Experience in the performance of tube sleeves has shown that the ability of a utility to achieve a desired steam generator lifetime once degradation has been detected and a sleeving repair has been selected will be determined principally by how well the degraded condition has been defined and how thoroughly the sleeving process has been qualified.

The latest sleeving report prepared by EPRI (the Electric Power Research Institute) covers the many lessons learned by the utility personnel who have experienced the challenges of this repair option.


Steam generator tubes in PWRs are subject to various types of degradation. Degradation can initiate from either the primary side (inside) or the secondary side (outside) of the tube surface. Corrective action must be taken if the degradation reaches some pre-established acceptance criteria, normally 40% or greater of the tube wall thickness. Corrective actions include removing the tube from service by plugging or leaving the tube in service by installing a sleeve. At the end of 1996, approximately 100 000 steam generator tubes had been repaired by sleeving. Figure 1 shows the number of tubes sleeved per year. Figure 2 shows this same information as a percentage of tubes in service. The major sleeving campaigns at Doel 4 in 1994 and Maine Yankee in 1995 account for the large increase in these two years.

Due to steam generator replacements and plant shutdowns, not all of these sleeves are still in service. By the end of 1996, a little over 34 000 of these sleeves were still in service. The table (page 38) provides additional information on these sleeves. The vast majority of the installed sleeves have performed as designed without any evidence of sleeve or parent tube degradation. However, there have been a few cases of sleeve and tube assembly degradation related to either difficulties encountered during the sleeve installation process, or in-service degradation of the parent tube at the sleeve-to-tube joint. The installed sleeve material has not shown any significant degradation.


Successful installation and performance of a steam generator tube sleeve requires the integration of a number of factors. These include sleeve material selection, qualified installation procedures, sleeve to parent tube joint design, and production controls. The sleeves should be long enough to bridge the degraded portion of the parent tube to ensure that the joints are made in sound parent tube material. The sleeve material should be resistant to both primary and secondary side degradation. The most commonly used sleeving materials have been thermally treated Alloy 600 and thermally treated Alloy 690. Currently, thermally treated Alloy 690 and Alloy 800 are being offered although Alloy 800 has not been licensed in the United States. In addition, nickel cladding and welded tube repair processes are being developed.

The sleeve joint is one of the most critical areas resulting from the sleeve installation process. Sleeve to parent tube joints are required at both the bottom and top of the sleeve. The primary sleeve joint designs that have been installed include the following:

• Hybrid Expansion Joint (HEJ).

• Laser Welded.

• TIG Welded.

• Kinetic Expansion.

• Hydraulic Expansion.

The process of forming a sleeve joint places an additional stress on both the sleeve and the parent tube materials. The additional stress in the joint area increases the parent tube susceptibility to environmentally induced cracking. Recent improvements in the sleeving process are aimed at minimising the residual peak stress caused by the sleeve installation process. In some cases it may be necessary to stress relieve the joint to reduce the residual stress levels.


Vendor qualification testing is performed for new sleeve designs and for modifications and installation changes to previously tested designs. A typical test programme is designed to accomplish the following:

• Confirm that the installation method produces the required joint strength and leak tightness.

• Confirm that the joint in both the sleeve and parent tube has sufficiently low stresses to provide acceptable resistance to stress corrosion cracking.

• Test the entire range of anticipated process variables.

• Confirm that the installation equipment and procedures perform satisfactorily.

A test programme normally consists of a number of different elements including mock-up testing, sleeve design, mechanical testing, structural and vibration analysis, stress evaluation, accelerated corrosion testing, installation verification, and sleeve inspection. Since in-service failures have been driven by high residual stress resulting in stress corrosion cracking, corrosion qualification testing is especially important.


Just as the steam generator sleeve designs have evolved, so to have the techniques used to inspect the sleeves. The original focus on sleeve inspection was to verify that the sleeve installation was properly performed and to provide a baseline inspection for parent tube integrity. The current emphasis is on providing the most sensitive and reliable inspection techniques to detect and size degradation in the parent tube behind the sleeve and at the sleeve joints. Current sleeve inspections in the United States are primarily performed with the CECCO probe, MRPC Plus Point® probe, or a Combination probe.

The CECCO family of probes are non-rotating probes with a fairly fast inspection speed. The CECCO-3 probe uses the driver-pickup (transmit-receive) technique with the coil connectivity providing axisymmetric compensation. CECCO-3 probes have been used for detection of circumferential cracks in the parent tubing but are not sensitive to axial indications. The CECCO-5 probe has been used for detection of both axial and circumferential cracks. This probe uses a number of pancake type coils for drivers and receivers. This technology has been enhanced by the development of new instrumentation to handle the multiple transmit and receive coils. Analysis software allows for additional data manipulation.

The MRPC Plus Point probe was developed to detect both axial and circumferential oriented flaws and takes advantage of two directionally wound coils differentially paired to cancel non-relevant indications. By differentially connecting the directional coils, a better signal to noise ratio is achieved. A magnetic bias MRPC Plus Point probe has also been used on several occasions to inspect welded sleeves. A magnetic bias is added to the probe to eliminate permeability effects caused by the welding process. The magnetic bias probe has typically not provided substantial improvement and has only seen limited use.

The Plus Point Sleeve Combination Probe contains two absolute bobbin coils with a plus point coil located between the absolute coils. The separation distance between the plus point coil and the absolute coils is the same in each direction. The absolute bobbin coils provide inner diameter profile measurements on the sleeve expansion. Additionally, the plus point coil can provide flaw position with respect to the transition as detected by the bobbin coils. Due to the geometry of the coils and a constant inspection speed, accurate flaw position with respect to expansion transition regions can be determined.


The options for dealing with degraded sleeved tubes include removing the sleeved tube from service by plugging or attempting to repair the sleeved tube. Extensive tube repairs have been performed at the Doel 4 plant in Belgium and the Kewaunee station in the US. At Doel 4 approximately 1700 sleeves were repaired in 1994 by hydraulically expanding the sleeves above the existing HEJ joint, performing a laser weld, and then performing a stress relief heat treatment. An additional 11 000 Laser Welded sleeves were also installed during this outage. The Doel 4 steam generators then operated one cycle before they were replaced.

An extensive sleeve repair campaign was initiated at Kewaunee in the fall of 1996. An initial attempt was made to place a laser weld in the centre of the existing HEJ hardroll joints. This proved unsuccessful and a second repair attempt was made to relocate the laser weld to an upper hydraulic expansion location. Approximately 1250 welds were made with an initial weld acceptance rate of over 80%. During preparation for startup, the secondary side of the steam generators were filled with water and a number of the repaired tubes were discovered to be leaking. A subsequent UT examination showed that a significant number of the welds had “debonded” after the post weld heat treatment. A number of the degraded sleeves were repaired by resleeving the tube. The lower section of the degraded HEJ sleeve was TIG relaxed, the sleeve was cut below the upper joint, and the lower section of the sleeve was removed. The upper sleeve joint was then expanded to allow the insertion of a new sleeve which was TIG welded into place.


Sleeving has been used as an effective repair process. Recommendations for achieving the highest probability of success include the following:

• Select the sleeve design and installation process most appropriate for the condition of the steam generator and the desired sleeve performance life.

• Develop a site specific qualification programme based upon mock-up and analytical testing which bounds the expected site conditions.

• Maintain accountability for, and closely monitor, all aspects of the sleeve installation process. All repair techniques should be qualified prior to implementation.

• For larger sleeving campaigns, consider an in-plant demonstration of a small number of sleeves.

• For all new sleeve installations, use a qualified NDE technique for the baseline examination.

• Keep closely informed and up to date on industry sleeve experience.

Although some sleeve problems have occurred, the performance of the vast majority of sleeves in service has been excellent. By following the above recommendations, utilities can build upon this experience and further enhance steam generator sleeve performance.