7EA community shares lessons learned on wrapper leak mitigation

Owner/operators attending the 7EA Users Group’s 2017 meeting in St. Augustine, Fla, October 24 – 26, likely will hear some colleagues discussing attempts at mitigating leakage from combustion-wrapper joints. “Attempts” is italicized because success has been elusive.

Jason Hampton, who chaired the steering committee last year, presented on the documented, deeply rooted frustrations experienced by users, since before the millennium, to eliminate leakage at casing joints—the four-way joints in particular.

Hampton generally attributed leakage to an OEM design issue, which, in his opinion is related to insufficient clamping force in suspect areas. A contributing factor is casing distortion suffered during operating regimes not envisioned by designers.

He invested a considerable amount of personal time to review the user group’s discussion-forum archives hosted on its website for registered users only and to survey owner/operators about their experiences. Here are a few of the comments, shared in half a dozen discussion threads on the forum from 2004 through 2012; most have been edited to reduce word count:

      • Split-line gaps on 7EAs are common and to my knowledge GE does not offer an effective means for eliminating wrapper leakage. Sealants have been tried with limited success.

      • We are experiencing split-line leaks on all six of our Frame 7s (five purchased in 1989, one in 2001).

      • We ensured that the OEM torque sequence/procedure is being done correctly. Result: Leaks slowed at first. Grooved the upper casing on one unit and installed a keyway in the groove; this helped for about eight months. Tried several different sealers, but they were effective only for a short period.

      • On our recent major inspection, we followed jacking instructions to the letter. Plus GE executed the work and the field TA (technical advisor) was one of the best in the business. We still have leaks.

      • Leaks on the turbine/wrapper horizontal joint were mitigated by welding straps on the outer casing. Further investigation revealed internal leaks by the combustion cans.

      • The original key-way mod was performed on one unit in 2006 but we still have a leak on one side at the horizontal joint. The bottom line: Money was not well spent.

      • We tried peening, re-torqueing, welding, and the seal-key mod with little success. The unit where the original seal key mod was installed leaked worse after the repair than before.

      • Peening over mating areas reduced the amount of air leaking by the joint but did not eliminate leakage.

Hampton then reviewed the various leak-mitigation methods attempted by the OEM, contractors, and users with marginal success. Here are the editors’ notes from that portion of the presentation:

    • Sealants, standard bolting practice.

Research suggested that a sealant and proper bolting should provide temporary leakage relief (Fig 1). However, expect fluid temperature and pressure to erode or blow out the sealant over time. Obtain appropriate torque values from GE spec 248A4158, “Bolt & Stud Torqueing.”

Among the sealing products suggested: Silver Seal II, a non-hardening fibrous paste that flows into rough or irregular surfaces and expands under heat, curing to a leathery-like consistency. Manufacturer IGS Industries, Meadow Lands, Pa, claims it will not crack from thermal cycling or vibration. Two additional high-pressure/high-temperature sealing compounds suggested by users include Esco Products Inc’s (Houston) Copaltite and ICS Industries’ Turbo-R and Turbo-50.

    • Weld wrapper horizontal joints together.

This is a relatively straightforward process with split-line cans 3 and 8 removed (Fig 2).

    • Weld strips of plate steel on the outside of horizontal joints.

This, too, is relatively straightforward as Fig 3 shows.

    • Peen horizontal joints after proper tensioning and removing combustion hardware.

The goal of peening is to spread metal over the joint area where leakage is occurring (Fig 4). Three ways suggested for doing this:

        • Use an air hammer with a punch attachment.

        • Strike the casing along the length of the joint, both above and below the joint, using the ball end of a ball peen hammer.

        • Hammer and chisel along the length of the casing both above and below the joint.

    • Perform “original” seal key mod.

This “solution” calls for machining slots in the outer wrapper (Fig 5). Lower-half horizontal joints are machined; the upper half is flipped and supported for the machining activities. Next, install seal keys into the lower-half horizontal joints. The time-consuming and costly mod was said to offer only marginal benefit. Industry consensus is that this is not an effective method for sealing the wrapper.

    • Install custom shims in horizontal joints.

Procedure is as follows:

        • Hone horizontal joints.

        • Install the upper-half wrapper and torque.

        • Measure gaps along the horizontal joints.

        • Remove the upper-half wrapper.

        • Install shims in gap areas.

        • Reinstall/torque the upper-half wrapper.

    • Retrofit unit with HYTORC tensioning hardware

The three-piece fastener from UNEX Corp’s HYTORC Div is a direct replacement for any type of helical nut. The main reason for its introduction to the 7EA fleet was to reduce the time for bolting/unbolting during outages. The precise bolt tensioning offered by this alternative also was believed to offer more precise torqueing and better flange tightness. However, clashing experienced between the tooling and casing in some areas did not allow for HYTORC nut installation.

    • Install angle iron at wrapper outer joints.

This method was implemented on the right side of the unit described in Fig 6 because it was the only area of significant leakage after installation of the original seal key mod. Think of this as a unique solution to a unique problem.

    • Perform “new” seal key mod to the compressor discharge case and wrapper.

This relatively complex mod is performed only by GE (Fig 7). It is said to be expensive—perhaps into the mid six figures. No users with first-hand experience were found by Hampton in his research. 

Posted in 7EA Users Group |

Comments are closed.