MACHINING PROCESS FOR MULTI-VANE NOZZLE

Abstract

The present invention relates to a method for machine finishing the shape of a blank casting for a multi-vane, in particular bi-vane, nozzle of a turbine engine, comprising a first vane and a second vane extending substantially in a radial direction between two walls that are radially inner and radially outer, respectively, the suction face of the first vane defining, together with the pressure face of the trailing edge of the second vane, a cross section of flow (SP), the method comprising measuring, by means of probing, the position of predefined points on said respectively radially inner and radially outer walls on the surface of the vanes and calculating the machining allowances (Δ1 and Δ2 respectively) on the first and second vanes with respect to the theoretical profile at said points, wherein the method comprises calculating said cross section of flow (SP) from the height of the duct between said radially inner and radially outer walls, and values of the machining allowances (Δ1 and Δ2), a correction of the machining allowance (Δ2) on one of the vanes being applied when the calculated value of the cross section of flow (SP) is outside predefined tolerances.

Claims

1. Method for machine finishing the shape of a blank casting for a multi-vane member of a turbine engine, comprising at least one first vane and one second vane extending substantially in a radial direction between two walls that are radially inner and radially outer, respectively, the suction face of the first vane defining, together with the pressure face of the trailing edge of the second vane, a cross section of flow, the method comprising measuring, by means of probing, the position of predefined points on said respectively radially inner and radially outer walls on the surface of the vanes and calculating the machining allowances on the first and second vanes with respect to the theoretical profile at said points, wherein said method comprises calculating said cross section of flow from the height of the duct between said radially inner and radially outer walls and values of the machining allowances, and correcting the machining allowance on one of the vanes when the calculated value of the cross section of flow is outside predefined tolerances.

2. Method according to claim 1 comprising measuring, by probing, the position of predefined points on the suction face of the first vane, points on the pressure face of the trailing edge of the second vane, and points on said respectively radially inner and outer walls, and calculating the cross section of flow from said points.

3. Method according to claim 1, the measurement of the position of points on the pressure face of the second vane being made on the basis of the measurement, by probing, of points opposite the suction face of the second vane, the value of the machining allowance on the pressure face being a function of, preferably equal to, the value of the machining allowance on the suction face.

4. Method according to claim 1, wherein the cross section of flow is calculated as the sum of basic cross sections Σ(Si) and a parameter Cm*Hm that is a function of a duct height Hm defined between the respectively radially inner and outer walls, Σ(Si) being determined over a plurality of transverse sections i distributed over different percentages of the height Hm between the respectively radially inner and outer walls, with Si=Ci*(Di+Δi1+Δi2+ε) where: Ci is a predetermined coefficient, the theoretical value of which is a function of the cross section (i), Di is the size of the basic cross section for the cross section (i), Δi1 is the value of the machining allowance on the suction face of the first vane for the cross section (i), Δi2 is the value of the machining allowance on the pressure face of the second vane for the cross section (i), the values of the machining allowances Δi1, Δi2 being obtained from machining allowances of said points of the pressure face of the second vane and of the suction face of the first vane, the position of which has been measured, and ε being a correction value to add to Δi2.

5. Method according to claim 1, wherein the value of ε is obtained by iteration from an initial increment value and from the calculation of the cross section of flow to which said increment value is applied as a positive or a negative value until the calculation of the cross section of flow is within a required tolerance range.

6. Method according to claim 4, wherein the value Δi2 of the machining allowance on the pressure face of the second vane is preferably selected so as to be equal for all basic cross sections of the same size i.

Description

DESCRIPTION OF THE FIGURES

[0024] The invention will be better understood, and other objects, details, features and advantages thereof will become clearer in the course of the following detailed explanatory description of an embodiment of the invention given by way of a purely illustrative and non-limiting example, with reference to the appended schematic drawings, in which:

[0025] FIG. 1 is a perspective view of a bi-vane high-pressure turbine nozzle;

[0026] FIG. 1a is a schematic cross section of the nozzle from FIG. 1 in a plane transverse to the vanes, comprising the direction III-III and substantially perpendicular to the direction II-II;

[0027] FIG. 2 is a cross section of the nozzle from FIG. 1 in a plane of the cross section of flow, passing through the trailing edge of the vane 2 and comprising the directions II-II and III-III;

[0028] FIG. 3 is a cross section of a vane of the nozzle from FIG. 1 before machining;

[0029] FIG. 3a is an enlarged view of the trailing edge of the vane from FIG. 3;

[0030] FIG. 4 is an enlarged view of the trailing edge of the vane from FIG. 3, showing the machining tolerances;

[0031] FIG. 5 is the flowchart of the sequences of the method of the invention.

DESCRIPTION OF AN EMBODIMENT OF THE INVENTION

[0032] The following description is applicable to a turbine engine member of the multi-vane type, such as a nozzle, and could also apply to another member of the same type.

[0033] FIG. 1 is a perspective view of a high pressure bi-vane turbine nozzle D at the end of “lost wax” casting operations, seen obliquely from its trailing edge side. The bi-vane nozzle corresponds to an annular sector around an axis (not shown in the figure). Said nozzle comprises two vanes 1 and 2 extending substantially radially with respect to said axis and connected by an inner ring sector 3 and an outer ring sector 4. The suction face of the second vane 2 can be seen between the leading edge 2BA and the trailing edge 2BF of said vane. The first vane 1, at the back of the drawing with respect to the vane 2, can only be seen in part, from the side of its trailing edge 1BF. The leading edge of vane 1 is hidden here by the vane 2.

[0034] In the case shown, the cross section of flow of the air circulating in the duct delimited between the first vane 1, the second vane 2 and the inner 3 and outer 4 ring sectors, corresponds to the surface area of the profile of said duct in a sectional plane passing through the trailing edge 2BF of the second vane. This sectional plane, comprising a direction II-II substantially radial to and a direction III-III substantially transverse to the profiles of the vanes at the trailing edge 2BF, crosses the first vane 1 far upstream of its trailing edge 1BF, as shown in FIG. 1a. This sectional plane, or plane of the cross section of flow, corresponds to a neck between the two vanes.

[0035] As can be seen in FIG. 1a, in the case shown, the first vane 1 is hollow in this sectional plane since this case corresponds to that in the prior art where the vanes are manufactured so as to provide a cooling circuit.

[0036] FIG. 2 shows the nozzle from FIG. 1 intersected by the plane of the cross section of flow between the two vanes 1 and 2 at the neck of the duct.

[0037] With reference to FIG. 2, in an embodiment of the method, this cross section of flow SP is defined in the figure by 10 points in the plane of the cross section of flow: [0038] P12, P32 and P52 are points on the pressure face of the trailing edge 2BF of the second vane 2 of the nozzle at three predetermined radii about the axis with respect to which the annular sector of the nozzle is defined; [0039] P11, P31 and P51 are points on the suction face of the first vane 1 opposite the aforementioned points and at substantially the same predetermined radii; [0040] P7 and P9 are points on the radially inner wall of the duct, on the side of the inner ring sector 3, and P8 and P10 are points on the radially outer wall of the duct, on the side of the outer ring sector 4.

[0041] FIG. 3 is a transverse section through the second vane 2, which section is perpendicular to the longitudinal axis of said second vane at the end of the casting process. Said figure again shows the core 2n having the strip 2nb that crosses the vane wall to form, after the core has been removed, the evacuation vents of the internal cooling circuit of the vane. The trailing edge also has a machining allowance 2ep of the strip 2nb that is to be removed by machining. The aim it to obtain a trailing edge profile corresponding to the outline 2f, which can be seen in the exploded view in FIG. 3a. This profile is obtained such that the cross section of flow SP is the desired cross section of flow, to within the tolerances.

[0042] To this end, the following is carried out:

[0043] The position of the above-mentioned points is measured using a probing tool that is suitable and known per se, [0044] P11, P31, P51 on the suction face of the first vane 1, [0045] P7, P8, P9, P10 on the radially outer and inner walls of the duct, [0046] P12, P32, P52 on the pressure face of the trailing edge 2BF of the second vane 2.

[0047] For the latter points, P12, P32, P52, the measurement is in fact carried out at points P12′, P32′ and P52′, which are not shown in FIG. 2 and correspond to Pi2′ in FIG. 4, where i=1, 3, 5, respectively. Points P12, P32 and P52 on the pressure face are opposite points P12′, P32′ and P52′ and the position thereof is obtained by extrapolation, knowing the thickness EP of the trailing edge (see the distance between Pi2 and Pi2′ in FIG. 4). Preferably, this thickness is the same for the three cross sections, i=1, 3, 5. This method is justified by the presence of strips 2nb and the machining allowance 2e before any machining.

[0048] Having obtained the coordinates of these points using the method described above, it is possible to determine the machining allowance ΔP11, ΔP31, ΔP51, ΔP12′, ΔP32′ and ΔP52′, ΔP7, ΔP8, ΔP9, ΔP10, denoted generally as A, existing at each of these points with respect to the position of said points in the known theoretical profile, as a result of being cast. The theoretical profile is defined by a three-dimensional theoretical digital model, whereas the actual component, at the end of the casting process, has a profile having a machining allowance at various points.

[0049] The theoretical cross section of flow SPt is known. The calculation of SPt can be set out as:

SPt=ΣCi*Di+Cm*Hm

where
Di is the size (in other words the width of the duct) of the cross section at the radius i at each predetermined point Pi2, i=1, 3, 5, and Ci is a predetermined cross section coefficient or integration coefficient, the theoretical value of which is a function of the radius i. As shown in FIG. 2, Di is calculated here as the distance between point Pi2 on the pressure face of the second vane 2 and point Pi1 on the suction face of the first vane 1 located at the same radius i, i=1, 3, 5. Each radius i corresponds to a cylindrical cross section of the nozzle based on a cylinder of radius i around the axis of the ring of which said cylinder is a sector. Reference will therefore be made in the following to cross section i or sectional radius i to reference the corresponding points or coefficients.

[0050] Cm and Hm are a coefficient and the height of the duct, respectively. They take into account the shape of the radially outer and inner walls of the duct.

[0051] Hm is an average duct height defined between the radially inner and outer walls. Moreover, the sectional radii i correspond to cross sections at various percentages of this average height Hm starting from the radially inner wall.

[0052] In practice, the sizes at three radii corresponding substantially to 10%, 50% and 90% of the height of the duct Hm are taken for this calculation.

[0053] By applying this formula to values measured by probing, the measurement of the actual cross section SPr is obtained:

SPr=ΣCi*(Di−Δi1+Δi2′+ε)+Cm*Hm(ΔP7,ΔP9,ΔP8,ΔP10)

where Ci is the same sectional coefficient as in the theoretical calculation

[0054] Di is the size of the section at the cross section i

[0055] Δi1 is the machining allowance relative to the theoretical profile and is determined by probing the point on the suction face of the first vane 1 at the cross section i.

[0056] Δi2′ is the machining allowance with respect to the theoretical profile and is determined by probing the point on the suction face of the second vane 2 at the cross section i. This value is considered to be equal to Δi2 which is the machining allowance extrapolated on the pressure face.

[0057] In order to determine Δi1 and, respectively, Δi2 from the measurements of the machining allowances ΔP11, ΔP31, ΔP51, and ΔP12, ΔP32, ΔP52, made by probing, the method for calculating the cross section of flow has been shown for three sectional radii i corresponding to three radii i where the probing measurements are carried out, but it is obvious that different numbers of points and sectional radii can be used. In particular, if the number of cross sections does not correspond to the number of probing points, it is possible, in order to use the formula for calculating the cross section of flow, to use interpolations, for example those used in the recalculated machining, in order to use values obtained at the points measured by probing.

[0058] Cm*Hm (ΔP7, ΔP9, ΔP8, ΔP10) is a value of the product of the coefficients Cm and Hm, taking into account the shape of the radially outer and inner walls of the duct and corrected taking into account the machining allowances measured by probing at points P7, P8, P9, P10. This corrected value is then no longer modified in the corrections of the cross section of flow SPr.

[0059] This value SPr is compared with the theoretical values SPt. If said value SPr is outside the tolerances, a correction ε is introduced, for example having the value of +0.01 mm or of −0.01 mm depending on whether the value SPr is greater than a predetermined maximum tolerance or less than a predetermined minimum tolerance.

[0060] The calculation is repeated N number of times until the calculation gives a SPr value within the predefined tolerance range.

[0061] In a last step, the pressure face is machined, preferably by recalculated machining, while applying the correction N*ε.

[0062] For each cross section i, the same value for A is applied: Δ12′=Δ32′=Δ52′.

[0063] It is preferably also verified that the A obtained makes it possible to remain within maximum and minimum manufacturing tolerances with respect to the thickness EP of the trailing edge, as shown in FIG. 4.

[0064] FIG. 5 shows the flowchart of the method.

100: Calculation of the cross section of flow SPr after probing.
101: Verification of whether SPr is within the tolerances.

102: Yes.

103: No.

[0065] 104: SPr is greater than the maximum tolerance.
105: SPr is less than the minimum tolerance.
106: Calculation of a new SPr from the Δi2 values to which an increment −ε has been applied.
107: Calculation of a new SPr from the Δi2 values to which an increment +ε has been applied.
108: Iteration of the calculation of SPr from the Δi2 values to which increments −ε have been successively applied until SPr is within the tolerance range; number N of iterations limited to 10.
109: Iteration of the calculation of SPr from the Δi2 values to which increments +ε have been successively applied until SPr is in within the tolerance range; number N of iterations limited to 10.
110: Machining of the pressure face profile of the trailing edge of the second vane 2 in order to remove a material thickness corresponding to N*ε.