ML17249A493
| ML17249A493 | |
| Person / Time | |
|---|---|
| Site: | Ginna  |
| Issue date: | 10/31/1979 |
| From: | ENERGY ENGINEERING GROUP |
| To: | NRC |
| Shared Package | |
| ML17249A491 | List: |
| References | |
| CON-FIN-A-6257 NUDOCS 8001220329 | |
| Download: ML17249A493 (15) | |
Text
STEAM GENERATOR WATER HAMMER TECHNICAL EYALUATION'INNA POWER STATION OCTOBER 1979 EGAG IDAHO, INC.
~ y CONTENTS I..'NTRODUCTION.
II.
FEEDWATER SYSTaaf.
1.
DESCRIPTION 2.
GEHERAL OPERATION III.
MEANS TO REDUCE THE POTENTIAL FOR WATER HAMMER.
1.
DESCRIPTION 2.
EFFECTIVENESS DURING TRANSIENTS AND CONDITIONS CONDUCIVE TO WATER HAMMER.
2.1 Reactor Trio 2.2 Loss of Main Fe dwater Flow.
2.3 Loss of Offsi'e Power.
2.4 Operator Error 2.5 Steam Line Br ak 2.6 Loss-of-Coolant Accident 1
2 2
3 5
~
e
~
~
5 7
7 7
8 8
9 10 IV.
CONCLUSIONS AHD RECOK".ENDA IQNS V.
REFERENCES'
~
~
~
~
~
ll 12
tp I.
INTRODUCTION An evaluation was performed for the Ginna feedwate.
systan.
Tne purpose of this evaluation was to assess the effectiveness of the existing means to reduce the potential for,steam generator water hammer in he feedwater sys an during norma'1 and hypothetical operating conditions.
The steam-water slugging 1
in the feedwater systems (specifically, the steam generator feedwater sparger rings and adjacent feedwater piping) was considered in this review.
No known water hamner event directly at ributable to feedwa-er ring draining has occurred at Ginna.
Two water banner transients+caused by feedwat r control valve malfunction did happen, however.
Tne potenrial for s~eam generator water hamner is avoided if the fe d-water system is maintained full of water.
Hence, this vaiuation was based on the effectiveness of he means utilized at Ginna to main-.in the fe dwater system full of wat r during normal and hypothetical operating conditions.
The information for this evaluation was obtained from:
- 1) discussions with the licensee,
- 2) licensee submittals of Augus-1, 1973
, July 17, C
1975~ j, October 31, 1975~ ~, January 30, 1976E
~, and June 15, 1978" ",
- 3) the "Ginna Nuclear Plant Final Safety Analysis Report"~ ~, 4)
"An evalu-ation of PMR Steam Generator Mater Hamner," NUREG-0291~,
and 5) 'Aestina-house Technical Bulletin, NSD-TB-75-7 [B1 A description of "he feedwater system at Ginna and
-;-.s gene. al operation
.is presented in Sect;on II.
The means to reduce the potential for s-.earn gen-equator water hamme.
are presented in Section III, including a discussion of their effec.iveness during operating conditions conducive o wat r hanmer.
Finally, conclusions and reconmendations are presented in Sec-.ion IV concern-ing the adequacy of the existing means to reduce the pot ntial for steam generator water hammer at this facili.y.
0 II.
F="-DWAT."R SYSTcB 1.
DESCRIPTIGN The feedwater system, or Ginna was designed to provide an adequate supply of feedwater to the secondary side of the two steam gene. ators during all operational conditions.
Feedwater is suppl'ied to the main feedwater pumps by the heater drain pumps and= by the condensate pumps via the low pressure heaters.
Feedwater from the main feedwater pumps is supplied to a main header via the high pressure heaters.
The main header splits in.o two 14-inch feedwater lines to supply a 10.75-inch diameter, hal,-inch wall 'thickness e dring inside each steam generator.
Fe dwater is discharged downward through inver ted "J" shaped tubes uniformly distributed on top of.e ch. =e dring.
The two main feedwater pumps are single stage, double flow, centrifugal
- type, each rated for a flow rate of 14,000 apm a
853 psig.
The two pumps, ach dr iven by a 5000 hp electric motor, share comnon suc-ion and discharge headers.
Tne oump motors are normally supplied;w'th power via the main auxiliary transformer.
In the event of a
.'rhine.: rio, of site power supolies the pump motors via the reserve auxiliary transformer.
The auxiliary fe dwater system (APrlS) provides feedwater to the steam generators for residual he t removal during reactor startup and shutdown, low power operation, and in the event of loss o-main feedwater flow.
The Ar:<S consis s of a main (M)ASS and a
s-.andby (SB)ASS.
The (N)ASS consists of 2 motor-driven
- pumps, each 200 gpm capacity nd turbine-driven, 400 gpm caoacity.
2 ch motor -dr.'ven pumo c
n supply ei=ner one or both s.
am gene.ators.
The turbine.-iriven pumo normally supplies both s-am generators.
The primary source of wa-.e.
is two 30,000 gallon condensa.e storage tanks which are cross-connec.ed through locked-open manually oper ed valves, while the backup supply is available from the servic water system.
Tne turbine-driven pump is supplied with s earn
=rom the main header s.
The (SB)
AFWS consists of 2 motor-driven
- pumos, each 200 gpm capacity.
This system, recently installed in a separate plant area from the (H)AFllS,was placed in service during August 1979, to provide independent AFHS capability following a steam or feedwater line break in 'the immediate vicinity of the (N)AFlJS pumps.
The primary water source is the service water system.
~
0 The two APrlS are each powered rom 2 redundant and independent AC emer-gency buses from plant emergency diesel gene.ators.
Tney are interlocked so that both are not simul.aneously loaded onto the vi=al AC buses.
The (M)ASS supplies feedwater to each steam gene. ator via 2 thre -inch lines connected to each main feedwater line just outside the containmen-building.
The (SB)ASS connections to each main feedwazer line are placed within the containment building via one line per, steam generator.
They ar capable of being cross-connected by the operation of, manual valves.
Z.
Gci<ERAL OP ERAT;OH, Ouring nodal power operation of the reactor, the main feedwater system supplies feedwater to the secondary side'oi the steam generators for heat removal from the reac.or coolant system.
The feedwater flow is regulated by individual regulating valves in the main feedwater lines to each steam generator.
Tne positions of the valves are automatically controlled based upon steam generator level, steam flow, and feedwater flow.
During plant shutdown, star.up, and for feedwat requirements up to about 3" of full reactor power, fe dwater is normally supplied by the main auxiliary feedwat r system.
Fe dwater low i:s manually regulated to main-tain adequate water levels in the stean generators.
As power is increased and sufficient high pr ssu.
e s
earn is available, a main feedwat pump is started and he auxiliary fe ~water system is shu" down.
For feedwater requirements of about 3
to I:" of full power, fe dwat r is manually controlled and supplied via low flow bypass lines wnich bypass the main feedwater regulating valve in eacn main fe dwa-r line.
The bypass regulating valve in each bypass line allows more accura;e and responsive feedwater flow control than would be oossible wi.h.he larger main regulat ng valves during low power (and Iow fe dwater flow) ooer.:;on.
Above feedwater requirements 0
about 15+ of uII reac or power, feedwater control is shifted to the main regula-..'ng valves.
As power is increased to 'i0-o0~ of full power, the second ma n feedwater pump is sorted and feedwaier flow is placed under au=".at c control.
3
After the loss of main feedwater flow to one or bot.", steam c=ne. a.crs, automatic ini.tiation of the main auxiliary ',o dwat r flow will re ul-u.""n receipt of one or more auxiliary feedwa:ar pumo startuo sicnal<s.
he.,=or driven auxiliary feedwater pumps start on:
I) the coinc;d nc of im ou= of three steam generator low-Iow water level (15" of narrow range, or 1.3 inches above the bottcm sur,ace of the feedring) signals from ei.her ste=-~ gene. ator,
- 2) the tripping of both main feedwater
- pumps, or 3) a saf y inje"tion signal (SIS).
The turbine driven auxiliary
~.'water pump starw on:
- 1) the coinc'.-
dence of two out o< three steam generator low-low water '.evel sic."a.ls fr m
both steam generators or 2) the coincidence of a turbine-gene.
tor t-..o nd
<-ss of offsite power.
one auxiliary feedwatar is subsequent:y manually ccn:".oII~ =o maintain proper water levels in all ste~",, gene.ators.
i.e motor "riven and turbine driven auxiliary feedwater pumos can also be started manually (local or remote).
The (SB)ALAS is manually initiated.
plant design speci ications allow f".r a maximum delay of one minute frt.
receipt of any auxiliary feedwater pumo star=up signals:o deI',ve~ of,,~in auxil iary feedwater to the steam generators.
A limit of ten minu~s is allcwed to get the (S8)AR<S on stream.
Operating procedures to administr tively limit auxiliary f~
uater flow during r covery of the s
earn generator faedrings frcm nor<al and abnormal transients have been implemented at Ginna.
in hese s'it aeons,
=he aux Iia-.I reedwatar flow rate o either steam generatcr is o
be m nual'.y I-:mi=w =o a
max'imum of 150 gpm.
Tnis Iimita-.ion i s:o apcly unenever s-earn cene. a-"r level is below the low-low level set poi,",.t, 15 of narrow rance, anc un:-.'I tne level is recovered to 25" (I inch bove the coo of
"..".e ring).
<nis Iim<itatlon is not applicable in the even: of safety injec=ion invclving +ate.
levels far below the fe dring.
3oth the main and standby auxiliary feMwater syst~ flow pa-M to the steam generators are not isolated auto;.a.cally as a resul-of a s-.e m
o".
feedwatar (main or auxi (iary) line br eaI<:.
The isolation '-
ace
<.=I isnec manua I Iy.
III.
MEANS TO REDUCE THE POTENTIAL FOR MAi=R HAWER 1.
OESCRIPTION The following are means currently employed at Ginna to reduce the potential
=or steam generator water hammer:.
1.
"J" shaped discharge 'tubes" on all s
earn generator feed-s 71 rings in conjunction with the prompt automatic initiation of auxiliary feedwater flow upon loss of main feedwater flow a nd/or steam generator feedr'. ng uncovery.
2.
Adminis rative controls to limit auxiliary,eedwate.
flow to less Chan. 150 gpm per steyn genera. or durino periods of s
e m generator fe dring uncovery.
3.
Tne reduction of the e=fecitve horizontal sec
-'on of main I
feedwater piping at the enCrance to all steam generators co less than eight feet" ".
[81 The "J" shaped discharge tubes were 'installed on C"p of.-che
<eedrings
'l and Che bottom holes were plugged to provide for Coo discharge of water ra her than bottom discharge.
Ouring periods of fear'ng
- uncovery,
".his arrangement increases the time for complete drainage of the
. ee rings and associated horizontal feedwater pioing from less than one minute to about 30 minutes.
Also, the maximum main auxiliary fe dwater flow (about
-",00 gp;,
per stean genera or) was no suf icien-. to maintain the =
drincs and fe d-wa-
." piping full of uacer when the feedrings had
- o cm discharge ho es.
he feedrings equioped wi.h "J" shaoed discharge
.ubes, however, pew;:
eedwater flow rates as low as about
<0 gpm per steam generator Co keeo
=he feedrings and,eedwater piping full of water until <eedring recovery occurs.
Substantial drainage of the feedrings and piping via h
fe dring fitting clearance does not occur for about five minutes which allows time for automatic ac:uation of the main auxiliary.fe dwat r sys-em f.e.
the loss of main feedwater
<low.
ine potentia! for wa+ar banner is avoided '-; -:".e
<eedrings and feedwater piping are keoc ful i of waca..
The prompt automatic'startup of any one main auxiliary fe dwater p
...p after the loss o-main feedwater low provides feedwate.
flow to keep t.-e fe drings and feedwater piping full of wat r.
because the "J" shaped d:s-charge tubes reduce the leakage from the feedring, the auxiliary feedwa=r flow from either of the motor driven pumps or the turbine driven pump is more than suf icient to keep the feedwater system full of water.'he present Ginna main fe'edwater piping geometry adjacent to each s:eaa generator consists of a horizontal run from the s
an genera or -to the first downward turning elbow in each line.
The horizontal runs are 2 feet 3 inches rom the nozzle to the center line of the downwa'rd leg of the lbow, we. 1 within the vendor's recommendations to minimize wate.
banner damage to =ne fe dwater piping system.
Prior to the decision to install "J" tubes an analysis of the Ginna ma.n feedwater piping using a preliminary, Mest nghouse-derived dynamic forcing function~
~ was performed.
Assumptions wer o that
=he steam-wat=-.
slug initiat d at the steam generator s; ha auxiliary feeowater was in ~se; and that the main feedwater check valves we. o closed.
The time depende..=
mathematical unction was modified -,or he Ginna piping configuration.
he time history of the acoustic shock wave generat d by the ste m-water slu=
was evaluated with respect to stress cri e, ia based on allowable stress ob-.ained from the original cons"ruction code.
one results" showed that
=he. e we.
several locations in both fe owazer piping sys-.ms which exceeded
-ne s~. ess criteria.
"J" tubes subsequently were ins-.alied to reduce the pc=en-
-.',al for water hammer.
No.est programs have been performed at Ginna :o."e~ermine whether any wa-er harrrner transient would occur as a resul= cf uncovering of the ste=-s generator feedrings.
- However, both fee'wat r lines inside of containme..-
were instrumented following the 1975 transien to provide the control rcom ooerator with piping vibration information during plant star up.
2.
".."".""ECTIVENESS OURING:RANSI:NiS ANO CONOI IONS CONOUCIVE TO MAT=R HP'%R The normal and hypothetical transients ard conditions conducive to st m generator water hammer are discussed in -.his section.
'~lith the excep-tion of subsection 2.4 entitled "Operator Er.or",
e ch subsec.-Ion describes a transient resulting from a single initiating event or failure with the unit in normal power operation." Potential component or system failures as a direct r suit of a hypothetical steam generator water harme.
ar accounted for in the analysis.
A single criterion was the basis for evaluatinc
. he effectiveness of
.he means to adequat ly reduce he po-enzial
-,or s-.earn cenerator water haamer.
'8=
~
ine criterion is to maintain the fe~water sys-.en full o.
wa:er during the time irom :he initia~ing event result ng in feedrinc uncovery to subsequent feedring recovery and stabilized steam generator water inventory.
2.1 Reac.or Trio A reactor t".ip with the plant in normali power cperation would result in a turbine trip and cause the water level in all steam oenerators to collapse to a level below the feedrings.
Mi:hin 60 seconds of the resulting s
earn cenerator low-low water level signals,
.he mo or driven and urbine driven main auxiliary feedwa-er pumos would automatically star.
and supply auxiliary feedwate.
-o the s-.earn generators.
I-, the in'-.-'aiirg event for the reactor
-rip did not close the main reedwara.
regulatinc va'ves,:he valves would ciose upon r
eipr. of:
1 ) low primary coolant aver ge
=.=..oe. a ur
- signals, 2) saba m generator high-high water level sicnais, or 3) an SIS.
Auxiliary feedwater would.hen be manually controlled to r estor the water levels in he steam genera ors and maintain the levels abov
-.he feedrings.
The potential
=or water hammer oc"urring
',n -he feedring.or feedwater piping after a reac-.or trip is very low beca se
.he main and auxiiiary feH-wa
~ ar keeps he, eedl irlgs and eedwaie.
pl p flc
. u 1 1 o
wa'..
2.2 Loss of 4)a n
.=o cwater Flow he main feedwa-ar supoly c"uid =e .n:err;o-.ed due
==
-..".e ;,'oss of offs'te oower,
- 2) miaifunction or tr'.ooing
=.".e ma n fe~wa-er
-umos,
".) loss of suction to the main feedwater
- pumps, or 4) closure of the main fe d-wa er regulating and/or isolation valves.
A reactor trio would occur upon r ceipt of the resulting steam/feedwater flow mismatch signals and low steam cenerator wave.
level signals.
The reactor trip would cuase the wa"er levels in all s.earn generators to collapse to a level below the feedrings.
The motor driven and turbine driven main auxiliary feedwater pumps would start upon ceipt of the subsequent low-low s earn generator water level signals.
Auxiliary feedwater would then be used to refill'. the steam generators and recover the eedrings.
The loss of main feedwater flow and the likely uncovery of the feedrings would not result in subs.antial fe dring and fe dwater pioing drainage since the main auxiliary feedwa.er pumps would start promptly to supply feedwater to "he s
earn generators.
Therefore, the. potential'or water hammer is significantly reduced.
2.3 Loss of Of si.e Power The complete in erruption ov offsite power would result"in a reactor trip and automatic startup of the emergency diesel gene. ators.
Automatic initiation of the motor driven and turbine driven main auxiliary fe dwater systems would occur to supoly feedwater to the steam generators.
The redundant auxiliary eedwater systems are fully functional without offsite power since the diesel "ene.
tors and DC batteries can supply all nec ssary elec:rical powe.
to both systems.
As was "he case for the loss of main feedwa.er flow, aux-iliary feedwa-.er
-. ~ow would maintain the
=
drings and fe dwater pioing fu; I of water until
-;e~ring re overy occurs and again the potential for wa.e.
nammer would be ve.y low.
2.-"
Operator =rror The ".ot ntial =or water hammer '.n:he, edwate.
system
',ncre ses ncovered feedrings are allowed to drain substance'ally af-'er an event c uses
- ne s:earn "enerat"r water levels
.o droo below ".he feedr'.ngs.dmission of wa-.er into the drained feedrings and horizontal fe dwa-e.
piping could
- ,";en
." suI-in wa-. r slugging and suosequent wa-. r hammer.
The unc"ve. y of "ne or ".oth feedrings is possible
-.nrough opera~or error linen -he riant is
operating at low power or during star up or while shutting down since
-.eed-watat is being regulated manually, rather than automat'.cally.
For this si:u-
- ation, an administrative l.imit of 150 gpm on auxiliary feedwat r flow has been impieoented in the operating procedures.
This limitation was recormended in Reference 7 based on tests at Indian Point.
2.5 Steam Line Break The potential for steam generator water hamner events resulting from or concurrent with the rupture of a s
earn line inside containment was conside. ed.
The sequence of events following such a failure was evaluated to determine i= the break could result in the 1) biowdown,'of~:he re.-.sining steam generator and/or
- 2) inability to supply auxiliary fe dwater to t.'",e unaffected steam genera. or.
The rupture of a stean line would automatically rosult in an SIS c using a reac or trio, a turbine trio, and isolation of ail main feedwat'er lines.
The loss of main reedwater flow to the ste m..generators would,.result in the automatic startup of the motor driven and turbine driven main auxiliary fe d-water pmps upon receipt of low-low steam gener ator ua r level signals.
Auxiliary feedwater would continue to be supplied -,or subsequent refill of the unaf. ec" d steam genera-.or and recovery or the re rina.
However, i. the rupture occurs in he immediate 'v',c'.nity of the main auxiliary feedwater pumps and renders them inoperable,
-.he operating or"cedures allow " n minu es for the ope. a-ion of switches in the ""n:rol room tha=
uill
.isolate :he (N)Ar'AS and aet the (SB)ArAS on stream.
A;thougn the <eedring might drain significantly in that t me, the administra:.'ve I',mit of 150 gran of auxiliary fe dwater flow will reduce he potential of wax r hammer occur. ence.
inis limiw ion is considered applicabl a
Ginna based cn he best available information obtained in tests at Indian Point.
one biowdown of a s
earn aenera-.or would 'not deprive -he turbine dr-.'ven auxiliary fe dwater pump of driving steam.
A check va;ve In each steam suooly line would prevent "crossover" blowdown throuah
-ne suzoi v lines from t."'.e unaffec:e" steam aenera"or to the associ.axed biowndown s-earn generator.
C
~
The potential for water hammer is low after a steam line break since prompt delivery of auxiliary feedwater in conjunction with the "J" shaped discharge tubes maintain full fe drings and feedwater piping in the
'unaffected s
earn generator.
2.6 Loss-of-Coolant Accident The potential for feedwater water banner during a postulated loss-of-coolant accident (LOCA) in either unit was examined, because 1) a water hamner could increase the consequences of a LOCA and 2) the plant protective actions during a
LOCA could result in conditions which are conducive to water hamaer if the feedwater system is not kept full of water.
A LOCA would result in an SIS,
- a. reactor trip, a turbine trip, and sub-P sequent isolation of the fe dwaier system.
The star.up of the motor dr'ven and turbine driven main auxiliary eedwater pumps would result and fe dwa-.er would be supplied to the steam generators within 60 se onds of the reac-or trip.
Refill of the steam generators and recovery cr tne feedrings would occur in a manner typical of a reactor trip or the, loss of offsite powe..
The conditions conducive to water homer in the fe drings and feedwa-.er piping resulting from a LOCA would be very similar to those resulting -rom a r actor trip.
The. efor e, the means to reduce the potential
-;or water ham;,er would be fully effective during a
LOCA.
7
I
LY ~
CONCLUSIONS ANO RECOt&ENDAT~GRS Tne assessment of the capability of existing means to r duce he potentia1 for steam generator water hammer during roti and hypothetica1 operating conditions was discussed in Section II..
7n's assessment has shown that under conditions which are most conducive t."water hanxner in
.he feedwater systans (specifically, uncovered and dra;ning fe drings and feef~ater piping subjected to admission of cold auxiliary eedwater),
the means available to reduce the.potential for water hammer at Ginna are ade-quate to maintain suf iciently full feedr ings and ;e dwater piping.
Keeping the feedrings and feedwater piping full of wa:er avoids the-potential for wate.
hamer Therefore, we conclude that tHe means
.". reduce the potential for st am generator water hamne.
- a. this facility ar adequate and we recommend acc ptance by the NRC staf=..
11
REFERENCES
- 3.
G.
"=. Green, Rocheste.
Gas and Electric Cor". (?G"-), 'i-.r to J. P.
O'Reilly, NRC, "R.
E. Ginna Nuclear Power ?lan-., Uni-No. 1, Failure o= Sensing Line on 'O'uxiliary Feedwater Pump, Abnor...al Occurrence 73-6 and 73-7," August 1, 1973.
K. M. Amish, RGE, l.r to R. T. Colson, NRC, "Feedwater Piping Transient Evaluation," July 17, 1975.
L. 0. Mhite, Jr.,
RGE, ltr to R. A. Purple, NRC', 'Se ondary System Fluid Flow Inswbility, R.
E. Ginna Nuclear Power Plant Unit No. 1,"
October 31, 1975.
K. M. Amish, RGE, ltr to R. A. Purole,
- NRC, "S"e i-,tc Changes
~zde at Ginna Relative to Mater Hattrier," January 30, 1976.
L. 0.
Mhi e, Jr.,
RGE, ltr to 0. liemann,
- NRC, Steam Genera or Mater Hammer Pr vention," June 15, 1978.
Ginna Nuclear Power Plant Unit No, 1 Updated Fi..ai
. ciii 'escription ana SaTe-v Anal sis Report, R -=,
NR ocket
,'lo. =.-
7.
J.
B. Block, et al, An Evaluation oi PMR S
=-m Ge..e. a:or Ma'= r Han~er, Crear
, Inc.
NUREG-OP9 Deca-oer i9 a).
8.
M.
<<". Bennett, Materhamme.
in Steam Genera-.or Fe~wa-e.
- Lines, Mestinghouse Technscai Bu
- exxon, NSD-TB-lv-I jvune 0, l97aj.