ML20197B917

Safety Evaluation Re Concrete Expansion Anchor Safety Factors for High Energy Line Break Restraints

ML20197B917
Person / Time
Site:	Dresden, Quad Cities
Issue date:	07/23/1997
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML17188A054	List: ML20197B914 ML20197B917
References
NUDOCS 9803130203
Download: ML20197B917 (5)

Text

'

.. a nog

'p \" UNITED STATES NUCLEAR REGULATORY COMMISSION l WASHINGTON, D.C. sete6 eeH

\.....)

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION REGARDING CONCRETE EXPANSION ANCHOR SAFETY FACTORS F03 HIGH ENERGY LINE BREAK RESTRAI)(Li CDPtiONWEALTH EDISDN COMPANY.AND I

MIDAMERICAN ENERGY COMPANY DRESDEN NUCLEAR POWER STATION. UNITS 2 AND 3 OVAD CITIEL h E EAR POWER STATION. UNITS 1 AND 2 DOCKET N05. 50-237. 50-249. 50-254 AND 50-265

1.0 BACKGROUND

Commonwealth Edison Company (Comed), the licensee of Dresden and Quad Cities nuclear power plants, stated in a document dated August 23, 1996, that a safety factor of two for concrete expansion anchors was used in the original design of the high energy pipe whip restraints at Quad Cities Nuclear Power plant. Although the document does not mention Dresden by name, it is understood that the issue also applies to Dresden. Comed also stated that it had used a minimum safety factor of four for the design of new concrete expansion anchors for high energy pipe whip restraints since the issuance of the NRC IE Bulletin 79-02, " Pipe Support Base Plate Designs Using Concrete Expansion Anchor Bolts," (Bulletin) which requires a minimum safety factor of four for concrete expansion anchors.

However, the licensee did not upgrade or re-evaluate those concrete expansion anchors for hinh energy pi>e whip restraints which were designed and installed prior to the issuance of tie Bulletin. Comed provided its reasons for not upgrading or re-evaluating those anchors as follows: (1) the Bulletin specifically addresses the safety factor of concrete expansion anchors for safety related pipe su) ports only, and does not address the safety factor of concrete expansion anciors for other types of applications, such as high energy pipe whip restraints, (2) the loading condition on high energy pipe wipe restraints is equivalert to a faulted loading condition and, therefore, a safety factor of two for cencrete expansion anchors is consistent with that of the structural steel and concrete design under similar loading conditions, and (3) the NRC allows a safety factor of two for wedge type concrete exnnnsion anchors to be used as operability criteria for safety related pipe supports at Quad Cities, and this justifies the use of a safety factor of two for concrete expansion anchors for high onergy pipe whip restraints because they are ENCLOSURE

~

9003130203 900212 PDR ADOCK 05000237 j P PDR J

f, l* .

l' .

1 l 1 i

subjected to a one-time accident load only, which is similar to the Safe Shutdown Earthquake load for the safety related pipe supports.

In addition to the above three reasons, the licensee indicated that the actual forces expected at the high energy pipe whip restraints would be less than those calculated for the design of concrete expansion anchors and, therefore, provide an additional margin of safety for the concrete expansion anchors.

-The licensee listed sources for which the calculated forces at the restraints were greater than the actual forces. These conservatisms included: 1 the forces at the high energy pipe whip restraints were calculated assumin(g)that the pipe break was'an instantaneous phenomenon which resulted in very large pipe break loads at the restraints rather than the actual pipe condition of leakage before the break which would result in much smaller loads at the restraints, (2) the pipe whis restraint forces were calculated by simple linear analysis techniques wiich resulted ic larger forces than the more realistic forces had the more rigorous nonlinear analysis techniques been used. (3) the pipe wall will deform locally when it breaks and this deformation will absorb energy and reduce the pipe whip restraint force - this phenomenon was ignored in the calculations of pipe break and pipe whip ,

restraint forces, (4) many relatively low pressure / temperature high energy lines with smaller diameters-can be shown to be unable to cause a pipe whip, and (5) NRC Generic Letter (GL) 87-11, ' Relaxation in Arbitrary Intermediate Pipe Rupture Requirements" would allow elimination of a significant number of the arbitrary intermediate breaks and, thus, a large number of high energy pipe whip restraints are no longer required.

The licensee argued that the safety factor of two for concrete expansion anchors for high energy pipe whip restraints should be acceptable because no licensing commitment on the safety factor for concrete expansion anchors for high energy pipe whip restraints was made for Quad Cities, and because of the technical reasons and discussions presented above.

2.0 U ALUATION AND DISCUSSION The staff commends the licensee's action for using the safety factor of four for designing its concrete expansi n anchors for high energy pipe whip after the issuance of the Bulletin, alti., A it believed that a safety factor of two should be adequate. The staff agrec with the licensee that the Bulletin specifically addressed the concrete expansion anchors for safety related pipe supports, as the title of the Bulletin stated. Since many safety related pipe supports had used concrete expansion anchors, and the staff was not aware that concrete expansion anchors were also used for high energy pipe whip restraints at the writing of the Bulletin, the title of the Bullet'n addressed pipe supports only. However, it has been the staff's position that the requirements in the Bulletin are acceptable for concrete expansion anchor design in all types of supports for safety-related piping or equipment.

The Updated Final Safety Analysis Report (UFSAR) for the Quad Cities and

-Dresden Station-in Section-3.8.4.6, " Concrete-Expansion Anchors", states; "A mixture of wedge- and self-drilling-type concreta expansion anchors have been

i used in safety-related areas (at Quad Cities Units 1 and 2). TLe minimum embedmont depth for wedge type expansion anchors is *-1/2 anchor diameters.

Self-drilling anchors were predominately used prior to 1977. All concrete expansion anchors were specified to be installed in accordance with ,

manufacturer's recommendations."

This implies that all expansion anchors were installed in accord 6nce with manufacturer's recommendations. In a conference call on June 19,1997, the licensee stated they did not follow the manufacturer's recommendations for j high energy pipe whip restraints. The licensee also stated that

, Section 3.8.4.6 is nearly word-for-word their response to NRC IE Bulletin 79-02.and that Section 3.8.4.6 applies only for pipe supports as addressed in the Bulletin. However, the UFSAR is silent with respect to the fact that i Section 3.8.4.6 is only for pipe sup> orts covered by the Bulletin and not other concrete expansion anchors suci as those used for the high energy pipe

whip restraints.

While the staff agrees with the licensee's assertion that the loading condition on high ener y pipe whip is equivalent to a faulted loading condition, the staff d sagrees with the licensee's statement that a safety factor of two for concrete expansion anchors was consistent with that of structural steel and concrete design under similar loading conditions. This is because the safety factor used in the structural steel and concrete vesign has a different meaning or definition from that of concrete expansion anchors.

This difference is explained below.

Steel is usually designed to fail in tension. Tensile yield stress is usually specified as a limit in structural steel design. Concrete is usually designed to fail in crushing. Compressive strength (crushing stress or strain) is 4

usually specified as a limit in concrete design. The allowable stress for steel and concrete for Category I structures during normal operating conditions, including Operating Basis Earthquake or wind, are set at about one-half of the steel yield stress and concrete crushing stress. Therefore, '

if the safety factor is defined by the ratio of failure stresses-(yield in steel and crushing in concrete) to stresses at plant operating condition, it ,

can be said that Category I structures posses a safety factor of roughly two

. to allow mainly for extreme loading conditions, such as Safe Shutdown Earthquake (SSE) or tornado loads. The yield stress of steel and compressive strength of concrete materials are usually obtained in a quite reliable manner, especially in nuclear construction where quality control and quality assurance programs are required. Therefore, the safety factor of two is quite reliable and, thus, judged to be adequate.

However, the failure load of concrete expansion anchors can not be precisely i and reliably predicted due to many factors. Concrete expansion anchors are used to resist tensile or/and shear forces at an anchcrage. The tensile force of a wedge type concrete expansion anchor depends on the friction force between the steel bolt and surrounding concrete. This type of frictional force is-influenced by the type of aggregate, especially its hardness =, used in concrete, and the-degree of compaction of concrete around the steel bolt.

--s_.i,__._.._,~--..--..,__~_..,...._--..,-. , , - , _ _ - , . , - - - - . . . . < , . . , ~ - , . m.m.,-ww,,----- _ _ _ _ - _ ~

T..

4 i

This kind of influence on anchor capacity is difficult to address in practice, and has been ignored by bolt manulacturers in their catalogs and by design engineers in their calculations. The tensile capacity of a concrete expansion anchor is reduced by concrete cracking near the wit,.and the amount of reduction depends on the size, length, location.-and amount of concrete cracking. Concrete cracking in the future is difficult to predict, especially as to the size, length, location, and the amount of cracking. The individual tensile capacity of concrete expansion anchors in an anchorage is also reduced if the steel bolts are closely spaced, and the amount of reduction depends on the spacing of the bolts. The same kind of influence and reduction also applies to the shear capacity of concrete expansion anchors although to a lesser degree. Safety factor in concrete expansion anchors was introduced to allow for variations in aggregates hardness, concrete compaction around the steel bolts, bolt embedmont length between specification and actual installation, and concrete strength between specification and actual condition, for possible future concrete cracking, for bolt spacing effect, and other effects, such as dynamic loading. Major anchor manufacturers, such as Hilti, Inc., have chosen one fourth of the tensile and shear load capacities from a single boit embedded in concrete with no cracking and pulled by a static loading device at their laboratories as the allowable loads.

Therefore, the 1/4 of the test capacities, or conversely the safety factor of four, simply reflects the need for covering the many ,ariabla physical conditions that may reduce the anchor capacity in the field from its test value in an " ideal" condition. The meaning of safety factor in concrete expansion anchors-is not to provide allowance for extreme loading conditions, such as SSE or tornado loads or the faulted loaj condition as indicated by the licensee.

The licensee has used the NRC allowance for a r.afety factor of two as operability criteria for safety related pipe sLpports at Quad Cities as a technical justification for the adequacy of using the safety factor of two for concrete expansion anchors for high energy pire whip restraints on a permanent basis. The licensee has misinterpreted the latent of the NRC allowance for the operability criteria. The provision of an operability criteria is intended to provide the licensee an opportunity to fix or upgrade those anchors meeting operability criteria by the next refueling outage for non-accessible supports. Meeting operability criteria does not mean those anchors are technically sound on a permanent basis, as the licensee indicated.

With respect to the first three sources listed by the licensee that could

, potentially reduce the anchor loading for the high energy pipe whip restraints, the staff recommends that the licensee perform the analysis to establish the new anchor loads. If the total reduction of anchor loads from these analyses can be shown to be equal or more than one half of their original values, then this would be equivalent to increasing the safety factor from two to four. However, statements alone without analyses can not justify the acceptance of the safety factor of two. With respect to the rest of the sources listed by the licensee that could eliminate many high energy pipe whip restraints for consideration or evaluation, the staff-encourages the licensee to do so. For those anchors the licensee can not eliminate, they need to be

o. ..

i l

upgraded to meet the manufacturer's recommendations. The staff does accost t se criteria listed in the Bulletin for the use of concrete expansion anciors for high energy pipe whip restraints. The anchorage criteria in the Generic Implementation Procedure (GIP , Revision 2, for resolution of Unresolved Safety Issue (USI) A-46 develo) ped by the Seismic Qualification Utility Group and approved by the staff is also acceptable.

3.0 CONCLUSION

The UFSAR fur both Quad Cities and Dresden indicates that all concrete expansion anchors were installed in accordance with the manufacturer's recommendations; however, the licensee said the high energy pipe whip restraint concrete expansion anchors were not installed in accordance with manufacturer's reconnendations.

Comed stated that it designed its new concrete expansion anchors for high' energy pipe whip restraints with a safety factor of four after the issuance of the Balletin, but believed that the criginal design with a safety factor of two prior to the issuance of the Bulletin should also be acceptable. The licensee submitted its technical arguments. The staff has reviewed the submittal and does not agree with the licensee's technical arguments as stated above.

The licensee must meet the manufacturers recommendations to confonn to the UFSAR. The criteria for concrete expansion anchors in Bulletin 79-02 and GIP, Revision 2, are both acceptable for meeting these recommendations for high energy pipe whip restraints; and the staff has determined that the licensee could perform additional analyses and/or upgrade its anchor capacity for high energy pipe whip restraints. The licensee should provide a schedule as to when they will either conform to the UFSAR, complete the calculations for their additional analyses or complete the anchorage capacity upgrade.

Principal Contributors: J. Ma R. Pulsifer Date: July 23, 1997

,