Investigation of a Railplug Ignition System for Lean-burn Large-bore Natural Gas Engines PDF Download

Author: Hongxun Gao
Publisher:
ISBN:
Category : Gas as fuel
Languages : en
Pages :

View

Book Description

Author: Michael Herbert Koenig
Publisher:
ISBN:
Category : Automobiles
Languages : en
Pages : 274

View

Book Description

Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :

View

Book Description
To meet the ignition system needs of large bore, high pressure, lean burn, natural gas engines a side pumped, passively Q-switched, Nd:YAG laser was developed and tested. The laser was designed to produce the optical intensities needed to initiate ignition in a lean burn, high compression engine. The laser and associated optics were designed with a passive Q-switch to eliminate the need for high voltage signaling and associated equipment. The laser was diode pumped to eliminate the need for high voltage flash lamps which have poor pumping efficiency. The independent and dependent parameters of the laser were identified and explored in specific combinations that produced consistent robust sparks in laboratory air. Prior research has shown that increasing gas pressure lowers the breakdown threshold for laser initiated ignition. The laser has an overall geometry of 57x57x152 mm with an output beam diameter of approximately 3 mm. The experimentation used a wide range of optical and electrical input parameters that when combined produced ignition in laboratory air. The results show a strong dependence of the output parameters on the output coupler reflectivity, Q-switch initial transmission, and gain media dopant concentration. As these three parameters were lowered the output performance of the laser increased leading to larger more brilliant sparks. The results show peak power levels of up to 3MW and peak focal intensities of up to 560 GW/cm2. Engine testing was performed on a Ricardo Proteus single cylinder research engine. The goal of the engine testing was to show that the test laser performs identically to the commercially available flashlamp pumped actively Q-switched laser used in previous laser ignition testing. The engine testing consisted of a comparison of the in-cylinder, and emissions behavior of the engine using each of the lasers as an ignition system. All engine parameters were kept as constant as possilbe while the equivalence ratio (fueling), and hence the engine load, was varied between 0.8, 0.9, and 1.0. The test laser was constructed with a 30% output coupler, 32% Q-switch initial transmission, and a 0.5% Nd concentration rod all pumped by approximately 1000 Watts of optical power. The test laser single mode output pulse had an energy of approximately 23 mJ, with a pulsewidth of approximately 10 ns, and an M2 value of 6.55. This output produced focal intensity of approximately 270 GW/cm2 with the modified on-engine optical arrangement. The commercial laser had similar output parameters and both laser systems operated the engine with similar results. Due to the shortening of the focal length of the on-engine optical setup both laser systems produced a spark well within the optical transfer cavity of the laser optics to spark plug adaptor. This shrouded spark led to a very long ignition delay and retarded combustion timing for all three values of equivalence ratio. This was evidenced by the in-cylinder pressure traces and the HRR waveforms. The emissions data indicate that both lasers produced very similar combustion. The ignition delay caused by the shrouded spark cause most of the combustion to happen after TDC which lead to poor combustion that produced high levels of CO and THC. The novelty of this work lies in the combination of the laser parameters to create a single high peak power laser output pulse for use as a spark ignition source. Similar configurations have been investigated in the literature but for different applications such as multiple output pulse trains for various industrial and communications applications. Another point of novelty is the investigation of the laser medium concentration on the output characteristics of a passively Q-switched laser system. This work has shown that lowering the Neodymium concentration in the active media within a passively Q-switched laser produces higher output energy values. This is significant because an actively Q-switched laser shows the opposite affect when the active ion concentration is varied.

Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :

View

Book Description
This Final Technical Report discusses the progress that was made on the experimental and numerical tasks over the duration of this project. The primary objectives of the project were to (1) develop an improved understanding of the spark ignition process, and (2) develop the railplug as an improved ignitor for large bore stationary natural gas engines. We performed fundamental experiments on the physical processes occurring during spark ignition and used the results from these experiments to aid our development of the most complete model of the spark ignition process ever devised. The elements in this model include (1) the dynamic response of the ignition circuit, (2) a chemical kinetics mechanism that is suitable for the reactions that occur in the plasma, (3) conventional flame propagation kinetics, and (4) a multi-dimensional formulation so that bulk flow through the spark gap can be incorporated. This model (i.e., a Fortran code that can be used as a subroutine within an engine modeling code such as KIVA) can be obtained from Prof. Ron Matthews at [email protected] or Prof. DK Ezekoye at [email protected]. Fundamental experiments, engine experiments, and modeling tasks were used to help develop the railplug as a new ignitor for large bore natural gas engines. As the result of these studies, we developed a railplug that could extend the Lean Stability Limit (LSL) of an engine operating at full load on natural gas from [phi] = 0.59 for operation on spark plugs down to [phi] = 0.53 using railplugs with the same delivered energy (0.7 J). However, this delivered energy would rapidly wear out the spark plug. For a conventional delivered energy (

Author: Henry Ward Beecher
Publisher:
ISBN: 9781163472019
Category :
Languages : en
Pages : 428

View

Book Description

Author:
Publisher:
ISBN:
Category :
Languages : en
Pages : 45

View

Book Description

Author: Roger Röthlisberger
Publisher:
ISBN:
Category :
Languages : en
Pages : 251

View

Book Description

Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :

View

Book Description
The primary objective of this cooperative research is to develop and verify models of internal combustion engine spark ignition devices in order to improve combustion chamber fuel ignition characteristics and to improve spark plug durability. As a direct result of this joint research, a novel spark plug design was improved. A theory of spark arc motion was developed that explains experimentally observed effects not explained by other published theories. The knowledge developed by this research will be used to further improve spark plugs as well as improve the ignition process in a combustion chamber. The predictive models developed here are compared with experimental measurements, including high-speed photographs, of the spark as it translates across the gap. Two different spark plug configurations were investigated: the conventional or J-gap plug, and a novel spark ignition device (the FANG plug) invented by Cummins, Inc., the CRADA partner. A description of the physics of arc dynamic motion in a spark plug gap, including the effects of an imposed transverse magnetic field, appears here in Appendix A as a result of the analytical effort. The theory proposed here does explain experimentally observed effects not completely explained by other research publications appearing in the scientific literature. These effects are due to pressure and ion, electron, and electrode interactions. A dominant mechanism for electrode erosion is presented for both spark plug configurations. Reversing the polarity of both types of spark plugs has verified this proposed erosion mechanism, according to data collected at Cummins. An extensive series of experiments measured the arc position, voltage, and current as a function of time during the approximately 2 millisecond spark discharge. FANG plug data, obtained with the fast-framing camera experimental apparatus operating at 200,000 frames per second, are presented that show the transverse arc velocity varying directly as the inverse square root of the elapsed time since arc initiation. At the request of Cummins, experiments were performed on three conventional spark plugs identical in design and having the same spark gap, but differing as follows: one was new, another had been used in an engine, and the third was new but had been sandblasted to simulate a used plug. Cummins had observed that only the used plug required a significantly higher breakdown voltage. Experiments at ORNL indicated that the used plug had a significantly higher breakdown voltage confirming the Cummins observations (although the sandblasted plug also exhibited a higher breakdown voltage than the new plug but lower than the used plug), and thus an apparent increase of the arc breakdown voltage results as the plug ages in use. Further analysis of this phenomenon is warranted.

Author: Arminta S. Chicka
Publisher:
ISBN:
Category :
Languages : en
Pages :

View

Book Description

Author: Ron Matthews
Publisher:
ISBN:
Category :
Languages : en
Pages : 23

View

Book Description
During the first two years of this project, four experimental subtasks were to have begun but only one of these was to have been completed. Additionally, six modeling subtasks were scheduled to begin, five of which were to have been completed. We are on schedule for all but two of these 10 subtasks. The experimental tasks are on schedule. A second round of durability tests has been developed and testing has begun. It is too early to draw conclusions from this second round of durability testing. The test engine has been moved to a new dyno with superior controls. The baseline tests have been repeated, revealing that the engine is somewhat more dilution tolerant than originally concluded. Railplug testing has begun, but it is too early to draw any conclusions from these tests. A new railplug design was generated. It is a hybrid between the coax and parallel designs that we refer to as the semi-parallel railplug. Development of a model for the railplug ignition process was scheduled for completion during the fourth 6-month period. This task consists of three elements. First, a railplug circuit model was developed and validated during the third 6 months. Second, an analytical model was developed for the effects of geometric and circuit parameters on the Lorentz force. From this model, it was concluded that thermal expansion is important to the performance of railplugs. Thermal expansion and other physical effects are incorporated in the numerical model that is the third element of Task 2.2. Although significant progress was made on this last model, unforeseen numerical problems were encountered due to the unusual nature of the boundary conditions for the electromagnetic force. We expect to find a solution to this problem in the near future. We delayed the development of a 3D model for the ignition process until near the end of the project because of the computational time requirements. We can learn most of the important lessons from the 2D model. Delay of this subtask will not affect the timely completion of the project. Progress has also been made in the technology transfer task. A third paper on the ignition process has been drafted. It will be submitted to a journal in the near future. Also, we have begun discussion with Stitt Spark Plug Company regarding commercialization of the railplug. Stitt makes spark plugs for large bore natural gas engines, and for other applications.