0
Research Papers

# Evaluating the Potential for Gasoline Geysering From Small Engine Fuel TanksOPEN ACCESS

[+] Author and Article Information
Todd M. Hetrick

Exponent, Inc.,
4580 Weaver Parkway,
Suite 100,
Warrenville, IL 60555
e-mail: thetrick@exponent.com

Suzanne A. Smyth

Exponent, Inc.,
4580 Weaver Parkway,
Suite 100,
Warrenville, IL 60555
e-mail: ssmyth@exponent.com

Russell A. Ogle

Exponent, Inc.,
4580 Weaver Parkway,
Suite 100,
Warrenville, IL 60555
e-mail: rogle@exponent.com

Juan C. Ramirez

Mem. ASME
Exponent, Inc.,
4580 Weaver Parkway,
Suite 100,
Warrenville, IL 60555
e-mail: jramirez@exponent.com

1Present address: Rimkus Consulting Group, Inc., 7501 S. Quincy Street, Suite 160, Willowbrook, IL 60527.

Manuscript received March 5, 2015; final manuscript received September 1, 2017; published online October 3, 2017. Assoc. Editor: Chimba Mkandawire.

ASME J. Risk Uncertainty Part B 4(2), 021001 (Oct 03, 2017) (4 pages) Paper No: RISK-15-1046; doi: 10.1115/1.4037866 History: Received March 05, 2015; Revised September 01, 2017

## Abstract

This paper explores an infrequently encountered hazard associated with liquid fuel tanks on gasoline-powered equipment using unvented fuel tanks. Depending on the location of fuel reserve tanks, waste heat from the engine or other vehicle systems can warm the fuel during operation. In the event that the fuel tank is not vented and if the fuel is sufficiently heated, the liquid fuel may become superheated and pose a splash hazard if the fuel cap is suddenly removed. Accident reports often describe the ejection of liquid as a geyser. This geyser is a transient, two-phase flow of flashing liquid. This could create a fire hazard and result in splashing flammable liquid onto any bystanders. Many existing fuel tank systems are vented to ambient through a vented tank cap. It has been empirically determined that the hazard can be prevented by limiting fuel tank gauge pressure to 10 kPa (1.5 psi). However, if the cap does not vent at an adequate rate, pressure in the tank can rise and the fuel can become superheated. This phenomenon is explored here to facilitate a better understanding of how the hazard is created. The nature of the hazard is explained using thermodynamic concepts. The differences in behavior between a closed system and an open system are discussed and illustrated through experimental results obtained from two sources: experiments with externally heated fuel containers and operation of a gasoline-powered riding lawn mower. The role of the vented fuel cap in preventing the geyser phenomenon is demonstrated.

<>

## Introduction

Gasoline-powered equipment has remained a centerpiece of the industrialized world since scaled manufacturing began in the early 20th century. The fundamental makeup of these systems has not changed in that a reserve fuel tank is mounted on the equipment and fuel is gradually depleted during operation as it is withdrawn for combustion in the engine. To provide for continued operation, the fuel tank must intermittently be refilled, which requires accessing the tank's interior by removal of a cap or other tank closure. This paper examines a potential hazard associated with superheating gasoline in a closed system.

Depending on the location of fuel reserve tanks, waste heat from the engine or other vehicle systems can warm the reserve fuel during operation. In the event that the tank is not vented, and if the fuel is sufficiently heated, the liquid fuel may become superheated and pose a splash hazard if the fuel cap is suddenly removed. Sudden depressurization of the superheated fluid will allow liquid fuel to rapidly flash into vapor. Depending on the fuel type, degree of superheating, tank geometry, and amount of fuel in the tank, a transient two-phase flow of flashing liquid can potentially erupt out of the tank.

There is one historical precedent for which it was alleged that this is a reoccurring problem [1]. Research on the Consumer Product Safety Commission website yielded no examples of a recall initiated because of the geyser effect in a fuel tank. A general internet search reveals anecdotal reports involving a variety of types of equipment [24]. Much of the information available in the public domain about these alleged incidents is substantially incomplete or unverified. In most of the instances reviewed, it could not be determined whether the equipment had been modified, if the equipment had been properly operated and maintained, or whether the fuel tank was vented through the use of a vented cap or other means. Therefore, from a scientific point of view, these anecdotal reports are informative, but not probative.

The authors note that classic system designs incorporating vented fuel tank caps are being phased out as a result of regulatory efforts to further limit fugitive fuel vapor emissions to the environment. The analysis and experimental work discussed in this work are limited to applications utilizing a vented cap to prevent superheating of the fuel.

Reports of a spray or geyser of gasoline venting from a small-engine fuel tank are rare events. The diversity of small engine operating characteristics and fuel tank geometries makes it difficult to generalize about the dynamics of these incidents. However, one factor that must be present for a gasoline geyser to occur is a flashing two-phase flow. In this paper, we describe the thermodynamics behind the gasoline geyser, explain why gasoline geysers are rare events, and demonstrate why vented fuel caps can prevent this hazard.

## Thermodynamics of Superheated Pressurized Fuel Tanks

When a liquid–vapor mixture is in thermodynamic equilibrium within a closed system, it can be stored at temperatures above its normal boiling point and remain a liquid–vapor mixture. The liquid in this context is often described as superheated. If the tank or containment device is suddenly depressurized, the liquid will partially flash into vapor. This is one of the key physical phenomena behind the so-called BLEVEs (boiling-liquid, expanding-vapor explosions).

By assuming that the energy stored in the superheated liquid is fully used to vaporize some fraction of the liquid, it can be shown (e.g., reference Crowl and Louvar [5]) that the fraction of liquid vaporized, fv, upon a sudden de-pressurization to atmospheric level, is given by Display Formula

(1)$fv=Cp(T0–Tb)hfg$
where Cp is the liquid's specific heat, T0 is the temperature of the liquid before depressurization, Tb is the normal (atmospheric pressure) boiling temperature, and hfg is the liquid's latent heat of vaporization. Equation (1) assumes constant physical properties across the (T0Tb) temperature range.

We present a sample calculation in this section to determine the fraction of liquid vaporized by using hexane as a simple analog for gasoline (hexane is typically present in gasoline on the order of 1% by mass). Commercially available gasoline is a complex mixture of many hydrocarbons, blending agents, and additives. Hence, a flashing calculation for this multicomponent mixture could not be accurately performed by a simplified model like that of Eq. (1). Instead, one would have to resort to empirical measurements. Nevertheless, the hexane calculations are reasonably expected to be representative of the order of magnitude behavior for gasoline flashing.

From Fig. 1, it can be seen that for a relatively modest overpressure inside the tank of say, 14 kPag (2 psig), the fraction of the liquid inventory that will flash into vapor is approximately 0.03 (3%). However, the temperature at which the tank has to be at the moment of sudden depressurization is approximately 73 °C (163 °F). For this phenomenon to occur with hexane (and by extension with gasoline), there has to be considerable heating of the tank, which itself has to be a closed (i.e., nonvented) system.

The manual removal of a screwed-on fuel tank cap by a small engine operator may not be directly analogous to a theoretical sudden depressurization (infinite rate of pressure change). Over the finite time, as the cap is removed, some vapor will be released and the system pressure will be reduced at a finite rate. A gradual reduction in pressure will have a mitigating effect on the potential and severity of flashing.

Ramirez, Smyth, and Ogle have demonstrated that the exergy is a useful way to determine the maximum thermodynamic work of a flashing system [6,7]. In this case, the work is performed by the flashing system by the upward expulsion of the vapor–liquid mixture. The exergy of the system increases with the degree of liquid superheat (the magnitude of the temperature elevation above the normal boiling point), which determines the fraction of liquid vaporized. Provided that vapor–liquid equilibrium prevails, a higher degree of superheat corresponds to a greater potential for a gasoline geyser.

## Gasoline Geyser Effect Caused by Superheating a Closed System

Even a small fraction of the liquid inventory flashing to vapor can cause a geyser effect because of the large change in specific volume. Indeed—for the conditions described in the paragraph above—the specific volume of hexane increases approximately 260-fold practically instantaneously upon flashing from liquid to vapor. This sudden expansion contributes to the rapid sloshing and expansion of the liquid volume due to interior bubble generation and can result in liquid ejection, if the liquid surface is close to the tank opening.

This discussion highlights the hazards of superheating a closed (i.e., nonvented) system containing a flammable liquid. Berry and Sevart have done experimental work showing precisely this phenomenon [8]. They immersed partially filled (reportedly 80% liquid full) closed containers with commercially available gasoline in a heated water bath for several hours. Bath temperatures ranged from 49 °C to 54.5 °C (120 °F to 130 °F). At these temperatures, they reported internal pressures of 41 kPa to 45 kPa (6 psig to 6.5) psig prior to the sudden depressurization. Depressurization was achieved by a rapid, mechanical removal of the cap on the container. The depressurization of the container resulted in a transient, flashing, two-phase flow (geyser) of gasoline expelled from the container. In one of their tests, they reported that approximately half of the liquid inventory was ejected from the container.

The flashing two-phase flow caused by the depressurization of a superheated liquid is a well-understood phenomenon [913]. This type of release is complex, especially for a multicomponent liquid like gasoline, and is predicated on bubble formation and disengagement from the liquid [12]. The determination if a release will be single- or two-phase is dependent on the liquid properties, liquid temperature, pressure in the vessel vapor space, vessel geometry, fractional liquid fill level, liquid surface area, and the properties of the hole or other release pathway [13].

Figure 1 also reveals that venting (providing means to prevent pressure buildup) eliminates the flashing hazard; that is, at 0 kPag (0 psig) internal tank pressure, the fraction of liquid that flashes is zero. This was demonstrated in another test by Berry and Sevart. A test was conducted with gasoline heated to a temperature of 32 °C (90 °F), and the headspace was pressurized with compressed air to a pressure of 93 kPa (13.5 psig). Depressurization of the system occurred by single-phase (gas) flow. No geyser was observed because the system was not superheated.

It has been empirically determined that flashing two-phase flow can be prevented by keeping the fuel tank pressure below 10 kPa (1.5 psi) above ambient [14]. This pressure threshold is simply a heuristic and is not a formal thermodynamic criterion. Depending on the design details of the fuel system, it is possible that the threshold fuel tank pressure to mitigate or prevent flashing two-phase flow may be greater than 10 kPag (1.5 psig).

## Normal Operation of Fuel Tanks Results in Heating of an Open System

While the discussion in the Gasoline Geyser Effect Caused by Superheating a Closed System section examined a closed system, this does not adequately represent the fuel systems of conventional, real world equipment. Fuel tanks are not closed systems; they are open systems. By necessity, fuel is drawn out of the tank and passed through filters, pumps, and into the carburetor during operation. Although each system will be specific to the equipment it is used with, the primary flow path will likely be similar and is depicted in Fig. 2.

Other components, such as a primer pump, a second fuel tank, fuel tank selector valves, and fuel return lines may also be present and the fuel tank itself may have multiple inlets and outlets.

The fuel tanks for each piece of equipment will differ considerably both in size and geometry. The fuel tank capacity can range between approximately 8 ounces for a small chainsaw to approximately 8 gallons for a single tank on a large riding mower. The amount of fuel in the tank will affect the absorption of heat and the head space available for expansion. The liquid volume and the height of the liquid interface decrease with fuel consumption. The tanks are often in complex geometries to fit around the other systems on the equipment. A wide variety of fuel tank shapes exist in the marketplace and some examples of fuel tanks from small gasoline engine powered lawn equipment are shown in Fig. 3. The geometry of the tank will affect the heat transfer pathways to the fuel, the path of a potential release, and the surface area of fuel available for evaporation.

Fuel tank caps are usually vented and allow both the release of fuel vapor and the inflow of air. The caps often include baffles to inhibit the release of sloshing liquid fuel. Based upon the wide variety of geometries, operating conditions, and equipment, it is not surprising that little information is publically available that outlines design parameters for the venting of small fuel tanks.

Since the full fuel system of small gasoline engine powered equipment is not adequately represented by the closed systems discussed in the previous sections, the actual conditions of the fuel system on a riding lawn mower were examined in situ. The mower was operated at varying loads, throttle positions, and environmental conditions, with vented and unvented caps. The experimental mower was equipped with two fuel tanks, where only one tank is selected at a given time. These tests illustrate the complex dynamic behavior of real fuel tanks.

Over all the test conditions, a maximum temperature of 53.5 °C (128 °F) was achieved in the fuel tank nearest the exhaust system. This temperature was not reached until over 1.5 h of operation and the corresponding pressure inside the fuel tank was measured to be 3.25 kPa (0.47 psig). This scenario corresponded to operation with 100% throttle, 100% load, full sun exposure, and an ambient temperature of 30 °C (86 °F). A plot of the liquid fuel tank temperature and headspace pressure in the selected fuel tank is shown in Fig. 4. The maximum fuel temperature was attained near the end of the test when the fuel level in the tank was down to a fractional fill level of approximately 20%.

The mower was operated with an unvented cap under similar conditions of 100% throttle, 100% load, full sun, 30 °C (86 °F) ambient temperature. The maximum temperature in the tank nearest the exhaust was 50.5 °C (123 °F), which was similar to the vented cap results. The measured pressure within the selected fuel tank was in vacuum (subatmospheric) for nearly the entirety of the test as shown in Fig. 5.

In order to achieve vacuum pressures, the effect of the outflow of fuel on the total pressure was likely greater than the effect of vapor production.

Data were also collected with the mower sitting in the sun with an unvented cap, but not operating. The temperature and pressure measured within the nonselected fuel tank are shown in Fig. 6.

This would be the closest approximation to a closed system possible on the real world equipment. A maximum temperature of 33.5 °C (92 °F) was reached in the tank and the maximum pressure was 15 kPa (2.2 psig). Using a nonvented cap, the final pressure was slightly above the pressure threshold of 10 kPag (1.5 psig). The higher pressure observed in this experiment demonstrates the resulting difference between an open system (vented fuel tank and/or a tank being drafted for engine operation) and an unvented tank that is allowed to progress toward thermodynamic equilibrium.

In a real fuel system, the potential of the fuel to become superheated is dependent on numerous parameters including, but not limited to, the quantity of fuel, the heat transfer to the fuel, the capacity and geometry of the fuel tank, and the inlets and outlets of the fuel tank. While a closed system is useful for simply demonstrating the potential effect of a release of superheated fuel, it neglects the complexity of a real system.

Heat transfer to the fuel inside the tank is complex and depends on the geometry of the tank, the layout of the engine components and their proximity to the fuel tank, the outflow of liquid and vapor fuel, and the ambient environmental conditions. It takes time for heat from the exhaust or other components to transfer into the fuel. As shown in the experimental results, the maximum temperature will likely occur after the mower has been operating and a significant amount of fuel has been consumed. In the mower tested, the liquid surface in the tank dropped to a fractional fill level of approximately 20% after 2 h of operation. The heat transfer in a real system is not well represented by a closed system at equilibrium.

## Safeguards to Prevent the Geyser Phenomenon

The hazard of a gasoline geyser is predicated on the attainment of superheating the liquid gasoline in the fuel tank. This hazard can be avoided by preventing the attainment of vapor–liquid equilibrium in the fuel tank. One strategy to prevent this equilibrium is by transforming the fuel tank from a closed system into an open system. A standard safeguard on many fuel tanks is a vented fuel cap. The function of the vented fuel cap is to prevent the excessive pressurization of the fuel tank by allowing a leakage flow rate of gasoline vapor into the atmosphere. The vapor leakage is controlled by the design of the vent orifice and cap design. An empirical criterion for avoiding the flashing two-phase flow hazard is to keep the fuel tank pressure below 10 kPag (1.5 psig). However, depending on the design details of the fuel system, it is possible that the threshold fuel tank pressure to mitigate or prevent flashing two-phase flow may be greater than 10 kPag (1.5 psig).

Another factor that causes a fuel tank to behave as an open system is the fuel tank outlet nozzle. This is the outlet port from which fuel is withdrawn from the tank and transported to the carburetor. The withdrawal of liquid fuel from the tank during engine operation creates a moving vapor–liquid interface within the tank. The descending liquid interface increases the vapor headspace volume resulting in a decrease in the headspace pressure.

Depending on the design of the fuel system, there may be other factors at work, which prevent the fuel tank from becoming a closed system.

## Conclusions

On rare occasions, small-engine fuel tanks have experienced gasoline spraying or venting when the fuel cap is removed from the tank. The expelled gasoline can become a fire hazard to bystanders. This hazard is fundamentally caused by the superheating of the gasoline inside the fuel tank. The potential to cause superheating appears to be the result of improper modification of the fuel system. A criterion for avoiding the flashing two-phase flow is to keep the fuel tank pressure below 10 kPag (1.5 psig); although, depending on the design of the fuel system, further testing may reveal that higher pressures can be tolerated. Vented fuel tank caps are being phased out in many applications to further limit emissions to the environment; however, existing equipment utilizing vented caps currently remain in widespread use. A vented fuel cap acts to reduce the headspace pressure and can control the hazard.

## References

FTC, 1984, “ Complaint: In the Matter of International Harvester Company, Final Order, Opinion, etc. in Regard to Violation of Sec. 5 of the Federal Trade Commission Act,” Federal Trade Commission, Washington, DC, Docket No. 9147, pp. 949–1088.
Thumper Talk, 2005, “ Gas Tank Geyser!,” ThumperTalk, Inc., Las Vegas, NV, accessed Apr. 14, 2014,
Can-AmTalk, 2006, “ Gas Tank Geyser…,” DOOTalk, LLC, Rochester, NY, accessed Apr. 14, 2014,
FJR Owners, 2010, “ Fuel Geyser! Help!,” Yamaha FJR Owners Forums, accessed Apr. 14, 2014,
Crowl, D. A. , and Louvar, J. F. , 2001, Chemical Process Safety, 2nd ed., Prentice Hall, Upper Saddle River, NJ.
Ogle, R. A. , Ramirez, J. C. , and Smyth, S. A. , 2012, “ Calculating the explosion energy of a Boiling Liquid Expanding Vapor Explosion Using Exergy Analysis,” Process Saf. Prog., 31(1), pp. 51–54.
Ramirez, J. C. , Ogle, R. A. , and Smyth, S. A. , 2011, “ Towards an Exergy-Based Explosion Energy Model for Boiling-Liquid Expanding-Vapor Explosions,” ASME Paper No. IMECE2011-62228.
Berry, T. , and Sevart, K. , 2013, “ Geysering Gasoline—The Hidden Hazard,” ASME Paper No. IMECE2013-63730.
AIChE, 1998, Guidelines for Pressure Relief and Effluent Handling System, American Institute of Chemical Engineers, New York.
Fisher, H. G. , Forrest, H. S. , Grossel, S. S. , Huff, J. E. , Muller, A. R. , Noronha, J. A. , Shaw, D. A. , and Tilley, B. J. , 1993, “ Emergency Relief System Design Using DIERS Technology: The Design Institute for Emergency Relief Systems (DIERS) Project Manual,” American Institute of Chemical Engineers, New York.
Darby, R. , Meiller, P. R. , and Stockton, J. R. , 2001, “ Select the Best Model for Two-Phase Relief Sizing,” Chem. Eng. Progress, 97(5), pp. 56–65.
Sheppard, C. M. , 1992, “ Disengagement Predictions Via Drift Flux Correlation Vertical, Horizontal and Spherical Vessels,” Plant/Oper. Prog., 11(4), pp. 229–237.
Van den Bosch, C. J. H. , and Weterings, R. A. P. M. , eds., 2005, Methods for the Calculation of Physical Effects—Due to Releases of Hazardous Materials (Liquids and Gases), Committee for the Prevention of Disasters, The Hague, The Netherlands.
Hillstrom, T. P. , 1981, New Gas Cap Improves Driver Protection on Old Tractors, American Society of Agricultural Engineers, Chicago, IL.
Techtronic Industries North America, Inc., Ryobi 26 cc String Trimmers and Burshcutters, Model Nos. RY26500, RY26520, RY26540, RY26901, RY26921, RY26941, Figure B, Techtronic Industries North America, Inc., Anderson, SC.
Homelite Consumer Products, Inc., Homelite SXLAO Chain Saw UT-10045 Parts List - Engine Internals , Homelite Consumer Products, Inc., Anderson, SC.
Deere & Company, John Deere JS60, JS60H 21-Inch Walk-Behind Mower, Fuel Tank, Diagram PU02307, Deere & Company, Moline, IL.
Husqvarna Group, 2011, Poulan PRO Model PB26H54YT Repair Parts Manual, P/N 532 44 51-50, Husqvarna Group, Stockholm, Sweden.
View article in PDF format.

## References

FTC, 1984, “ Complaint: In the Matter of International Harvester Company, Final Order, Opinion, etc. in Regard to Violation of Sec. 5 of the Federal Trade Commission Act,” Federal Trade Commission, Washington, DC, Docket No. 9147, pp. 949–1088.
Thumper Talk, 2005, “ Gas Tank Geyser!,” ThumperTalk, Inc., Las Vegas, NV, accessed Apr. 14, 2014,
Can-AmTalk, 2006, “ Gas Tank Geyser…,” DOOTalk, LLC, Rochester, NY, accessed Apr. 14, 2014,
FJR Owners, 2010, “ Fuel Geyser! Help!,” Yamaha FJR Owners Forums, accessed Apr. 14, 2014,
Crowl, D. A. , and Louvar, J. F. , 2001, Chemical Process Safety, 2nd ed., Prentice Hall, Upper Saddle River, NJ.
Ogle, R. A. , Ramirez, J. C. , and Smyth, S. A. , 2012, “ Calculating the explosion energy of a Boiling Liquid Expanding Vapor Explosion Using Exergy Analysis,” Process Saf. Prog., 31(1), pp. 51–54.
Ramirez, J. C. , Ogle, R. A. , and Smyth, S. A. , 2011, “ Towards an Exergy-Based Explosion Energy Model for Boiling-Liquid Expanding-Vapor Explosions,” ASME Paper No. IMECE2011-62228.
Berry, T. , and Sevart, K. , 2013, “ Geysering Gasoline—The Hidden Hazard,” ASME Paper No. IMECE2013-63730.
AIChE, 1998, Guidelines for Pressure Relief and Effluent Handling System, American Institute of Chemical Engineers, New York.
Fisher, H. G. , Forrest, H. S. , Grossel, S. S. , Huff, J. E. , Muller, A. R. , Noronha, J. A. , Shaw, D. A. , and Tilley, B. J. , 1993, “ Emergency Relief System Design Using DIERS Technology: The Design Institute for Emergency Relief Systems (DIERS) Project Manual,” American Institute of Chemical Engineers, New York.
Darby, R. , Meiller, P. R. , and Stockton, J. R. , 2001, “ Select the Best Model for Two-Phase Relief Sizing,” Chem. Eng. Progress, 97(5), pp. 56–65.
Sheppard, C. M. , 1992, “ Disengagement Predictions Via Drift Flux Correlation Vertical, Horizontal and Spherical Vessels,” Plant/Oper. Prog., 11(4), pp. 229–237.
Van den Bosch, C. J. H. , and Weterings, R. A. P. M. , eds., 2005, Methods for the Calculation of Physical Effects—Due to Releases of Hazardous Materials (Liquids and Gases), Committee for the Prevention of Disasters, The Hague, The Netherlands.
Hillstrom, T. P. , 1981, New Gas Cap Improves Driver Protection on Old Tractors, American Society of Agricultural Engineers, Chicago, IL.
Techtronic Industries North America, Inc., Ryobi 26 cc String Trimmers and Burshcutters, Model Nos. RY26500, RY26520, RY26540, RY26901, RY26921, RY26941, Figure B, Techtronic Industries North America, Inc., Anderson, SC.
Homelite Consumer Products, Inc., Homelite SXLAO Chain Saw UT-10045 Parts List - Engine Internals , Homelite Consumer Products, Inc., Anderson, SC.
Deere & Company, John Deere JS60, JS60H 21-Inch Walk-Behind Mower, Fuel Tank, Diagram PU02307, Deere & Company, Moline, IL.
Husqvarna Group, 2011, Poulan PRO Model PB26H54YT Repair Parts Manual, P/N 532 44 51-50, Husqvarna Group, Stockholm, Sweden.

## Figures

Fig. 1

Fraction of liquid flashed. Hexane liquid–vapor in thermodynamic equilibrium in a pressurized tank.

Fig. 2

Schematic of a fuel system

Fig. 3

Four examples of fuel tanks from a:(a) string trimmer, (b) chainsaw, (c) push mower, and (d) riding mower [1518]

Fig. 4

Temperature (solid) and pressure (dotted) inside the fuel tank over time for a riding lawn mower at 100% throttle and 100% load running in full sun with a vented cap

Fig. 5

Temperature (solid) and pressure (dotted) inside the fuel tank over time for a riding lawn mower at 100% throttle and 100% load running in full sun with an unvented cap

Fig. 6

Temperature (solid) and pressure (dotted) inside the fuel tank over time for a riding lawn mower sitting in full sun, while not operating, with an unvented cap

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Articles from Part A: Civil Engineering
Related Proceedings Articles
Related eBook Content
Topic Collections