This invention relates to gas turbine combustion engines. In particular, this invention relates to an assembly for transporting expanding gasses to the first row of turbine blades.
Gas turbine combustion engines with can annular combustors require structures to transport the gasses coming from the combustors to respective circumferential portions of the first row of turbine blades, hereafter referred to simply as the first row of turbine blades. These structures must orient the flow of the gasses so that the flow contacts the first row of turbine blades at the proper angle, to produce optimal rotation of the turbine blades. Conventional structures include a transition, a vane, and seals. The transition transports the gasses to the proper axial location and directs the gasses into the vanes, which orient the gas flow circumferentially as required and deliver the gas flow to the first row turbine of blades. The seals are used in between the components to prevent cold air leakage into the hot gas path, and to smooth flow during the transition between the components.
Configurations of this nature reduce the amount of energy present in the gas flow as the flow travels toward the first row of turbine blades, and inherently require substantial cooling. Gas flow energy is lost through turbulence created in the flow as the flow transitions from one component to the next, and from cold air leakage into the hot gas path. Cold air leakage into the hot gas path through seals increases as seals wear due to vibration and ablation. Significant energy is also lost when the flow is redirected by the vanes. These configurations thus create inefficiencies in the flow which reduce the ability of the gas flow to impart rotation to the first row of turbine blades.
The cooled components are expensive and complicated to manufacture due to the cooling structures, exacting tolerance requirements, and unusual shapes. Layers of thermal insulation for such cooled components may wear and can be damaged. For example, vane surfaces and thermal insulation layers thereon are prone to foreign object damage due to their oblique orientation relative to the flow. Such damage may necessitate component repair or replacement, which creates costs in terms of materials, labor, and downtime. Thermal stresses also reduce the service life of the underlying materials. Further, the vanes and seals require a flow of cooling fluid. This requires energy and creates more opportunities for heat related component damage and associated costs.
Vanes are produced in segments and then assembled together to form a ring. This requires additional seals between the vane components, through which there may be more cold air leakage into the hot gas path. Further, these configurations usually require assembly of the components directly onto the engine in confined areas of the engine, which is time consuming and difficult.
The invention is explained in the following description in view of the drawings that show:
The inventors of the present system have designed an innovative arrangement, made of multiple, modular, interchangeable, flow directing assemblies. One such assembly is identified by the trademark NOVA-Duct™ by the assignee of the present invention. The combustor cans of the gas turbine combustor have been reoriented to permit the use of an assembly of components that direct individual gas flows from the combustor cans of a can annular combustor of a gas turbine combustion engine into a singular annular chamber immediately upstream and adjacent the first row of turbine blades. The inventors of the present system observed that prior configurations for delivering flows of can-annular combustors to the first row of turbine blades kept each flow separate and distinct from the other flows all the way to the first row of turbine blades. As a result, between each flow about to contact the first row of turbine blades there is a gap, or trailing edge, where there is reduced flow delivered to the blades. These trailing edges, which vary in magnitude from design to design, create flow disturbances and associated energy losses. Consequently, as the first row blades rotate, they alternately see regions of a high volume of very hot flow, and cooler regions of reduced flow. The blades thus experience rapidly changing temperatures and aerodynamic loads as they rotate through these regions, and these oscillations shorten blade life. The assembly eliminates walls between adjacent flows in the annular chamber. Eliminating the walls between adjacent flows eliminates trailing edges associated with the walls, and the accompanying energy losses. A recent design innovation, as disclosed in commonly assigned U.S. Pat. No. (7,721,547) to Bancalari et. al., incorporated by reference herein, replaces the conventional transition, seals, and vanes with an assembly of one piece transition ducts that transport expanded gasses from the combustion chamber directly to the first row of turbine blades, while simultaneously orienting the gas flow to properly interface with the first row of turbine blades. This orienting is achieved by curving and shaping each duct, and consequently each respective gas flow, along its length. By using fewer seals, aerodynamic losses due to seals are reduced, as are flow losses through the seals. The newer design uses the entire length of the duct to properly orient the flow, while the designs of the prior art used vanes at the end of the duct to orient the flow, which resulted in a relatively abrupt change in the flow direction, and associated energy losses. Further, this newer design reduces costs associated with manufacturing, assembly and maintenance.
Another recent design innovation, as disclosed in pending and commonly assigned U.S. patent application Ser. No. 12/190,060 to Charron filed on Aug. 12, 2008,and incorporated herein by reference, orients the combustor cans of the gas turbine combustor to permit the use of an assembly of components that form a straight path between each combustor can and a respective circumferential portion of the first row of turbine blades. In the Charron configuration, gasses flowing from each combustor can flow along an individual straight path, without mixing with any other flows, exit the assembly, and flow into the first row of turbine blades. As a result of these straight paths, there are fewer aerodynamic energy losses, and thus a greater amount of energy is delivered to the first row of turbine blades. The current arrangement improves upon the ideas presented in the incorporated documents.
The arrangement comprises multiple sets of flow directing structures, one for each combustor. Each flow directing structure may include a cone and an associated integrated end piece (“IEP”). A combustor in a conventional gas turbine engine may be oriented radially inward and axially downstream with respect to a gas turbine engine longitudinal axis. However the combustions cans in a gas turbine engine that uses the present arrangement may be oriented circumferentially and downstream with respect to the gas turbine engine longitudinal axis.
Combustion gas exits the combustor along a straight gas flow path longitudinal axis and is constrained discretely from other combustion gas flows emanating from other combustor cans until all gas flows reach a common annular chamber. Once in the annular chamber the gas flows may deviate from respective straight gas flow longitudinal axis, and the gas flows are no longer separated by structural walls. The gas flows then exit the annular chamber through the annular chamber outlet. The annular chamber outlet comprises a plane perpendicular to a downstream end of the annular chamber, where the gas enters the turbine first stage section.
Upon exiting the combustor, a cone directs gasses from the combustor can to the IEP. It is possible, however, that a cone not be used and the can itself discharge into an IEP. The associated IEP receives a gas flow from the cone and ultimately delivers the gas to the blades of a turbine first stage blade section. The IEP may deliver a portion of, or all of the gas flow it receives from the combustor to an adjacent and downstream IEP. The adjacent and downstream IEP may deliver a portion of the gas flow it receives from the IEP to the annular chamber outlet. It may also deliver a portion of the flow it receives from the IEP to another IEP further downstream from it, which may in turn deliver some or all of the gas flow to the annular chamber outlet. A gas flow that enters an IEP may flow through a total of two, three, or more IEPs before making its way entirely through the annular chamber.
How many IEPs a gas flow traverses between exiting a cone and fully exiting an annular chamber outlet depends in part on the angle between a combustor longitudinal axis and a plane perpendicular to the gas turbine longitudinal axis. The angle between the combustor longitudinal axis and the plane perpendicular to the gas turbine longitudinal axis may be influenced by the number of combustor cans present in the gas turbine engine. A smaller angle means the combustor is oriented more circumferentially with respect to the gas turbine longitudinal axis, and a larger angle means the combustor is oriented more axially with respect to the gas turbine longitudinal axis. A shallower angle will require more circumferential travel for every unit of distance traveled along the gas turbine longitudinal axis. Conversely, a larger angle will require less circumferential travel for every unit of distance traveled along the gas turbine longitudinal axis. A gas flow path through an annular chamber that requires more circumferential travel will necessarily span more IEPs than a gas flow through an annular chamber that requires less circumferential travel. An axial length of the annular chamber will also determine how many IEPs a gas flow will traverse before entirely exiting the annular chamber outlet and entering the turbine first stage blade section. In an embodiment, fillets resulting from transitions of flow defining walls can be tapered in a downstream direction to reduce a runout length of the fillet. This allows for an annular chamber 18 of a shorter axial length.
A stage of a conventional gas turbine engine may include a small length of space upstream of vanes, the vanes themselves, a gap between the vanes and blades, the blades themselves, and a length of space downstream of the blades. However, since the present arrangement eliminates the need for vanes, the first stage section of a gas turbine engine using the present arrangement does not include the vanes, but is instead and will be considered herein a small length of space upstream of the blades, the blades themselves, and a length of space downstream of the blades.
A perimeter of each gas flow exiting from a combustor can is fully bounded by structural walls of the flow directing structure. The gas flows exiting the annular chamber outlet are not separated from each other by structural walls, but are instead only partially bounded to keep them flowing within the walls of the annular chamber. Consequently, at some point between exiting the combustor and exiting the annular chamber outlet the flow must be transitioned from having a completely bounded perimeter to having only a partially bounded perimeter.
Furthermore, conventional combustor cans have a circular cross section, but configuring gas flows to abut but not intersect adjacent gas flows while entering the annular chamber and producing a single, annular flow to the first row of blades necessitates gas flows with non-circular cross sections. Consequently, the gas flow must be morphed from a gas flow with a circular cross sectional shape to a gas flow with a non-circular cross sectional shape as it travels along the overall gas flow longitudinal axis.
In addition, once adjacent gas flow paths actually abut, it is preferred that there be no wall to separate the gas flow paths. As a result, it is important that each flow be properly oriented so that no gas flow path intersects or overlaps another gas flow path. Each gas flow path must also be formed such that it comprises a collimated flow (i.e. non diverging or converging, where all molecules of the gas are flowing parallel to each other and to the longitudinal axis of the gas flow path, but may be flowing at different speeds) so that once a gas flow path's perimeter is not fully bounded by walls the gas flows do not diverge into adjacent gas flows. Such divergence would result in loss of aerodynamic efficiency. Even more ideal would be a flow comprising a uniform flow profile, where all molecules of the gas are flowing parallel to each other and to the longitudinal axis of the gas flow path, and where all the molecules are flowing at the same speed.
It is the innovative design of the IEP disclosed herein that permits it to: transition the gas flow from a fully bounded perimeter to a partially bounded perimeter; morph the flow from a circular cross section to a non-circular cross section; properly orient each gas flow so that no gas flow paths intersect or overlap; and generate a collimated flow within the gas flow prior to transitioning it to a partially bounded gas flow.
As used herein, a gas flow path refers to a gas flow path defined by walls when walls are present, and where walls are not present, the boundary of the gas flow path is defined by the plane created by a downstream projection of the wall where it exists upstream. In other words, if a wall ends at some point along the gas flow path, the boundary of the gas flow path is considered to be an extension of that wall. The IEP attempts to define gas flow paths such that when gas flows are not physically separated from each other they will not, in theory, mix. However, fluid dynamics, particularly in such a dynamic environment as a working gas turbine engine, make it essentially impossible to ascertain exactly what actual flow path a gas flow will actually take once partially unbounded. For example, perturbations in the gas flow downstream of the annular chamber may impart transient changes to the actual path the gas flow takes while flowing through the IEP. Furthermore, the interaction of the gas flows with each other and with walls in the IEP may cause the gas flows to flow in a manner other than through the path defined for it by the structure. In addition, changes in load levels in the gas turbine as well as atmospheric conditions etc. may influence the actual gas path the gas takes through the IEP. Hence, the disclosure focuses on the gas path as defined by the structure, not by the actual gas path the gas takes while in the IEP.
As best understood by the inventors, but not meant to be limiting in theory, gas flows entering the annular chamber adjust volume as necessary to fill the entire volume of the annular chamber, the annular flow exiting the annular chamber is a single annular flow, and a circumferential motion is imparted to some degree to every part of the single annular flow exiting the annular chamber outlet. It is believed that the single annular flow may not be uniform in nature, but it is more uniform than a plurality of discrete flows with walls there between. This uniformity increases aerodynamic efficiency and reduces the range of oscillatory mechanical loads on the blades. Furthermore, with a substantially straight gas flow path from the combustor to the annular chamber, aerodynamic losses resulting from excessive gas flow redirection are reduced, increasing engine efficiency.
Turning to the drawings,
The arrangement 10 is composed of multiple sets of flow directing structures 12. There is a flow directing structure 12 for each combustor (not shown). The combustion gasses from each combustor flow into a respective flow directing structure 12. Each flow directing structure includes a cone section 14 and an IEP 16. The IEPs 16 together form an annular chamber 18. Each gas flow enters the annular chamber 18 at discrete intervals circumferentially at an orientation that includes a circumferential component and an axial component with respect to the gas turbine engine longitudinal axis 20. Each gas flow originates in its respective combustor can and is directed as a discrete flow to the annular chamber 18. When discrete, each flow is separated by walls, but in the annular chamber 18 the flows are not separated by walls. The flows are still constrained to the annular chamber 18, but they are not separated from each other. Each IEP 16 abuts adjacent annular chamber ends at IEP joints 24.
Immediately downstream of the annular chamber 18 is the first row of turbine blades (not shown). In conventional can annular gas turbine combustion engines each flow is discrete until it leaves a transition immediately upstream of the first stage which, in the conventional gas turbine engine includes flow directing vanes and then a row of blades. The transitions keep the flows discrete until just before encountering flow directing vanes. The flow directing vanes may further divide the discrete flows prior to each flow reaching the blades. As such, the blades see varying amounts of combustions gasses as they rotate through the divided flows. The annular chamber 18 eliminates any walls that separate the flows, and also eliminates the first row of flow directing vanes that divide the flows. As a result, the flows are not divided, but rather are essentially a single, annular flow immediately prior to entering the first row of turbine blades. Each gas flow path enters the annular chamber 18 along an overall gas flow longitudinal axis 22. Once in the annular chamber 18 the walls that defined the top and bottom of each flow upstream cease to do so. In addition, the walls that define the inner and outer sides of the flow transition from straight walls to arcuate walls that partially define the annular chamber 18. As the gas flow path continues circumferentially through the annular chamber it simultaneously advances along the gas turbine longitudinal axis. As a result, the bottom of the gas flow path first reaches the annular chamber outlet (not shown) and at a circumferentially downstream location the top of the gas flow path then reaches the annular chamber outlet.
Conventional can combustors comprise a circular cross section, as does the combustion gas flow emanating from it. Were the discrete flows to remain circular as they entered the annular chamber the rotating blades would encounter arched oval arcs with hour-glass shaped areas devoid of combustion gas flow there between, and thus the blades would still encounter a significant range of mechanical loads as they rotate. Overlapping the circumferential ends of adjacent circular cross section flows would induce aerodynamic inefficiency in the flows and is therefore less preferable. In order to present an annular flow path a non-circular cross section for the gas flow path was chosen such that when combined in an annular chamber 18, they could unite into an annular flow with a cross section where it is believed every portion contains combustion gasses. Such a cross section is more uniform and thus the blades see a more uniform gas flow as they rotate. This in turn reduces the range of mechanical loads on the blades, thereby increasing their service life. The geometry required to do this, however, is somewhat complex.
Each flow directing structure 12 defines part of each overall gas flow path; it does not define the overall gas flow path a particular combustor's gas flow takes through the arrangement 10. In an embodiment, but not meant to be limiting, the overall gas flow path actually spans three flow directing structures 12 before entirely exiting the annular chamber 18. This can be seen in
A side view of the overall gas flow path 50 delineated by the structures but without the structures blocking the view is shown in
For sake of clarity,
For sake of clarity,
In and embodiment the IEP second flow path 112 is also used to transition the overall gas flow path 50 from being straight as it enters the IEP second flow path 112 to an overall gas flow path 50 that will be helical after traversing the annular chamber outlet 46. (The entire overall gas flow 50 may or may not exit the annular chamber outlet 46 while in the IEP 16 where it transitions from straight to non straight.) While the overall gas flow longitudinal axis 22 may itself transition from straight to helical at some point in the IEP second flow path 112, the overall gas flow path 50 is still bounded on the top by the overall gas flow top boundary plane 56, and on the bottom by the overall gas flow bottom boundary plane 58 from entering into the IEP second flow path 112 until exiting the annular chamber outlet 46. Since a helix is a curve, and the top and bottom boundaries are defined by planes when within the annular chamber, the top and bottom of the overall gas flow path 50 cannot be helixes when within the annular chamber. In fact, because the overall gas flow top boundary plane 56 and the overall gas flow bottom boundary plane 58 are not parallel, but instead converge on a radially inner side in an embodiment, and because the annular chamber curves radially inward so to speak, the overall gas flow top boundary plane 56 and the overall gas flow bottom boundary plane 58 would actually meet at a point in the annular chamber sufficiently downstream, were the annular chamber 18 lengthened along the gas turbine engine longitudinal axis 20. This would effectively end the theoretical overall gas flow path 50 and the combustion gasses would have no choice but to breach the boundaries of the overall gas flow path 50, which would defeat the purpose of having discrete flow paths. As a result, the overall gas flow path 50 is transitioned from having planar top and bottom boundaries to having helical top and bottom boundaries, and this transition begins in the IEP second flow path 112. Helical top and bottom boundaries will enable the theoretically discrete gas flow paths to remain discrete once transitioned to the annular chamber 18, thus reducing mixing of adjacent flows.
The overall gas flow top boundary 52 in the IEP second flow path 112 transitions from straight to curved while still remaining within the overall gas flow top boundary plane 56. The intersection of the overall gas flow top boundary 52 with the annular chamber outlet 46 defines the theoretical helical top boundary of that flow downstream from the intersection. That helical top boundary would be defined by a helical top boundary outer edge helix and a helical top boundary inner edge helix. The helical top boundary outer edge helix is defined by an outer tangent 118 of an overall gas flow top boundary outer edge at an outer tangent intersection point 120 with the plane of the annular chamber outlet 46. The helical top boundary inner edge helix is defined by an inner tangent 122 of an overall gas flow top boundary inner edge at an inner tangent intersection point 124 with the plane of the annular chamber outlet 46. The helical top boundary would be a helical plane between the helical top boundary outer edge helix and the helical top boundary inner edge helix. The same geometry applies to the overall gas flow bottom boundary 54 and a resultantly formed helical bottom boundary, since the overall gas flow top boundary 52 is the bottom of an upstream adjacent flow etc. The overall gas flow bottom boundary 54 transitions to helical earlier along the overall gas flow path longitudinal axis 22 than does the overall gas flow top boundary 52. The overall gas flow path longitudinal axis 22 transitions to helical at some point in between when the overall gas flow bottom boundary 54 transitions to helical and when the overall gas flow top boundary 52 transitions to helical.
It is also worth noting that it does not matter if an overall gas flow top boundary 52 traverses the annular chamber outlet 46 in the IEP second flow path 112 it entered, or a downstream IEP. The theory of the transition is the same for different configurations, the geometry will simply adapt to a shallower or steeper overall gas flow path 50 with respect to the gas turbine longitudinal axis 20. Furthermore, it is also worth noting that in another embodiment, transitioning the overall gas flow path 50 from being straight may begin to occur in the IEP first flow path 40. In such instances the transition of the overall gas flow path from straight is governed by the same principles, but the transition simply begins at some point in the IEP first flow path. For example, the overall gas flow path 50 is still bounded on the top by the overall gas flow top boundary plane 56, and on the bottom by the overall gas flow bottom boundary plane 58 from entering into the IEP second flow path 112 until exiting the annular chamber outlet 46. Whether the transition occurs in the IEP second flow path 112 or the IEP first flow path 40 is a matter of design choice, and may be driven in part by the number of combustors, or the angle between the combustor longitudinal axis and the plane perpendicular to the gas turbine longitudinal axis.
In order that adjacent overall gas flow paths not intersect or overlap each must be properly oriented when with respect to the adjacent upstream overall gas flow path and the adjacent downstream overall gas flow path. The geometry in an embodiment permits an overall gas flow to have a portion with a non-circular cross section, where the overall gas flow top boundary plane 56 and overall gas flow bottom boundary plane 58 are planar. As a result, as can be seen in
Also visible in
In addition to properly orienting each gas flow so that no gas flow paths intersect or overlap each other, each flow directing structure 12 may do any or all of the following: morphing the flow from a circular cross section to a non-circular cross section; generating a collimated flow within the gas flow prior to transitioning it to a partially bounded gas flow, and transitioning the overall gas flow path from a fully bounded perimeter to a partially bounded perimeter before delivering each gas flow to the annular chamber 18.
In an embodiment where all requirements are executed, and the cone has only circular cross sections, the IEP then must morph the cross section from a circular cross section to a non-circular cross section, and since morphing must be completed before a flow can be made to have a collimated profile, the morphing must occur when the entire perimeter of the flow is bounded, i.e. upstream of any partially unbounded regions. As can be seen in
In order to generate a collimated gas flow a converging gas flow path with circular cross sections can follow a convergence profile known in the art as the Witoszynski formula for convergence. The Witoszynski formula provides a uniform radius (or diameter) convergence for circular cross-sections as a function of normalized distance. The Witoszynski formula is as follows: R/Rout={1−(1−1/AR)(1−x2)2/(1+x2/3)3}−0.5, where R/Rout is the radius at length x divided by the outlet radius; AR is the (inlet area)/(outlet area) ratio; and x is the normalized distance from the inlet. The Witoszynski formula can be found in the following reference: “Witoszynski, C. 1924: ber Strahlerweiterung und Strahlablenkung. In: Vortrage aus dem Gebiet der Hydro- und Aerodynamik, Hrsg, Th. von Karman und T. Levi-Civita, Innsbruck, Springer Verlag, Berlin, S. 248-251.” However, when an overall gas flow path morphs to a non-circular cross section, the Witoszynski formula no longer directly applies because the non-circular cross section has no diameter (or corresponding radius) for the Witoszynski formula, which requires one. More particularly, the Witoszynski convergence profile inherently requires a known relationship between the radius of the cross section and the area of the cross section, (as well as the shape of the cross section), and this is accomplished when all cross sectional areas are limited to circular shapes. Consequently, the converging region with non-circular cross sections must follow a uniform convergence rate some other way. In an embodiment, in an area of convergence with a non-circular cross section, an equivalent diameter for the non-circular cross section may be derived and the equivalent diameter for the non-circular cross section conforms to the Witoszynski formula. In an embodiment, an area of the non-circular cross section may be used as an area of an equivalent circular cross section, and an equivalent radius/diameter of the equivalent circular cross section may conform to the Witoszynski convergence. In another embodiment the equivalent radius/diameter may be a hydraulic diameter of the non-circular cross section. Alternately, an equivalent radius/diameter may be something other than a diameter of a circular cross section of the same area as the non-circular cross section, or a hydraulic diameter; it may be another parameter of the non-circular cross section found to work better with the Witoszynski formula in such a configuration. For example, a diagonal length of a non-circular cross section, such as a trapezoid, may be used to determine the equivalent diameter, when a relationship between the length of the diagonal and the cross sectional area is known. Furthermore, ratios or conversions of a parameter may be used to reach an equivalent diameter, such that an equivalent diameter is proportional to the parameter. Additionally, a formula for determining an equivalent diameter may incorporate one or more parameters of the non-circular cross section. This allows for flexibility in the application of the Witoszynski formula to non-circular cross sections, as there may be differences in the convergences of a circular cross section and a non-circular cross section that can be accommodated with such ratios/formulas/conversions etc.
In yet another embodiment, the convergence may use the Witoszynski profile, but may use parameters of the non-circular cross section without regard to any relationship between the parameter used and the cross sectional area, to produce a collimated flow. In such an embodiment, a largest dimension 100 of a non-circular cross section such as that shown in
However, given the various configurations possible with non-circular cross sections, other restrictions may be imposed in an effort to reach a collimated flow in the flow downstream of the morphing. For example, as shown in
These requirements may be imposed because there exist circumstances when a morphing non-circular cross section could decrease in an equivalent diameter, such as a hydraulic diameter, but could actually diverge in one dimension. In this case the convergence of the non-circular cross section would conform to the Witoszynski formula but may still diverge. For example, at an upstream end a square cross section with a given area may converge to a rectangular area with a smaller area downstream, but if the rectangle were to be very thin and very long, the long dimension of the rectangle could be larger than the diameter of the smallest circular cross section upstream, which means some of the flow would actually diverge although the equivalent diameters of the non-circular cross sections were following the Witoszynski formula. Since this divergence is to be avoided the additional restrictions may be imposed.
There may be circumstances when a convergence that follows a uniform convergence profile such as that called for by the Witoszynski formula does not produce the desired collimated flow. For example, manufacturing tolerances and dynamic operating conditions may work against a collimated flow. In addition, when a non-circular cross section converging area follows the Witoszynski formula for convergence by using equivalent diameters, the flow produced simply may not be the ideal collimated flow desired. This may occur because the Witoszynski formula for convergence assumes circular cross sections. In view of the possibility of such circumstances or other unforeseen circumstances, a throat region may also be used.
A single IEP 16 is shown in
From this it can be seen that an associated IEP 28 may receive a gas flow from a cone 14. The received gas flow will have a circular cross section as it enters the IEP first flow path 40. The IEP first flow path 40 may have an IEP first flow path upstream portion 106 in which the overall gas flow path 50 is fully bounded, and an IEP first flow path downstream portion 108 where the overall gas flow path 50 is partially bounded. These two may meet at the upstream end of the partially bounded region 90. Within the IEP first flow path upstream portion 106 the overall gas flow path 50 may: morph from having a circular cross section to having a non-circular cross section, and while doing so it may follow a uniform convergence to produce a collimated flow; and also comprise a throat region. Should the IEP first flow path upstream portion 106 have a throat region 94, morphing from circular to non-circular cross sections must finish at some point upstream of the throat region 94, though that point can be the throat region upstream end 116.
In an embodiment the cone joint 86 may be located far enough upstream of the IEP first flow path downstream portion 108 that any cold air leakage into the cone joint 86 not interfere with the formation of the collimated flow to be developed prior to the IEP first flow path downstream portion 108. Further, in an embodiment, upstream end of the partially bounded region 90 may be located downstream of an IEP second flow path upstream end 110. This may impart mechanical strength and reduce fluctuations in the shape of the annular chamber 18 induced by mechanical loads and thermal gradients.
It has been shown that the inventors of the innovative present arrangement have created an assembly that directs combustion exhaust gas from a combustor to a first row of turbine blades along a mostly straight overall gas flow path, while dispensing with the first row of vanes present in the first stage of conventional can annular gas turbine engines. The uniformity of the flow is increased because each discrete flow is no longer separated by walls upon delivery to the first row of blades. This reduces the range of the mechanical load oscillations the first row of blades sees, thereby increasing their service live. Furthermore, the flow is already aligned, so aerodynamic losses associated with the first row of flow redirecting vanes are eliminated, as are the costs of producing and maintain those blades. Finally, the flow directing structures are modular, so individual flow directing structures can be replaced, and if made with components, any component can be individually replaced.
While various embodiments of the present invention have been shown and described herein, it will be obvious that such embodiments are provided by way of example only. Numerous variations, changes and substitutions may be made without departing from the invention herein. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims.
This application is a continuation-in-part of U.S. patent application Ser. No. 12/420,149 to Wilson et al., filed 8 Apr. 2009 (attorney docket 2008P18832US01), which in turn claims priority to U.S. provisional application No. 61/100,853 filed 29 Sep. 2008, both of which are incorporated by reference herein.
Number | Date | Country | |
---|---|---|---|
61100853 | Sep 2008 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 12420149 | Apr 2009 | US |
Child | 13096041 | US |