Blade row arrangement for turbo-engines and method of making same

[0002] This application claims the priority of German Patent Document No. 100 53 361.2, filed in Germany, Oct. 27, 2000, the disclosure of which is expressly incorporated by reference herein.

[0003] The invention relates to a blade row arrangement for turbo-engines of an axial-flow coaxial construction. Preferred embodiments of the invention relate to a blade row arrangement for turbo-engines, particularly for gas turbines, in an axial-flow coaxial construction with two guide blade rows situated in a fixed axial and circumferential position relative to one another, having a different number of blades and each having a constant pitch angle between their blades, as well as having a moving blade row rotatably arranged between the guide blade rows, the upstream guide blade row having a flow-off direction with an axial and circumferential component comparable with respect to the size.

[0004] Promising starting points for optimizing the efficiency of turbo-engines by fluidic measures exist in the form of a fixed defined assignment of the circumferential positions of successive guide blade rows or of successive, synchronously rotating moving blade rows. This principle, which in technical terminology has become known as “clocking” or, more concretely, as “stator or rotor clocking”, has the object of leading the wakes originating from the individual blades of a first row of blades in a defined fluidically optimal circumferential position to a similar row of blades which is next in the downstream direction. If two “clocked” rows of guide blades are involved, it should be taken into account that the wakes are considerably influenced and changed by the moving blade row rotating between the guide blade rows, particularly because of displacements, deformations and separations. The complexity of these flow patterns has the result that so far there are no unambiguous reliable rules for a constructive “clocking”.

[0005] European Patent Document EP 0 756 667 B1 (corresponding U.S. Pat. No. 5,486,091) protects a “clocking” method in which the wakes of a first blade row are directed by a second blade row with a relative motion to the blade inlet edges of a third blade row stationary relative to the first, in which case a maximal circumferential deviation between the wake and the inlet edge of plus/minus 12.5 percent of the blade pitch should be permissible.

[0006] Tests have not confirmed that this type of “clocking” would generally increase the efficiency.

[0007] Irrespective of how the optimal relative circumferential position of the blade rows is selected, it is a prerequisite of “clocking” according to the above-mentioned prior art arrangements that the coordinated blade rows pertaining to the same relative system (stator or rotor) have the same number of blades when the blade pitch is circumferentially constant.

[0008] It is an object of the invention to suggest a blade row arrangement with two guide blade rows and one moving blade row arranged between the latter which, despite different blade numbers of the two guide blade rows, permits a fluidically advantageous relative circumferential positioning of the guide blade rows in the sense of a “clocking”.

[0009] This object is achieved in certain preferred embodiments by providing a blade row arrangement for turbo-engines, particularly for gas turbines, in an axial-flow coaxial construction with two guide blade rows situated in a fixed axial and circumferential position relative to one another, having a different number of blades and each having a constant pitch angle between their blades, as well as having a moving blade row rotatably arranged between the guide blade rows, the upstream guide blade row having a flow-off direction with an axial and circumferential component comparable with respect to the size, wherein the blades of the upstream first guide blade row, in one of a first cohesive partial area T1 of the row and a partial area T1 distributed in several separate sectors along the row circumference, successively have an axial offset Δm of the same amount as well as in the same direction, wherein the axial offset Δm, as a function of the blade number ratio Z1/Z2 of the first and the second guide blade row is selected such that, at Z1>Z2, the axial offset Δm increases an effective flow-off cross-section Aeff between the blades, and such that, at Z1<Z2 reduces the flow-off cross-section, and wherein the blades of the first guide blade row, in one of a second cohesive partial area T2 of the row and a partial area T2 of the row distributed in several separate sectors along the row circumference, successively have an axial offset Δn which has the same size or varies and is oppositely directed in relation to Δm.

[0010] According to the invention, the upstream guide blade row—despite a constant pitch angle of the blades along the circumference—is constructed with two different partial areas which are individually cohesive or distributed in several separate sectors along the row circumference, in both areas each blade being axially offset in a defined manner with respect to its neighboring blade. Thus, the stacking axes of the blades are no longer—as customary—situated in a common radial plane but on screw surfaces with a constant or varying pitch, in which case concrete blade points are correspondingly situated on helical lines. The first partial area with Δm describes, for example, a “forward screw”; the second partial area with Δn describes a “backward screw” connecting the ends of the Δm area, or vice versa. In the sense of a “clocking”, only the first partial area acts with a constant defined axial offset Δm from blade to blade; the second partial area is used only for the return of the entire added-up axial offset in a linear or non-linear manner by means of Δn while avoiding relevant fluidic disadvantages. Since the guide blade rows have a diagonal flow-off with a strong circumferential component, the axial offset between adjacent blades effectively causes an enlargement or reduction of the outlet-side flow cross-section. In the first partial area, the axial offset Δm is constant and is selected as a function of the blade number ratio of the two guide blade rows. If the blade number Z2 of the second guide blade row is smaller than that of the first guide blade row (Z1), the effective flow-off cross-section of the first guide blade row is enlarged by means of Δm; if Z2 is larger than Z1, the flow-off cross-section of the first row is reduced by means of an opposite axial offset. In the second partial area of the row with the axial offset Δn, the opposite will in each case apply correspondingly; here, no targeted “clocking effect” occurring at the second downstream guide blade row.

[0011] By the variation of the effective flow-off cross-sections of the first guide blade row, the invention results in a certain asymmetry of the flow distribution and thus of the mass distribution in the ring-shaped flow duct cross-section. This has, among others, the advantage that instabilities and disturbances which, in the case of symmetrical or periodic conditions, may expand further over the circumference, can be displaced and partially prevented. Furthermore, by means of the invention, reactions can take place in a targeted manner to certain asymmetries in the afflux.

[0012] The “clocking effect” primarily endeavored by means of the invention, because of its angular limitation may, for example, also be called “partial clocking” or “sector clocking”.

[0013] Further features of preferred embodiments of the invention are described below and in the claims.

[0014] Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings.

[0015] FIG. 1 is an explanatory, not-true-to-scale representation of a blade row arrangement with two guide blade rows and one moving blade row arranged in-between, constructed according to preferred embodiments of the invention;

[0016] FIG. 2 is an explanatory, not-true-to-scale representation of four blade profiles of a guide blade row with an axial offset;

[0017] FIG. 3 is a diagram with the course of the axial offset over the guide blade row circumference; and

[0018] FIG. 4 is a diagram comparable to FIG. 3 but with a course of the axial offset periodically varying in four sectors.

[0019] For a better understanding, it should first be pointed out that FIGS. 1 and 2 show the blade rows as if they were plane rows—without any curvature with parallel blades—, in which case only a concrete profile is shown for each blade. This type of representation is much simpler, clearer and more easily understandable than a realistic spatial representation with radial three-dimensional blades, etc.

[0020] In FIG. 1, the flow through the blade row arrangement takes place from the left to the right; a guide blade row 2 being situated upstream (left); a moving blade row 5 being situated in the center; and another guide blade row 3 being situated downstream (right). The blades of the rows 2, 5 and 3 have the reference numbers 6, 9 and 7. The rotating direction of the moving blade row is indicated below the latter by means of an upward-pointing black arrow. Above the moving blade row 5, a horizontal double arrow is indicated by a broken line and points out that the row may be constructed axially displaceably in order to additionally influence the course of the flow. The colors gray and black indicate the—so-called—wakes 10 of the guide blade row 2, the wakes 11 of the moving blade row 5, and the change of the wakes 10 on their path through the rows 2 and 5; the dotted curves and straight lines describing the paths of the wakes in relation to the unmoved stator system. The axial offset of the blades concerns only the upstream guide blade row 2 and is not shown in FIG. 1. FIG. 1 also does not show that the guide blade rows 2 and 3 have different numbers of blades.

[0021] FIG. 2 therefore shows a guide blade row 4 which is comparable to the row 2 in FIG. 1 and has axially offset blades 8 according to the invention. The pitch angle between all blades 8 is constant, so that the vertical offset is in each case constant in the figure. See the statement 2π÷Z1 on the left, which corresponds to the radian measure divided by the radius r, that is, to the radius-related radian measure from one blade to the next. From above, the first, second and third blade are axially (here, horizontally) offset with respect to one another in each case by an amount Δm, in which case the blades move from above in the downward direction farther to the right, that is, downstream. The flow-off from the guide blade row 4 takes place at an angle β of approximately 45° diagonally to the right upward, that is, with a comparatively large axial and circumferential component. This diagonal flow-off has the result that an axial offset between two blades necessarily results in a change of the effective flow-off cross-section Aeff. In the present geometry, the flow-off cross-section is enlarged in comparison with an arrangement of the blades without an axial offset Δm. See in this regard the position of the second blade from above indicated by a broken line without an axial offset in relation to the uppermost blade. The enlargement of the flow-off cross-section can also be recognized by the fact that the vertical distance between the flow lines originating from the blade trailing edges, here, the radius-related radian measure 2π÷Z2, is larger than the measure 2π÷Z1, specifically by the added value Δm÷(r.tanβ). In this regard, see the equation at the right-hand top in the figure. This corresponds to an effective adaptation of the guide blade row 4 to a guide blade row which is situated downstream, is not shown here and has a larger spacing of the blades; that is, a smaller number of blades Z2>Z1. Because the blade numbers Z1, Z2 in the respective row are constant along the duct height, that is, they are independent of the radius r, tan β should at least along the largest portion of the radial duct height be selected to be inversely proportional the radius r.

[0022] For an adaptation to a downstream guide blade row with a larger number of blades, that is, Z2>Z1, the flow-off cross-sections of the blades 8 would have to be reduced in relation to a row without any axial offset Δm. In the figure, the upper three blades would then have to be moved from above in the downward direction farther to the left, in each case, by a constant axial offset Δm to the left. This principle is easily understandable and is therefore not shown separately.

[0023] It should be noted that the lowermost blade in FIG. 2 relative to the blade situated above that blade has no longer moved by Δm to the right but by an axial offset Δn to the left. In reality, it is fluidically not useful to arrange all blades of a guide blade row in the sense of a helical line with a continuous axial offset, in which case a large axial jump with very negative fluidic consequences would exist between the first and the last blade of such a row. The invention therefore provides that a first partial area T1 of the guide blade row be equipped with a continuous axial offset Δm, and in a second partial area T2, the sum of all Δm be completely canceled again by means of opposite axial offsets Δn.

[0024] This principle can be best understood on the basis of FIG. 3 which illustrates the course of the axial offset ΣΔm, Δn along the circumference U of the guide blade row, the concrete blade positions being marked by small circles. A first partial area T1 is shown; here, a partial area T1 extending over 270°, with a linearly rising axial offset, from blade to blade in each case by Δm. This is followed by a second partial area T2; here extending over 90°, in which the axial offset decreases again successively, either linearly (broken line) or according to an S-cure, for example, a cosine curve. With respect to the S-curve, it is shown that the axial offset Δn may vary from blade to blade. Which type of a curve would be more favorable here, will have to be determined by tests, among other things. The blade (small circle) at the ordinate 0 is identical with the blade at the ordinate 2π, because the row circumference closes here. The present diagram therefore outlines 16 different blade positions. In reality, the blade numbers will, as a rule, clearly be larger. The ratio of sizes of the partial areas T1 and T2 is indicated only as an example, in which case T1>T2 should be endeavored. Since in practice, the blade numbers Z1 and Z2 differ only a little, relatively small axial offsets Δm are sufficient for applying the invention.

[0025] FIG. 4 shows the course of the axial offset ΣΔm, Δn along the circumference U of a guide blade row, whose partial areas T1, T2, in contrast to the embodiment of FIG. 3, are not arranged in an individually cohesive manner but are each distributed in four separate sectors T1÷4, T2÷4 along the row circumference, so that a quadruply periodic course is obtained in each case with a positive and negative axial offset Δm, Δn. The division into four sectors is used as example; they may also be two, three, five or more sectors. The course of the partial area sectors T2÷4 is linear here in each case. Naturally, S-curves can also be used instead, as illustrated in FIG. 3. As a result of the division of the “clocked” partial area T1 and of the partial area T2 into, in each case, several separate sectors, asymmetries of the flow field along the duct cross-section—as in an embodiment according to FIG. 3—can be avoided, in which case, these may, however, also be desirable.

[0026] The foregoing disclosure has been set forth merely to illustrate the invention and is not intended to be limiting. Since modifications of the disclosed embodiments incorporating the spirit and substance of the invention may occur to persons skilled in the art, the invention should be construed to include everything within the scope of the appended claims and equivalents thereof.