Information Notice No. 94-54: Failures of General Electric Magne-Blast Circuit Breakers to Latch Closed


August 1, 1994



All holders of operating licenses or construction permits for nuclear power


The U.S. Nuclear Regulatory Commission (NRC) is issuing this information
notice to alert addressees to a condition in which certain General Electric
(GE) medium-voltage Magne-Blast circuit breakers may begin to randomly fail to
latch closed.  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 are not NRC requirements; therefore, no specific action or written
response is required.

Description of Circumstances

The Maine Yankee Atomic Power Station (Maine Yankee) and the Millstone Nuclear
Power Station (Millstone) have recently experienced failures of safety-related
GE Magne-Blast 4160-Vac circuit breakers to latch closed (also called "going
trip-free").  Failures occurred during post-overhaul, preoperational, and
in-service surveillance testing.  At Maine Yankee, failures occurred shortly
after an overhaul of the breakers, during troubleshooting of an indirectly
related problem with limit switches.  The affected breakers have failed to
latch on a random basis during up to about two-thirds of attempted closures,
both electrically and manually.  At both plants, the affected breakers were
among those recently overhauled by the GE Apparatus Service Division of
Philadelphia, with quality assurance coverage provided by GE Nuclear Energy
(NE) Power Delivery Services (PDS) of King of Prussia, Pennsylvania.  Failure
analysis, inspection, and testing of one of the affected breakers from Maine
Yankee were performed at the GE Apparatus Service Division under the direction
of GE NE PDS and the design engineer from the breaker manufacturer, GE
Specialty Breaker Plant.  The results were consistent with onsite testing at
both Millstone and Maine Yankee.  The failure analysis activities, observed by
NRC and licensee representatives, identified the primary failure mode,
apparent principal root causes, and the most effective remedies.

9407270127 .                                        IN 94-54
                                        August 1, 1994
                                        Page 2 of 4


The breaker models considered to be potentially affected at the present time
include (1) Type AM-4.16-350-1H and (2) Type AM-4.16-250-6, -7, -8, or -9HB.
GE is currently evaluating other Magne-Blast models for potential
susceptibility.  Among these models, only breakers equipped with one prop
reset spring (prop spring), and in which all "Tuf-Loc" (Teflon-coated
fiberglass) bushings (especially the prop bushings) have been replaced with
aluminum-bronze bushings, have recently experienced unreliable latching.
However, it should be noted that this failure ultimately results from a
combination of contributing factors and not all breakers of the potentially
affected type and configuration are prone to failure.  In addition, there have
been a few instances (although not recently) of one-prop spring breakers with
Tuf-Loc prop bushings that have failed to latch due to other unsatisfactory
conditions within the breaker such as excessive wear, hardened or insufficient
lubrication, and/or being severely out of adjustment.  No failures are known
to GE or the NRC to occur in breakers equipped with two prop springs (included
as original equipment since 1968, or added during maintenance or repair).  The
failure analysis testing appeared to confirm the GE position that breakers
with two prop springs should not experience this problem.  The details of the
failure mode are discussed in Attachment 1 to this information notice.
Attachment 2 shows the prop springs.

It should be noted that, depending on the severity of the various contributing
conditions, the onset of failure to latch reliably in these breakers may vary,
and a breaker that has successfully passed rigorous post-maintenance testing
and further receipt inspection and preoperational testing may subsequently
start failing to latch reliably in service.  In recent instances, failures to
latch have begun to occur after about 35-50 operations (cycles) following
overhaul or repair of the operating mechanism.  Once started, the failure
frequency has been observed during testing, to vary from 1 failure to latch in
as many as 50 consecutive operations to 16 failures in 24 attempts.  The
operating characteristics of a given breaker (in particular, adequacy and
consistency of prop action as explained in Attachment 1) can indicate the
likelihood of unreliable latching.  According to GE, the desired prop action
positions the prop fully forward against its stop, thus providing maximum
engagement under the prop pin (called prop wipe).  However, affected breakers
that start (or are likely to start) failing to latch reliably typically
exhibit inconsistent (and often less than the design minimum) prop wipe even
when they do latch.

The NRC and licensees have expressed concerns regarding increased
suscepti-bility to unwanted tripping with less than minimum design prop wipe
caused by vibration or shock (e.g., during a seismic event).  The Magne-Blast
design engineer has stated that because of the geometry and magnitude of
forces acting on the prop, once the mechanism has successfully latched (even
if only barely), it would be extremely unlikely for the shocks or vibration
associated with seismic events to cause the prop to move backwards and out
from under the prop pin.  Nevertheless, a breaker that is latched, but with
less than the design minimum prop wipe, is not in a condition for which it
was analyzed or in which it was tested during seismic qualification..
IN 94-54                                         August 1, 1994
Page 3 of 4

Normal prop wipe can be confirmed with the breaker in its cubicle and closed
(in test position if necessary) in most installations by visual inspection
(using an inspection mirror and light) without having to get too close to
energized components or hazardous mechanism parts if the switchgear cabinet
door can be opened with appropriate safety precautions.  Pending issuance of a
service advice letter (SAL) by the manufacturer on this problem or a safety
communication by GE NE, Attachment 2 illustrates an inspection hole in the
right side of the mechanism frame through which the prop and prop pin position
may be viewed.  While the fully forward prop position (to the stop) is the
desired condition (maximum wipe), the absence of failures to latch along with
consistent and greater than minimum allowed prop wipe is, according to GE,
indicative of continued reliable latching.  However, because minimum prop wipe
is not a published design specification, and because certain other factors
(e.g., opening spring adjustment) may influence latching reliability, GE has
requested that any licensee experiencing Magne-Blast breaker latching problems
or noticing changed, inconsistent or apparently abnormally low prop wipe,
particularly following overhaul or repair including prop bushing replacement,
contact GE NE PDS at (610) 992-6049.

Pending installation of a second prop spring in affected breakers, Maine
Yankee has shifted, to the extent possible, its available dual prop spring
breakers to locations where operability requires assured latching reliability
for closure or reclosure during a design-basis event.  For the remaining
potentially affected breakers, Maine Yankee is planning expedited inspections
with the assistance of GE NE PDS to aid in its operability determinations.

Related Generic Communications

Information Notice 84-29, "General Electric Magne-Blast Circuit Breaker
Problems," dated April 29, 1984.

Information Notice 90-41, "Potential Failure of General Electric Magne-Blast
Circuit Breakers and AK Circuit Breakers," dated June 12, 1990.

Information Notice 93-91, "Misadjustment Between General Electric 4.16-KV
Circuit Breakers and Their Associated Cubicles," dated December 3, 1993.

Information Notice 94-02, "Inoperability of General Electric Magne-Blast
Breaker Because of Misalignment of Close-Latch Spring," dated January 7, 1994.

.                                        IN 94-54
                                        August 1, 1994
                                        Page 4 of 4

This information notice requires no specific action or written response.  If
you have any questions about the information in this notice, please contact
one of the technical contacts listed below or the appropriate Office of
Nuclear Reactor Regulation (NRR) project manager.


                        Brian K. Grimes, Director
                        Division of Operating Reactor Support
                        Office of Nuclear Reactor Regulation

Technical contacts:  Stephen Alexander, NRR
                     (301) 504-2995

   Kamalakar Naidu, NRR        Sikindra Mitra, NRR
   (301) 504-2980            (301) 504-2783

1.  Details of Failure Mode and Contributing Factors
2.  Magne-Blast Prop Mechanism Diagram
3.  List of Recently Issued NRC Information Notices
.                                        Attachment 1
                                        IN 94-54
                                        August 1, 1994
                                        Page 1 of 2


The cause of the failures to latch cited in the body of the Information Notice
is related to a component of the Type ML-13 breaker operating mechanism called
the prop (see Attachment 2).  The breaker will fail to latch closed when the
prop does not travel forward fast enough (relative to the motion of the
closing linkage) to be in the proper position under the prop pin of the
closing linkage as it rises above then descends onto the prop during the
closing cycle (whether manually or electrically initiated).  Absent a trip
condition upon closing, the prop is permitted (by the rotation of the main
cam) to be pulled forward under the prop pin by the prop reset spring (or
springs when two are fitted).  If the slower speed of the prop relative to the
pin is such that the pin passes down in front of the prop instead of landing
on top of it (called latching) as it is supposed to, the linkage will
"collapse" causing the moving contact arms to immediately drop down to the
open position.  This condition or action is also called the trip-free
condition because it occurs in the same manner as if a trip signal were
present at the breaker tripping mechanism when the breaker attempts to close,
thus preventing the prop from moving forward and latching.

According to GE, and as apparently confirmed by testing, several conditions
must be present to render a given breaker susceptible to this failure mode.
Most fundamental are the individual breaker model and configuration, mechanism
type, and vintage.  Within these constraints, certain modifications or
upgrades that have (or have not) been performed on a given breaker influence
its susceptibility.  Finally, the condition of the breaker mechanism in terms
of age and type of moving parts, wear, lubrication, and adjustment can
ultimately be the determining factor with regard to latching reliability.

Due to their particular combination of operating (opening and closing) spring
configurations, all 4.16-kilovolt (Kv)-rated, vertical-lift, air-magnetic
(Magne-Blast) breakers of 350-million volt-ampere (MVA) interrupting rating
(Type AM-4.16-350-1H) that are fitted with only one prop spring are considered
by GE to be potentially affected.  Also potentially affected, owing also to
their particular operating spring configurations, are those 4.16-Kv,
250-MVA-rated Magne-Blast breakers with one prop spring that have a
78-kilo-ampere (kA) close-latch rating (called a "high momentary" rating),
which includes Models AM-4.16-250-6, -7, -8, or -9HB (the "H" suffix
indicates the ML-13 type mechanism and the "B" indicates the high momentary
rating). Within this population, the major factor determining susceptibility
is the number of prop springs installed.  With one anomalous exception (a
breaker suspected of being grossly out of adjustment), only breakers with one
prop spring have thus far been reported to experience unreliable latching.
Breakers of these models built before about 1968 were originally fitted at the
factory with a single prop spring.  After this time, breakers with close and
latch ratings of 77 Ka or 78 Ka were factory fitted with two prop springs, in
order, according to GE, to make the breakers less sensitive to fit and
adjustment tolerances and to facilitate factory setup.  The failure analysis
testing indicated that breakers built with or modified subsequently to add a
second (upper) prop spring, and that are not otherwise in severely out of
tolerance condition, should not be susceptible to this failure mode..
Attachment 1                                         IN 94-54
                                        August 1, 1994
                                        Page 2 of 2

The breaker manufacturer, in GE Service Advice Letter (SAL) No. 073-348.1,
issued in December 1990, recommended replacement of the main (lower) prop
spring with an improved spring (called the "gold" prop spring because of the
color of its cadmium plating).  NRC Information Notice 90-41, "Potential
Failure of General Electric Magne-Blast Circuit Breakers and AK Circuit
Breakers," also addressed this problem.  The gold prop spring has been fitted
in breakers built since 1971, due to fatigue failures reported in the original
spring.  Although SAL 073-348.1 did describe the dual prop spring
configuration, it did not address the addition of a second prop spring.  See
Attachment 2.

Recent breaker overhauls/repairs that included replacement of the Tuf-Loc
(Teflon-coated fiberglass) prop bushings with aluminum-bronze prop bushings,
in conjunction with cleaning and relubrication of the closing linkage and
renewal of worn parts have not been demonstrated conclusively to be the sole
cause of the recently reported failures to latch.  However, they appear to
have been the common precipitating events.  In breakers of the affected models
with one prop spring that have had all the Tuf-Loc bushings (except for the
prop bushings) replaced with aluminum-bronze, the recently reported failures
to latch have begun to occur after about 35-50 operations (cycles) after
completion of overhaul or repair that included replacement of the Tuf-Loc prop
bushings with aluminum-bronze.  Since 1975, these breakers have been fitted
with aluminum-bronze sleeve bearings or bushings in their mechanisms (except
for the prop bushings) for improved wear characteristics.  Many of the older
Magne-Blast breakers, in accordance with GE SAL 073-318.1 (and .1A), issued in
1977, and 318.2, issued in 1979, have had their original Tuf-Loc bushings
replaced with the aluminum-bronze bushings.  Information Notice No. 84-29, "GE
Magne-Blast Circuit Breaker Problems," also addressed this issue.  The factory
bushing replacement kits, supplied under catalog no. 0156C9403G001, did not
include replacement bushings for the prop.  Hence, most prop bushing
replacements have been performed only recently.

When other conditions in a breaker are conducive to marginal latching, it
appears that the new bushings, in conjunction with reduced friction in the
closing linkage from the overhaul, cleaning and new lubricant, may shift the
force and speed balance in the mechanism.  The shift can be sufficient to
speed up the motion of the closing linkage and prop pin relative to the prop
(which also may actually be slightly retarded) beyond design margins.  Under
these conditions, a single prop spring can no longer move the prop forward
fast enough to latch reliably, or if it does latch, to ensure the desired prop
wipe.  Where the second prop spring has been installed in conjunction with the
mechanism overhaul, failure to latch reliably in the manner described herein
has not been reported, and according to GE, as confirmed by testing, would not
be expected to occur.  Disassembly and detailed inspection of the Maine Yankee
breaker operating mechanism after testing revealed no additional significant
factors contributing to the failures to latch experienced by the test breaker.
Therefore, the test results appear to have confirmed the factors to which
close-latch reliability is most sensitive.
.                                        Attachment 2
                                        IN 94-54
                                        August 1, 1994
                                        Page 1 of 1


Page Last Reviewed/Updated Thursday, March 25, 2021