ML20031C930
| ML20031C930 | |
| Person / Time | |
|---|---|
| Issue date: | 05/27/1981 |
| From: | Barnes I, Ellershaw L NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20031C889 | List: |
| References | |
| REF-QA-99900746 NUDOCS 8110090138 | |
| Download: ML20031C930 (9) | |
Text
,
U. S. NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT REGION IV Report No.
99900746/81-01 Program No. 51300 Company:
Weir Pumps Limited Cathcart Works Newlands Road Glasgow, Scotland G44 4EX f
Inspection Conducted:
Mar F
- April 3, 1981 Inspectcc:
_L']t. Ellershaw, Contractor Inspector, Date Q
f-4 7-O Reactive Inspection Section Vendor Inspection Branch c
Approved by:
he J - M -&' l G arnes, Chief Date Reactive Inspectica Section Vendor Inspection Branch l
Summary Inspection conducted March 31 - April 3, 1981 (99900746/81-01) f Areas Irsnected:
Implementation of 10 CFR Part 50 Appendix B Criteria, and applicaole codes and standards, including:
a construction deficiency report; welding material control; welding procedure specifications; weloing heat treatment and control of nonconformances.
hours on site by one NRC inspector.
The inspection involved 28 inspector-Results:
In the five areas inspected, one nonconformance was identified.'
There i
were no unresolved _ items.
3 i
i Nonconformances:
Control of Nonconformances - certain nonconformi'ng parts were 4
not tagged ant could not be identified or located (Notice of Nonconfor,mance).
t 4
?% _
+ >
3 t
b m*
t
/
8110090138 810702
~
PDR GA999 EEC*****
e
-99900746 PDR r+-,..
v---e--+---
- - - - - - - * " - - - - ^ ' ' ' " ' ^ ~ ~ - - ~ - ~ " ' "
2 DETAIL SECTION (Prepared by L. E. Ellershaw)
A.
' Persons Contacted W. Anderson, Welding Foreman W. Buchanan, Manager, Production Commercial Nuclear Products R. Carrick, Manager, Quality Control D. Downie, Contracts Administrator A. B. Duncan, Chief Engineer S. Gair, Welding Engineer W. Gracia, Engineer in Charge, Nuclear Pumps N. Lee, Sr. QA Engineer - Commerical Nuclear Control I. H. Lundie, Manager, Quality Assurance T. Nellis, QLality Control, Welding I. Strain, Qeclity Engineer - Commercial Nuclear control T. Weir, Chief Inspector B.
General Information Weir Pumps Limited (WPL), part of the Weir Group, are holders of ASME Certificates N-1678 (N) for Class 2 and 3 vessels and pumps, and N-1679 (NPT) for Class 2 and 3 vessel and pump parts and appurtenances, piping subassemblies and component supports.
The Certificates of Authorization apply to the Cathcart facility only, and expire on March 11, 1983.
Approximately one percent of their work is devoted to commercial nuclear applications destined for the United States.
They currently are Working
~
on two U. S. contracts:
four boron injection pumps for the South Texas Project, and eight auxiliary feedwater pumps for TVA's Yellow Creek 1 and 2 sites.
WPL has previously supplied 32 essential raw cooling water pumps to TVA's Bellefonte, Hartsville, and Phipps Bend sites.
WPL's Cathcart fa'cilities contain 664,404 square feet floor area and 35,573 square feet o; yard area.
There are approximately 2549 emplcyees at the Cathcart works.
The current Authorized Inspection Agency (AIA) is Royal Globe.
At the time the TVA Bellefonte Unit 1 pumps were manufac-tured, Kemper Insurance Company was the AIA.
C.
Constructicii Deficiency Report (CDR)
===1.
Background===
TVA initiated a CDR to the NRC on August 19, 1980, pertaining to excessive vibration during testing of the first two essential raw cooling water (ERCW) pumps at Bellefonte Nuclear Plant Unit 1.
Subsequent interim reports were generated by TVA on September 17 and December 2, 1980, with a final report issued on February 4, 1981.
3 TVA and WPL have evaluated this condition and agreed that the vibrations were the result of resonant amplification caused by the operating speed being too close to the natural frequency of the pumps.
Although not conclusively demonstrated, TVA has ascribed this problem to be related to the pump design.
WPL has determined that the problem can be attributed to modification of the pump natural frequency resulting from inadequate foundation stiffness in the area where the pumps are installed.
2.
Objectives The objectives of this area of the inspection were to determine any generic implications and to verify that WPL had taken the necessary steps to assess, correct, and preclude recurrence of the problem.
3.
Method of Accomplishment The preceding objectives were accomplished by:
a.
Review of Contract 11514/001-008 documentation including:
(1) Design Specification BNP-OS.1925-2913-02.
(2) TVA Purchase Order No. 77K35-820122 dated September 27, 1976.
b.
Review of Seismic Stress Analysis Report No. ME-361, " Seismic-Stress Analysis of ASME Section III Class 3 Pumps," as performed by Mcdonald Engineering Analysis Company, Birmingham, Alabama, and dated December 3, 1976.
c.
Review of the Design Review Report No. DRR/NED/18 dated March 1, 1977, performed by WPL during which the calculaticns of ME-361 were verified.
d.
Review of " Hydraulic Performance Test Procedures" Revision 2, dated July 14, 1978.
e.
Review of WPL ERCW Pumps Pe.-formance Test results.
f.
Review of correspondence between TVA and WPL after the identifica-tion of the frequency problem.
g.
Discussions with cognizant personnel.
i l
1
+
9 t
WPU/RIS RIS VIB IES DD-RIV LEllershaw/dsm IBarnes UPotapovs JGagliardo
.JCollins 5/26/82 1
5/ /81 5/ /81 5/ /81 5/ /81 l
4.
Findings a.
Within this area of the inspection, no nonconformances or unresolved items were identified.
b.
Four pumps; 1Al-A. lA2-A,181-8, and 182-8 were shipped from WPL to TVA (Bellefonts 'Joit 1) between September and December 1978.
The punt were s"lisequently installed and testina was performed.
The recorded maximum vibration amplitudes in miis, peak to peak, were as follows:
Pump 1Al-1, 25.5; Pump 1A2-1, 6.6; Pump 181-8, 0.7; and Pump 182-B, 1.0.
The TVA specification allows a maximum of 4.5 mils, peak to peak.
Both IAl-A and 1A2-A exceed specifi-cation limitations.
The excessive vibration has been determined to be the result of the occurrence of the pump natural frequency close to the pump running speed.
The running speed of the pumps is 1185 rpm.
Natural frequency measurements showed these to be as follows:
1Al-A, 1145-1181 Hz; and 182-B, 1271-1293 Hz.
The Mcdonald analysis predicted that the natural frequency of the pumps would be 1440 Hz, which was supposed to be a value conservatively below that actually expected.
All physical characteristics were checked and reviewed, including the torque of bolting materialm Different combinations of motors and stools were used on 1Al-A with no appreciable difference in results.
Vibrations were measured at different component locations, with two apparent results:
a.
A significant rigid body rotation of the headgear took place at the base, indicating potentially inadequate floor stiffness.
b.
Very little flexure of the pump headgear and motor, indicating that,the headgear natural frequency is considerably higher than the measured natural frequency.
TheNRCinspector'sreviewshowedthattheMcDonalddesigkanalysis was based on Computer Code ICES - STRUDL, a program developed by the Department of Civil Engineering at the Massachusetts Institute of..
Technology.
This program computes natural frequency as being a function of mass, pump head stiffness, motor reed frequency, and foundation stiffness.
WPL's evaluation after identification of the problem included theo^retical calculations, varying both motor reed frequency and pump headgear stiffness.
The results showed that the variance in these parameters would have to be far,a excess of the motor manufacturer's stated motor reed frequency values, or the Mcdonald headgear stiffness values to produce the observed conditions.
As a result of the above, an evaluation was performed of foundation stiffness, the only remaining variable in the ICES-STRUDL program.
j s
r 5
During the evaluation, it was noted that concrete in the area where the four pumps were installed had been poured 2" too high.
The concrete was chipped out to a depth of 4", a coating of epoxy applied and 2" of concrete repoured.
Review of correspondence between WPL and TVA showed that TVA had noted that the anchor bolts had not been removed during the repair operations.
WPL had questioned the adequacy of the anchor bolt concrete bond, as a result of floor vibration measurements taken around the 1Al-A pump foundation ring and knowledge of the repair operations performed in this location.
It should be noted that TVA did not consider the WPL evaluation as being sufficiently conciusive to demonstrate that foundation stiffness was the primary cause of the vibration problem.
Several possibilitias were explored in an attempt to correct the excessive vibration condition in pumps 1Al-A and 1A2-A.
The most practical solution, which has been successfully demonstrated at the l
site, was the addition of mass at the top of the motor.
By increasing l
mass, the natural frequency is shifted away from the running speed.
l l
The addition of 1500 pounds to the motor on 1Al-A reduced vibration from 25.5 mils peak to peak, to 1.1 mils peak to peak.
The addition l
of 1700 pounds to the motor on 1A2-A reduced vibration from 6.6 mils peak to peal, to 3.95 mils peak to peak.
The permanent addition of weight to these motors would require seismic requalification.
WPL stated that the seismic qualification of the motor had been reviesced and would not be invalidated bet that a re-analysis of the pump head top flange by finite element analysis would j
be required to show that stresses are within the allowable.
i TVA has concurred that this is an acceptable solution but a final determination would not be made until August 1981.
4 o
e E
i.
3 h
S l
A s
,__y m
,,._..._.e_..m.,-.-_.,c._r
6 D.
We! ding Material Control 1.
Objectives The objectives of this area of the inspection were to verify that WPL had implemented the requirements for the contro! of welding material in accordance with the QA Manual and applicable NRC and ASME Code requirements.
2.
Method of Accomplishment The preceding objectives were accomplished by:
a.
Review of QA Manual Section 14.0, "Yelding," revisica 4.
b.
Review of Welding Material Authorization Index, revision 67 dated December 11, 1980.
c.
Observation of welding material storage.
d.
Review of Welding Consumables Issue Forms.
Review of certified material test reports pertaining to welding e.
materials observed in storage and,;elding materials that had been issued for use in the nuclear welding area.
f.
Discussion wilh cognizant personnel.
3.
Findings a.
Nonconformances
^
None 7
b.
Unresolved Items Nace E.
Welding Procedure Specifications (WPS) 1.
Objectives 3
The objectives of this area of the inspection were to verify that WPL had implemented the requirements for qualification, issvar,ce, and control of WPSs in accordanco with the QA Manual and applicable NRC e
and ASME Code requirements.
j A
s M
^
y a
7 2.
Method of Accomplishment The preceding objectives were accomplished by:
a.
Review of QA Manual Section 14, " Welding," revision 4.
b.
Review of the following SMAW WPSs and associated procedure qualification records (PQR) denoted in parenthesis:
100/002:
(009); 100/005, (002,009); 100/004, (009, 013, 018); 100/292, (001); 100/200, (001); and 100/207 (001).
c.
Review of GTAW WPS 101/204 and its associated PQR No. 006.
d.
Discussion with cognizant personnel.
3.
Findings a.
Nonconformances o
None b.
Unresolved Items None 1
F.
Weldina Heat Treatment 4
1.
Objectives The objectives of this area of the inspection were to verify that WPL had implemented the requirements for the control of welding heat treatment in accordance with the QA Manual and applicable NRC and ASME Code requirements.
6 l
2.
Method of Accomplishment j
The preceding objectives were accomplished by:
Review of Section 14, Revision 4 of the QA Manual, "'helding.',
d a.
b.
Review of Section 16, Revision 2 of the QA Manual, " Heat Treating, Hydrostatic Testing and other Processes."
c.
Review of welding 'and postweld heat treatment records, inciuding heat treat charts, Tor two TVA essential raw cooling water pumps j
(Serial Nos. 3 ana 4).
e g
f N
.. - _ - - _. _ -, _ - - -, _~ _
}C, 8
d.
Review of certified material test reports associated with the identified weld material used in the two pumps.
e.
Discussions with cognizant personnel.
3.
Findings a.
Nonconformances None b.
Unresolved Items None G.
Control of Nonconformances 1.
Objectives The objectives of this area of the inspection were to verify that WPL had implemented the requirements for identification and control of nonconformances in accordance with the QA Manual and applicable NRC and ASME Code requirements.
2.
Method of Accomplishment The preceding objectives were accomplished by:
a.
Review of Section 19, Revision 9 of the QA Manual, "Nontanformities."
b.
Review of nonconformance report (NRC) registers, which are initiated and maintained for each job.
c.
Review of open and closed NCRs, and the Nuclear Route Cards associated with the selected NCRs.
d.
Observation of designated areas used for segregation of nonconform-ing materials / parts.
e.
Observation of nonconforming parts to assure that they are identified as being nonconforming.
f.
Discussions with cognizant personnel.
I
E e
-9 3.
Findings a.
Nonconformances See Notice of Nor :anformance.
Twelve diffusers. Serial Nos.
1-12, for Job No. 11732-005/8 were identified as being noncon-forming.
A review of the Hold area in which they were identified as being held, revealed certain other parts including diffusers which were not tagged.
Because they were not tagged, identifica-tion of the 12 specific nonconforming diffusers could not be made.
b.
Unresolved Items None H.
Exit Meeting A meeting was held at the conclusion of this inspection on April 3, 1981, with the following management representatives:
R. Carrick, Manager Quality Control A. B. Duncan, Chief Engineer R. Garrick, Managing Director M. B. Leiper, Executive Director Production T. O. Leith, Director, Contracts I. H. Lundie, Manager, Quality Assurance M. L. Ryall, Director, Technical Division P. T. Syme, Executive Director, Manufacturing The scope and findings of this inspection were summarized.
Management acknowledge the statements relative to the findings.
i