How clocking of compressor blades may impact wear in some vane rows

Have you ever thought about why people remember you, and how little what they remem­ber you for might reflect about your professional abilities?

An example: Many graying read­ers probably recall Bowzer, the gan­gly greaser of Sha Na Na, a popular rock ‘n’ roll band of the 1970s and 1980s.

The editors believe rela­tively few doo-wop lovers knew that under the black muscle shirt and Converse® Hi-Tops was a child prodigy and gifted classical pianist who began attending The Jul­liard School at age 12.

Many gas-turbine (GT) users know Rodger Anderson, manager of GT technology for DRS Power Tech­nology Inc, Schenectady, NY, as “the guy who pins GT compressor vanes.” Good reason for that: Anderson has made numerous presentations on the subject at user-group meetings over the last five years and participates in many industry vendor fairs. Plus, his technology and methods have been featured in several magazine articles (access www.combinedcyclejournal.com/archives.html, click 2Q/2008, click CTOTF on the magazine cover and scroll to p 94 in the second article).

However, most who know Ander­son for his success in mitigating the effects of compressor hook-fit wear in Frame 6B, 9E, 7EA, and 7FA mod­els by pinning together individual vanes into segments, are unaware of the depth and breadth of his knowledge and past success both as a compressor designer for a GT manufacturer and as a chief engineer for a boiler and machinery insurer.

Anderson has pinned vanes in more than five dozen frame engines over the last several years, including more than 30 7FAs. Fret­ting wear on the bases of the 7FAs’ S15 vanes and damage to casing grooves was a common finding. A puzzling observation was that some machines exhibited much more dam­age than others. In fact, at 2 × 1 com­bined-cycle sites, one GT might have relatively little fretting wear in S15, the other extensive fretting.

The latter can be particularly problematic: Excessive fretting can make vanes so loose they tilt into and contact rotating blades in the adja­cent row. Damage is a given. In the extreme, vanes can liberate.

Anderson thought about the varia­tion in degree of 7FA S15 fretting wear for a long time. One day while staring at a compressor rotor he noticed that the R14, R15, and R16 blades were not aligned the way Anderson believed they should be. Each of the three rows had an equal number of blades (64) and he thought they should be in line as shown in Fig 1. But in this case they were “fully clocked” (shifted by a blade width from perfect alignment)—as shown in Fig 2.

Anderson began reviewing his notes and asked users to check the orientation of S15 and S16 blades the next time they did borescope inspections (Fig 3). Fig 4 is one such photo taken that shows what he considers “good” orientation of R15 and R16 airfoils. As information came in from the field, Anderson retrieved pho­tos of vanes and casing grooves for machines he had repaired. He com­pared those with “good” blade orien­tation to those from machines where airfoils were clocked.

The forensic evidence gathered clearly revealed a consistent pattern of very little fretting on the bases of the S15 vanes and their casing groove when R15 and R16 airfoils had “good” orientation and a consistent pattern of higher fretting wear when rotor blades were clocked. When blades were fully clocked as shown in Fig 2, the most severe fretting wear was in evidence. Examples are presented in Figs 5-9.

Anderson thinks clocking prob­ably can be traced to the compressor-wheel manufacturing process. It is his understanding that after the wheels are drilled to accommodate the long bolts that hold rotor compo­nents in place, the slots that retain the airfoils are broached.

Apparently, broaching is initiated with no consistent orientation with respect to disk bolt holes. Thus the position of R15 and R16 blades with respect to each other can range from perfect alignment (Fig 1) to fully clocked (Fig 2) and include every­thing in between.

Anderson suggests that all 7FA owner/operators check R14/R15 and R15/R16 during their next borescope examination to see if clocking exists, and if it does, how much. Where “bad” orientation of airfoils is identified, he thinks vanes should be checked regu­larly—perhaps even as frequently as quarterly—for lift, shim protrusion, and evidence of fretting wear.

Users in the market for new GTs, he continues, may want to write “no clock­ing” into their specifications. Those buying preowned equipment or an existing plant might want to include clocking in their due diligence.

Vane pinning

There are at least two schools of thought on how to deal with loose vanes and associated shim loss and casing-groove wear. One is the DRS method, in which vanes are robustly pinned together in segments (typi­cally six or eight vanes per segment in rows 14 to16). Where shims are required, they are pinned as part of a vane segment.

Most original vanes can be reused with the DRS method. Replacements generally are required only where airfoils had been damaged during operation or when they were removed from severely worn casing grooves.

Benefits of the DRS approach, Anderson says, are speed, simplicity, and relatively low cost. Proven result: Vane rock is virtually eliminated even where there is heavy casing groove wear like that shown in Fig 5.

Another approach used is to weld-repair and finish-grind severely worn casing grooves to their original configurations and then reinstall vanes. Where shims are needed, they are pinned to adjacent vanes only. All vanes remain as individual air­foils—a few with shims attached, the large majority without. As Ander­son sees it, one problem with this approach is that nothing has been done to mitigate a recurrence of the rocking/fretting problem.

The alternative solution also requires replacement of vanes with fretting (Fig 7) and/or hook-fit wear (Fig 8). In sum, this approach can take considerably more time than the two or three days Anderson requires for pinning; plus it is more expensive.

To better understand the operation­al dynamics associated with vanes, Anderson offered this explanation: During initial startup of a new GT, or of a rebuilt unit with new vanes, rapid thermal expansion crushes the line contact edge of the vane bases as shown by the red line in Fig 10.

From this point in time onward, he says, all vane rows run with a cir­cumferential looseness of 150 to 250 mils, which is conducive to fretting wear. Thus shims pinned to an indi­vidual vane only still are subject to loosening, protrusion, and liberation, Anderson concludes. ccj