Information Notice No. 85-10: Posttensioned Containment Tendon Anchor Head Failure

                                                            SSINS No.: 6835 
                                                            IN 85-10        

                                UNITED STATES
                        NUCLEAR REGULATORY COMMISSION
                    OFFICE OF INSPECTION AND ENFORCEMENT
                           WASHINGTON, D.C. 20555

                              February 6, 1985

Information Notice No. 85-10:   POSTTENSIONED CONTAINMENT TENDON ANCHOR 
                                   HEAD FAILURE 

Addressees: 

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

Purpose: 

This information notice is provided to alert recipients of current 
information relating to a potentially significant problem regarding recent 
failures of 170-wire posttensioned containment tendon anchor heads at Unit 2 
of the Farley Nuclear Station. It is expected that recipients will review 
the information for applicability to their facilities and consider actions, 
if appropriate, to detect a similar problem at their facilities. However, 
suggestions contained in this information notice do not constitute NRC 
requirements; therefore, no specific action or written response is required.

NRC is continuing to obtain and evaluate pertinent information. If specific 
actions are determined to be required, an additional notification will be 
made. 

Description of Circumstances: 

Farley Unit 2 

On January 28, 1985, while conducting a preintegrated leak rate test 
walkdown of the exterior of the containment structure at the Farley Unit 2 
facility, an alert utility worker noted grease leakage and a deformed 
vertical tendon anchor grease cap on the top of the containment ring beam. 
When the grease cap on the same tendon was inspected in the tendon access 
gallery, it also revealed a deformed grease cap. Removal of the grease cap 
showed that the field anchor head had broken into seven pieces. The 
posttensioning force (approximately 1.5 x 10-6 pounds) also had been 
released and numerous broken wires from the 170-wire tendon were found. 

On the basis of this finding, the utility removed some additional tendon 
anchor grease caps. Of the first eight anchor heads uncovered for 
inspection, one was found to be cracked. Inspection was curtailed until the 
cracked anchor head can be detensioned. The tendon associated with this 
anchor head is still transmitting posttensioning force to the containment. 
The utility determined from their records that the broken anchor head and 
the cracked anchor head have the same fabrication lot control number. 

8502060136
.

                                                           IN 85-10        
                                                           February 6, 1985
                                                           Page 2 of 3     

Additionally, the utility has sent some of the pieces of the broken anchor 
head to two laboratories for a series of tests for failure analysis, 
including tests on metallurgical, mechanical, chemical and other physical 
properties. Testing of the corrosion inhibitor grease is under way. 

Currently, the utility has personnel from the architect-engineer and the 
material supplier on-site in the continuing investigation. Neither the NRC 
nor the licensee has yet fully integrated the information regarding the 
results of previous tendon surveillance activities at the Farley site into 
this information notice. Oral information from the licensee indicates the 
tendons at Farley Unit 2 were posttensioned in early 1977. The unit has been
operational since May 1981. The specific tendon whose anchor head failed and
the one found with a cracked anchor head were not included in the sample of 
tendons that were subjected to surveillance activities since the plant began
operation. Thus, there is no definitive information currently available on 
the time of occurrence of the breakup of the one anchor head or the crack 
formation in the other anchor head. 

While no specific conclusions have been reached at this time regarding the 
cause of the failures, the NRC believes that based on the conversations with
the supplier, INRYCO, that all material from the same fabrication lot 
control number as the failed heads was utilized exclusively at Farley Unit 
2. 

The previous history of anchor head failures before the event at Farley Unit
2, in nuclear applications, has been confined to occurrences during the 
construction phase (during or shortly after posttensioning). It is during 
this time that the tendon system, including the anchor head, undergoes the 
maximum loading force. 

As background information,, previous 170-wire tendon anchor head failures 
during construction at other facilities are briefly summarized below. 

Bellefonte Units 1 and 2 

During 1975 and 1976 a series of eight rock anchor heads, supplied by INRYCO
for the containments at Bellefonte Units 1 and 2, failed during construction
installation. In the phased construction process these 170-wire assemblies 
were sealed for long periods in a highly alkaline water environment. These 
anchor heads were to be coupled to the posttensioned containment vertical 
tendons to serve as a direct tie between the containment and the rock 
foundation material. In these instances the anchor head also broke into 
several pieces. The licensee's investigations completed on these failures 
cited several possible contributors. These included: (1) high anchor head 
stress as a result of a 1.4-inch-diameter hole in the head for grout 
passage, (2) inclusions in the steel found oriented parallel to the final 
failure plane, (3) bending of shims and anchor plate, and (4) unknown 
environmental conditions which facilitated stress corrosion cracking. The 
NRC had an independent study made that concluded possible stress corrosion 
cracking as the initiator. 

The resolution of the problem resulted in the removal of all the anchor 
heads and replacement with new anchor heads made from a vacuum degassed 
(cleaner)
.

                                                           IN 85-10       
                                                           February 6, 1985
                                                           Page 3 of 3    

steel with the center grout hole eliminated and the anchor head coated for 
temporary environmental protection. The NRC is aware of no further failures 
at Bellefonte after this corrective action. 

Byron Units 1 and 2 

In November of 1979 two 170-wire anchor heads on horizontal tendons were 
reported to have failed during construction of the Byron containments. One 
failure occurred one day after stressing and seating the tendon and the 
other occurred 13 days after stressing and seating. By the end of January 
1980, two additional anchor heads had been reported as having failed. The 
supplier of the anchor heads was INRYCO. Investigations were made by INRYCO 
on the material from seven separate fabrication lots. It was found that the 
basic steel material used in several of the batches had been manufactured by 
a process that utilized vanadium grain refinement causing an incompatibility
with the postfabrication heat treatment. This resulted in a different steel 
chemistry that would have required a higher temperature for proper heat 
treatment. As a result of this conclusion all anchor heads that had received
improper heat treatment for the basic steel chemistry were removed and 
replaced. The NRC is not aware of any failures at Byron since the corrective
action. 

Discussion 

Because the integrity of the posttensioned concrete containment structure is
based on a highly redundant system of numerous tendon elements (several 
hundred), the failure of one such element in a family of tendons does not 
jeopardize containment structural capability. It does, however, necessitate 
a determination that a mechanism or systematic problem has not arisen under 
service conditions when one such failure in a tendon is revealed. Specific 
tendon geometry, tendon size, containment design details, and location of 
individual tendons with lost or lowered strength properties would dictate 
the critical number of tendons that could be lost before containment 
integrity is jeopardized. 

No specific action or written response is required by this information 
notice. If you have any questions about this matter, please contact the 
Regional Administrator of the appropriate regional office or this office. 


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

Technical Contact:  R. Shewmaker, IE
                    (301) 492-7432

Attachment: List of Recently Issued IE Infomation Notices
.