Information Notice No. 86-81, Supplement 1:Broken External Closure Springs on Atwood & Morrill Main Steam Isolation Valves
UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
January 11, 1988
Information Notice No. 86-81, SUPPLEMENT 1: BROKEN EXTERNAL CLOSURE
SPRINGS ON ATWOOD & MORRILL
MAIN STEAM ISOLATION VALVES
Addresses:
All holders of operating licenses of construction permits for nuclear power
reactors.
Purpose:
This notice is provided as a supplement to Information Notice No. 86-81 to
alert recipients of information about additional closure spring failures on
main steam isolation valves (MSIVs) that have been reported at Fermi Unit 2.
The MSIVs were manufactured by Atwood & Morrill and the springs that failed
were manufactured by Duer Spring and Manufacturing. It is expected that
recipients will review the information for applicability to their facilities
and consider actions, as appropriate, to avoid similar problems. However,
suggestions contained in this information notice do not constitute NRC
requirements; therefore, no specific action or written response is required.
Background:
At Fermi 2, in May 1986, the licensee observed that four external closure
springs (the inner members of the affected pairs of external springs) were
broken into several pieces. The licensee analyzed two of the broken springs
and determined the failure to be quench cracking caused by the heat
treatment process during manufacturing. The spring manufacturer, Duer
Spring and Manufacturing, also performed laboratory metallographic
examination of the failed springs and verified that the failure was the
result of quench cracking. Both Atwood & Morrill and Duer recommended that
all the external closing springs on all MSIVs be cleaned and subjected to
magnetic particle testing at the earliest opportunity. Duer provided an
inspection procedure. In addition, General Electric (GE) issued Service
Information Letter No. 422, dated July 18, 1986, recommending a visual
inspection and in some cases load tests. It also stated that springs
passing the 105 percent load compression test would be expected to provide
satisfactory service at normal loads. The broken springs at Fermi were
replaced with springs that had been compressed to 105 percent normal load.
No magnetic particle inspection was conducted at that time.
8801050394
.
IN 86-81, Supplement 1
January 11, 1988
Page 2 of 3
Description of Circumstances:
At Fermi 2, on March 21, 1987, the licensee found two additional broken
springs of the same type and location. These springs had been inspected
according to the GE recommendations. The licensee attributed the failures
to pre-existing seams and quench cracks (induced during heat treatment) that
had propagated as brittle fracture failures. The spring material, which is
the same as that of the springs that had previously failed, had undergone
temper embrittlement. The licensee plans to replace all currently installed
valve springs in early 1988. After heat treatment, but before they are
painted, the replacement springs will be subjected to the following tests:
magnetic particle, full compression (108 percent of the maximum operating
stress), and spring constant.
Discussion:
The licensee performed a fracture mechanics analysis of the data, taking
into account temper embrittlement. Their calculated value for critical
crack size in temper embrittled material under 100 percent maximum operating
stress is 0.03 inch. Their estimate for critical crack size in temper
embrittled material under 105 percent maximum operating stress is 0.027
inch. These results are essentially equal. Measurements of critical crack
size on 7 actual fractures varied from 0.044 inch to 0.077 inch. These
crack sizes are smaller than those visually detectable, particularly when
the surface is coated. Consequently, the licensee concluded that the 105
percent load test followed by a visual examination would not suffice for
separating defective springs. It should be noted that a previous analysis
contracted for by the licensee did not take into account temper
embrittlement effects and did lead to an earlier conclusion that the 105
percent load test would suffice.
The failures of springs which had been subjected to 105 percent load tests
and visual inspection appear to confirm the licensee's conclusion that the
root cause is temper embrittlement, quench cracking, and surface
imperfections. The licensee's corrective action of conducting full
compression testing followed by magnetic particle testing thus appears to be
appropriate.
Springs from Duer Spring and Manufacturing have been supplied to both
pressurized-water reactor and boiling-water reactor nuclear power plants on
MSIVs manufactured by Atwood & Morrill. Atwood & Morrill has identified
those plants that would be affected by the problem. This list is attached.
.
IN 86-81, Supplement 1
January 11, 1988
Page 3 of 3
No specific action or written response is required by this information
notice. If you have any questions about this matter, please contact the
technical contact listed below or the Regional Administrator of the
appropriate regional office.
Charles E. Rossi, Director
Division of Operational Events Assessment
Office of Nuclear Reactor Regulation
Technical Contacts: R. L. Cilimberg, NRR
(301) 492-9656
Vern Hodge, NRR
(301) 492-8196
Attachments:
1. List of Affected Plants as Identified
by Atwood & Morrill
2. List of Recently Issued NRC Information Notices
Page Last Reviewed/Updated Tuesday, March 09, 2021