Applied Energy 156 (2015) 129137

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier .com/locate /apenergy

Design considerations for an engine-integral reciprocating natural gascompressor

http://dx.doi.org/10.1016/j.apenergy.2015.07.0020306-2619/ 2015 Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +1 (970) 491 3539.E-mail addresses: mmalakou@engr.colostate.edu (M. Malakoutirad), Thomas.

Bradley@colostate.edu (T.H. Bradley), chris.hagen@osucascades.edu (C. Hagen).1 Tel.: +1 (541) 322 2061.

Mohammad Malakoutirad a, Thomas H. Bradley a,, Chris Hagen b,1a Colorado State University, Department of Mechanical Engineering, Campus Delivery 1374, Fort Collins, CO 80523-1374, United Statesb Oregon State University Cascades, School of Mechanical, Industrial and Manufacturing Engineering, 2600 NW College Way, Bend, OR 97701, United States

h i g h l i g h t s

An engine-integral natural gas compressor was developed under contract to ARPA-E. System is novel in that an engine powers its 6th cylinder as a multi-stage compressor. A structural and functional description of the system is presented. Dynamic and thermal characteristics of the system dictate the design.

a r t i c l e i n f o

Article history:Received 5 November 2014Received in revised form 10 June 2015Accepted 4 July 2015

Keywords:Natural gasEngineCompressorAutomotive

a b s t r a c t

Conventionally, compressed natural gas (CNG) vehicles are refueled using a high-cost, centralized, andsparse network of CNG fueling stations that has primarily been developed for the use of fleet customers.An engine-integral reciprocating natural gas (NG) compressor has the capability to disrupt the incumbentCNG market by enabling the use of NG for personal transportation, fueled at home, from the preexistinglow-pressure NG infrastructure, at low parts count, using conventional components, and therefore at lowincremental costs. The principal objective of this paper is therefore to describe and analyze the dynamicand thermal design considerations for an automotive engine-integral reciprocating NG compressor. Thepurpose of this compressor is to pressurize storage tanks in NG vehicles from a low-pressure NG sourceby using one of the engine cylinders as a multi-stage reciprocating compressor. The engine-integral com-pressor is developed by making changes to a 5.9 l displacement diesel-cycle automotive engine. In thisnovel design and implementation, a small tank and its requisite valving are added to the engine as anintermediate gas storage system to enable a single compressor cylinder to perform two-stage compres-sion. The resulting maximum pressure in the storage tank is 250 MPa, equivalent to the storage and deliv-ery pressure of conventional CNG delivery systems. Dynamic simulation results show that the highcylinder pressures required for the compression process create reaction torques on the crankshaft, butdo not generate abnormal rotational speed oscillations. Thermal simulation results show that the temper-ature of the storage tank and engine increases over the safety temperature of the natural gas storage sys-tem unless an active thermal management system is developed to cool the NG before it is admitted to thestorage tanks.

2015 Elsevier Ltd. All rights reserved.

1. Introduction

Transportation is a major contributor to the productivity of theUS economy, but the air pollution costs, and fuel import costs of apetroleum-fueled transportation system are high. Transportation

produces 79% of the criteria air pollutants, 50% of total nitrogenoxide emissions, 26% of greenhouse gas (GHG) emissions, and 42%of total volatile organic compound emissions in the U.S [1].Although regional air quality and the specific GHG emissions(kg/km) of transportation have been improving with advance-ments in transportation technologies, many regions of the US donot meet federal requirements for air quality [2], and total GHGemissions continue to increase [2]. Transportation fueled by natu-ral gas (NG) engines has been demonstrated to produce fewer CO2,CO, and HC emissions than gasoline vehicles [3,4], and it is a

http://crossmark.crossref.org/dialog/?doi=10.1016/j.apenergy.2015.07.002&domain=pdfhttp://dx.doi.org/10.1016/j.apenergy.2015.07.002mailto:mmalakou@engr.colostate.edumailto:Thomas.Bradley@colostate.edumailto:Thomas.Bradley@colostate.edumailto:chris.hagen@osucascades.eduhttp://dx.doi.org/10.1016/j.apenergy.2015.07.002http://www.sciencedirect.com/science/journal/03062619http://www.elsevier.com/locate/apenergy

130 M. Malakoutirad et al. / Applied Energy 156 (2015) 129137

near-term feasible technology for improving the sustainability ofthe transportation sector [5].

Compressed natural gas powered vehicles (CNGVs) have notbecome mass-market vehicles in the US for a variety of reasons[68]. Consumer and fleet adoption of natural gas powered vehi-cles are hampered by sparse fueling infrastructure, limited drivingrange, and the incremental cost of NG vehicles relative to conven-tional petroleum vehicles. NG fueling infrastructure primarily con-sists of a sparse network of non-public fueling stations. There areapproximately 1300 NG fueling stations in the US [9], as comparedto >150,000 gasoline stations. The scarcity of NG fueling infrastruc-ture is exacerbated by the limited driving range available fromconventional CNGVs. For example, the driving range of aChevrolet Cavalier dual-fuel compressed natural gas (CNG) vehicleis 110 miles, approximately 1/3 of the driving range of a conven-tional vehicle. Finally, NG-powered vehicles have a several thou-sand dollar2 higher purchase price than conventional petroleumpowered vehicles. CNGVs are therefore subject to the conundrumin which inadequate infrastructure limits the market for theCNGVs, and the limited market for the vehicles discourages invest-ment in a costly CNGV fueling infrastructure.

Many researchers believe that the solution to the NG infrastruc-ture and cost conundrum is the development of localized (asopposed to centralized) business or home NG compressor and refu-eling systems [1012]. A localized refueling system would com-press the low pressure NG that is delivered to many businessesand homes through the preexisting, low-pressure NG infrastruc-ture and would transfer the compressed NG to the vehicle. Thisconcept has several advantages over the more conventional cen-tralized CNG infrastructure solutions [13]. First, the allocation offueling infrastructure costs to a single vehicle means that the costand availability of the fueling infrastructure is independent of thestate of CNG vehicle market penetration. Second, by placing fuelinginfrastructure at the locations where the vehicle regularly parks,refueling can be performed daily or more often without inconve-nience to the driver. The primary challenge that must be overcometo improve the feasibility of localized refueling infrastructure is itscost. There exist a variety of localized (home) NG compression sys-tems that will fuel a CNG vehicle by locally compressing the NGavailable from low-pressure residential-type infrastructure. Thesesystems have a high incremental cost because they are not subjectto mass-production, they have tight tolerances and parts count,and they require high-power, high-torque power supplies.3

Engine-integral air compressors are commercial products4, althoughno similar systems have been developed to produce compressed NG[14]. The most comparable systems design effort that the authors areaware of is in the field of integral free piston engine/compressordesign [15].

This article proposes the development of an engine-integral,low-parts count, reconfigurable NG compressor as a disruptivetechnology to address the cost barrier to localized NG compressionand fueling. During normal driving, all 6 cylinders of the engineoperate under a conventional Otto internal combustion mode.During a refueling event, this concept allows the engine controllerto reconfigure one of the engines combustion cylinders (cylinder1) to function as a reciprocating multi-stage compressor thatdirectly refuels the vehicles CNG tank using the low-pressure res-idential NG infrastructure. This functional change is achievedthrough a change in valve control algorithms. Such a system allowsfor the advantages of home CNG refueling while minimizing the

2 The Natural-gas alternative. The pros and cons of buying a CNG Powered car(http://www.consumerreports.org/cro/2012/03/the-natural-gas-alternative/index.htm).

3 BRC FuelMaker, http://www.brcfuelmaker.it/eng/casa/phill.asp.4 Dunn-Right Inc., Volks-Air Air Compressor, http://www.dunnrightinc.com.

incremental cost of materials, components and systems. To charac-terize this disruptive technology, this article describes the struc-tural and functional concept, and provides a set ofthermodynamic and mechanics models and results to characterizethe design requirements of the system. Finally, we propose ourunderstanding of the barriers that must be overcome to enablethe technical and commercial viability of the technology.

2. Structural description

This section describes the vehicle components and mechanicalsystems that make up the engine-integral compressor system.The primary components of the engine-integral compressor sys-tem are the CNG tanks, the CNG internal combustion engine(ICE), the intermediate pressures storage tanks (IPST), and the sys-tem of control valves, as shown in Fig. 1.

CNGVs store natural gas in high-pressure on-board tanks thathave rated pressures of 25 MPa (in the US) [16]. Conventionally,these tanks are filled from a high-pressure connection to ahigh-pressure CNG refueling system. In this concept, we seek to fillthe on-board CNG tank by compressing NG from line pressure(1.02 bar absolute) to tank pressure using an engine-integralcompressor. Fueling the on-board tank using low-pressure NGminimizes the need for a high-pressure CNG infrastructure.

CNG-powered engines are similar in structure and control toconventional ICEs. Modified NG injectors and emissions controlequipment are often used to convert conventional ICEs toNG-powered ICEs. In this concept, by using the internal compo-nents of the ICE as a reciprocating compressor, we can reducethe parts count and therefore manufacturing costs of the integralcompressor and engine systems. In this concept, a reciprocatingcompression cylinder is powered mechanically from the enginecrankshaft as the other 5 cylinders perform conventional ICE oper-ations. A modified diesel-cycle engine is chosen for this conceptdescription to allow the mechanical components of the engine totolerate repeated in-cylinder pressures of up to 25 MPa.5

Conventional diesel engines have compression ratios (CR) ofbetween 14 and 22, primarily limited by clearance volume require-ments, and diminishing returns on engine efficiency with increas-ing CR [18]. Because of this CR limitation, in this concept thepressurization of the NG from line pressure to tank pressure mustbe performed in multiple stages. In this concept we use interme-diate pressure storage tanks (IPST) to store intermediate pressureNG between the stages of compression to allow for multi-stagecompression of the NG to storage pressure using a single compres-sion piston. The IPST are envisioned as rigid, metallic cylinders thatare connected to the cylinder with a system of control valves.Under each stage of compression, the cylinder can achieve a com-pression ratio of between 14 and 22, which implies that between 2and 3 stages of compression (and therefore between 1 and 2 IPST)are required to generate a tank pressure of 250 bar from thenear-ambient supply pressure. In this concept, a Cummins 5.9LDiesel engine is chosen for detailed consideration. Its CR of 17allows for the use of a single IPST.

The conversion from a conventional engine to anengine-integral compressor system requires the development andinstallation of a custom engine head that incorporates the valvingand flow passages illustrated in Fig. 1. The conventional exhaustand intake valves for cylinder 1 are disabled (closed) during com-pression operations.

5 Heavy-duty diesel engines maximum pressure has been increasing since 1970and it is predicted to reach 25 MPa in the near future while future incrementalimprovements in specific power are predicted [17].

http://www.consumerreports.orghttp://www.consumerreports.orghttp://www.brcfuelmaker.it/eng/casa/phill.asphttp://www.dunnrightinc.com

Fig. 1. Schematic of engine-integral reciprocating natural gas compressor.

M. Malakoutirad et al. / Applied Energy 156 (2015) 129137 131

3. Functional description and analysis

Using the components described in the previous section, thissection describes the function and operation of theengine-integral reciprocating NG compression system. In this sec-tion, we describe the process of in-cylinder NG compression, theprocess of transfer of intermediate pressure to and from the IPST,and the process of transfer of high pressure NG to the on-boardstorage tank.

3.1. In-cylinder natural gas compression process

Upon connection of the vehicle to the home NG fueling line, oneof the cylinders of the ICE begins to operate as a compressor pow-ered by the other 5 cylinders of the ICE. Low-pressure gas is admit-ted into to this compression cylinder by way of the fuel controlvalve, a normally-closed, spring return, solenoid valve. As the pis-ton moves upwards in the cylinder, the NG is compressed. Theaction of the one-way valve shown in Fig. 1 are such that the cylin-der will only transfer pressurized NG from the compression cylin-der to the IPST when the in-cylinder NG pressure is greater thanthe pressure in the IPST. As the cylinder moves past its top deadcenter (TDC) position, the in-cylinder pressure will reduce until itis below the pressure in the IPST, closing the one-way valvebetween the compression cylinder and the IPST. The compressioncylinder then expands the remaining in-cylinder NG until thecylinder reaches near ambient pressure, at which pointlow-pressure gas is admitted into the compression cylinder fromthe home NG fueling line. The cycle repeats to provide compressionof the NG from the fueling line, and to and from the IPST, in turn.

3.2. IPST tank pressurization process

As described in the preceding paragraph, the NG in the com-pression cylinder increases in pressure as the cylinder approachesTDC, and the 3 position, normally-closed, spring return, solenoidvalve between the compression cylinder and IPST allows the move-ment of NG from the cylinder into the IPST. Details of the processare shown in Fig. 2. For this conceptual design study, the compres-sion of NG is simulated using a polytropic process (Eq. (1)) with apolytropic exponent (k) of 1.3, whereas the ratio of the

volume-constant specific heat to the pressure-constant specificheat for NG is 1.4 [19].

PiPf 1

CR

k1

As the compression cylinder passes through TDC, the pressurein the compression cylinder begins to decrease and the solenoidvalve between the compression cylinder and IPST is closed by itssolenoid actuator. The cylinder expands the remaining NG untilnew NG is admitted to the cylinder at near-atmospheric pressurenear BDC. This process repeats itself until the pressure in theIPST reaches its pressure setpoint (which varies from between0.7 MPa and 2.8 MPa, depending on the storage tank pressure).Each cycle of compression moves incrementally less mass to theIPST, and increases the pressure within the IPST incrementally lessuntil the pressure in the IPST reaches its pressure setpoint. For theexample illustrated in Fig. 2, between two and four rotations of thecompression cylinder are required for the IPST to reach its pressuresetpoint.

As the pressure in the IPST rises over its setpoint, the compres-sion system begins to implement its second stage of compression.As the cylinder nears TDC, the control valve between the IPST andthe compression cylinder opens and the high-pressure NG is trans-ferred from the IPST back to the compression cylinder. This step iswhat causes the drop in IPST pressure in the cycle after the pres-sure setpoint is reached (as illustrated in Fig. 2). The solenoid con-trol valve between the IPST and the cylinder closes again when thecompression cylinder reaches BDC. At this point in the example,the state of the compression cylinder is that it is at intermediatepressure and temperature and is at BDC. The compression cylindertherefore compresses from intermediate pressure as it moves fromBDC and generates high pressure NG to fill the vehicle storage tank,as described in detail in the next section.

3.3. Storage tank pressurization process

The vehicles NG storage tank is a simple storage vessel con-nected to the compression cylinder using a one-way solenoid con-trolled valve. This valve simply transfers NG to the storage tankwhenever the pressure of the compression cylinder is greater thanthe pressure in the storage tank. The volume of the storage tank ismuch greater than the volume of the cylinder or IPST, so that thepressure of the storage tank builds slowly with each cycle of themulti-stage compression process toward a 25 MPa rated pressure.This procedure is shown in various levels of detail in Fig. 3. Early inthe compression process, the storage tank is at low pressure andeach cycle of pressurization increases the tank pressure by approx-imately 0.05 MPa. The tank continues to receive compressed NGand increase its pressure until it reaches a final pressure of25 MPa.

The functional result of this system and process is anengine-integral, low-parts count, NG compressor as a disruptivetechnology to address the cost barrier to localized NG compressionand fueling.

4. Detailed design considerations

In practice, the function of this complicated system will be con-strained by the detailed design considerations surrounding a fewkey tradeoffs among the systems components and capabilities. Inthis section of the paper, we detail these key tradeoffs withintwo problem domains that were identified through peer review,commercialization workshops, and discussion with ARPAe projectmanagers: dynamic torque loads and thermal considerations.

Fig. 2. Pressure inside the IPST during the compression process. Figure A shows the IPST pressure during the first few seconds of the compression process. Figure B shows theIPST pressure during the final seconds of the compression process as the storage tank pressure reaches 25 MPa.

Fig. 3. Pressure inside the storage tank during the compression process. Figure A shows the storage tank pressure during the first few seconds of the compression process.Figure B shows the storage tank pressure during the final seconds of the compression process as the storage tank pressure reaches 25 MPa.

132 M. Malakoutirad et al. / Applied Energy 156 (2015) 129137

4.1. Dynamic torques and rotational energy requirements

Generating high pressures in the compression cylinder requiresthat high torques be applied and received by the engine crankshaft.For the example engine considered in this study, the engine torquepulses due to the compression and expansion of NG inside theengine compression cylinder will cause instantaneous torquespikes up to 1100 Nm. The dynamics of applying large torques tothe compression cylinders piston, connection rod, and crankshaftjournals will have important implications for engine noise, vibra-tion and component lifetime.

One means to analyze these considerations is to model thedynamics of the engine-integral compressor with the objective ofquantifying the magnitude of any increase in engine speed fluctu-ations. The selected engine for this design has a stock flywheelwith an inertia of 2.1 kg m2, and the instantaneous velocity fluctu-ation at standard engine operation at 1000 rpm is 10 rpm. If the

engine speed fluctuations are significantly increased by thedynamics of the simultaneous compression and Otto combustioncycles, then the behavior will have to be mitigated to meet durabil-ity, noise, vibration, and harshness requirements.

As discussed earlier, during refueling operations, 5 of the cylin-ders of this engine operate under the normal pressure cycle associ-ated with an Otto cycle internal combustion engine. Relative to theother cylinders, the compression cylinder develops a large torquepulse during its second-stage compression and subsequent expan-sion processes. As all cylinders share a common, conventionalcrankshaft, this high torque pulse makes the engine operate withhigher speed fluctuation during the NG compression cycle.

To determine the magnitude and direction of this speed fluctu-ation, we can model the rotational dynamics of the example engineas its cylinder 1 undergoes the compression cycle, and its cylinders26 undergo Otto combustion cycles. For this purpose, we use asimple, dynamic model of 6 slider bar linkages as in [20]. This

Fig. 4. Crankshaft torque and in-piston pressure as a function of crankshaft position (negative torque represents an extraction of work from the piston).

M. Malakoutirad et al. / Applied Energy 156 (2015) 129137 133

calculation assumes that the cylinder 1 (the compressor cylinder)and 6 operate out of phase to show the worst case scenario inwhich combustion torque in cylinder 6 and expansion torque incylinder 6 both accelerate the crankshaft. As shown in Fig. 4, thepeak pressure in cylinder 1 occurs at 540 of crankshaft rotation(TDC), as is expected for a polytropic compression cycle. The peakpressure in cylinder 6 occurs at 550 of crankshaft rotation, as isexpected for a pure combustion cycle with conventional ignitiondelay.

The 10 misalignment in peak pressure between cylinders 1 and6 means that for cylinders 1 and 6, the sum of the integrated tor-que in expansion is larger than the sum of their integrated torquein compression. Fig. 4 shows the detail of the engines torque andpressure calculation, and Fig. 5 shows the velocity fluctuation inthe crankshaft. As shown in Fig. 5, operating the engine in a com-bined combustion and compression mode generates a positive

Fig. 5. Crankshaft velocity fluctuations during the NG compression process.

crankshaft velocity fluctuation of between 950 rpm and1000 rpm. Although this represents an increase in engine velocityfluctuations relative to the stock engine model, a 100 rpm speedfluctuation is not abnormal for automotive applications [21].

These results demonstrate that although there may be somedurability and NVH mitigating systems required to achieveOEM-level requirements, the high inertia and I-6 architecture ofthe conventional engine mean that the engine will be able to runwithout stalling or malfunction while performing compressioncycles.

4.2. Thermal considerations and requirements

The CNG will increase in temperature as it is compressed, andmust therefore be cooled to enable it to be stored in compliancewith regulations [22]. In practice, the rapid compression of NG thatthis engine-integral compression system makes available wouldrequire the active cooling of the NG before it can be placed intothe NG storage cylinder. To develop an understanding of the ther-mal requirements of this cooling system we can model the thermo-dynamics of the NG compression and heat transfer processes. Forthis study, heat transfer is considered (1) from the NG under com-pression to the engine components, (2) from the NG in the IPST tothe ambient air, and (3) from the NG in the storage cylinder to theambient air.

To model the thermal behavior of the NG under compressionwithout the engine cooling system, we can consider the compres-sion as a polytropic process. Under the assumptions of a polytropiccompression process (shown in Figs. 6 and 8), the CNG tempera-ture in the compression cylinder and storage tank increases from300 K at near-atmospheric pressure to more than 900 K at250 MPa, higher than the materials capability of the CNG tanks.As shown in Fig. 10, IPST temperature does not increase signifi-cantly in comparison to the compression cylinder and storage tankbecause its maximum pressure is smaller than that of the storagetank and compression cylinder.

To consider the effect of the stock engine cooling system, wemust consider conductive and convective heat transfer to the

Fig. 6. NG temperature within the compression cylinder under polytropic compression conditions.

Fig. 7. NG temperature within the compression cylinder including heat transfer to the engine cooling jacket.

134 M. Malakoutirad et al. / Applied Energy 156 (2015) 129137

engine components. The first thermal system that the compressedNG comes in contact with is the engine itself including cylinderwalls, head, and exhaust valves. These components are thermally

regulated during normal engine operation by the engine coolingjacket and heat exchanger systems which have a heat exchangecapacity of tens of kW of thermal power. For modeling the thermal

Fig. 8. NG temperature within the storage tank under polytropic compression conditions.

Fig. 9. NG temperature within the storage tank including heat transfer in engine cooling jacket, IPST, and storage tank.

M. Malakoutirad et al. / Applied Energy 156 (2015) 129137 135

behavior of the gas under compression, we must consider that thewall and NG are not at equilibrium. For this exercise, we considerconvection between the NG inside the cylinder and the cylinderwall (havg = 10 W m2 K1), zero radiative heat transfer (no

combustion), and steady state conduction, using the methodsdetailed in [23]. The temperature results for the compression cylin-der including heat transfer to the engine cooling jacket are shownin Fig. 7. Including the engine coolant system model makes the

Fig. 10. NG temperature within the IPST under the polytropic compression conditions.

Fig. 11. NG temperature within the IPST under compression conditions including heat transfer in engine cooling jacket and IPST.

136 M. Malakoutirad et al. / Applied Energy 156 (2015) 129137

temperature excursions of the NG smaller, but the heat transfer tothe engine is lower than would be characteristic of ICE operation,suggesting that high temperature gas will be transferred to theIPST and storage tank components of the compression system.

Thermal regulation of the IPST and storage tank is achieved bydeveloping a NG-to-air cooling system. These components aremore amenable to air cooling than is the compression cylinderbecause the storage tank has high volume and surface area, and

M. Malakoutirad et al. / Applied Energy 156 (2015) 129137 137

the IPST has lower maximum pressure and therefore lower maxi-mum temperature than the compression cylinder. A model hasbeen developed to show the results of the NG-to-air cooling systemincluding high temperature NG transfer to the tanks, conductionthrough the tanks (kavg = 60 W m1 K1) and external forced airconvection (Tamb = 300 K, havg = 20 W m2 K1). The temperatureresults for the storage tank including heat transfer from the tankto ambient air are shown in Fig. 9. The temperature of the IPSTwithout and with cylinder and IPST convective cooling is shownin Fig. 11. For both components, the low power NG to air convec-tive cooling system is able to control the NG temperature. The tem-perature of the IPST including cooling fluctuates from between350 K and 270 K depending on the state of the compression cycle,and the temperature of the storage tank is controlled to within10 K of the ambient.

5. Conclusions and future work

This study has analyzed the design considerations required toenable the development and commercialization of anengine-integral reciprocating NG compressor. The structural andfunctional architecture of the concept show that theengine-integral reciprocating NG compressor can be constructedwith minimal additional parts-count and control system develop-ment from conventional diesel ICEs. The rotational dynamics, andthermodynamics of the NG compression system have been identi-fied as design considerations of note and this study has addressedthese potential barriers to the realization of the concept. The rota-tional dynamics considerations can be solved by using an IPST tosplit the compression forces into two stages, a large-scale I-6engine where opposing cylinders are undergoing compressionand combustion cycles simultaneously, and an application whereengine inertia is high enough to overcome any induced rotationaldynamics problems. The thermal considerations can be solved byremoving only a fraction of the high-pressure andhigh-temperature NG from each compression cycle, by maintain-ing the presence of the conventional engine cooling system to per-form thermal regulation of the cylinder and piston, and by enablingsimple natural convection around the IPST and storage tanks.

Disruptive technologies are those that can reach customers thatare not served by the mainstream product. They are generally tech-nologically straightforward, consisting of off-the-shelf componentsput together in a new product architecture that can realize differ-ent benefits that the incumbent technologies [24]. At present, com-pressed NG (CNG) vehicles are refueled using a high-cost,centralized, and sparse network of CNG fueling stations. As thesesystems were designed for the use of fleet customers (an incum-bent and high profit margin customer), they are often not availablefor public refueling.6 At the same time that the cost of NG to fuelCNGVs has gone down in recent years, a consumer demand forlow-cost, utilitarian, and environmentally preferable personal trans-portation has emerged with the production, and sales of electrifiedand hybridized vehicles. As such, an engine-integral reciprocatingNG compressor has the capability to disrupt the incumbent CNG

6 US Department of Energy, Alternative Fueling Station Locator, 2013, available athttp://www.afdc.energy.gov/locator/stations/.

market by enabling the use of NG for personal transportation, fueledat home, from the preexisting low-pressure NG infrastructure, at lowparts count, using conventional components, and therefore at lowincremental costs. By enabling low-cost NG fuel to enter the per-sonal transportation market, the potential for this or similar technol-ogy to disrupt the NG fueling energy sector is high.

References

[1] Wang M, Huang H. A full-fuel-cycle analysis of energy and emissions impactsof transportation fuels produced from natural gas. Transp Technol R&D Center1999;1:134.

[2] Parrish DD, Sign HB, Moline L, Madronich S. Air quality progress in NorthAmerican megacities: a review. Atmos Environ 2011;45(39):701525.

[3] Aslam M, Masjuki H, Kalam M, Abdesselam H, Mahlia T, Amalina M. Anexperimental investigation of CNG as an alternative fuel for a retrofittedgasoline vehicle. Fuel 2006;85(56):71724.

[4] Yang M, Kraft-Oliver T, Yan GX, Min WT. Compressed natural gas vehicles:motoring towards a cleaner Beijing. Appl Energy 1997;56(34):395405.

[5] Hekkert Marko P, Hendriks Franka Hjf, Faaij Andre Pc, et al. Natural gas as analternative to crude oil in automotive fuel chains well-to-wheel analysis andtransition strategy development. Energy Policy 2005;33(5):57994.

[6] Flynn Peter C. Commercializing an alternate vehicle fuel: lessons learned fromnatural gas for vehicles. Energy Policy 2002;30(7):6139.

[7] Engerer H, Horn M. Natural gas vehicles: an option for Europe. Energy Policy2010;38(2):101729.

[8] Peterson MB, Barter GE, West TH, Manley DK. A parametric study of light-dutynatural gas vehicle competitiveness in the United States through 2050. ApplEnergy 2014;125:20617.

[9] Xuping Li, Ogden Joan M, Kurani Kenneth S. An overview of automotive homeand neighborhood refueling. EVS24 international battery, hybrid and fuel cellelectric vehicle symposium, Stavanger, Norway, May 1316; 2009.

[10] Cao W, Lu X-S, Lin W, Gu A. Parameter comparison of two small-scale naturalgas liquefaction processes in skid-mounted packages. Appl Therm Eng2006;26(89):898904.

[11] Axworthy RT. The home-rechargeable natural gas commuter car. Methane1982:1318.

[12] Wegrzyn J, Gurevich M. Adsorbent storage of natural gas. Appl Energy1996;55(2):7183.

[13] Yeh S. An empirical analysis on the adoption of alternative fuel vehicles: thecase of natural gas vehicles. Energy Policy 2007;35(11):586575.

[14] Waldrop JL. Integral gas compressor and internal combustion engine. USPatent US5203680A; 1993.

[15] Mikalsen R, Roskilly AP. A review of free-piston engine history andapplications. Appl Therm Eng 2007;27(1415):233952.

[16] Golovoy, A, Braslaw J. On-board storage and home refueling options for naturalgas vehicles. SAE technical paper 830382; 1983. doi:http://dx.doi.org/10.4271/830382.

[17] Megel Marc, Westmoreland Barry, Jones Guy, Phillips Ford, Eberle Douglas,Tussing Mark. Development of a structurally optimized heavy duty dieselcylinder head design capable of 250 bar peak cylinder pressure operation. SAEtechnical paper 2011-01-2232; 2011.

[18] Kirkpatrick Allan T, Fergusen Colin R. Internal combustion engines: appliedthermosciences. 2nd ed. Wiley; 2000.

[19] Lyons William C, Plisga Gary J. Standard handbook of petroleum & natural gasengineering. 2nd ed. Amsterdam: Elsevier; 2005.

[20] Norton Robert L. Design of machinery: an introduction to the synthesis andanalysis of mechanisms and machines. New York: McGraw-Hill; 1992.

[21] Davis Roy I, Lorenz Robert D. Engine torque ripple cancellation with anintegrated starter alternator in a hybrid electric vehicle: implementation andcontrol. IEEE Trans Ind Appl 2003;39:10.

[22] Menon VC, Komarneni S. Porous adsorbents for vehicular natural gas storage: areview. J Porous Mater 1998;1:16.

[23] Pulkrabek Willard W. Engineering fundamentals of the internal combustionengine. 2nd ed. Upper Saddle River, N.J.: Pearson Prentice Hall; 2004.

[24] Christensen Clayton M. The innovators dilemma: when new technologiescause great firms to fail. Boston, Mass: Harvard Business School Press; 1997.

http://refhub.elsevier.com/S0306-2619(15)00831-4/h0005http://refhub.elsevier.com/S0306-2619(15)00831-4/h0005http://refhub.elsevier.com/S0306-2619(15)00831-4/h0005http://refhub.elsevier.com/S0306-2619(15)00831-4/h0010http://refhub.elsevier.com/S0306-2619(15)00831-4/h0010http://refhub.elsevier.com/S0306-2619(15)00831-4/h0015http://refhub.elsevier.com/S0306-2619(15)00831-4/h0015http://refhub.elsevier.com/S0306-2619(15)00831-4/h0015http://refhub.elsevier.com/S0306-2619(15)00831-4/h0020http://refhub.elsevier.com/S0306-2619(15)00831-4/h0020http://refhub.elsevier.com/S0306-2619(15)00831-4/h0025http://refhub.elsevier.com/S0306-2619(15)00831-4/h0025http://refhub.elsevier.com/S0306-2619(15)00831-4/h0025http://refhub.elsevier.com/S0306-2619(15)00831-4/h0030http://refhub.elsevier.com/S0306-2619(15)00831-4/h0030http://refhub.elsevier.com/S0306-2619(15)00831-4/h0035http://refhub.elsevier.com/S0306-2619(15)00831-4/h0035http://refhub.elsevier.com/S0306-2619(15)00831-4/h0040http://refhub.elsevier.com/S0306-2619(15)00831-4/h0040http://refhub.elsevier.com/S0306-2619(15)00831-4/h0040http://refhub.elsevier.com/S0306-2619(15)00831-4/h0050http://refhub.elsevier.com/S0306-2619(15)00831-4/h0050http://refhub.elsevier.com/S0306-2619(15)00831-4/h0050http://refhub.elsevier.com/S0306-2619(15)00831-4/h0120http://refhub.elsevier.com/S0306-2619(15)00831-4/h0120http://refhub.elsevier.com/S0306-2619(15)00831-4/h0125http://refhub.elsevier.com/S0306-2619(15)00831-4/h0125http://refhub.elsevier.com/S0306-2619(15)00831-4/h0055http://refhub.elsevier.com/S0306-2619(15)00831-4/h0055http://refhub.elsevier.com/S0306-2619(15)00831-4/h0065http://refhub.elsevier.com/S0306-2619(15)00831-4/h0065http://dx.doi.org/10.4271/830382http://dx.doi.org/10.4271/830382http://refhub.elsevier.com/S0306-2619(15)00831-4/h0080http://refhub.elsevier.com/S0306-2619(15)00831-4/h0080http://refhub.elsevier.com/S0306-2619(15)00831-4/h0085http://refhub.elsevier.com/S0306-2619(15)00831-4/h0085http://refhub.elsevier.com/S0306-2619(15)00831-4/h0090http://refhub.elsevier.com/S0306-2619(15)00831-4/h0090http://refhub.elsevier.com/S0306-2619(15)00831-4/h0095http://refhub.elsevier.com/S0306-2619(15)00831-4/h0095http://refhub.elsevier.com/S0306-2619(15)00831-4/h0095http://refhub.elsevier.com/S0306-2619(15)00831-4/h0100http://refhub.elsevier.com/S0306-2619(15)00831-4/h0100http://refhub.elsevier.com/S0306-2619(15)00831-4/h0105http://refhub.elsevier.com/S0306-2619(15)00831-4/h0105http://refhub.elsevier.com/S0306-2619(15)00831-4/h0110http://refhub.elsevier.com/S0306-2619(15)00831-4/h0110http://www.afdc.energy.gov

Design considerations for an engine-integral reciprocating natural gas compressor1 Introduction2 Structural description3 Functional description and analysis3.1 In-cylinder natural gas compression process3.2 IPST tank pressurization process3.3 Storage tank pressurization process

4 Detailed design considerations4.1 Dynamic torques and rotational energy requirements4.2 Thermal considerations and requirements

5 Conclusions and future workReferences