IR 05000423/1985032

Insp Rept 50-423/85-32 on 850708-12.No Violations Identified.Major Areas Inspected:Followup on Licensee Actions Re Violations,Unresolved Items,Const Deficiency Rept & Verification of FSAR Commitment Per Reg Guide 1.94

ML20134H002
Person / Time
Site:	Millstone
Issue date:	08/14/1985
From:	Jerrica Johnson, Kamal Manoly, Varela A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20134G984	List: IR 05000423/1985032 ML20134G982
References
RTR-REGGD-01.094, RTR-REGGD-1.094 50-423-85-32, NUDOCS 8508280191
Download: ML20134H002 (12)

Text

, . __ . - . .

,

o

~

_

.. O

!

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

Report No. 50-423/85-32 Docket N License No. CPPR-113 Priority --

Category A Licensee: Northeast Nuclear Energy Company P. O. Box 270 Hartford, Connecticut 06101 Facility Name: Millstone Nuclear Power Station, Unit #3 Inspection At: Waterford, Connecticut Inspection Conducted: July 8 - 12, 1985 Inspectors: YOr b ~ O *T i K. Man ly, d Reactor Engineer date j G 8- &~ SE A. Varela, Lead Reactor Engineer date Approved by: -h k*

J.' Johnson, Chief, Operational date Programs Section, OB, DRS Inspection Summary: Routine Inspection on July 8 - 12, 1985 (Report No. 50-423/85-32)

Areas Inspected: Routine, unannounced inspection by two region-based inspectors to follow-up on licensee actions related to the following:

violations, unresolved items and construction deficiency reports the majority of which were identified during the Construction Team Inspection (CTI)

No. 423/84-06. In addition, the inspection included a follow-up on a request by the Office of Inspection and Enforcement (IE) regarding verification of FSAR commitment to Regulatory Guide 1.94 for exemptions of concrete testin The inspection involved 51 hours5.902778e-4 days <br />0.0142 hours <br />8.43254e-5 weeks <br />1.94055e-5 months <br /> on site and 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> in the regional office.

!

Results: No violations were identified.

l l 8508280191 850823 i

'

PDR ADOCK 05000423 G PDR

,

- . . _ . - - - . - - _ .

-_ _ -. - - . - - . . - - - . - _ .

~

.

f

.

.

O u DETAILS Persons Contacted Northeast Utilities Service Company (NUSCO)

• K. Gray, Staff Assistant

• P. Quinlan, Staff Engineer

• L. Nadeau,- Assistant Project Engineer

• D. Hoisington, Senior Engineer J. LaWare, QA Engineer Technologist K. Lakshmi, Structural Engineer S. Orifice, QA Engineer Stone and Webster Engineering Corporation (SWEC)

G. Milley, Lead Engineering Mechanics R. Bain, Millstone #3 Project Engineer Mechanics Staff

• G. Basilesco, Millstone #3 Project Engineer Mechanics Staff C. Kuhns, Assistant QA Program Administrator

• W. Vos, Senior FQC Engineer

• M. Matthews, Assistant Superintendent, FQC

• P. Nelson, Lead Engineering Assurance M. Sinha,

. Principal Structural Engineer J. Capozzoli,Jr. , Supervisor of Construction R. Currier, Structural Engineer R. Zawacki, Lead QC Engineer P. Reilly, Structural Designer L. Peterson, QC Engineer U. S. Nuclear Regulatory Commission (NRC)

• T. Rebelowski, Senior Resident Inspector

• Indicates persons present at the exit meetin . Licensee Response on Previous Identified Items 2.1 (Closed) Unresolved Item (423/83-21-04):

This item is related to the evaluation of feedwater piping at the steam generator (S.G.) inlet nozzl Concerns were identified re-garding the qualification of the piping reducing elbow fitting at the S.G. nozzle on the basis of actual geometry and wall thickness. Four specific items of concern were identified in inspection report N /84-12.

l The piping stress re-analysis of the feedwater piping was performed by SWEC to reconcile the actual as-built configuration and geometry of the reducing elbow and nozzle junction at the steam generator,

~

l i

, ,. --wy%..- . - - - - -..-_my- , _ w ,-m-._ . - . w4 ,-w .--., - ,. _ .. - ~ , , # -. +_--...,_9-., y w w. +- ,.,s --- p w ,.- - +

- ...

.

Revision #3 of the re-analysis (which is documented in stress calcu-lation No. 12179-NPCF-X1701) was reviewed during this inspectio The re-analysis addressed the previously identified concerns as follows:

Consideration for effects of stress intensification (SIF) of nozzle stresses on both sides of the junction has been addresse The guidance regarding stress indices for abutting products in ASME section III sub-section NB 3683.2(a) was found to be appli-cable to SIF's for two piping products (reducers) at the nozzle junctio *

The actual as-built configuration and wall thickness of the nozzle reducing elbow were obtained by taking the outside diameter and UT thickness measurements at 0 and 180 along four locations on the elbow centerline._ These measurements were used for the comp-utation of the flexibility and stress intensification factors as required in the pipe stress re-analysis for steam generator C & *- Consideration of load transients resulting from the isolation condenser valve closure inside containment was addressed in the analysi The transient analysis was performed using time depen-dent forcing function at various locations along the piping (at change of direction points).

The maximum computed stresses at the nozzle junction to the steam generator were determined as follows:

s'

--

Sustained loads (EQ.8): 7,768 psi

--

Occasional loads (Normal + upset) (EQ.9): 13,733 psi

--

Occasional loads (Faulted) (EQ.9) : 15,383 psi

--

Thermal Expansion (EQ.11) : 33,240 psi All of the above stresses were below the corresponding ASME code

-

allowable limits in section NC 3650. It was also determined that the licensee has included the inlet nozzle of the steam generator in the ISI program. This item is close .2 (Closed) Violation (423/84-04-05) and CDR (423/84-00-05)

The violation (423/84-04-05) was related to the deficiency identi-fied in ITT Grinnell hanger strut No. 3-CCP-2-PSR039 in which the rod end was found to be jammed against the weld on the inside of the rear bracket and thus did not allow the required angular movement of i 5 degrees of the rod. The licensee issued a potential significant deficiency (CDR 423/84-00-05) indicating that the five degree cone of action required between the strut and the rear bracket was not available due to an interference which only allowed a swing angle of two degrees from norma _

_

c- .

-

(G

_

.

Q,)

The CDR also indicated that the interference was due to a bent rear bracket in conjunction with a bent end of the sway strut. This determination differed from an earlier finding in the Report of a Problem (ROAP) which identified the weld on the rear bracket as the cause for interference with the rod end of the stru The inspector reviewed the licensee's response which concluded that construction damage had caused movement of the pipe clamp 2 1/2 inches from its installed location. The following documents were also reviewed in conjunction with the identified deficiency and corrective action:

• Nonconforamnce and Disposition Report (N&D) No. 4741 which authorizes and documents the rework by replacing the existing defective rear bracket and moving the hanger to its original condition. The completed rework was signed off on April 16, 198 *TheQAinspectionreport(IRNo.P4A91977)docEmentingthe sampling inspection of two hundred and two supports randomly selected from tnirteen systems for similar end bracket deficiency as identified on hanger No. 3-CCP-2-PSR-039. All of the inspected hangers were found to have acceptable paddle to the end bracket assembl * Memoranda from superintendent of construction to construction personnel on January 19 and July 20, 1984 emphasizing the necessity for adherence to Quality Standard (QS) 14.1, " Post Acceptance Work Control" which controls unplanned work performed on an item com-ponent, equipment or structure which had previously been accepted by FQ The identified hanger deficiency was found acceptable to close the violation and the construction deficiency repor .3 (Closed) Violation (423/84-04-09)

This violation was concerned with the identification of two beams (No. 34 B1 & 33 B3) in the Main Steam Valve Building (MSVB) which were not properly seated such that full bearing on the stiffened beam seat was not achieved as required in detail "AJ" of drawing ES-31J- The inspector reviewed the licensee's response to the violation which addressed the original corrective action in addition to that undertaken as a result of the agreement between SWEC, NUSCO and the NRC on August 3, 1984. The licensee 's original corrective action included the inspection of all structural beams in the MSVB with end connection details similar to those with the identified deficienc "

L

.

-

' .I ..

The reinspection by SWEC field QC resulted in the identification of a third improper bearing of beam No. 33 81. The inspector reviewed.

, the QA reinspection record (IR No. S4A03661) and Non-Conformance &

l Disposition Report (N&D) No. 4768 which provided the disposition for the repair by using shims to provide full-bearing for three noncon-forming beam seats. The shimming was completed and inspected in July 13, 1984.

l Additionally, the inspector reviewed the QA records (IR No. 54A04575-and S4A04576) which document the reinspection of fifty more similar beam seats for proper installation (41 in the Reactor Bldg. and 9 in the Control Bldg.). These connections were selected, from a sample of 119 beam seats of a similar design function, and were provided by the structural group for FQC inspection on a sample basis. The in-

.spection was performed in accordance with the structural steel in-stallation specification (C970). This effort was conducted to meet the new commitment for additional sampling reinspection dated August 3, 1984. No other findings of unsatisfactory beam seat installations were identified in the field QC reinspection effort. The licensee's action was considered adequate to close this violatio .4 (Closed) Construction Deficiency Report (423/82-00-13)'

This item invcived the reduced capacity of pipe straps with " reduced ears" and the licensee's action regarding the modification of four-pipe anchors which utilized these straps. The supports were orig-inally designed as six way restraints and were modified as three way restraints. The concern was related to the capability of the straps to restrain the piping along its axi The inspector reviewed the licensee's written response which was provided by SWEC. The response was incorrect as it stated that the pipe straps were designed to restrain the piping in two lateral directions only and that the third or axial restrained direction (for the strap and other support structures associated with it) was de-signed to accommodate the frictional forces resulting from the normal loads in the two lateral direction As was identified during the inspection and from the review of the drawing, the supports are three way restraints for which tne axial load components exceeded those in the plane of the strap. Despite the inaccuracies in SWEC's response, generic calculations were per-formed for the qualification of the straps in three directions using

"STRUDL" computer program. The qualification was performed for a two directional in plane load of 100 lbs and an out-of plane load of 1000 lb _

-

, . -

,

.

h

4 The straps stiffness were determined to be equal to 5 x 10 lb/in in the out-of plane direction. The inspector also reviewed the four calculation packages involving the modified anchor supports. The packages reviewed are:

• Calculation No. Z88A - 031 for pipe support No. 3BRS-2-PSR 031

• Calculation No. Z88A - 066 for pipe support No. 3BRS-2-PSR 066

• Calculation No. Z88A - 019 for pipe support No. 3BRS-2-PSR 019

.

• Calculation-No. Z88A - 065 for pipe support No. 3BRS-2-PSR 065 No further concerns were identifed during this review. This item is close .5 (Closed) Unresolved Item (423/84-04-06)

This item is related to the approach used by SWEC for the combination of piping seismic support loads of intermediate anchor points located between piping runs. Anchor seismic loads are determined using the square root of the sum of the squares (SRSS) of the respective pipe support loads obtained from the pipe stress analyses of piping in both sides of the ancho A review was performed of SWEC's response to the unresolved item, which included specific analyses of three representative piping systems to demonstrate the adequacy of the loading combination tech-nique being used. In addition, the response included other support-ing studies which demonstrate the inherent conservatism in the pre-sent approach for piping analysis. The following studies were presented in SWEC's response:

• A study by SWEC based'on three piping models, each containing one or more intermediate anchors. The models were separated at their intermediate anchor points and each sub-model was analyzed using the response spectrum modal analysis technique. The three models were also analyzed using the more accurate seismic time history method. The time history analysis utilized the floor time history from which the response spectrum is base A comparison of the results from both analyses indicated that the approach utilizing the SRSS summation of the sub-model results had overpredicted the intermediate anchor loads by a factor of 1.24 to more than three over that which utilizes the time history approac The analyses covered pipe sizes from 4" to 14" inclusiv * A study by G. Hubert (ASME Piping and Pressure Vessel Conference, Denver, June 15-16,1981) to compare the dynamic response of LMF8R coolant loop piping using the response' spectra and time modal superpostition techniques. Both analyses were based on the assum-tion of linear conditions. The modal responses in the response

,

..

- . . -

o

'

.

. O

spectrum method were combined in accordance with NRC Reg. Guide 1.92. In'the time history method, the maximum respenses were determined utilizing four possible algebraic combinations of three directional earthquakes. The results at the reactor nozzle from the piping analysis using the time history method were between 11 percent to 42 percent lower then those determined using the response spectrum metho The conservatism resulting from removal of valleys in the response spectra method was evaluated in the study done by Hubert. Examin-ation of the analytical results at a particular mode (12.31 HZ),

indicated that the contribution to nozzle loads using the response spectrum method were approximately 24.5 percent higher than those obtained using the time history metho * A study of the effects of pipe supports non-linearities by L. Severud et al (ASME Piping and Pressure Vessel Conference, June 1982, Vol. 67, "Special Applications in Piping Dynamic Analysis").

A comparison of small bore piping responses was done using:

dynamic test simulating seismic accelerations, linear response spectrum analysis, and non-linear time-domain analysis. The results indicated that when a more exact piping model (including support non-linearities) is introduced, the piping responses were signi-ficantly lower than those obtained using the linear response spectra approach. The study also showed that the time domain non-linear analysis overpredicted the piping response when compared to the results from the pipe testin * Studies by Barte et al (ASME Piping and Pressure Vessel Conference, August 1980, vol. 40, " Seismic Analysis of Piping with Non-linear Supports") were performed to evaluated the effects of supports non-linear characteristics such as free play and local damping at snubber locations. The study was performed using three piping models and utilized the linear response spectrum and non-linear (direct integration) time history method Results from the analyses of the 4", 16" and 28" piping models indicated that the linear response spectra method predicted more conservative piping response and support loads than the more exact non-linear time history method when various support non-linearities were included.

! The above data were found to provide a reasonable confidence that

the SRSS technique for combination of loads at intermediate anchor points in piping runs is an appropriate analytical approac This l item is closed.

i .-

-

n p

,

-

U d

2.6 (Open) Unresolved Item (423/84-04-08)

This item is related to the assumption of decoupled piping response when performing seismic analysis of piping systems supported from structural steel beams. Four cases of piping hangers supported from structural steel in the reactor building were provided to the licensee as examples of this concern. The response provided by SWEC addressed the bases for performing de-coupled seismic analysis when piping systems are compared to the structural steel (as a primary system) with regard to mass and stiffness characteristics. The response, however, did not address the coupling effect between the floor system, such as the annulus steel rack framing with attached piping and equipment, and the building structure (as the primary system) at the specific locations for which the response spectra are specified. Effects of floor amplifications of piping system response as a result of floor mass and flexibility characteristics (as sub-system) was not addressed either for the four cited examples or gen-ericall This item will remain open pending the licensee evaluation and NRC revie .7 (0 pen) Unresolved Item (423/84-04-10)

This item is related to the lack of specific evaluation for local stresses at web attachments of structural steel, introduced by piping or conduit hanger supports. Review of design calculation for the structural steel floor framing in the reactor building annulus area, at elevation 14'-10" between column lines 2 & 3, indicated that no specific evaluation was performed for piping hanger No. 3CCP-1-PSST-092 support attachments to the web of supporting wide flange floor beam SWEC's response to the unresolved item was that " envelop" loads were considered in the design of structural steel framing to account for anticipated piping and other support attachments. It was also stated in the response, that the effects of local stresses would have been evaluated as hanger loads and attachment configurations were determined. However, no specific procedure was available for this evaluation of local attachments to webs of structural steel shapes. As was indicated in the unresolved item and confirmed by SWEC's response, Guideline NETM-27 was being utilized for the eval-uation of eccentric attachments to the flanges of structural steel W shapes. Discussion of the above concern with the licensee and SWEC's representatives identified that SWEC has recently developed a procedure for the evaluation of attachments to webs of structural shapes. This procedure along with NETM-27, will be used for the evaluation of final attachment loads to building steel during the stress reconciliation effort of piping systems. It was also identi-fied that a procedure No. NETM-57 was developed to provide technical direction for the verification of Category I structural steel. The procedure outlined a sampling program for the verification and doc-umentation of original steel design for final loading conditions.

i

_

- -- - - - - - .n . - -

.

"

.

.

o O

The inspector requested to review the generic web attachment proce-dure. A partial submittal was provided after the exit meeting. This item will remain open pending the evaluation of the entire procedure and its implementation in the load verification effort of Category I structural stee .8 (Closed) Unresolved Item (423/84-04-07)

This item is related to the identification of inaccuracies in the de-sign calculation of pipe support No. 3-CCP1-PSA-152. Improper appli-cation of loads was noted during the review of the above design docu-ment. A review of the revised calculations was performed during in-spection (423/84-12) and it was verified that the proper design inter-face loads were used for the evaluation of the embedded plate at which the piping support was anchore The licensee had also committed to performing additional samplirg re-view of piping support calculations to adequately address the extent of the identified design inaccuracy. The inspector examined the re-sults of the licensee's sampling review of 109 design packages pre-pared between October 1 and December 12, 1984. All Engineering -

Mechanics Division (EMD) designers were required to fill out a special load verification form prior to the revision of any EMD cal-culation of duct supports, instrumentation supports and large and small bore piping supports. In addition, the inspector performed a sample review of the following pipe support design load verification identified abov Support Mark N l Calculation Package N I CP-374092-H001 l 12179-NP(F)-Z2074-S092-H001-1 CP-374039-H011 l 12179-NP(F)-Z2074-S039-H011 CP-374524-H018 l 12179-NP( F)-Z074-S524-H018-0 CP-372053-H002 l 12179-NP(F)-072-R053-H002-1 CP-374524-H004 l 12179-NP(F)-Z74-S524-H004-1 The licensee's corrective action was adequate, and no discrepancies were identified during the review of the above design document This item is close .9 (0 pen) Unresolved Item (423/83-17-01)

This item is concerned with the generic approach to the analysis of piping systems which is based on nominal values of piping and fitting thicknesse The concern was identified at another facility, when the failure to consider heavier walled and higher schedule pipe fittings in pipe stress analysis had resulted in increased piping and support loads. The licensee presented the report prepared by SWEC which addresses the review of piping analysis including effects of heavy elbow e - -

.

.

.

O O

The licensee was informed that the office of Nuclear Reactor Regu-lation of the NRC will -issue a safety evaluation for this generic concern.' In addition, the licensee was also informed that, based on preliminary discussions with NRR, a specific evaluation for effects of actual over-thickness of piping fittings (in particular, rigid piping systems which operate at high temperature) will be required to resolve this concern. The licensee agreed to provide the results

.for.the evaluation of the diesel generator exhaust piping system for NRC review, to assess the potential overstress in a typically sus- '

ceptible piping system. This item will remain open pending the issu-ance of the Safety Evaluation by NRR, the licensee's evaluation of the diesel generator exhaust piping considering actual piping / fitting thicknesses and the NRC review of the licensee's evaluatio .10 (Closed) Significant Deficiency (423/81-00-05), Fire Damage to the Containment Liner:

The licensee's final report, submitted to the NRC, dated August 30, /

1982 presented SWEC engineering and construction specifications and techniques for corrective actions to repair the fire damage. The final completion of the prescribed repairs and the required veri-fication and testing by QA/QC was accomplished in February, 198 Documentation of the corrective actions, QA/QC records and the close out of NCRs were reviewed by the inspector and discussed with cogni-zant personnel and responsible QC engineers.. The actions taken to

'

correct the significant deficiency adequately fulfill the require-ments of 10 CFR (50.55.e) and warrants the close out of CRD (423/

81-00-05). This is supported by work observations identified in previous NRC inspection raport . Veri ication of FSAR Commitment to Regulatory Guide 1.94 In the Millstone Unit 3 FSAR, amendment 10, two exceptions are taken in the degree of compliance to Regulatory Guide 1.94, Revision of 1976, re-garding inprocess sampling of concrete in construction and qualification testing of a concretc admixture. Tne licensee states that the degree of

, compliance with RG 1.94 which endorses ANSI N. 45.2.5 (1978) is as follows: '

' Exception #1 The frequency of correlation testing of structural concrete is re-quired by ANSI N45.2.5 to be performed in accordance with the applicable paragraphs of section 6.11 of this standard. The degree

! of compliance to Section 6.11 involves an exception to the frequency '

of inprocess testing for correlation tests between point of delivery and the placement point. This is explained as follows: Documentary evidence of inprocess tests on concrete were reviewed by the in-spector. QC records for concrete sampling of pumped concrete at the l

_

L_

y

'

n p

,

i V d

.

- 10

/

,A ,f-

• • ' / , truck discharge identify that strength tests were taken at the truck l 7 discharge. Tests of air content, slump and temperature taken every 50 cy. at truck discharge were correlated with the same tests every 200 cy taken at the point of placement. SWEC specifications on Concrete Testing' Services (C282) and Concrete Mixing Delivery and

. Placement (C299) further require that any changes in slump, air content and temperature from the truck discharge to the placement point are considered acceptable as long as the test results at the point of placement are within specification limits. If correlation

'a r tests results are unacceptable, the specifications require that correlation tests shall be made at the truck discharge and at the placement point until the problem is alleviated. The compressive strength test reports were observed to clearly identify the correla-

' e t

tion tests for evaluation purposes. The inspector considers the above exception adequately satisfies ANSI N 45.2.5, 1978, Section 6.11 by the more stringent controls required to alleviate unaccept-able inprocess correlation test Exception #2

'

The admixture manufacturer shall submit certified test data confirming air-entraining admixture complies with ASTM C260 when tested in accordance with ASTM C233. For each production lot shipped, the manufacturer shall certify that the admixture is similar to the material represented by the test dat This is an exception to Re ' Guide 1.94/ ANSI N45.2.5 which requires admixture be inprocess tested T- ,'

by a composite of each shipment for chemical composition, Ph and Nt

'

specific gravity per ASTM C-494. The exception is explained as follows: Documentary evidence of the supplier's certified initial report and subsequent quality assurance certification of individual lot shipments were reviewed by the inspector. The admix manufac-turer, Master Builders tested their airentraining product MBVR to meet ASTM C260 requirements when tested in accordance with ASTM C23 Records of each lot shipped certified that the admixture, Vinsol

,

Resin Standard, conforms with all the requirements of ASTM C260, AASHO M154 and Corps of Engineers CRD C13 specifications for airen-training admixture' for concret The records identify that ASTM C233 section 3.1 blend of.three different cements and aggregates qualified the admixture to the ASTM C260 requirement Based on the above documentation and observations during construction by the inspector,

• he concludes that the suppliers control in his manufacture of the

< liquid admixture that was shipped to Millstone Unit 3 Is adequate to

,' support taking exception to RG 1.94/N 45.2.5 of testing the composite of each shipment without testing its' affect on other concrete ingre-dient The licensee exceptions taken to FSAR commitment to RG 1.94, Installation,

, ; f ~ lf it Inspection and Testing of Structural Concrete, were found acceptabl I

'

, - - - . - ---

---

m---.-------------- --------------y---- ---------------- - - _ _ _ _ _

.

-

O O 11 Unresolved Items Unresolved items are matters about which information is required in order to ascertain whether they are acceptable items, violations or deviation Items remaining unresolved in this inspection are discussed in Section 2 of this repor . Exit Meeting An exit meeting was held on July 12, 1984 with members of the licensee staff and contractors as denoted in Section 1 of this report. The inspector discussed the scope and findings of the inspection. At no time during this inspection was written material provided to the licensee by the inspecto l I

i i

{

i I

'

i

_

b '

. . . . - . -

i

_ _ . _ _ . . _ _ _ _ _ _ _ _ _ - - .