ML20212M770
| ML20212M770 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 10/31/1986 |
| From: | Boseman J, Hanson G GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML17055C894 | List: |
| References | |
| JJB86-74, NUDOCS 8703120179 | |
| Download: ML20212M770 (14) | |
Text
- _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - _ _ _ _ _ - _ _ - _ _
o .
.. v.
GENERAL 4 ELECTRIC KKL IN-SERVICE EXPERIENCE WITH EPG - BALL VALVES (NIAGAPA MOHAWK TASK NO. GE-1) __
OCTOBER 1986
~
PREPARED BY: @ h da _ w /d//5/86 -
g.A.805EMAN,PRINCIPALENGINEER MECHANICAL EQUIPMENT / VALVE DESIGN PREPARED BY: - /#NT/f[
G. HANSON, SENIOR ENGINEER
~
MATERIALS APPLICATION ENGINEERING
- NOTICE Neither the General Electric Company nor any of the contributors to this -
document makes any warranty or representation (express or implied) with respect to the accuracy, completeness, or usefulness of the information contained in this document, or that the use of such information may not j' infringe privately owned rights; nor do they assume any responsibility for i
liability or damage of any kind which may result from the use of any of the
! infnrmation contained in this document.
I B70312 PDH A Q $$$o to PDR P
JJB86-74
Purcose The purpose of this report is to document the in-service experience provided l to GE by KKL on EPG designed ball valves used at Leibstadt. This report is in l suoport nf the Niagara Mohawk Power Corporation task 'nrce efforts regarding the current concerns associated with the Main Steam Isolation Valves (MSIVs).
Background
On October 5, 1986, Mr. J. Boseman (Valve Design Engineering) and Mr. G.
Hanson (Materials Engineering) from General Electric Nuclear Engineerina Business Operations were dispatched to meet with KKL personnel to obtain the Leibstadt experience with EPG designed ball valves. GE and KKL oersonnel met on the 6th and 7th of October, 1986. for this purpose. The following KKL personnel were in attendance:
-(*) Mr. Ulrich Frick; Mechanical Engineering Department Head Mr. Peter Buhlmann; Mechanical Engineering Group Leader Mr. Walter JAK; Mechanical Engineer Mr. Arthur Oberle, Consultant to KKL (retired BBC engineer - previously acquainted with Leibstadt ball valve efforts) ,
(*) - Part Time The following information was provided by KKL regarding their experience (s) on this type of ball valve design.
^
?
JJB86-74 . .
1.0 LEIBSTADT BALL VALVE SYSTEM APPLICATION (Si KKL uses.large ball valves in two system apolications as noted on Figure dl, attached. A 202 inch size ball valve was originally used as a shuto "
valve in each of the two feedwater lines located in the auxiliary build-ing. A 26-inch size ball valve is used as a turbine building shutof' valve on'each of the ' cur mainsteam lines. These valves are also located in the auxiliary building.
?.0 BALL VALVE MANUFACTURER AND REFERENCE INFORMATION Mnufacturer: . Gulf & Western Energy Products Group (EPG)
Warwick, Rhode Island Typical Drawing: E20-1500-1 (Feedwater Valve)
Actuator Type: Efcomatic Assembly Series 600 (same size used on both feedwater and mainsteam shutoffvalves)
Installed Ball Valve Stem / Bonnet Orientation: _
Feedwater valve - mounted in the horizontal axis Mainsteam valve - mounted in the vertical axis 3.0 BALL VALVE REOUIREMENTS 3.1 Feedwater Shutoff Valves The feedwater shutoff valves are required to close automatically within 30 -seconds after .receiot of an RHR initiating signal. -The valves are required tn open within 60 minutes after receiving an open signal. The~ -
maximum valve leakage criteria is to be consistent with proper RHR system operation. The actual maximum criteria was not available during the meeting. The original criteria used for the production units was identi-fied as being 2cc per inch of valve size diameter.
3.2 Mainsteam Shutoff Valves .
KKL identified that the original opening, closing and leakage require-ments imposed on this valve for procurement purposes was the same as those imposed for tha inboard and outboard mainsteam isolation valves.
In actual practice 9 wever the ~ valve need only be capable of beinq
- _ manually closed ai+* mechanically er electrically within 20 minutas
! after the inboard 3M outboard MSIV's have closed. The leakage criteria applied need only be : onsistent with safe and prudent maintenance prac-tice when sarvicinq 'he turbine buildino equipment.
JJBP,6 74 7-
l .
l a.0 INSTALLATION AND OPEPA NG HISTORY 4.1 Histnev of Feedwater Shutoff Valvas 1978-1980 -
Design and manufacture of valves by EPG and subsequent storage at the-site (approximately two years of storage). _
1981-1982 Installation May-October 1982 - pressure Test of Feedwater Lines - During the hydro-test of the feedwater system it was found that these valves leaked so badly that they could not build up sufficient pressure to complete the _
test. ---
{
November-December 198? - Disassembly of Valves - The major leakage was due to the spool / seals being frozen into the spool bore due to corrosion of the carbon steel spool bore and the subsequent buildup of corrosion product. This probably resulted in low seat to ball pressure. It was believed that leakage through the packing area also contributed. KKL designed a special tool to apply enough pressure to remove the spool / W4 seals which were " frozen" in place from the corrosion. .The pitting ~ of the carbon steel in the packing area was approximately 10-20 mils deep.
There was also some scoring and removal of carbide from the surface of the ball in the same area as the scoring of the Nine Mile balls, KKL contacted Union Carbide regarding the damage to the carbide. Union Carbide stated that the Tungsten Carbide would stabilize. KKL locally ground the edge when the carbide was spalled. KKL cleaned up the corroded areas by grinding and reassembled the valves. They changed packing from Chesterton 1500 to Chesterton 1000 which was lower in chloride. The Chesterton 1500 had 16-46 ppm chloride. The subsequent January,1983, Pressure Test at 169 bars for 30 minutes was successful.
KKL did not cycle the valves under pressure. KKL initiated a study of packing materials to determine the effects on room temperature corrosion .; . ;:4 of carbon steel. They found that some packing performed better than ; .i.-l others. One of the better performing ones was Graflex 6501, a graphite .
coated asbestos. A Swiss packing (Titan) was ultimately chosen which 0.fl.
G also gave low corrosion. ';.g,p .
February 1983 - Hydraulic actuator pump motor bus was found broken. The~ @M.y f hydraulic cylinder was not properly centered resulting in seizure. The cylinder was replaced with a spare of the same size and properly aligned. M
$ {W s.y 9 NRh 103 l/Feedwater valve 52 was disassembled because of July high leakage. to November The scen seal seating surface, on the pressure ,side was EIMM found to have 5? "d41 cracks in the stellite overlay. The other seats TV.f were not cracked. Now seats were installed. $hW Xli L.
All of the small sorinos in the seats plus every #curth large spring was removed to reduce the sprina load on the ball seating surface. }Q-rf The corrosion was ground out on the spool / seals and the spool / seals were j%'
7 9 g, electroless Nickel plated to maintain the proper gap between the spool /-
seal and the spool bore (body). }"%p
(:'
JJB86 74 NM
e --
The bonnets -were modified on both valves to overccme a leakage problem and to resolve binding and scoring of the bonnet to valve body surfaces :
on disassembly. The original design required retensioning of bolts after !
the first pressurization. This was not practical so the bonnet and seal was redesigned to assure that the bolt preload was adecuate to remain sealed without _retorquing after pressurization.
Scored areas on the balls were locally blended by grinding.
On one valve actuator, the hydraulic fluid drained from the hydraulic cylinder to the reservoir due to elevation of the reservoir being lower than the cylinder.
- This resulted in a shock loading of the valve and actuator structure and probably caused the bracket fracture previously observed. KKL relocated the fluid reservoir to prevent the draining potential.
The roller bearings were found to have more clearance than the packing lantern rings (0.5m for bearings versus 0.3 for lantern rings). KKL increased the clearance in the lantern ring to assure that the bearing would take the load. -
November 1983 - Modifications Complete, Testing Resumed - Problems were '
encountered with the latch mechanism. The problem was scoring and KKL modified all solenoids by removing the
'ing(of
/e Sealthe Ring trip solenoid.
No. 28) which caused the binding.
February 1984 -
Problems with FW52 - Trip solenoid defective, needle valve damaged, filter replaced, pump coupling rubbing on sides. Mis-alignment of hydraulic cylinder bearings caused seizure of the cylinder-and would not open the valve.
~
May 1984 - Plant Startuo - Could not trip or close FW51 valve due to jamed solenoid. Could not open FW52 because of pump seals. KKL re-placed the seals and freed the solenoid.
June 1984 - Plant Shutdown - Could not trip FW51. Latch mechanism had to be readjusted again.
uns e
I JJB86-74
- o -
. s e
- October 1984 - External leakace through the packing. Pesolved by tight-ening the packing.
Summer 1985 =- Feedwater valves were removed and replaced with check valve plus a motor. operated gate valve in each line.
4.2 History of-Main' Steam Shutoff Valves !
July 1983 --All inner spool / seals springs were removed. The modified designed bonnets were installed. Weld overlayed the bonnet sealing area with 13Cr 4Ni weld deposit. Restellited spool / seal seating surface (s).
Steam valves were exposed to steam and temperature from an auxiliary boiler. -
Spool / seal bores were cleaned and ground to remove pitting. Spool / seals were Kanigen Nickel plated to provide for proper clearance (packing).
November 1983 - MSIV 22 - Wouldn't close during functional test due to failure of the solenoid trip.
Pressure test at 1.1 bars. All valves leaked greater than 50 liters per minute.. They couldn't tell how much higher.- MSIV 22 was dismantled and lapped. This reduced the leakage on this one valve to 30. liters per hour (0.9 cu ft per hour).
~
1985 Overhaul and Test - New manual mechanical trip levers were supplied by BBC and _ installed for ease in tripping the valves. If the trip solenoid does not operate electrically when actuated, then the manual mechanical trip lever assures tripping. _
MS21 hatch roller bearing was found to have a crack. -
Valves were tested for closing with no pressure with the following results:
MS21 - Closed within 2 days (not sure when)
MS22 - Closed after. 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, would not close on hard trip (prior to insalling the new mechanical manual trip).
MS23 - Closed properly (less than two minutesi MS24 - Closed only after manually tripping via the solenoid.
~
August 1986 - Valves were actuated with no pressure and only MS22 failed to close. The trio solenoid was frozen in place due to the lubricant which had hardened.
Valve was disassemblad and examined.
Some scoring and carbide was missing on the ball in high pressure surface areas. Also some upstream edge corrosion /spalling. Ball and seat relapped and valve reassembled with new packing.
JJB86-74 _ _ _ _ _ - _ _ - _ _ _ _ - _ _ _ _ _ _ _
5.0 MAINSTEAM SHUT 0FF VALVE MODIFICATIONS.
The KXL mainsteam ball valve design was modified as described belnw:
5.1 All of the - small inner springs were remoud from the nested springs located in the spool / seal. assembly. Removal of these springs was recom-mended by the manufacturer to minimize scoring and thereby the leakage potential by reducing the imposed spring -load from 600 lbs to approxi-mately 400 lbs or approximately 40%.
5.2 The ball is coated 'with Nickel NEVER-SEEZ after maintenance.and relapping in order to minimize friction and thus- the potential for ball scoring.
The vendor recommended this practice to enhance operability.
5.3 The bonnet pressure seal was redesigned / modified to minimize the poten-tial for scoring the bonnet sealing surface during bonnet. removal. This modification was recommended by BBC.
5.4 The body and bonnet guide surfaces were hardfaced with a 13 Cr 4Ni alloy to minimize corrosion between those surfaces and thus the scoring poten-tial of the sealing surfaces for use with the pressure seal. This modification was a BBC recommendation. -
5.5 The originally supplied Chesterton Packing 1500 was replaced with a 3/8 inch square Titan (Swiss supplier) packing set. This modification was -
incorporated due to the concern for corrosion of the bore surfaces and the quality control of the original type packing for chlorides, flourides and contaminants which could act as a catalyst for corrosion.
5.6 The spring retainer bore diameter was -increased in order to assure that the spring (s) woulc not' bind-up shen compressed.
^
5.7 The lantern ring diamete.r was changed to assure that the roller bearing would be the effective load bearing member and to minimize eccentric rotation.
5.8 All trip solenoids were modified by removing the sleeve (PC #28) which
.had been the cause for malfunction on the feedwater-valve. _
_ 5.9 The manual operator (actuator) trip lever was modified to permit manual mechanical tripping from the above flooring. In addition, the modified trip lever design incorporated a greater mechanical advantage to assure mechanical tripping of the actuator. .
5.10 In addition to the above, KXL uses a specially (B'E) desigred ore load fixture for boltinc up the body to bonnet closure. This fixture was designed to minimize the potential for leakage through that joint and to minimize axial mosf erent of the ball due to stud relaxation after the valve is pressurized.
5.11 Attachment 1 identifies the originally provided materials list for the various spare parts items which can be used to compare to the NMP-2 MSIV ball valves.
JJ886-74 - _ _ _ _ - . - _ -, . . _ . - _ . - - '...-.-__ __ -__--- . _ _ _ - __ _ _ _ _
a.
- ?. -
5.12 KKL noted that after 'inding the radial cracks on the feedwater valve the - j spool / seal'on all. of the valves were re-hardfaced with stellite to assure a quality spool / seal seat.
60 LEIBSTADT EPG-BALL VALVEfS) OPERATING STATUS 6.1 KKL' identified that. the two feedwater stop valves were removed from the system and replaced with a third swing check valve (inside the auxiliary building) and a motor-operated gate valve in the turbine building. The i decision to remove these valves was based primarily upon the inability of the valve to close reliably within the 30 second requirement for compati-bility with operation of the RHR system.
6.2 With regard to the main steam shutoff ball valves, KKL identified that the modified valves are still installed in the system. Since the. valves are not required to perform a safety-related function, there - are no imediate plans to replace those steam valves.
7.0- OTHER ITEMS DISCUSSED /NOTED 7.1 KKL identified that numerous problems were experienced with the EFCOMATIC Series 600 actuator. The problem areas experience are highlighted on the attached figures 2 and 3. The types of problems ranged from leaking seals, broken bracket, cracked roller bearing to misalignment of parts.
Due to the short time schedule for the visit, specific details were not discussed for each problem area. Essentially, KKL replaced or repaired parts or re-aligned features depending upon the cause. Modification efforts were limited to beefing up the bracket and changing the mechan-ical manual trip lever.
7.2n KKL indicated that they have not specifically noted chattering of ' the
~ ~
bal1 valve (s) per. se.
7.3 With regard to scoring of the thrust washers, KKL did not observe .any evidence of contact of the thrust washer on the feedwater shutoff valves but did note some light scoring on the main steam shutoff valves.
7.4 KKL identified that should Niagara fohawk Power Corporation have need for
- ~ an engineering test valve, that the removed feedwater valves are avail-able for their use. KKL also identified that Sultzer Y-pattern MSIV's
~ are available from the Swentendorf cancelled power plant should Niagara K Mohawk be interestad. If interested, KKL suggested contacting the U.S.
L based KSB agent, Mr. Bradford H. Robinson, who is located at the (KS8 Inc.) Commerce Cew,175 Commerce Drive, Hauppauge, N.Y. d11788 (Tel:
516 31-0303, Exto e on 06).
7.5 Since Dr. Varga frer "C was on vacation and not available, Mr. A. Oberla (retired) forma?ly ' rem the BBC Enoineering Deoartment and a corsultant tn KKL, was askeri to raet with us since he was knowledgeable with some of the history and problem areas concerning this ball type valve design.
During the discussinn, it was noted that some informal seat load calcu-
!ations (not available at the meeting) had been per#ormed and that high seating stresses were noted when a static differential cressure was applied across the ball to seat contacting surfaces. Recollection by Mr.
JJB86-74 _____ _ _ _ - _ _ _ _ _ _ - - _ _ _ _ _
f Oberle, indicated that . :he highest stress condition occurred 4ust prine to valve closure assuming concentric valve rotation. High contact stress could also be expected when closing the valve should eccentric ball rotation occur which:was evidenced by the ball scoring loca'tfons..
7.6 During the meetjng it was noted that alignment and tolerance stack-ups of-the ball to soool/ seal is considered important to improve valve operation / performance.
7.7 Both of the balls from the feedwater valves were visually examined. Both-of the_ balls were covered with a high temperature red oxide on the-~inside
. diameter of the ball.. One of them was.also coated with red oxide on-the outside surface of the ball indicating that some water had been present on the _outside surface of this ball. There-were heavy burnish marks on the Tungsten Carbide starting from the bore and'eminating 8 or 10 inches toward the seat. These burnish marks were present in a band' approximate-ly 8 inches wide. There were only small amounts of carbide removed )less than t inch diameter spots) inthe same high stress area as the Nine Mile Point valves.
Visual examination of the photographs of the balls from the steam valves showed similar wear marks and spalling on the ~ carbide surfaces except that . the wear pattern was not as wide as the wear pattern on the feedwater valves. Another difference noted was the steam valve had what
- appeared to be corrosion of the carbide coating at the leading edge of the hole in.the ball on the upstream side of the valve (pressure side).
All of the Leibstadt balls (both steam and feedwater valves) showed much less spalling of the carbide than the Nine Mile balls examined by GE.
General Electric, on behal' of. Niagara Mohawk Corporation, thanked the KKL personnel and Mr. Oberle for their time and cooperation in assisting us, on such a short meeting notice, with their~ experiences and information.
1 JJB86 74 o -
, , .
- A. d
. s r
'~
!s i f!5
- e 8
og [ 5 l' ! :
g ol E
I
- I
~3 l0
- I ! 2g 2
- e $ $ "
G. 3 '
g5 - 3 -- 1 ,l. -
Ia 8 "
33 u 3
,E ' '-
I
~
r
'l l* '
l l $3 i ,,
t - -
r g dk
% W 1 l-gxwwwuuun m
NJ tuwwuuuu l ' i f.
um __
sy1 -
e_ ~ -
NE W . I
]
d I
4.
> . e >
g i i - t I
. t<
- l<
I !
t !f yC l 37 Et g M .s .-
1
~
h'\
1,-
s l
~
i
- ,> ,i i
) >
I
! s
. Iu KKL xsauxamwinx tasuaru QG and SC Classification Aeolicable to Power Conver-sion System Components in S'.iR/ 6 Plants SAR FO. mi
)
. s ., '
l
~
scosst nata 2NEA 1,cADTNG RZ288 szar, manzxta-
~ ##C '3 9 e . . . . l JW= 3 / 3 -. 5 3 7,.; l passsazz sus.- i
[/M /% .C.s=' , .
- -W -L....,.,
z u ; " w..,sN.=.3-azaaInorun / @- . .u
- A== .
3C1057 4-44.e..
- 3" 4 ROILEX BEAAING =
anAImaasunn .
i e
RAIL
- latye.l TERUST ICLSMEX INSTAILED CN
~
ECTI SIDES EURING
{fM., m,,
~
- L % .9.a? '
5700L/$ EAT gg SPRI'E. AZ*AINEA
SPRIMC Cov1A 3EAAIEE s
>Cory( , p4 E.j. .$-r-.
.- e--.....,
,.a
- W = .
n- W*
ed l t
\
! k6ad // e,a'. any adb;44. y'.,)..I ( .,,, . ;
VALVE AS$t>gLT 50'*g %..,,.s,..a E4 D'A e kra.sgs
- e, f f '
(**)
EXPLocco vitu RING
- #*f $ > * * $" ' ' *i ** * *J FIGURE 2 d
- S CARI'!C PLATE l
' ~ - -.. _
.o; -
s s
CLEVIS, DRIVE SHAFT LATCH ARM (BLOCKING LEVER ARM)
,I -
' l -
SOLENOID TOWER
.I O c-l s ) '
l' SOLENOIl l N g .
PLuncza:
1
%. v 1 .
l .
I k -
s .
s
\ .Ph 4
ps MANUAL
,,, TRIP LEN s
' y.*i
~
t1 el l
5 8F I429 - a PIVOT PLATE
~* $9 N N(9 Ae<Y LATCH ARM AND SOLENOID TOWER ,
FIGURE >4.b ,
~
. s .
SWITCH DECg BEARING L N ASSEMBLY ASEENDLY
\
/ , ,' ,". ,' Q
,,, ^ UPFSA TRACK
/ , ' ,
/ . . -
/ .
I g' SPRING SYSTEM ASSEMBLY l COVER CLEVIS DRIV @
SHAFT
,N em N(-
Lowra TRACK
- g. ,
M[
1 i l c o*
1
/ %> cs l .
- t N
W
@ T
.l _
I
/
,. ' / / BLOCKING LEVE
' g f'. . .
- .gdg!.f! .. J' 'N \
-@.$@',j)'I4 i SOLENOID TOWER l
'..g._ . 1 --r
- W- .$L)
L,,, ; } i e
;1l-f.*,.
kIS, ' ' [r?. mn,:- - ?s N x s ,a
, s + MOTOR-s pangg h33 !..
y k/)' - t
-1..
OPERATE:
- PUMP e 1 HYDRAULIC
,g.#
- A PIPING SYSTEM \
! SERIES 600 ACTUATOR \ RESERVOIR EXPLODED VIEW ,. fj , TOWER g, FIGURE h 4 HYDRAUL:C
-6 fa 6 C3 )
.9 -
t 6 s l l
~
i
' t .
- e
( s
=.
~ -. l >
* - 8 f %"
., D \
- ar,%)
i
~
.e rk, N
a, j.
,,6,
- e I
I n
/.s' / F l ['
#~
di$-h Am .Z' s
. \s r ;
j
~- m , >fQ ,
- _ ._+__- . l g 3 1i $Yrkl i A V :
( n
~ n SERIES 600 ACTUATCR f O FIGURE 3=e,3
~
_s. .
)
y- - - . - - _ .
- e. .
G3 w i- '. h
.,9 -
e n G
' t .
-v g
k. 8 s' * , p u l a
- s' I
* - t e1
^
P.[ W
~ .1 l ,
s .
) >R
,!f
) ~
I' ff l
' ~
g s-..., d ~-, s rs p
' ' g- 4 .
a s p t J
)
. ~
~ , >f L w
..~..
o 7'
/
\: n .
"" f
** ll b( -
V I i
\ N f e SERIES 600 ACTUATOR FZGURE M ,f 6
-w. . - , - , - , , . ,w-,.mw.-- -
-,-,.,4,--... ,,-y,,,_,,,e-.m-,,
-,,w-m-,-,,,._.,_. . - - - ---,,,,,,,,..,,v__-.---_,,,-.--,-.-----.,-,,-,m_,mm.-.,
-}}