Information Notice No. 84-49: Intergranular Stress Corrosion Cracking Leading to Steam Generator Tube Failure

                                                            SSINS No.: 6835 
                                                            IN 84-49       

                               UNITED STATES  
                        NUCLEAR REGULATORY COMMISSION
                    OFFICE OF INSPECTION AND ENFORCEMENT
                            WASHINGTON, DC 20555

                                June 18, 1984

Information Notice No. 84-49:   INTERGRANULAR STRESS CORROSION CRACKING 
                                   LEADING TO STEAM GENERATOR TUBE FAILURE 

Addressees: 

All pressurized water power reactor facilities holding an operating license 
(OL) or construction permit (CP). 

Purpose: 

This information notice is provided, as a notification of potentially 
significant problems pertaining to operation and inservice inspections of 
steam generators in pressurized water reactor systems. It is expected that 
recipients will review their facilities and consider actions, if 
appropriate, to minimize similar problems occurring at their facilities. 
However, suggestions contained in this information notice do not constitute 
NRC requirements and, therefore, no specific action or written response is 
required. 

Description of Circumstances: 

In February 1984, 3 weeks before a scheduled refueling outage, Fort Calhoun 
detected a primary leak rate of approximately 0.2 gpd in steam generator B. 
In a concerted effort to locate the leak during the outage, the licensee 
conducted helium mass spectroscopy indicator tests before and after sludge 
lancing. Both tests were unsuccessful in identifying the location of the 
leak. A hydrostatic test with a dye indicator also was unsuccessful in 
locating the leak. 

During the outage, extensive eddy current testing (ECT) was conducted as 
part of the licensee's planned inservice inspection program and in support 
of a rimcut modification program. The Fort Calhoun Station has two steam 
generators, each containing 5,005 Inconel-600 tubes which are 0.75 inch 
outside diameter and 0.048 inch minimum wall thickness. Full length 
examinations were made of 1,454 tubes in steam generator A and 1,034 tubes 
in steam generator B. At the time of the testing, data evaluation detected 
only one previously known flaw in steam generator B, A total of nine tubes 
were plugged because they would not pass the 0.540 inch ECT probe. 

On May 16, 1984, the unit was conducting a hydrostatic test in preparation 
for returning to power operation. The cold-leg temperature was 398F. 
The reactor coolant system pressure was 1,800 psi and the steam generator 
pressure was 200 psi. While plant personnel were closely watching steam 
generator B for indications of the small leak experienced before shutdown, 
an unanticipated increase 

8406180360 
.


                                                              IN 84-49     
                                                              June 18, 1984 
                                                              Page 2 of 3  

in water level indicated a tube failure. The maximum leak rate was later 
estimated at 112 gpm. A high leak rate persisted for approximately 10 
minutes, while the RCS pressure was decreased and the main steam line 
isolation valve associated with steam generator B was closed. 

The failed tube was found in the second peripheral row from the outside. The
failure was a 1 1/4-inch-long- axial "fishmouth" opening along the tube 
bottom on the hot-leg side of the horizontal run at the top of the "U". It 
was located between the scallop bars in the vertical batwing support. 
Sections of the failed tube and adjacent tube were removed for laboratory 
analysis. 

Analysis revealed the failure mode to be intergranular stress corrosion 
cracking (IGSCC) from the outside, through 95% of the wall thickness, with 
the remaining 5% evidencing ductile tearing. The tube cross section was 
ovalized, with elongation by 0.046 to 0.122 inch on the major axis (along 
the plane of the fracture) and compression of 0.045 to 0.070 inch on the 
minor axis. An additional defect, through approximately 50% of the wall, was 
found 1/4 inch from the hot leg end of the fishmouth failure. This was 
similar to the first defect, except that it was oriented 45 to the tube 
axis. Modified Huey tests indicated that the material was not sensitized. 
Microstructure was typical of mill annealed Inconel-600. Scanning electron 
microscope energy dispersive spectrometry failed to reveal corrosive 
chemical deposits, even in the crack tips. There was no evidence of fretting 
or wall thinning. 

The failed tube was one that had been the subject of eddy current testing 
(ECT) in both 1982 and 1984. Review of the ECT tapes of those tests showed 
no flaw in 1982 but revealed an indication of a defect through 99% of the 
wall in 1984. Although this indication was unambiguous and not affected by 
interference, it was missed by the analyst who evaluated the 1984 tapes 
before the hydrostatic test. The second defect also was apparent in the 1984
ECT tapes and also was missed. 

Prior to restart, the licensee is performing ECT of all tubes in both steam 
generators which are accessible with the remote probe insertion machine and 
which were not tested in 1984. The licensee will reevaluate, with 
independent verification, the ECT data tapes for the tubes already tested in 
1984. The licensee has presented test results which indicate that tubes 
sufficiently ovalized to obscure serious defects from detection by ECT are 
sufficiently restricted to prevent passage of the 0.540 inch ECT probe. 
These tubes would be plugged on the basis of their restriction. 

Fort Calhoun has always operated with an all-volatile-treatment secondary 
chemistry program. ECT examinations were conducted in 1975, 1976, 1977, 
1978, 1981, and 1982. Very few degraded tubes were detected over this 
period, and the failed tube is the first defective tube detected. ECT 
conducted after the tube failure has revealed another tube in steam 
generator B with a defect through 42% of the wall in the batwing section on 
the hot-leg side of the horizontal run at the top of the bundle. In 
addition, two tubes in steam generator A were found to have defects on the 
cold-leg side, near the tube sheet: one showed a defect through 39% of the 
wall, about 10 inches above the tube sheet; 
.

                                                              IN 94-49     
                                                              June 18, 1984 
                                                              Page 3 of 3  

the other showed 2 defects 27% and 50% through the wall, 4 inches above the 
tube sheet. 

Although it is likely that the failed tube is the one which was leaking 
before the outage, this cannot be known with certainty, until the reactor 
returns to power operation. Investigations by Combustion Engineering  and 
the licensee are continuing in an effort to identify the cause of the IGSCC. 
The Nuclear Regulatory Commission is continuing to review the results. 

If you have any questions regarding this matter, please contact the Regional
Administrator of the appropriate NRC regional office or this office. 



                                   Edward L. Jordan Director 
                                   Division of Emergency Preparedness 
                                     and Engineering Response 
                                   Office of Inspection and Enforcement 

Technical Contact:  S. Long, IE
                    (301) 492-4791

Attachment:
List of Recently Issued IE Information Notices