration for PCC-B. for PCC-B. Decreasing Decreasing of combustion of combustion stability stability for pre-combustion for pre-combustion chamber chamberlocated in located port B in does port notB does affect not the affect THC the emission, THC emission, because because overall overall emission emission decreases decreases for both for bothinvestigated investigated cases, cases, when thewhen ignition the ignition occurs later.occurs Further later. analysisFurther ofanalysis Figure9 of shows Figure that 9 combustionshows that combustionstability is worse, stability when is worse, additional when air additional is delivered air to is the delivered prechamber. to the The prechamber. value of variance The value of peak of variancepressure coefﬁcientof peak pressure increases coefficient from 1.3% increases up to 1.5%. from Additionally, 1.3% up to 1.5%. decreasing Additionally, of equivalence decreasing ratio of in equivalencePCC causes signiﬁcantratio in PCC increase causes of significant NO emission increase and slightof NO reduction emission of and THC slight amount reduction in the of exhaust THC amountgases. As in mentioned the exhaust earlier, gases. during As mentioned the experiment earlier, it was during impossible the experiment to characterize it was the impossible air–fuel ratio to characterize the air–fuel ratio equivalence value in PCC. Therefore, in the final step of research the

Energies 20182018,, 1111,, x 336 11 of 15 pre-combustion chamber without supply of additional air was used and future investigation are equivalence value in PCC. Therefore, in the ﬁnal step of research the pre-combustion chamber without needed. supply of additional air was used and future investigation are needed.

Figure 9. Combustion stability (COVC) versus ignition time. Figure 9. Combustion stability (COVC) versus ignition time. 3.3. Analysis of Engine Operation with PCC Ignition System 3.3. Analysis of Engine Operation with PCC Ignition System During the last step of the investigation, the engine was equipped with an ignition system basedDuring on one the pre-combustion last step of the chamberinvestigation, located the inen Portgine Bwas in equipped each of twelve with an cylinders. ignition Thesystem selection based onof configurationone pre-combustion was done chamber based onlocated results in of Port one cylinderB in each work of twelve in which cylinders. the lowest The NO selection and THC of configurationemission as well was as done the lowest based valueon results of COV of one were cylinder observed. work The in angle which of the ignition lowest was NO set and at − THC5.4◦ emissionbefore TDC. as well Experiment as the lowest has been value prepared of COV for we fivere rotational observed. speeds The angle from of 290 ignition to 330 rpm.was set During at −5.4° the beforeinvestigation, TDC. Experiment the analysis has of been main prepared engine operating for five rotational parameters speeds was performed. from 290 to The 330 valuerpm. ofDuring peak pressurethe investigation, coefficient the variance analysis COV ofE,II mainwas engine calculated operating as a ratio parameters of standard was deviation performed. of the The 12 cylindersvalue of peak pressure andcoefficient average variance IMEP of COV the engineE,II was aftercalculated modification. as a ratio of standard deviation of the 12 cylindersFigure peak 10 showspressure the and overall average emission IMEP ofof toxicthe engine compounds after modification. for the whole engine equipped with a pre-combustionFigure 10 shows chamber. the overall It can emission be observed of toxic that compounds contents offor nitricthe whole oxide engine and carbon equipped monoxide with a pre-combustionsignificantly decrease, chamber. while It thecan THC be observed share rises, that compared contents to of engine nitric equipped oxide and with carbon 2-SP. Reductionmonoxide significantlyof NO emission decrease, has changed while the from THC 33% share at 290rises, rpm compared up to 60% to engine at 330 rpm.equipped Reduction with 2-SP. of nitric Reduction oxide ofis dueNO toemission the lack has of highchanged temperature from 33% peak at 290 region rpm up in theto 60% main at combustion 330 rpm. Reduction chamber, of induced nitric oxide by the is duespark to plug the lack ignition. of high Additionally temperature the peak pulse region jet injected in the from main PCC combustion causes the chamber, entrainment induced and mixingby the sparkof fresh plug charge ignition. in the Additionally main combustion the pulse chamber jet inje andcted ensuresfrom PCC more causes uniform the entrainment distribution and of air-fuelmixing ofmixture, fresh charge reducing in the areas main of enriched combustion mixture chamber inducing and increasedensures more NO formation.uniform distribution of air-fuel mixture,The reducing emission areas of THC of enriched rose by aroundmixture 25%inducing for all increased examined NO rotational formation. speeds and causes the increaseThe ofemission the specific of THC fuel consumptionrose by around value 25% parameter for all examined to 0.291 Nmrotational3/kWh. speeds In comparison and causes to tests the 3 withincrease 2-SP of ignition the specific configuration, fuel consumption the gi increased value parameter by around to 0.291 seven Nm percent,/kWh. which In comparison means a roughly to tests with2.2% 2-SP reduction ignition of fuelconfiguration, conversion the efficiency. gi increased Similar by around results seven were pe obtainedrcent, which by Roethlisberger means a roughly and 2.2%Favrat reduction [24], where of thefuel comparison conversion ofefficiency. operating Simi parameterslar results of were natural obtained gas powered by Roethlisberger engine equipped and withFavrat prechambers [24], where the and comparison spark plugs of wasoperating realized. parameters Increase of of natural THC emission gas powered can suggestengine equipped the drop withof combustion prechambers stability. and spark Although plugs was the realized. calculated Incr valuesease of of THC COV emissionE,II coefficient can suggest decreased the drop for allof combustioninvestigated rotationalstability. speeds.Although The the value calculated of COVE,II valuesis in the of range COV fromE,II coefficient 9.9 up to 12.4 decreased with an averagefor all valueinvestigated of 10.8 rotational (11.7 for engine speeds. with The 2-SP). value Thisof COV meansE,II is thatin the PCC range ignition from system9.9 up to improves 12.4 with combustion an average valuestability of 10.8 by eight (11.7 percent for engine compared with 2-SP). to 2-SP This system. means Inthat Figure PCC 11 ignition, a comparison system improves of maximum combustion peak of pressurestability by for eight the whole percent engine compared with two to 2-SP investigated system. In ignition Figuresystems 11, a comparison is presented. of maximum peak of pressure for the whole engine with two investigated ignition systems is presented.

Energies 2018, 11, x 12 of 15 EnergiesEnergies 20182018, 11, 11, x, 336 1212 of of 15 15

Figure 10. Overall emission for gas engine GMVH-12 with pre-combustion chamber ignition system. FigureFigure 10. 10. OverallOverall emission emission for for gas gas engine engine GMVH-12 GMVH-12 wi withth pre-combustion pre-combustion chamber chamber ignition ignition system. system.

Figure 11. Peak of pressure for GMVH-12 engine with SP and PCC ignition arrangements. Figure 11. Peak of pressure for GMVH-12 engine with SP and PCC ignition arrangements. Figure 11. Peak of pressure for GMVH-12 engine with SP and PCC ignition arrangements. TheThe explanation of of increased increased THC THC emission emission can can be bethat that thethe back back parts parts of prechamber of prechamber and near and The explanation of increased THC emission can be that the back parts of prechamber and near nearpiston piston wall wallregions regions are the are place the place of unburned of unburned methane methane formation formation and andincreased increased emission. emission. The piston wall regions are the place of unburned methane formation and increased emission. The Thetemperature temperature rise riseis repressed is repressed due due to toheat heat tran transfersfer to to the the cylinder cylinder walls causing incompleteincomplete temperature rise is repressed due to heat transfer to the cylinder walls causing incomplete combustion.combustion. LeanLean mixtures sittingsitting beyond the flammabilityflammability limitslimits disrupt flameflame propagation andand partpart combustion. Lean mixtures sitting beyond the flammability limits disrupt flame propagation and part ofof thethe mixturemixture remainsremains withoutwithout burning,burning, resultingresulting inin increasedincreased hydrocarbonhydrocarbon emission emission [ 25[25].]. of the mixture remains without burning, resulting in increased hydrocarbon emission [25]. 4.4. Conclusions 4. Conclusions TheThe currentcurrent investigationinvestigation on on influence influence of of ignition ignition system system in ain two a two stroke, stroke, large large bore bore stationary stationary gas The current investigation on influence of ignition system in a two stroke, large bore stationary enginegas engine has shownhas shown that that adding adding a prechamber a prechamber ignition ignition system system provides provides better better combustion combustion stability stability and gas engine has shown that adding a prechamber ignition system provides better combustion stability decreasesand decreases toxic toxic compounds compounds emission emission such assuch nitric as ni oxidetric oxide and carbon and carbon monoxide monoxide compared compared to a spark to a andplug decreases arrangement. toxic Forcompounds both configurations emission such the overallas nitric emission oxide and decreases carbon withmonoxide the rise compared of engine to load. a

Energies 2018, 11, 336 13 of 15

The retroﬁtting of investigated engine by change of ignition system led to several positive phenomena such as:

decrease of nitric oxide emission up to 60%; reduction of carbon monoxide in the range from 22–26%, for 290 and 330 rpm respectively; improvement of combustion stability by eight percent compared to 2-SP system; increase of maximum peak pressure value.

Experimental study of separate cylinder work has shown that the location of the prechamber in the cylinder head has a significant influence on the engine operational parameters. During the study, two ports were investigated, one located in the axis of the cylinder (port A) and the second one in port B. Combustion stability improvement (lower value of COVC) as well as NO and CO content reduction were observed only for port B, while for port A the overall emission was similar to emission for 2-SP ignition but more engine misfire was measured (COVC = 3.6). Further analysis including the influence of ignition time and stoichiometry in PCC allows us to make the following observations:

ignition time delay improves the combustion stability and reduces the NO and THC emission in case of PCC-A, while for PCC-B this trend is the opposite; nitric oxide emission and combustion instability increase with the decrease of equivalence ratio.

The results conﬁrm that the pre-combustion ignition system can be successfully used as an upgrade for a two-stroke gas engine equipped with a spark plug ignition. It ensures reduction of combustion instability and toxic compounds emissions with a slight decrease in efﬁciency.

Author Contributions: Rafał Slefarski´ and Jacek Wawrzyniak conceived and designed the experiments; Michał Goł˛ebiewski, Paweł Czyzewski,˙ Jacek Wawrzyniak, Rafał Slefarski´ and Przemysław Grzymisławski performed the experiments; Michał Goł˛ebiewski, Rafał Slefarski,´ Paweł Czyzewski˙ and Przemysław Grzymisławski analyzed the data; Rafał Slefarski´ and Jacek Wawrzyniak contributed reagents/materials/analysis tools; Paweł Czyzewski,˙ Rafał Slefarski´ and Michał Goł˛ebiewski wrote the paper. Conﬂicts of Interest: The authors declare no conﬂicts of interest.

Nomenclature

IMEP indicative mean effective pressure COV coefﬁcient of variation TDC top dead center PCC precombustion chamber 2-PCC two precombustion chamber PCC-A precombustion chamber located in port A PCC-B precombustion chamber located in port B SP spark plug 2-SP two spark plugs THC total hydrocarbons EGR external gas recirculation SCR selective catalytic reduction SNCR selective non catalytic reduction C cylinder E,I engine with spark plugs ignition system E,II engine with prechamber ignition system gi speciﬁc fuel consumption φ equivalence ratio rpm revolutions per minute Energies 2018, 11, 336 14 of 15

References

1. Rojewski, J.; Slefarski,´ R.; Wawrzyniak, J. Analysis of combustion process in industrial gas engines powered with high methane and nitrified low-calorific gases. Siln. Spalinowe 2013, 52, 42–50. 2. Bhandari, K.; Bansal, A.; Shukla, A.; Khare, M. Performance and emissions of natural gas fueled internal combustion engine: A review. J. Sci. Ind. Res. 2005, 64, 333–338. 3. Hill, S.C.; Smoot, L.D. Modeling of nitrogen oxides formation and destruction in combustion systems. Prog. Energy Combust. Sci. 2000, 26, 417–458. [CrossRef] 4. Knop, V.; Benkenida, A.; Jay, S.; Colin, O. Modelling of combustion and nitrogen oxide formation in hydrogen-fuelled internal combustion engines within a 3D CFD code. Int. J. Hydrogen Energy 2008, 33, 5083–5097. [CrossRef] 5. Raine, R.R.; Stone, C.R.; Gould, J. Modeling of nitric oxide formation in spark ignition engines with a multizone burned gas. Combust. Flame 1995, 102, 241–255. [CrossRef] 6. Krishnan, S.R.; Srinivasan, K.K.; Singh, S.; Bell, S.R.; Midkiff, K.C.; Gong, W.; Fiveland, S.B.; Willi, M. Strategies for Reduced NOx Emissions in Pilot-Ignited Natural Gas Engines. J. Eng. Gas Turbines Power 2004, 126, 665–671. [CrossRef] 7. European Union. Directive (EU) 2015/2193 of the European Parliament and of the Council—Of 25 November 2015—On the Limitation of Emissions of Certain Pollutants into the Air from Medium Combustion Plants; Official Journal of the European Union: Luxembourg, 2015; Volume 58. 8. Königsson, F.; Dembinski, H.; Angstrom, H.-E. The Influence of In-Cylinder Flows on Emissions and Heat Transfer from Methane-Diesel Dual Fuel Combustion. SAE Int. J. Engines 2013, 6, 1877–1887. [CrossRef] 9. Sorathia, H.S.; Rahhod, P.P.; Sorathiya, A.S.; Engineering, M.; College, G.E.; Kutch, B. Effect of Exhaust Gas Recirculation (EGR) on NOx Emission from c.i. Engine”—A Review Study. Int. J. Adv. Eng. Res. Stud. 2012, 1, 223–227. 10. Rudkowski, M.; Dudek, S.; Wołoszyn, R. Test results of an external exhaust gas recirculation system of a Cooper Bessemer GMVH-8 engine-compressor. Siln. Spalinowe 2010, 49, 74–81. 11. Ma, F.; Wang, Y.; Liu, H.; Li, Y.; Wang, J.; Ding, S. Effects of hydrogen addition on cycle-by-cycle variations in a lean burn natural gas spark-ignition engine. Int. J. Hydrogen Energy 2008, 33, 823–831. [CrossRef] 12. Arcoumanis, C.; Hull, D.R.; Whitelaw, J.H. An Approach to Charge Stratification in Lean-Burn, Spark- Ignition Engines; SAE Paper 941878; SAE International: Warrendale, PA, USA, 1994. 13. Ma, F.; Wang, Y. Study on the extension of lean operation limit through hydrogen enrichment in a natural gas spark-ignition engine. Int. J. Hydrogen Energy 2008, 33, 1416–1424. [CrossRef] 14. Tozzi, L.P.; Salter, D.W. Pre-Chamber Spark Plug. U.S. Patent 7,922,551 B2, 7 June 2005. 15. Olsen, D.B.; Adair, J.L.; Willson, B.D. Precombustion Chamber Design and Performance. In Proceedings of the ASME 2005 Internal Combustion Engine Division Spring Technical Conference, Chicago, IL, USA, 5–7 April 2005; pp. 415–428. 16. Olsen, D.B.; Lisowski, J.M. Prechamber NOx formation in low BMEP 2-stroke cycle natural gas engines. Appl. Therm. Eng. 2009, 29, 687–694. [CrossRef] 17. Tozzi, L.; Sotiropoulou, E.; Zhu, S. Improving the Efficiency/Emissions Trade-off with a Novel Lean-Burn Precombustion Chamber. In Proceedings of the 10th Dessau Gas Engine Conference, Dessau-Rosslau, Germany, 6–7 April 2017; pp. 165–176. 18. Le Conseil fédéral Suisse. Ordonnance sur la Protection de l’air (OPair), du 16 décembre 1985 (Etat au 12 octobre 1999); Le Conseil fédéral Suisse: Berne, Switzerland, 1999. 19. Roethlisberger, R.P.; Favrat, D. Comparison between direct and indirect (prechamber) spark ignition in the case of a cogeneration natural gas engine, Part II: Engine operating parameters and turbocharger characteristics. Appl. Therm. Eng. 2002, 22, 1231–1243. [CrossRef] 20. Heyne, S.; Meier, M.; Imbert, B.; Favrat, D. Experimental investigation of prechamber autoignition in a natural gas engine for cogeneration. Fuel 2009, 88, 547–552. [CrossRef] 21. Rojewski, J. Optimazation of Combustion Process in Industrial Gas Engines. Ph.D. Thesis, Poznan University of Technology, Poznan, Poland, 2014; p. 7. 22. Litak, G.; Syta, A.; Yao, B.F.; Li, C.X. Indicated mean effective pressure oscillations in a natural gas combustion engine by recurrence plots. J. Theor. Appl. Mech. 2009, 47, 55–67. Energies 2018, 11, 336 15 of 15

23. Wang, J.; Chen, H.; Liu, B.; Huang, Z. Study of cycle-by-cycle variations of a spark ignition engine fueled with natural gas-hydrogen blends. Int. J. Hydrogen Energy 2008, 33, 4876–4883. [CrossRef] 24. Roethlisberger, R.P.; Favrat, D. Comparison between direct and indirect (prechamber) spark ignition in the case of a cogeneration natural gas engine, Part I: Engine geometrical parameters. Appl. Therm. Eng. 2002, 22, 1217–1229. [CrossRef] 25. Youseﬁ, A.; Birouk, M. Investigation of natural gas energy fraction and injection timing on the performance and emissions of a dual-fuel engine with pre-combustion chamber under low engine load. Appl. Energy 2017, 189, 492–505. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).