1CAN118610, Requests Relief from Requirement for Volumetric Exam of Reactor Coolant Pump (RCP) Flaw Indications,As Part of First 10-yr Inservice Insp Program.Exam of RCP a Showed No Growth in Flaw Length Over 12 Yrs of Operation.Analyses Encl

From kanterella
(Redirected from 1CAN118610)
Jump to navigation Jump to search
Requests Relief from Requirement for Volumetric Exam of Reactor Coolant Pump (RCP) Flaw Indications,As Part of First 10-yr Inservice Insp Program.Exam of RCP a Showed No Growth in Flaw Length Over 12 Yrs of Operation.Analyses Encl
ML20214K294
Person / Time
Site: Arkansas Nuclear Entergy icon.png
Issue date: 11/24/1986
From: Enos J
ARKANSAS POWER & LIGHT CO.
To: Stolz J
Office of Nuclear Reactor Regulation
Shared Package
ML20214K296 List:
References
1CAN118610, NUDOCS 8612020250
Download: ML20214K294 (4)
v • d • e


Text

. .

Ei]

ARKANSAS POWER & LIGHT COMPANY POST OFFICE BOX 551 UTTLE ROCK. ARKANSAS 72203 (501) 371-4000 November 24, 1986 1CAN118610 Mr. John F. Stolz, Director .

PWR Project Directorate No. 6 Division of PWR Licensing - B U. S. Nuclear Regulatory Commission Washington, DC 20555

SUBJECT:

Arkansas Nuclear One - Unit 1 Docket No. 50-313 License No. DPR-51 Relief Request for RCP Weld Flaw Indications

Dear Mr. Stolz:

This letter is provided to document the results of our meeting with your staff conducted at ANO on November 20, 1886 and supersedes our letter dated November 11, 1986 (1CAN118604).

During our current refueling outage, we conducted a volumetric examination of the "A" Reactor Coolant Pump (RCP) welds as required by our first 10 year ISI Program (based on the requirements of the 1980 Edition through Winter 1981 Addenda of Section XI of the ASME Code), by performing a radiographic (RT) examination of the pump casing weld. The RT indicated the presence of a flaw which exceeded the allowable indication standards of IWB-3000. The indication is best described as a series of slag inclusions having an effective length (per ASilE Section XI Criteria) of 5.66 inches. The indication is located in the vertical weld which ties together the upper and lower scroll welds of the pump casing (see Figure 1).

Radiographic parallax techniques indicate that the top of the flaw is 1.5 inches below the outside surface of the weld. The weld is approximately 2.6 inches thick in this area. Application of special UT techniques indicated that the flaw indication does not extend to the internal diameter of the pump casing.

1

$%A2o20,ggggggga @d G

\

MEMHER MsOOLE SOUTH UTILITIES SYSTEM

Nsvemb3r 24', 1986

.The original construction radiographs were then reviewed to determine if any-flaw existed at this location preservice. Our review was performed with two Level.III Inspectors and found five small inclusions that are part of the current flaw indication of 5.66" in length. 'These inclusions on the original radiograph-were determined.to be acceptable per the Code during the -

preservice examinations. Because of_the quality of the preservice

-radiograph in the area of the. indication, equipment was brought on-site to

_ perform computer enhancement of the area of the_ flaw. This process allowed us to characterize the flaw on the. original. film more clearly and allowed us i to determine conclusively that the current: flaw indication' and the original flaw were identical.

l After determining that no growth in flaw length has occurred over the past 12 years lof pump service, we then performed fracture mechanics and pump case stress analysis on the weld flaw in accordance with ASME Section XI. The analysis was done on a model of a continuous flaw of 5.66" in length and (conservatively) 1.1" in depth to the internal diameter (ID) of the' casing.

(UT and visual examinations performed indicated that the flaw did not extend through the ID.) .The transient-analysis that was.used was 240 heat-up and cooldown cycles (40 years) which was determined analytically to be the most severe transient in contributing to flaw growth. From the analysis, the flaw indication was found acceptable in accordance.with IWB-3612 criteria of

-ASME Section XI. The analysis shows negligible flaw growth over-the 40 year design life of the plant'and our comparison'of the' current ISI radiograph to the original construction radiograph confirms no increase in flaw length over 12 years' service.

The original construction radiographs for the remaining three pumps were then reviewed (by a Level III Inspector) searching for any preservice. flaw indications or weak areas in film density. Identified areas were then

-computer enhanced (by another Level III Inspector) in an attempt to identify any unacceptable flaws that were previously unidentified. Portions of approximately 20% of all preservice radiographs were computer enhanced.

From this review, "C" and "D" pumps were determined to have no unacceptable preservice flaw indications, however the computer enhancement on "B" pump

-did-indicate an unacceptable flaw indication in the same general weld area as the "A" pump.

The flaw indication on "5" pump thrcugh the computer enhancement process was shown as 1.5" in length. The original construction radiograph of this area shows a flaw of 0.625" in length which was acceptable per Code requirements

at that time. The wall thickness in the area of the flaw indication is 1

3.1". Ultrasonic techniques (UT) were used in an attempt to better characterize the flaw indication (see Attachment 1 for more detail). Due to the material of the pump casing (coarse grained statically cast stainless) and the small size of the indication, UT was not able to specifically characterize the flaw. However, from these examinations, we were able to

-determine that the flaw size _was no bigger than 1.5" long by 1.5" deep. A discussion of this is provided in Attachrant 1.

Navimb2r 24, 1986 To conservatively bound this flaw analytically, 2 calculations were perfo rmed.

Calculation 1:

A fracture mechanics analysis was performed in accordance with IWB-3122.4. This analysis assumed a flaw length of 1.5" and (conservatively) through wall (3.1"). 240 Heatup/Cooldown cycles was assumed and a Center' Cracked Panel was used as a model. This analysis demonstrated that a crack of this size would not significantly propagate over the life of the pump.

Calculation 2:

A fracture mechanics analysis (IWB-3122.4) and a pump case stress analyses were also performed. These analyses assumed a continuous flaw length of 4.6" and 2.3" deep (75% through wall). The analyses further assumed 240 Heatup/Cooldown cycles and a semicircle flaw model. These analyses demonstrated that no significant flaw growth would occur over the life of the pump and the pump casing is capable of maintaining its pressure retaining capability with this conservatively sized flaw for the life of the pump. These analyses demonstrated compliance with the acceptance criteria of IWB-3612.

The UT. technique applied was such that it is extremely unlikely that the flaw could be larger than 1.5" x 1.5" without being readily detectable. The actual readings obtained by the UT examination indicated that the flaw is substantially less than 1.5" in depth. Thus there is very high confidence that the above calculations very conservatively bound the existing flaw.

The flaw discovered in "A" RCP during the inservice inspection requires augmented inservice inspection in accordance with IWB-2420(b). AP&L intends to rarform the required inspection of the flaw area using techniques (currently under development) which will not require disassembly of the RCP.

IWB-2430(a) requires an examination of an additional component (RCP).

Currently techniques do not exist to perform a volumetric examination of a RCP without disassembly of the pump. Disassembly, inspection, and reassembly of a RCP cequires several weeks of outage time, is very costly (million dollars), and results in considerable man-rem exposures. Our disassembly of "A" RCP this outage resulted in greater than 60 man-rems of accumulated exposure. Such a job requires a substantial amount of very careful planning to minimize exposure.

The ANO-1 refueling outage is currently scheduled to end on December 3, 1986. There is not sufficient time to allow for disassembly of an additional RCP without significantly extending that outage (months) nor has the necessary planning to minimize man-rem exposure been conducted.

Therefore we request NRC grant relief to IWB-2430(a) (in accordance with 10CFR50.55a(g)(6)(i)) and allow examination of a second RCP ("B") during our next scheduled refueling outage. We intend to perform inspection of "B" RCP

, . , , - _ , . , - , , . _ . _m -,

Novemb:r 24, 1986 during the next scheduled' refueling ~ outage using techniques that will not require disassembly of the RCP. AP&L has consulted with several vendors knowledgeable in this area and has determined that substantial development.

efforts are underway to use UT and'other techniques to perform examinations

'of RCPs. We will be pursuing these efforts and believe a suitable technique will be available for that outage. '

We believe the relief request is in the best interest of the public as it would alleviate the need for extending the outage through the coldest winter months and can result in_ lower man-rem exposure due to the availability of non-disassembly techniques. It is further justified as the inservice examination of "A" RCP did not. indicate that there is any mechanism acting inservice (the flaw is a preservice flaw) that is causing the initiation.

and/or propagation of flaws thus there is no cause to believe that new flaws are occurring on other RCPs. The reviews of'preservice radiographs by the Level.III' inspectors were thorough. 100% of "B" RCP preservice radiographs have been. computer enhanced and did not identify.any other unacceptable flaws. Application of the state of the art computer enhancing technique provided assurance beyond that required by the Code. All evidence indicates that the existing flaw on "B" RCP is very small and has been very conservatively bounded by analyses which indicate that the flaw will not significantly grow or jeopardize the pressure retaining capability of the RCP.

Due to the startup schedule of ANO-1, we request NRC grant this relief by

December 3, 1986.

Very truly yo s,

. Ted Enos, Manager Nuclear Engineering and Licensing JTE/GWW/sg Attachments: 1.' Description of UT Techniques

2. Fracture Mechanics Analysis of ANO-1 "A" Pump Case Indication
3. ANO-1 Pump Case Stresses ("A" RCP)
4. Fracture Mechanics Analysis of ANO-1 "B" Pump Case Indication - Calc 1
5. ANO-1 pump Case - Crack Growth Due to Transient ("A" & "B" RCP)
6. Fracture Mechanics Analysis of ANO-1 "B" Pump Case Indication - Surface Flaw - Calc 2
7. Fracture Mechanics Analysis of ANO-1 "A" Pump Case Supplementary Topics 4

l L_

l

(-FIGURE 1 )

A4 l

I l I

i l

l .

l CASEMENT CIRCUMFERENTIAL l WELD 1

j A4 l

l "B" PUMP FLAW "A" PUMP FLAW _

INDICATION 3 /

/ INDICATION l

1 CASEMENT i CIRCUMFERENTIAL

) WELD

> d ARKANSAS NUCLEAR ONE REACTOR COOLANT PUMP UNIT I

.m .

-- y a,--

i l-

=*

  • , ~.

h ATTACHMENT 1 a,

Description of UT techniques applied to ANO-1 RCP Casing Welds on "A" & "B" Pumps. ,

i

~

The'following techniques were applied to the A &,B C'asing Welds to provide further characterization of indications detected by radiography.

I. 0* longitudinal wave technique Objective: To determine' depth to'the top of the indication.as

[ - measured from the 0D surface.

Technique: A 0* transducer is applied to the OD surface of the-casing in the area of the RT-indications. A sound wave is transmitted

, normal to the surface and is reflected back to the transducer when it encounters a defect or the back (ID) surface. Several different /'

frequencies and sizes were used for these investigat, ions. Frequencies 1 - ranged from .50 to 2.25 MHz, sizes ranged from .5" to 1.0" in diametdr.,

. Both single and dual elements were used.

Results: Several reflectors were observed in the are'as of the RT:

indications'and seemed to be confined to the mid-suction of the casing '

wall. No indications reported in the upper 1/3T (00).

'~

Illustration:

TRANSDUCER ,

OD '

v y CASING WALL ID M "

II. 45* side by side dual element technique 4

Objective: To complement the 0*~ examination and try to measure thru-wall extent.

Technique: Two 1.0" diameter transducers of 1.0 MHz frequency were mounted on a shoe to produce a 45* refracted longitudinal wave in the casing and weld material. One of the transducers acts as the transmitter and the other as a receiver. The sound wave is transmitted through the material, if it encounters a defect, the sound is reflected

and transmitted back to the' receiver. A calibrated time base is used to measure the travel time of the sound wave and therefore the depth of the reflector can be determined.

w.. * , - - - - + - . - .y .,-+--_r-e__ --

-.-.,,m= .r 3-- w - - - , , - m. --r, ..m,,___s .r . .,,,_,n .- , . - - , . , . , , - . - - . , - . - . .

Jig .

af

,1., .- #

J7-J-Results: 'Several reflectors were observed in the mid-wall section of

<> the casing in the areas of the RT indications. Due to the affects of the coarse grained material in the casing, poor signal to noise ratio's.

resulted and no estimate of the'thru-wall depth could be provided. No

-indications were observed in the upper 1/3T (00).

Illustration:

TRANSMITTING TRANSDUCERS TRANSDUCER OD

{ RECEIVING TRANSDUCER CASING ID 5

/ i , .

SIDE VIEW n

7 III. 45* opposed dual element technique ,

, Objective: _ This technique provides a bounding measurement of flaw

/ extent (i,e., flaw is bigger than or smaller than).

! ~O gj Technique: ,

Two transducers, 1.0" diameter and 1.0" MHz frequency, 1

are mounted in a fixture to transmit the sound into the material at a 45" angle (longitudinal. mode) and'the receiving transducer is positioned at the appropriate distance to receive the reflected signal from the back (ID) surface. If no obstruction of the sound beam is

. encountered then a strong received signal results. As the surface area of the flaw increased, with respect to the area of the sound beam, the

'1- amplitude of the received signal decreased. Based on the beam width associated with these transducers a flaw with a dimension of 1.5" by 1.5" would be large enough to completely obstruct any of the

. transmitted energy from reaching the receiver.

Results: This technique was applied to the area containing the RT indications as well as adjacent ateas of weld and base material. No significant changes in the received amplityde were observed. The conclusion is.that the RT indication is suostantially less in area than 1.5" by 1.5".

Illustration:

TRANSMITTING RECEIVING

TRANSDUCER TRANSDUCER OD SIGNAL CASING WALL 45, RECEIVED ID _

TRANSMITTING RECEIVING TRANSDUCER TRANSDUCER WELD OD SIGNAL CASING WALL 45 BLOCKED ID

Surface Condition The surface conditions in the areas where the ultrasonic transducers were placed had been ground (by hand) in the factory. This resulted in some localized areas where contact was not sufficient for reliable examination.

While this condition is not optimal it appeared sufficient for the bounding type of characterizations performed. Obviously, improving the entire surface would provide some improvement in characterization. However, this would likely result in a small bounding condition and would probably not provide any added value with the technology available at the time of these examinations.

l t

I

- . , - . ., , - -

Retrieved from "https://kanterella.com/w/index.php?title=1CAN118610,_Requests_Relief_from_Requirement_for_Volumetric_Exam_of_Reactor_Coolant_Pump_(RCP)_Flaw_Indications,As_Part_of_First_10-yr_Inservice_Insp_Program.Exam_of_RCP_a_Showed_No_Growth_in_Flaw_Length_Over_12_Yrs_of_Operation.Analyses_Encl&oldid=3270457"