ML20216G711

Insp Rept 99900005/98-01 on 980218-20.Violations Noted. Major Areas Inspected:Review Manufacturing & Testing Activities for Specialty Metal Plan & Determine If Activities Meet NRC Requirements

ML20216G711
Person / Time
Issue date:	03/18/1998
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20216G625	List: ML20216G621 ML20216G711
References
REF-QA-99900005 NUDOCS 9803200068
Download: ML20216G711 (10)

Text

.

! U. S. NUCLEAR REGULATORY COMMISSION l OFFICE OF NUCLEAR REACTOR REGULATION l;

Report No.: 99900005/98-01 Organization: Westinghouse Electric Company Specialty Metals Plant Commercial Nuclear Fuel Division Blairsville, Pennsylvania 15717-8904

Contact:

Mr. Joseph H. Ewing, Manager Product Assurance Department l Nuclear Industry Westinghouse Specialty Metals Plant manufactures zirconium alloy Activity: fuel clad tubing for use in the nuclear power industry.

Inspection Dates: February 18-20,1998  !

Inspectors: Robert Pettis, Jr.,HOMB/DRCH Gregory Cwalina, HOMB/DRCH Donald Naujock, ECMB/DE Approved: Robert A. Gramm, Chief Quality Assurance and Safety Assessment Section Quality Assurance, Vendor inspection, and Maintenance Branch Division of Reactor Controls and Human Factors i

i l

l 9803200068 900318 PDR GA999 EMVWEST 99900005 PDR ENCLOSURE

I' I

1 INSPECDON

SUMMARY

The purpose of the inspection was to review the manufacturing and testing activities for the Specialty Metals Plant (SMP) and determine if these activities were conducted in accordance with NRC requirements. The inspection bases were as follows:

Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants," to Part 50 of the Code of Federal Reaulations (CFR),10 CFR Part 50.

l Part 21, " Notification of Failure to Comply or Existence of a Defect," of 10 CFR. 2 STATUS OF PREVIOUS INSPECTION FINDINGS The previous inspection, 99900005/95-01, did not identify any findings.

3 INSPECTION FINDINGS AND OTHER COMMENTS 3.1 Backaround The Westinghouse SMP, part of Westinghouse Commercial Nuclear Fuel Division (CNFD),

located in Blairsville, Pennsylvania, manufactures zirconium (Zr) alloy (zircaloy) tubing for use in

! the nuclear power industry. Although production at the plant in Blairsville began in 1955 (e.g.,

l nuclear fuel pellets through 1960, stainless steel turbine blades, and forged bar and strip products), the manufacturing of zircaloy tubing started in 1967, and the manufacturing of inconel steam generator tubing began in 1968. In 1985, the manufacturing of inconel was discontinued and the CNFD SMP was committed completely to zirconium-alloy-based nuclear-grade (a) l tubing for fuel rod cladding, (b) tubing for discrete burnable absorber rod cladding, and c) tubing for thimble tubes, instrumentation tubes, sleeves, spacers and connectors. According to CNFD SMP, it has produced over 70 types and sizes of zirconium alloy tubing. Typical products l produced at SMP are Zircaloy-2 for boiling-water reactors, Zr4 for PWRs, ZlRLO for longer operating cycles and higher bumups, guide thimbles, and burnable absorber tubes.

L 3.2 Fuel Rod Manufacturina Process  ;

The inspectors evaluated the adequacy of the fuel rod manufacturing process (see Figure 1),

including process control, manufacturing operations, and in-process checks and inspections.

Specifically, the team observed pilgering, cleaning, pickling, straightening, grit blasting, cutting and finishing of the rods. The insrectors reviewed applicable procedures and interviewed

. several manufacturing process operators and in-process technicians during the Inspection.

Tube Sortin9 Tube Tube Tube a Finishing Forming Inspection Weighing l

[  !

Figure 1 - Zircaloy Tube Manufacturing Process 2

l t

e

f l -

1 r .

I 3.2.1 Follow Cards and Inventory Control The inspectors reviewed SMP procedure PE (Process Engineering)-200," General Manufacturing Requirements," and observed SMP processes and procedures for assuring material traceability and process control. Prior to manufacture, each manufacturing lot is given a ,

lot number for process tracking and inventory control. Each production lot is segregated and will be accompanied by a follow card for the entire manufacturing process. The follow card contains i the lot number (which is also etched on the stock as it moves through the manufacturing operationa in accordance with PE-200), original ingot number (from the Western Zirconium facility), customer identification and applicable drawing number. The follow card also identifies 4 each manufacturing step (operation) and the appropriate PE procedure for performing the step, and any other requirements (e.g., weighing and recording). The lot number and each individual manufacturing step are also identified by a unique bar code, included on the follow card, which is read into the computer tracking system by the operator at the beginning and end of each operation. The operator is required to enter into the computer the bar code of each lot, the process operation number, the stage of the process (e.g., start or stop), and the operator's identification which is read off the operator's badge. The computer tracks the manufacturing l

progress and prevents a step from being started before the previous step is complete. The ,

follow card also requires the operator to initial for the completion of each step and to record the l

date and amount of material processed and scrapped in accordance with PE-200. I The NRC inspectors observed operators using the follow cards during several manufacturing operations. Operators logged in all pertinent information and utilized the bar code readers with no difficulty. In response to an inspector questions, an operator logged incorrect information into the system. The inspectors noted that the computer tracking system identified the errors and  ;

required the operator to provide corrections before allowing the operator to proceed. The inspectors inquired as to the difficulty of using the system. Operators indicated that the current system worked very well and did not express any concerns. Some operators did identify that a few problems had existed in the past, when pencil style bar code readers were used. However, the switch to pistol grip readers approximately a year ago has significantly reduced any problems in using the bar code system.

The inspectors concluded that the current manufacturing process provides sufficient controls to assure proper material traceability throughout the manufacturing process and to track manufacturing progress of the Individuallots.

3.2.2 Sortina The inspectors reviewed PE-201, " TREX Sorting and Weighing," and observed TREX (Tube Reduced Extrusions) that had been sorted and identified for manufacturing to assure SMP's method for identifying material for manufacture was adequate. SMP receives Zr4 and ZlRLO TREX from Westinghouse's Western Zirconium facility located in Ogden, Utah. After receipt, SMP sorts the TREX into production lots, the lot size depending upon the final product. For example, fuel rods are typically manufactured from lots in the 550-650 pound range, while thimble tubes lots usually weigh approximately 250 pounds. The inspectors toured the fuel sorting area and noted the TREX had been sorted into specific manufacturing lots, lot numbers were etched on the TREX and the weight of the lot had been noted. In addition, the lots had been sorted by material, Zr4 and ZlRLO , and placed on color coded skids. Severallots had been tagged with the follow card and were ready to be moved into manufacture. Lots for which 3

the follow card had not yet been prepared were identified by color coded tracking cards until the follow cards could be attached. Based upon sorting area observations, the inspectors concluded that SMP's method for lot preparation was adequate.

3.2.3 Tube Formina The inspectors reviewed procedures and observed SMP forming operations, including tube reduction (by pilgering), deburring, cleaning, pickling and annealing. SMP receives hollow TREX approximately 2.5 inches in diameter from Western Zirconium. Fuel clad tubing is produced by reducing the outside diameter (OD) and the TREX wall thickness through three basic reduction steps (described below) to form fuel rods. The production steps include cold pilgering, deburring, cleaning, pickling and annealing, as well as in-process Ultrasonic (UT) and dimensional checks (see Figure 2). The final rods are then sent through the tube finishing and finalinspection processes. All activities were performed in accordance with Follower Cards (travelers) and written procedures.

cold * ==

pg Deburr Clean Pickle Anneal M - In process uT and Dimensional Checks ww - Not performed for each step Figure 2 - Zircaloy Tube Forming Process The inspectors were provided a typical follow card for producing .360" OD ZlRLO fuel tubes and followed the process used for that item. In some cases, depending upon the product, the manufacturing process or sequence varied slightly.

Pilaerina Fuel rods are produced by cold reducing TREX in three successive steps through cold pilgering until the final product size is reached. This process, performed in accordance with PE-211,

Cold Pilgering," accomplishes tube elongation and wall reduction by rolling TREX back and forth between two grooved dies. During this process, the tubes are rotated and advanced in small increments over a stationary mandrel. These tube hollows are then cut to a specified length, deburred and engraved with the lot number. The cut tubes are placed in a wet holding tank until the run lot is completed. The holding tank keeps the tube lubricant moist, making it easier to remove in the cleaning process. The inspectors observed the pilgering process and discussed the process with pilger operators. The process was being performed in accordance with the specified procedure. The inspector had no concerns in this area.

4 l

i w

In-Process Testino During the pilgering process, several samples are taken to assure the pilger machines are operating properly and producing tubes within specified acceptance ranges. SMP personnel use the test results as a go no-go test for releasing the material to the next processing step.

The number and sizes of the samples are given in Parameter Sheet 5.1 of PE-211. There are two basic sample sizes: 1) the first is a long sample (full rod length) which is sent for UT examination,2) the second is a short sample consisting of a an approximately 3-inch section and another 1-inch section. The pilger operator cuts and prepares a 3-inch short sample and performs an initial look by measuring the ID with an air gage and the OD with a laser micrometer. The sampla is then placed on a tray marked with the pilger machine identifying number. An in-process technician is responsible for measuring and recording the sample information. i The 3-inch sample is checked for ID and OD by using the air gage and laser micrometer and the outside surface is visually examined. The 1-inch sample is cut (by the operator) lengthwise and is visually examined for inside surface defects. The inspector observed an in-process technician reject a 3-inch sample due to outside surface indications. The pilger operator changed pilger dies on the machine in question, before continuing the production run, after being informed of the problem by the technician.

The long sample is checked by UT and the UY results are checked by the in-process technician and plotted on the same control chart as the short samples. The inspectors determined that the UT machine is calibrated for ZlRLO , although Zr4 is also examined using the same machine.

Discussions with the technicians determined that there is a very slight difference between the materials. Therefore, the technicians must be aware of that when making their determinations.

However, the technicians informed the inspector that the control chard are plotted using the actual measurements, not adjusted values.

l The inspectors observed the pilger operator cut, prepare and measure samples. The inspector

)

also observed the in-process technician measure, examine and plot the results for the short <

samples and review and plot the UT results on the long samples in accordance with PE-469, '

"Pilger Control Chart Forms." The inspectors determined that samples were taken, measured and recorded in accordance with applicable procedures. The inspectors reviewed a completed control chart for a .374-inch diameter fuel tube and noted that the ID and OD variance was greater for the long sample than the short sample. The operators explained that the long sample UT results were taken for an entire tube length as opposed to the short sample length of approximately 3 inches. Therefore, it is expected that the variance would be greater. The inspectors examined several other charts and noticed similar results. In all cases examined, the long sample results bounded those taken on the short samples. The inspectors concluded that the in-process testing performs an adequate function for provid;ng the process operators with timely input on process quality. The inspectors had no concerns in this area.

Cleanina Following the pilgering process, the tube hollows were transferred to the pickle house for cleaning and pickling (if necessary). Cleaning is accomplished in accordance with PE-205,

! " Alkaline Cleaning Pilgered Tubes." First, the tube hollow bundles are rinsed with service water I

for several minutes. The bundles are then lifted in slings and immersed in a heated alkaline

! 5

l solution for several minutes, periodically being lifted and drained to remove most of the loose dirt. The bundles are then rinsed again. Final cleaning is accomplished after the tubes are built into a carrier ( an array which separates the tubes allowing more thorough cleaning of each tube). Again, the tubes are immersed in an alkaline solution for several minutes, being raised to drain and redipped several times. The tubes are rinsed to remove most of the soap from the

< tubes, then are rinsed in a second tank. Two more rinses are performed, one with service water and one with deionized water. The tubes are then hung to drain before being placed in a dryer.

The cleaning process is performed after each pilger reduction. The inspectors observed several bundles being cleaned, of various size tube hollows. The cleaning was accomplished in accordance with the requirements of procedure PE-205. The inspectors did not identify any concerns in this area.

1 Picklina After cleaning, the tube hollows are pickled. The pickling process consists of acid etching the tubes to remove a small amount of material and provide a smooth, uniform finish, both inside and out. SMP utilizes two pickling methods, depending upon the material and process requirements. For the example follow card provided to the inspectors, the tubes are to be ID flush pickled after the first pilger reduction. This process is controlled by procedure PE-210

" Flush Pickling." The tubes are mounted in a plastic-lined tank and attached to a manifold. The tubes are then flushed with water, acid etched and then flushed again. Wall thickness !s measured before and after pickling to assure stock removalis within specified limits. The inspectors observed the flush pickling process performed on tubes following their first reduction and discussed the process with the pickle operator The operator measured and recorded the wall thickness at the same location before and after pickling, in addition to flush pickling the ID, all tubes are subjected to bright pickling of both the ID and OD after the final pilger reduction and cleaning. PE-209, " Bright Pickling," specifies the requirements and acceptance criteria for this pickling process. In this process, the tubes are dipped into the acid solution to remove stock material from both the inside and outside surfaces and then immediately dipped in a rinse tank. (Note: when performing bright pickling, the tubes are not rinsed with deionized water or dried first). The number of dips is determined by the strength of the acid solution. Measurements are taken and plotted before and after dipping. The inspectors observed the bright pickling process performed on tubes following their final reduction and discussed the process with the pickle operator who performs and records all measurements as required. The inspectors did not identify any concems in this area.

Annealino Following each cleaning / pickling operation, the tube bundles are placed in a furnace for annealing. The annealing process heat treats the tubes to relieve stresses caused by the pilger process, permitting the tubes to continue through the process, either the next pilger reduction or tube finishing. Annealing is done in either cold or hot-walled furnaces at a specified temperature and time. The inspectors had no concerns with the annealing ~ process.

l The inspectors determined that the tube forming process was accomplished in accordance with approved procedures. The inspectors noted some minor discrepancies with some procedures.

Those discrepancies were discussed with SMP personnel who agreed to make necessary changes. No other concerns were identified.

6 i

l

. 1 3.2.4 Tube Finishina l

l The inspectors observed SMP tube finishing operations to assure they were being accomplished in accordance with approved procedures. At SMP, tube finishing includes tube straightening, ID blasting, cutting, polishing and final cleaning (see Figure 3).

l Final I

Straighten ID Blast Length Polish )

Clean Figure 3 - Zircaloy Tube Finishing Process Straiahtenina Tube straightening was performed by machine. The inspectors observed that each tube is run through a series of offset rollers which slightly bend then straighten the tubes. Each tube is monitored by in-process checks. The inspectors did not identify any concerns in the tube straightening area. '

Inside Diameter Blast The inside diameter of the tubes are then blasted to provide a final internal polish and remove any residual manufacturing imperfections. The tubes are blasted using a specified size grit.

Blasting can be done in either or both directions. The inspectors observed the grit blasting process and discussed the process with the operator. The machine automatically monitors the grit flow and shuts down when flow is insufficient. The inspectors were present during an automatic shutdown and observed the operator adjust the machine and perform some surface cleaning. The machine did not experience any further problems. The inspectors did not identify any concerns with the grit blasting process.

Cuttina The tubes are then cut to a specified length, and the ends were faced, deburred, and checked for squareness. The cutoff machine number was recorded, the weight measured and recorded, and the follow card signed by the operator. The inspectors observed the operators set up and operate the machine to perform automatic cutting and measuring of end squareness. Cutting is accomplished by removing a specified length from one end and the rest of the material from the opposite end. The inspectors did not identify any concems in the cutting area.

Finishina At this point tubes are sent to the finishing cell. This cell cleans the tubes using a specified cleaning solution, rinses and dries the tubes, inserts polyurethane end plugs and polishes the outside surface. The inspectors observed an operator load and operate the finishing cell and also observed another operator prepare and operate the surface polisher. The operator 7

l l

adjusted the pressure of the polishing belts (the machine includes three belts of specified grit).

The pressure is adjusted to remove a specified amount of material from the OD. A test piece is measured, run through the polisher, and remeasured to determine the amount of material removal. During this inspection, the inspectors observed the first test piece was processed and found to have had too much material removed. The operator readjusted the machine and polished another test piece within acceptable limits. Finally, the tube is passed through an alloy verification system to assure that the proper alloy was being supplied to the customer. ZlRLO tubes are marked with their trade name at the end of each tube.

Based upon observations and discussions with operators, the inspectors determined that the manufacturing and inspection activities were performed in accordance with established SMP procedures. The inspectors did not identify any concerns in this area.

3.2.5 Tube inspection ,

i The inspectors observed SMP final tube inspection operations to assure they were being accomplished in accordance with approved procedures, including Procedure PA-212,

" Dimensional Standards for Zirconium Alloy Tubing," Revision 5, dated August 22,1994 and the UT portions of Procedure OC-301, " Final Inspection - Ultrasonic Dimensional Setup and Calibration," Revision I, dated July 6,1989, and Procedure OC 302, " Final Inspection -

Ultrasonic Flaw Setup," Revision 24, dated October 15,1997.

Final tube inspection involves UT examinations, and inspections of ID, length, end squareness I and OD surface (see Figure 4). The inspectors reviewed nondestructive examinations (NDE) of l fuel clad tubing. SMP performs UT for OD, ID, wall thickness, and flaw detection on 100% of the {

Zr4 and ZlRLO finished tubing produced for nuclear application. j UT ID/ Ends Length & End 00 Surface Pack &

Inspection Inspection Squareness Inspection Ship Squre 4 - 7.ircaloy Tube Inspection Process UltrasonicTestina (UT)

SMP performs final UT inspection to the recommendations of the 1991 Edition of the American Society for Testing and Materials (ASTM) B 353, " Wrought Zirconium and Zirconium Alloy 1 Seamless and Welded Tubes for Nucleer Service," Annex A3, " Recommended Procedure for Ultrasonic Testing of Zirconium and Zirconium Alloy Tubing for Nuclear Service," and ASTM B 811," Wrought Zirconium Alloy Seamless Tubes for Nuclear Reactor Fuel Cladding," Annex A3, l

" Procedure for Ultrasonic Flaw Testing of Zirconium Alloy Nuclear Fuel Cladding Tubec." The inspectors examined UT equipment installed at high speed UT Inspection Station 12. The

, equipment consisted of a personal computer connected to two test units, a fiber optics i l positioner, and a strip chart recorder. One test unit examines the tube for flaws using 45 degree i shear wave,10 MHZ, cylindrically focused transducers. These transducers are positioned to ,

examine the tube in two transverse directions and two longitudinal directions. The other test unit  !

8 l

l

I examines tube dimensions with 0 degree,10 MHZ transducers located on both sides of the passing tube; a third transducer measures the effects of water temperature change on the UT signals.

The inspector observed a SMP Level 11 UT examiner demonstrate setting up UT inspection Station 12 for a run of 0.360 OD ZlRLO fuel tubes. The demonstration started with verifying the calibration of " dimensional standard" 31495Z (0.360 OD ZlRLO tube). The examiner used calibrated micrometers and UT to determine wall thickness, a laser micrometer for OD measurements, and a calibrated air gage for ID measurements. After verifying the dimensional standard, the examiner checked the dimensions of a " reference standard" tube. The reference standard is used every two hours or more frequently, as needed, during the run to verify the dimensional accuracy of the UT equipment.

For the demonstration, the Level ll examiner used Inspection Station 12. The setup was verified under dynamic conditions by running the dimensional standard, reference standard, and flaw standard through the inspection station. The flaw standard tube has one transverse and one longitudinal electrical discharge machined notches located according to SMP specifications.

Following the demonstration, the inspectors observed the operators at inspection Station 4 inspecting ZlRLO . The operators inserted the reference standard and flaw standard into a lot of ZlRLO , a part of the routine dynamic calibration checks required by SMP's procedures.

The inspection station automatically culls rejected tubes from the lot and separates them by defect type. After running the lot, the rejected tubes were checked for ID moisture and retested using a slowerinspection speed.

Final Inspection After UT inspection, the inspection station operator performs a visual examination for camber, stains, and pin hole defects (pitting). Pin hole defects are usually too small for the high speed automatic UT to detect. The inspectors observed two tubes that a visual examiner had rejected for surface flaws, one for small pits and the other for spiral stains. Both tubes had successfully passed the UT examination.

i The tubes are then passed on for further visual and other examinations. The ID surface was i examined by visually examining the tube ID from both ends against a lighted background. The finish on the ends was also examined. Length and end squareness were checked, as well as straightness and ID at the ends. The OD surface was visually examined and a rnechanized OD surface examination was performed to check for surface roughness. Throughout the process,  !

the number of pieces accepted, reworked, and scrapped was recorded and the Follow Card was l signed by the operator.

Handlina. Storaae, and Shiocina The inspectors evaluated the packaging and shipping of tubing with respect to the protection of the metal surface condition during shipping. Full sheets of Styrofoam contoured to match the geometry of the tubing were used to separate the fulllength of each layer of tubing within heavy wooden boxes lined with thick brown paper. The packaging appeared effective and had not resulted in any reported shipping damage. The inspectors concluded that the 100% final UT and visual inspection process provides an adequate inspection to insure product quality and reliability.

9

3.2.6 UT Technician and Pilaer Operator Trainina The NRC inspectors reviewed SMP Procedure PA 204, " Inspection - NDE Certification,"

Revision 10, July 18,1995, which specifies the training, qualifications and certification of personnel required to perform nondestructive examination methods and techniques.

l The inspectors observed the training records of one UT Level ll examiner. UT training followed the guidelines in American Society for Nondestructive Testing SNT-TC-1 A,1988 Edition. The Level 11 examiner received his Level I training in-house under a program developed by SMP's Level lli examiner. For Level 11 training, SMP's Level lli examiner reviewed and approved a training program administered by US Air, Pittsburgh, PA. The inspectors determined that SMP training records satisfied the criterion contained in the guidelines.

The inspectors reviewed a sample of training records for several pilger operators to determine if they were trained in the various pes necessary to perform their job responsibilities. For

- example, a review of the records of one pilger operator indicated he was trained to PE-460,

" Instructions-Cold Pilger," dated June 1994; PE-469, "Pilger Control Charts," dated July 1995;

{

PE-469, Revision 051, dated September 1995; PE-469, Revision 055, dated January 1996, and PE-211," Cold Pilgering," Revision 26, dated January 1997. Based on the sample selecad, the inspector determined that the training was satisfactory.

3.3 - Entrance and Exit Meetinas l

An entrance meeting was held on February 18,1998, in which the scope of the inspection was discussed with SMP management and staff. On February 20,1998, an exit meeting was held with SMP to discuss the inspection.

4 PERSONNEL CONTACTED The following represents a partiallist of persons contacted during the inspection:

Specialty Metals Plant Joseph Ewing, Manager, Product Assurance Brian Jones, Manager, Business Process improvement

. Richard Kaiser, Manager, Production Services Jerry Leysock, Manager, Product Assurance Engineering Chris Mitchell, Manager, Human Resources Ron Weisser, Manager, Product Assurance Operations Chris Skuplen, Manager, Materials Management Mark Troxell, Audit Coordinator Carrie Monaco, Customer Projects William Jacobsen, Process Engineering 10 l

l