The present invention relates to non-uniform stator vane spacing in a compressor and particularly relates to non-uniform blade counts of stator vanes in the upper and lower compressor casing halves of a compressor stage to minimize or eliminate vibratory response of adjacent rotating blades.
In axial flow compressors, stator vanes alternate with rotating blades or buckets in the various stages of the compressor. The stator vanes are circumferentially spaced one from the other about the compressor axis and are secured to the upper and lower compressor casing halves. The upper and lower casing halves are joined one to the other at the compressor midline and provide a complete circumferential array of stator vanes for each compressor stage. As each rotating blade mounted on the rotor completes each revolution at a given rotational velocity, the rotating blade receives aerodynamic excitation pulses from each stator vane. This pulse can be generated from the wake of the upstream stator vane or the bow wave of the downstream stator vane. It is also possible to generate excitations in the rotating blade from differences between the upstream and downstream stator vane counts. These pulses induce a vibratory response in the rotating blade which can be deleterious to the rotating blade causing failure due to high cycle fatigue.
Typically the stator vane or blade count in the upper and lower halves of the compressor casing for a given stage are equal in number to one another. For example, in an initial stage S0 of a given compressor, the blade count for the stator vanes in each of the upper and lower compressor casing halves is 24/24. In the next stage S1, the blade count is 22/22. The first number represents the number of stator vanes in the upper casing half and the second number represents the number of stator vanes in the lower casing half of the same stage. The total stator vane count in S0 and S1 is therefore forty-eight and forty-four stator vanes respectively. However, because of the vibratory responses of the rotating blades, non-uniform vane spacings between upper and lower casing halves have been used in the past. Thus, different and alternative upper and lower blade counts in succeeding stages have been provided to reduce or eliminate the vibratory response. For example, in one compressor, stages S0 and S1 have had vane counts of 24/23 and 23/24, respectively. These non-uniform blade counts have been used in original equipment manufacture.
There are, however, a significant number of compressors in use in the field where there is an equal number of stator vanes in the upper and lower compressor halves for given stages. Certain other compressors in the field have an unequal number of stator vanes in the upper and lower compressor halves with adjacent stages, e.g. S0 and S1, having equal numbers of blades but alternate blade counts between the upper and lower halves of the compressor casing. Changing blade counts in the field was not previously considered practical since costly removal of the rotor in the field was required. Consequently there developed a need to retrofit compressors in the field with non-uniform blade counts among upper and lower compressor halves of the same stage to reduce vibratory response and without the necessity of removing the rotor.
In accordance with a preferred aspect of the present invention, there is provided a method of installing stator vanes in the field which enables a change of blade counts in the upper half of the compressor casing permitting compressors in situ or in the field to be upgraded to compressors with non-uniform upper and lower compressor casing blade counts to reduce the vibratory response of the rotating blades. For example, for a particular compressor in the field with a given count of stator vanes, the adjacent stator stages in the upper half of the compressor casing are provided increased stator blade counts, e.g., 26/23 for S0 and 24/23 for S1 yielding blade counts of forty-nine for the S0 stator vanes and forty-seven for the S1 stator vanes. Consequently, only the upper half of the compressor casing requires removal in the field to alter the stator vane count while the same number of stator blades in the lower compressor half for each stage is maintained. Significant advantage accrues to this alteration in stator vane count since removal of the rotor and access to the lower casing half are not required to alter the blade count. By altering only the count of stator vane blades in the upper compressor half, and changing the blade count of adjacent stator stages, the rotating blades cannot lock into a synchronous vibratory response and consequent high cycle fatigue is minimized or avoided.
In a preferred embodiment hereof, there is provided a method of retrofitting a compressor comprising the steps of (a) removing an upper half of the compressor casing in situ to open the compressor; (b) removing a first set of stator vanes of the array thereof having a first blade count from the removed upper half of the removed compressor casing; (c) in place of the removed first set of stator vanes, installing in the removed upper half of the compressor casing a second set of vanes with a second vane count different than the vane count of the first set of stator vanes; and (d) closing the compressor by securing the upper half of the compressor casing with the second set of vanes to the lower half of the compressor casing. Preferably, the compressor is retrofitted in situ to reduce vibratory response of one set of rotating compressor buckets to aerodynamic excitation pulses generated by at least one array of stator vanes adjacent to the one set of rotating compressor buckets.
Referring to
Referring to
In the prior compressor stator vane arrangement illustrated in
The installation procedure is schematically illustrated in
It will also be appreciated that the stage S1 stator vanes are altered in their count. Preferably, the third set of original stator vanes of stage S1 are changed to provide a fourth set 35 of twenty-four stator vanes in the removed upper casing half while retaining the original twenty-three stator vanes in the lower casing half. As a consequence of the foregoing, stage S0 after modification has a blade count of 26/23 for a total blade count of forty-nine blades while stage S1 has a blade count of 24/23 for a total count of forty-seven blades. The unequal blade counts in the upper and lower casing halves and the adjacency of the stages S0 and S1 reduce the vibratory response of the buckets or vanes 22 of the rotor. Additionally, the blade counts of forty-nine and forty-seven were selected based on the fact that they were prime or near prime numbers and that they are not whole order of multiples of typical engine order excitations, 2/revolutions, 3/revolutions and 4/revolutions. This typically comes from the shape of the air at the inlet. Engine air typically has a large content of these engine orders. By using prime numbers, harmonics of these excitation orders are avoided.
While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
3006603 | Caruso et al. | Oct 1961 | A |
3194487 | Tyler et al. | Jul 1965 | A |
3849023 | Klompas | Nov 1974 | A |
3918832 | Shuttleworth et al. | Nov 1975 | A |
4135857 | Pannone et al. | Jan 1979 | A |
4193568 | Heuvel | Mar 1980 | A |
4967550 | Acton et al. | Nov 1990 | A |
5141391 | Acton et al. | Aug 1992 | A |
5340271 | Freeman et al. | Aug 1994 | A |
5690469 | Deal et al. | Nov 1997 | A |
5765993 | Weiss | Jun 1998 | A |
6416278 | Caddell et al. | Jul 2002 | B1 |
6543997 | Donnaruma et al. | Apr 2003 | B2 |
6752589 | Vogan et al. | Jun 2004 | B2 |
6869270 | Bunker et al. | Mar 2005 | B2 |
6905303 | Liu et al. | Jun 2005 | B2 |
6984108 | Anderson et al. | Jan 2006 | B2 |
7097420 | Cormier et al. | Aug 2006 | B2 |
7234914 | Usab et al. | Jun 2007 | B2 |
7434313 | Dasilva et al. | Oct 2008 | B2 |
7458219 | Herzog et al. | Dec 2008 | B2 |
20050191177 | Anderson et al. | Sep 2005 | A1 |
20060045772 | Slovisky et al. | Mar 2006 | A1 |
20060198726 | Schirle et al. | Sep 2006 | A1 |
20070077148 | Bird | Apr 2007 | A1 |
20070079506 | Gautreau et al. | Apr 2007 | A1 |
20070231155 | Burdgick et al. | Oct 2007 | A1 |
20080282541 | Anderson et al. | Nov 2008 | A1 |
Number | Date | Country |
---|---|---|
1576611 | Feb 2005 | CN |
1772596 | Apr 2007 | EP |
2007100700 | Apr 2007 | JP |
Number | Date | Country | |
---|---|---|---|
20070079506 A1 | Apr 2007 | US |