ML20210U319
| ML20210U319 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 09/09/1996 |
| From: | Epstein M, Hammersley R, Henry R FAUSKE & ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML20198H322 | List:
|
| References | |
| ENG-ME-301, FAI-96-77, FAI-96-77-R, FAI-96-77-R00, NUDOCS 9709180002 | |
| Download: ML20210U319 (26) | |
Text
_
suse -
" i 3 so samense
- % ,",,,",,_ s TRMN *B" l
I m _L --.c w . m -
,nr 18 y - -i.D.I (g "
g #21 CENA erggt a-a-efm-a-
- A9 s Al an G Eye l l ,
m- a-, m- . . s I
-Filam t s '*-'*-'
1I >=
f=1 c -e-. ,
g' o an casa
.7 o = aan == euw n
If
[= -a. ,28
..wi g;; ar*sm , _ . <=4 MN n.= W e g
" A.G H , _
. _ ._ .g "d'/*;g*l t --8.y,,q== i-m, , . swe T,, I seu ! '"'a -
- ' I ! 'ni= - ms Q l -
M 5*/."h .. i l.. . - ,= =. m, .
l
'a y q l aM' **. ,'
ag- g.a
- a. a. -
a si 2.4 s 24s ,..
- s a [) ;
-a a- i.. ..
g a al 7 . A
^
.- MB p q o', "
uan tenn "A" Tu - ~ ** "l , , , _ _
- 2. ____ _ . . _ .
'~~~
,_ _ D r_ r "a ~o._,= " Uy Northern States Power Company Cooling Water System REV: 4 B35-01 Prairie Island Training Center 8131/95 O.
9709100002 970915 PDR ADOCK 05000282 PDR.
$ 2
,A y ,r, .
P ,gr
'J
t * -' #
AUX furs W -32132 IN DG hlDG Q CV-39701 FO It CNfutif . .
7 W-w,323,f Ku '
~3940s w,'32 483 W-32332 7g p l e COct eNG :s:
e.
,,,g
a
. gg WATER PuuP - -
4 pg
-p a na ; --> W-32329: lg A $ ;
Eul
~
i
' W-3203I XE uv-32844 *^N' fh -
W -32378 '
uv-32838 W-32139 U"39' )N '
fIURB S* c c f[
22 k 24 22 & 24 il FCu'S # Cu's l@ W-31032
usef 2 -l II & 28 Utef 2 li & 21 . g , 11 & 28 1i Ctm LIN fCu COOLING CH21ED Ost t t D WlR *., PLY WAffa = 12 + wagg g ,, fCu,
, ,coolest pgg COND'S eM>=PS SHROUD C1C Pue#5 j X-) CV-39449 -t 28 & 23 Cost ASSu'Y 2s & 23 FCu"5 FCu'S CV*3940 CV-394iG TUR8 e CV-334 )l- CV-39408 )l-)
W-32o33 gg W-32 e9 c trl 1_.
W-3238o n pg;= h
- -32 iu *-3n 42 b ),-l * -32322: :E I i 28 CCotec 3:
4}: ' '
l ** Slaras WAl[R PuuP cv. 39403 cv- 39409 uv-32334 . i Fat j FO Il u u [ TO 4[Lu]
1 % ,; '2 Cp;unt I [A] (v pro 3 uv-323n W-32,35 W-32,3s l
88 & 21 Use tl h
. mas i
Nerth;rn States Power Company Fan Coil Unit Chilled Water Supply REV: 2 B35-23 Prairie Island Training Center 8/29t*5
PINGP 1063, Rev.1 Front Page 1 of 1 Retention: Ute NORTHERN STATES POWER COMPANY 4
PRAIRIE ISLAND NUCLEAR GENERATING PLANT CALCULATION COVER SHEET Calculation Number: Ntf - M6 - 3 '
i Calculation Rev. No.: 4 Addenda No.: e _
Calculation
Title:
_/luennA e ce hu ho Sw Cm-se
'RP%t & 4oAD3 Safety Related?: V Calculation Verification Method (Check One):
Design Review Alternate Calculation Qualification Testing Scope of Revision:
( Attu u Anca U AS PCMcEnfB uaDEL A YG~%~L GA Vlow w . Tan cat SH% 'Dcwakn ^f]? &<wrwed l
l l
Documentation of Reviews and Approvals:
Originated By: g Date: 9.,g . 7g Checked By: g Date: p,g,g Verified By: Date: f _ y .7g 4-Apprcved By: [g _
Date: j,yf py
FAUSKE & ASSOCIATES, INC. FA1/96-77 Page / of J F Rev. O Date:
CALCULATION NOTE COVER SilEET SECTION TO BE COhtPLETED BY AUTIlOR(S):
Page.
Calc Note Number Fallo 6-77 Revision Number 0
\
Title Assessment of Prairie Island Fan Cooler Pinine Loads Project Prairie Irland Shop Order NSP 001
Purpose:
Estimate piping loads following void collapse and refdl of partially voided fan cooler cooling water supply and return headers.
Results Summary: Dynamic waterhammer pressure loads of approximately 145 psi or less are estimatal for void collapse and the impact of the reful water slug upon another stagnant water slug. 'Ihe refill velocity is high enough to prevent stratified flow in the horizontal pipe segments so condensation induced waterhammer would not occur, Author (s): Completion Name (Print or Type) Signature Date R J. Hammerslev -
/ September 5.1006 y
mmw SECTION TO BE COh!PLETED BY VERIFIER (S):
Verifier (s): , Completion -
. Name (Print or Type) Signature Date Michael En=teie N Mbb \
- l\ tk9l SECTION TO BE COh!PLETED BY MANAGER:
Responsible Manager: Approval Name (Print or Type) Signature Date Robert E. Hentv .,
d ,A- September 13, 1096
. 6
FA!/96-77 P:ge 2 of 2C Rev. O D:te: 9/5/96 0
TABLE OF CONTENTS Page LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii LIST OF TABLES ...........................................iv 1.0 PURPOSE ............................................ 1 2.0 INP UT D ATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3.0 ASSESSMENT OF LOADING CONDITIONS . . . . . . . . . . . . . . . . . . . . . 5 4.0 RESULTS ............................................ 20
5.0 REFERENCES
....................... ....... ,....... 21 il
FA!/96-77 Page 3 or a 5-Rev, O Date: 9/5/96 LIST OF FIGURES Figure Pace 1 FCLR#11 Piping Configuration .................... .......... 6 2 FCLR#12 Piping Configuration ..............................,7 3 FCLR#13 Piping Configuration .............................. 8 4 FCLR#14 Piping Configuration ...................... . .....9 5 FCLR#21 Piping Configuration .............................. 10 6 FCLR#22 Piping Configuration .............................. 11 7 FCLR#23 Piping Configuration .............................. 12 8 FCLR#24 Piping Configuration .............................. 13 9 Piping System Branching for Fan Cooler Unit . . . . . . . . . . . . , , . . . . . . . . 14 i
Y
. ]
- w. ...i FAI/%77 PJg3 Y of a S Ray. O D:te: 9/5/96 LIST OF TABLES b E EASA 1 Prairie Island, Unit 1 Drawing List . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Prairie Island, Unit 2 Drawing List . . . . . . , , . . . . . . . , . . . . . . . . . . . . 4 3 Drainage Calculation for 8" Vertical Pipe at 900 gpm . . . . . . . . . . . . . . . . . 18 .
4 Refill Velocities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 t
kV
_ _ . _ _ _ _ _ . _ _. _.._. J
FAI/96 77 Pcge 6 of c2.5~ l Rev. O Dete: 9/5/06 1
1.0 PURPOSE A postulated loss of coolant accident coincident with a loss of offsite power is a potential design basis accident sequence which would result in the ! css of cooling water flow through the fan coolers. The duration of the flow interruption would be approximately five seconds given that the diesel-driven supply pumps were aligned or twenty-six seconds given that the motor-driven supply pump was aligned to the fan cooler units. Void formation in the fan cooler units, supply header or return header for the cooling water could occur during the interval of interrupted cooling water flow. The automatic restart of the cooling water flow upon pump restart will refill the piping and could lead to void collapse and potentially cause dynamic loads on the piping and its supports.
The purpose of this calculation is to estimate the magnitude of the potential loading condition which could be produced during the refilling of the voided fan cooler piping sub-systems. The method used to estimate the potential piping load employs the following steps:
- 1. Review of Prairie Ishnd fan cooling piping configuration regarding the possibility of void formation.
- 2. Assess the nature of the piping and fan cooler refill and its potential for producing a stratified configuration of steam and subcooled water.
- 3. Estimate the dynamic waterhammer load magnitude given void collapse and impingement of two water columns.
G 1
a n . -A
FAl/96 77 P:ge b cf O 5 Rzv. O D:te: 9/5/96
'2.0 INPUT DATA The input data used in this calculation is as follows:
- 1. The supply and return headers for e.ach fan cooler arrangement. l The elevation of the piping including the fan cooler was assembled for each of the four fan cooler units in each of the two Prairie Island containments. Tables 1 and 2 summarize the drawings used
- to derive this information.
- 2. The design flow rate through each fan cooler unit sub-system is 900 gpm (Ref.1). The piping local to each fan cooler is reduced down from the eight inch supply and discharge header dimension as it is branched to four separate faces for each fan cooler.
- 3. The diesel-driven cooling water supply has a five second recovery interval and the motor-driven cooling water supply has a twenty-six second recovery interval (Ref, 2). The recovery interval represents the time to re-establish cooling water flow at the normal steady state flow rate.
2
FAI/06 77 Pa68 7 of 2 S' Rev. O Date: 0/5/96 Table 1 PRAIRIE ISLAND. UNIT 1 DRAWING LIST a
Dneription Drawine No.
Unit 1 Return Side - Outside Containment X-HIAW-106-98 Rev. G 14 Fan Cooler Return X-HIAW-106-359 Rev. B 11 Fan Cooler Supply X-HIAW-106-332 11 Fan Cooler Return X-HIAW-106-319 12 Fan Cooler Supply X-HIAW-106-330 12 Fan Cooler Return X-HIAW-106-318 13 Fan Cooler Supply X-HIAW-106-333 Rev. A 13 Fan Cooler Return X-HIAW-106-329 Rev. A 14 Fan Cooler Supply X-HIAW-106-360 Rev. C Unit 1 Supply Side - Outside Containment X-HIAW-106-97 Rev. K O
3 i
. _ _ _ _ b
FAI/06 77 Page 8 of M Rev. O Dete: 9/5/96 Table 2 PRAIRIE ISLAND. UNIT 2 DRAWING LET Descriotion Drawine No.
Fan Cooler 22/24 Supply & Return Outside Containment X-HIAW-1106-46 Rev. E Fan Cooler ,
X-HIAW-106-96 Rev. C Fan Cooler X-HIAW-1106-30 Rev. D Fan Cooler X-HIAW-106-98 Rev, G Fan Cooler 21/23 Supply & Return Outside Containment X-HIAW-1106-45 Rev. G Fan Cooler 24 Return X-HIAW-1106-2575 Rev, A Fan Cooler 23 Return X-HIAW-1106-2574 Rev. B Fan Cooler 22 Return X-HIAW-1106-2573 Fan Cooler 22 Supply X-HIAW-1106-2559 Rev, A Fan Cooler 23 Supply X-HIAW-1106-2560 Rev. C Fan Cooler 21 Return X-HIAW-1106-2569 Rev, B Fan Coolci 21 Supply X-HIAW-1106-2558 Rev. A Fan Cooler 24 Supply X-HIAW-1106-2561 Rev, A 4
FAl/96 77 Page 9 of 2S Rev. O Dde: 9/5/96 3.0 ASSESSMENT OF LOADING CONDITION Tne plant drawings of the fan cooler supply and return piping were used to prepare Figures 1 through 9. These figures illustrate the relative elevation of the fan cooler and its headers for each of the eight Prairie Island fan cooler units. Figure 9 provides an example of the piping branches that connect the supply and return headers to each of the four faces of a fan '
cooler unit.
The eight fan cooler unit piping configurations were reviewed and it was determined that there are no check valves in the supply line between the cooling water supply pump and each fan cooler unit. It is seen that piping segments of 2.5 inch, 4 inch, 6 inch, 8 inch and 10 inch are included in each fan cooler's cooling water sub-system. Additionally, the high points in both the supply and discharge piping for seven of the eight fan cooler ur.its are at higher elevations than those of the fan coolers. Therefore, following supply water pump trip and depressurization of the piping system, cooling water will drain under gravity from both the supply and discharge lines for all eight fan coolers. The drainage rate for cach line will be controlled by the available hydrostatie head (piping elevations), flow resistances and pump coastdown. The drainage will lead to column separation in portions of both the discharge and supply line for all eight fan cooler units. Stagnant water will be trapped in seven of the eight fan coolers and the attached piping between each fan cooler and the high point in its supply and return headers. The drainage induced column separation and voiding could lead (3 both vertical and horizontal piping sections being voided for each of the eight fan coolers supply and return headers. Thus, the supply and return headers for each fan cooler will experience celumn collapse upon system refill.
The postulated accident sequence being considered in this calculation includes a loss of coolant accident as the initiating event. This will lead to an increase in the steam mass fraction in the containment regions which will lead to energy being transferred to the fan cooler tubes and the stagnant water within them. The energy transfer willincrease substantially as the steam partial pressure increases in the local vicinity of each of the fan cooler units. The thermal response of the stagnant water within the fan cooler tubes depends upon the rate of energy transfer from the steam in the containment and the duration before cooling flow is re-established to the fan coolers. For initial plant configuration with the diesel-driven supply pumps aligned to the fan coolers cooling water flow will be re-established in five seconds. A short interval for stagnant water conditions would exist allowing for pump coastdown following loss of AC power.
The peak water temperatute within each fan cooler at the end of the five second interval will be significantly less than the local boiling point within those tubes. Thus, for the diesel-driven alignment no net boiling of the stagnant water trapped in the fan cooler units would occur. If the initial plant configuration at the anst of .ne accident included the motor-driven cooling water 5
FAI/96-77 Page /O_,of 4 S' Rev. O D:te: 9/5/96
_____,_s
/ y
/ \
l %
- \
/ \
,/ \
s
, i Pra..ine s, i
/ Island i, 780 - Z ? --
E _ l FCLR #11 l .
_ i , _
= I i :
770E- l .
l 9
- I ; -
_ i l
i 5 _
760 E- l l E
- l l :
= i i :
C I l 750 !
I
- i :
C : s
.9 i g
i E>
740: _
i i i Q C 10' M i 3
w 8i , _
_ i -
2 730 - l - _
! E
- i i :
1 i
1 2
- i i :
i 720E- l l -E
- i i :
= -
I I
. I
. I
_ i i _
~
7102- l l
24' 10* C I
i _ J i i
l l _
i l
l T i -
- i i :
700 ' '
Return Supply JR95o117.COR 8 30 95 Figure-4 FCLR#11 Piping Configuration.
6 J
FAl/96 77 Page M of J 5' Rev. O Date: 9/5/06 I
l
- ~~,
- s %
\
/
\
\
/ g i
Pra..ine s,
/ Island i 780 -- k h --
E
l FCLR #12 i E l
i i
7702- l .
l c
- i i :
_ i , _
= l
_ i 1
, =_
7605-_
l i
l E ,
i _
= i :
_ i ,i _
= i i 2
750 ! l i
i ,i :
g E l l 5
. . _ i i :
E 740=:_ i i _-
> _ i i _.
.a : i i :
m E l i
l 5
- i :
730 - '
l _ .
i , =
=
= M, _
Jl i =
7202-:
i i _ _
i i :
_ i , _
=
i I
1- _
,i
= -
- i i =
710 - ac 30-j ; _: ,
= I i i I I :
_ i _ _
i i,
i i
- i i :
700 i '
Return Supply JR96o121.CDR 8 30 96 Figure -2 FCLR#12 Piping Cm , urc 7
4 FAI/96-77 Page la of .D S Rev, _p_ Date: 9/5/96-
.-...s s
/ g
/ \
/ \
/ \
/ e . g
/
Prairie s
\
/ Island i 7802- h h --
E.
l FCLR #13 l .E
= i i :
770E--
l .
l -E i i :
1 l t c
- i ! =
760E - l 1 l
=
i i
t _5
_ i i :
i -
- i i :
2 ! 2 750 I i
l
- i :
lM c . ! .E O 2 I I 5> 740E_ _
l i
i i
_E _.
.e LLI E
r iop @'i s
l -
r E
i i -
- 730 E- l l =
_ i i -
- i i :
i i ~
. E l l =
720~-:
i i
i
- i i :
C -
I i .
i , _
- - i i =
7105- l l
- 24 i i l l :
r %- i i
i i
i i
i i
700_ i ' -
Return Supply JRG5o119.CDR 8 30 GS Figu E-3 ! C ':"13 Piping Configuration.
8
FAI/96 77 P:ge ,/ 3 of J .S' Rev. O Drte: 9/5/96
, s,
/ g
/- g
- ./ \
/ \
/ e e x
/ Prairie s
- Islar1d
,/ i
~ ~,
780 - + ~
i ! FCLR #14 j E i
i i =
i 770E- l .
l
- i i :
. = l i 1
l :
l l
, 760 =5- i , =
- i
l .i
= ,
l =
750-: i i
i
., i i :
4 C 2 .
I ,i 2 O 2 l 8
=s-740:- ,
i s
E l t f l w
i l
, 2-730E- E
- i l 4 li 1 :
E
+4c+-
i l 5 -
i 7202- :
i i
i i
c
- i i -
1 I C
- , ,i -
, ,
710: 24- 30- l i
E i i
l i
- L_ 1 E 2
l l
i i :
'700:-
i '
Return Supply JR96o123.00R 8 30-96 d
Figure-4 FCLR#14 Pipinr, Configuration.
9
FAI/96 77 P:ge /Y of d#
Rev. O D:te: 9/5/06
,- - - ~,,
/ %
\
/
\
/
\
/ g
/ \
l \
I 1 780=- i Prairie $ :
E.
l Island l _E.
= i i :
/70 -- FCLR #21 l
, i -5_
E i l E
- i i :
760:- _
i i
l :
i _
= i i :
- i ; -
1 i
750E-- !
i l
- i :
i, i _
e .
i =
.O i
I i
=
w, 740.=.=_
i i
i m E- ! l E w i i
= i 7302- l i '. ' .
- i ) :
~
i i :
l so -c>__r a i- i 720j- l ) 7-- p l _g i
- _J i :
l l
i i
710 - 10' -
l l
-E 24' '
a
' E- !
i l 10' " 24- 5
. : i i :
i _
700_ ' '
Return Supply JR95o111.COR 8 30 95 Figuar5 FCLFi21 Pipir.t C0:. '.p:ntion.
10
1 FAl/9647 Page '8 of JS Rev. O Dite: 9/5/96
- g
/ g
/ \
/ \
/ \
/ \
/ g i \
l i i
~ ~i 7803 - Pra..irie -
E i
l Island l .5_
i
= i i 7705-
=
li FCLR #22 li -5
=
0_
I i C i i
- ie0 -
E l i
l
=_
i
=_
i i
=
= i i :
- i l 750-: i i
i i
O C E l l x. B 5
C m
ec 740__
i i
i l.
i i
- n. 10 _
a u;
io A.i i i
i E
=
730:_
i i l
- i i :
- i i
=_ i i :
i i 720- i -
i :
, 5
=
l I 1 l _
=
5 l
i i
' ~
710:- 1 o' '
5 2,. l l ,o paa- E
_ i , -
i i :
. = i i
i =
i -
700_ ' '
Return Supply JR96o109.CDR B 30 96 Figue 6 FCLR#22 Piping Configuration.-
11 m --
4 pal /96 77 Page e' 0 of =2 S' Rev. O Date: C/5/90
/ %
/ g
/ \
/ \
/ \
/ 4 9 g
/ Praine s
/ Island i 780=- k h i l FCLR #23 l j
= i i
i i _
770E- l .
l -5
= i i :
E- ! E
_ i !, _
760E- _
l i
l
_5 i
i i
_ i i _
i i
l 2 750 i l
_= i ,i :
g E l l 5
._ i i :
w, 740=__ i i
_ _ i
=_
i ..
o 2 l E tu l i
i i :
i 730 -
l - h E
=
i i
i i
3' 8
=
10' i i 720:=-
i i
i :
i E
=
l l 5 i i 710:- t o' -
l l -E
=~
24- i i - =
~~~
l 10*
- 24' I
- i i :
= .
i i 700 i '
Return Supply JR96o115.CDR 8 30-96 Figure 7 FCLR#23 Pi ir.;- r Cs.9i. :um .
12
PA!/96 77 Pap / 7 of JS Reve 0 Dzte: 9/5/96
- ..~,
- g
/ g
- \
/ \
/
\
/ e e Praine
\
s I \
/ Island i 780=- k h =
g, l
FCLR #24 l .i
i i
770E- l .
! -E
i i
E- l l
_ i I f i =_
7605- _
-5_
i, ,i
_ i , _
= i = -
i i, _
750- i i
_=
=
i i ,i _-
=
= ,
r , =
.9 Tu 740 s .
E l i
i !
i s8 E
_=
+
h i o-i , f_
730[--
=
l l
l i
4=
= ,
l =
i,
i, -
720:- i i :
1 i
_ i i -
i i
i
=_
i
, ,
7102- 10' l -3
.=
=
- u. i l
i i
ji % =
=
l
= .
i, l =
, =
700= ' '
Return Supply JR95o113.COR 8 30 96 Fipre'2 FCLPJ24 Piping Configuration.
13 1
FAl/96 77 Page / 9 of D 6 Re O D:te: _ 9/5/06 Prairie Island - FCLR >l21 4*O O
g 6"O IM 4ga G -
4"O L-I O 8"O
_Y Top View 4"O Supply Return Elevation Elevation 2.5"O g O *- 715.5' 2.5"O k --
S'
^~"
JRGCA313. coa 6 30 96 Side View Figure 9 Piping System Branching for Fan Cooler Uni 14
, ~, -,. - --
i FA1/96 77 Page /'7 of M Rev. o Date: 0/5/96 supply pumps then the duration of the stagnant water condition for the fan coolers would be 4
approximately twenty six seconds. The energy, transferred to the stagnant water in the fan coolers during this interval could be sufficient to initiate boiling. Shortly after the onset of boiling the stagnant water could be forced from the fan cooler units by the exiting steam. Thus, fan cooler units with stagnant water could be volded by steam generation within their tuoes before the motor driven pump could re-establish cooling water now.
Therefore, two potential voiding conditions could be encountered following recovery of the coo!1ng water supply pumps and initiation of a refill of the piping system. The conditions depend upon the pump configuration at the accident initiation. For those accidents initiated with the diesel-driven water supply pumps aligned the volded sections of the supply and discharge piping would be separated by water trapped in the fan coolers and some of its attached piping.
For the plant configuration with the motor driven supply pump aligned, a single volded region without water traped in the fan coolers could be anticipated. In either case, upon the recovery of the cooling watei Gow the refilling of the supply header for each fan cooler would be initiated
- and soon result in a water Illied configuration. The voided portions of the system will contain steam and likely some air. The steam will result from boiling in the fan coolers as well as Dashing of the cooling water as the system pressure is reduced during column separation. The source of air would be that dissolved in the cooling water which then comes out of solution at the negative (gauge) system pressure produced by column separation. This void must be collapsed or forced from the system to re-establish a single phase configuration. The water fill rate dictates the means whereby the steam void would be condensed and Daid transients (including waterhammer) may be experienced that ultimately lead to the return of the cooling water now to steady state conditions.
The most important element of the refill rate is the Froude (Fr) number given by Fr = (1)
D where Uw is water refill velocity g is acceleration of gravity D is the pipe diameter.
If this Froude number approaches or exceeds unity the pipe will run filled with liquid.
The refill initiated upon pump recovery will first start to fill the supply headers for the fan coolers. Given the relative elevation the supply pump and the configuration of the supply headers, the verti ' elements Will be refilled from the bottom up. The refill rate wil: be 15
FA!/96 77 Page J 0 of M Rev. o Date: o/5/96 dictated by the pump flow rate as the vold is condensed or pushed through the balance of the piping system.
Horizontal pipe segments present the possibility of producing a stratified flow condition during refill. The possibility of establishing and maintaining a sepsated steam water flow regime has been assessed (Ref 3) for horizontal pipes. Assessments (including experiments) of condensation induced waterhammer have demonstrated that a separated steam water flow regime is required to initiate such conditions in a horizontal pipe. Experiments have shown that a water velocity (U) through a horimntal pipe which exceeds a value:
U = (0.25 g D)t/2 (2) will result in a water filled condition. A water filled condition (entire pipe cross section filled with water) does not lead to condensation induced waterhammers. If this is the case, the dynamic loads on the piping system and its supports would be thow related to refilling the piping sub systems. The refill velocity is dictated by the pump and piping system configuratiom The potential pressure load associated with stagnating the refill column on the stagnant water column in the partially drained headers is given by the waterhammer equation AP = k (p,/g.) a, U, (3) where AP is the waterhammer induced pressure k is a factor that reflects the compressibility of the impacted surface a,= is the speed of sound in water g, is the gravitational conversion factor The value of k would be 1 for those situations where the water column inducing waterhammer loads is stagnated on a perfectly rigid surface such as a valve face. The value of k is 0.5 for the situation where the moving water column is stagnated by impacting ano'her water column giver, the compressibility of water.
Another aspect of the refilling rate is the behavior of the vertical piping in the return headers with the water being added at the top of those piping segments. In this regard, the drainage behavior of the piping segment can be related to the refilling rate to determine if the vertical pipe can run full in a downward direction or whether the drainage (film and/or rivulets) as well as water falling through the central region of the pipe determines the flow regime. The drainage rate for the pipe dimensi'ons included in the return header would S expected to exceed 16 t
FAl/96 77 P:ge 2 / of M Rev. O D:te: 9/5/96 the refill rate as produced by the cooling water supply pump (see Table 3 for example). This would lead to a situation where water films would fall through the vertical return header and cause it to refill from the bottom up. This would result in fmal void collapse at the top of such a vertical header. Alternately, the refill rate of water falling into the vertical header could be limited by the void rise rate as the void rnoves upward due to buoyancy forces. Thus, the possibility of vertically upward or downward waterhammer induced loads should conservatively be considered when applying the loading condition.
Table 4 presents the refill velocities in the different diameter pipe segments incorporated in the fan cooler supply and discharger headers. The velocities are calculated based upon the steady state pump capacity and the branchlug within the system. Additionally, Equation (1) has been applied to each piping segment diameter to determine the minimu,m flow rate required to assure that it runs full during refill. These results are also presented in Table 4. Comparison of these velocities for each pipe segment diameter leads to the conclusion that all of the horizontal pipe elements will run full upon refill. Therefore, based on this experimentally derived criterion, no condensation induced waterhammer loads will be produced upon the refill of these horizontal pipe segments.
Eventually the refill water will impact upon a stagnant water column which may be in either the supply or discharge header. The estimated magnitude of the waterhammer load is 145 psi based on applying Equation (3) with a value of k equal to 0.5, a water density of 62.3 3
lbm/ft , a water velocity of 5.7 ft/sec in the 8 in, piping and the speed of sound in water of 3800 ft/sec. This value for the speed of sound in water is a representative number that generally includes the effects of the pipe elasticity and a srnall (less than 0.05 %) amount of air entrainment in the water (Ref 5).
The actual loads produced during refill may not be as large as the predicted value due to the presence of air in the piping system or the compressibility provided by steam bubbles which are not entirely condensed but are swept out by the refilling water flow (Ref. 6).
17
~
l
s FA!/96 77 P ge O 3 of O S*
R:v. O Date: 9/5/96 Table 3 Dralnage Calculation for 8" Venical Plpe at 900 spm i
)
Assume Nusselt's analysis for laminar film flow (Ref. 4)
Flow Rate per Unit of Wall Length I' (lb/sec/ft) = P 8 I' 3
6= 34 I = water film thickness P* g I' = 112 lb/sec
= 53 lbm/sec/ft er (.67 ft) p = 62.4 lbm/ft3 p = 5.8 x 104lbm/ft/sec g = 32.2 ft/sec2 6 = 0.012 ft (0,14 in)
This is much less than a thickness that would fill a pipe, l.c. the
- 8 inch discharge would not run full.
18 -
I FAl/96 77 Page 4 3 of a S' Rey, J. Dcte: 9/5/96 Table 4 REFILL VEI.OCITIES Minimum (D Pipe Segment Flow Refill Required Refill Diameter (in) Rate (com) Velocity (fos) _ Velocity (fps) .
10 1800 7.4 2.6 8 900 5.7 2.3 6 450 5.1 2.0 4 225 5.7 1.6 2.5 113 7.4 1.3 5/8 (tube) 5.63A 5.9 0.65 (U
Minimum required velocity for full pipe flow as given by 0,5ED.
%wenty tubes per fan cooler face.
19
FAl/96 77 Page J E of,,2 f Rev. O Date: 9/5/96 4.0 RESULTS The assessment concludes that the ren11tng of the partially volded fan coolers and their attached piping would not result in condensation induced waterhammer loads due to the magnitude of the ren11 velocity A conservative estimate of the potential waterhammer loads for the impact and stagnation of a moving water column impacting a stagnant '. vater column has been calculated. A value of 145 psi during the refill has been estimated for the postulated accident scenario for the Prairie Island piping configuration. This load should be considered for those pipe elements which could experience void collapse during refilling by the recovered cooling water flow.
20
4 FAl/96 77 Page 46 of Js Rev, o Due: 9/5/06 3 5.0 REFERILNCES
- 1. DDD SYS 35, " Cooling Water Design Basis Document."
- 2. Personal communication with hit. Steve Thomas.
- 3. Bjorge, R. W. and Griffith, P.L.,1984, " Initiation of Waterhammer in Horizontal and Nearly Horizontal Pipes Containing Steam and Subcooled Water," Trans. AShiE, Ir. of IIcat Transfer,106, pp. 835 840.
4 Krieth, P.,1960, Principles of Heat Transfer, International Textbook Company, Scranton, PA.
- 5. EPRI, "Waterhammer Prevention, hiitigation, and Accommodation Volume 3:
Experimental and Engineering Data," EPRI NP-6766, July 1992.
- 6. Izenson, hi. G., Rothe, P. H. and Wallis, G. B., " Diagnosis of Condensation Induced Waterhammer (hiethods and Background)," NUREG/CR 5220 (Vol.1).
S 21