ML18030A047
| ML18030A047 | |
| Person / Time | |
|---|---|
| Site: | Susquehanna  |
| Issue date: | 05/14/1981 |
| From: | Curtis N PENNSYLVANIA POWER & LIGHT CO. |
| To: | Youngblood B Office of Nuclear Reactor Regulation |
| References | |
| PLA-758, NUDOCS 8105190388 | |
| Download: ML18030A047 (9) | |
Text
REGULAT INFORMATION DISTRIBUTION TEi4I (BIDS)
ACCESSION N'dH'8105190386 Di)C ~ DATE: 81/05/ill NOTARlZED; NO DOCKET FACIL:50-387 Susquehanna Steam Electric Station< Unit 1< Pennsylva 05 87 50-358 Susquenanna Steam Electric Stations Unit 2g Pennsylva v
8~
AUTH'AiME AU THUR AFFILIATION CUH T IS i N. 8 ~
Pennsy l vani a Power 3, Light Co ~
Rc C IP e VANE 'EC II'It;NT OFF ILIATION YOUNGBLOODiB.J.
Licensing Branch 1
SU6JECT:
Forwards methodology for calculating submerged structure oraq loaos to close SER Outstanaing Issue 29.
DISTRIBUTION CODE:
9001S COPIES RECEIVED:LTH ENCL SIZE:
TITLE: ?SAR/FSAR AROTS and Related Cor respondence NOTES:Send 18E 3 copies f-S~R 8 al l amends.l cy:BviR-LRG P4l(L.RIB)
Send ICE 3 conies FSAR 8 al l amends
~ 1 cy:6'iiR LHG P "f(L~ RIB) 05000387 0500038<
ACT ION;
<ECIP IENT Iu CODE/NA~E, A/~
COPIES iTfR ENCL 0
A HECIPIEii T I 0 CODE/NAi"iE LIC PH o2 8C STARK~RE VW COPIES LTTR ENCL l) 1 1
Ii>>TERNAL:
ACC ID EVAL BH2b CHER E;lG BR Od CORE PERF oR lv Ei41c,HG PREP 22 GEJSCIENCES iii HYD/GEO BH lv I R,i Ot LIC QUAL HH MECH E'JG BH 15 4HC PDR 02 OP I I C BH PROC/TST HEV 20 RAi)
S BR22 FILE Ol i>>G BR2s 1
1 1
1 1
1 0
1 2
3 1
1 1
1 1
1, 1
1 1
1 1
1 1
1 AUX SYS BR 07 CONT SYS BH O9 EFF TR SYS BR12 EQUIP QUAL BR 13 Hu~<
F AC T Ei>>G bR I AC SYS BH 1 b LI C GUID 6R i)ATL ENG BR 17
%PA OEI D PO'r'lEf>
S YS BH 19 QA BR 21 REAC SYS BR 23 SIT ANAL BR 2O EXTERNAL:
05 lb ib 1
LPOR v3 1
1 TOTAINUHcsEH OF COPIES RE~VI,RED:
LT'l'R ~
f'i CL
TWO NORTH NINTH STREET, ALLENTOWN. PA.
18101 PHONE:
t2 I S) 821-S IS I NORMAN W. CLIRTIS Vice President Engineering & Construction 821.5381 May 14, 1981 B. J. Youngblood, Chief Licensing Branch No.
1 U.S. Nuclear Regulatory Commission Washington, D.C.
20555 Docket Nos.
50-387 50-388 SUSQUEHANNA STEAM ELECTRIC STATION SER OUTSTANDING ISSUE 29 ER 100450 EILZ 841-2 PLA 758
Dear err. Youngblood:
Attached is the methodology for calculating submerged structure drag loads.
This response completes our action to close SER Outstanding Issue 29.
Very truly yours, N.
W. Curtis Vice President-Engineering and Construct'on-Nuclear CTC/mks At"achment cc:
R.
M. Stark NRC 8 t 8519ga Qg PENNSYLVANIA POWER LIGHT COMPANY
Provide methodology for calculating submerged structure drag loads.
~Res onse:
The following write-up will be included in Subsection 4.2.2.5 of Revision 6 to the SSES 'Design Assessment Report.
Condensation Oscillation and chugging induce flow fields in the suppression pool causing drag loads on the submerged structures (i.e.,
SRV lines, downcomers, etc.).
The methodology fox calculating these drag loads to be combined with the otner design basis loads (see Section 5.0) is presented below.
Zn 1904, Prandtl (Reference
- 1) enunciated a theory describ-ing the flow of a fluid past a body as (i) the viscous flow in the thin boundary-layer, and (ii) the classical inviscid flow about a new body consisting of the original body enhanced by the thickness of a boundary layer plus possibly a wake (Reference
- 2).
- Thus, the force exerted on a submerged structure by a fluid moving relative to it is ecrual to the surface integral of the normal pressure and tangential shear stresses acting on the structure (Refexence 3).
The tangential shear stresses are important for Reynolds numbers less than one (Reference 4), but for the flow fields due to CO and chugging and the SSES submerged stx'uctures, the Reynolds numbers are much greater than one.
- Thus, the tangential shear stresses will not be considered here.
Thus F
= pdS where p is the fluid pressure acting on the area increment dS
=
Rds and 8 is an inward-pointing normal unit vector at the centroid of dS.
', pacae 2
The flow pattern of the fluid about the structure is as follows.
At large Reynolds numbers the field of flow may be subdivided into an external region and a thin boundary-layer near the structure together with a wake behind it.
In the external region potential flow theory can be used to evaluate the flow field while in the boundary-layer and wake, the Navier-Stokes equation must be used (Reference 5 and 6).
Therefore, the surface integral in Eq. (l) can be divided into an integral over the structure surface outside the wake region, and an integral over he surface inside the wake.
Hence F = pe ds
+
pwds Sy W
(2) where Sy and S
are the time-dependent areas adjacent to the W
potential and wake flows.
The pressure p< corresponds to the potential flow region and for a linear isentropic fluid is given by the equations of potential flow combined with Euler's Equation (Refe ence 7).
p By adding and subtracting the integral+>dS Eq.
(2) becomes SW F
=
p dS
+
(pW p )dS 4
W where S is the boundary-3.ayer surface S~ plus the surface of the structure adjacent to the wake.
Since the boundary-layer is thin, the thickness being inversely proportional to the square root of the Reynolds number, negligible, error is incurred by approximately S with the structures surface.
Thus the total force is the sum of an acceleration force FA and an unsteady drag force FD.
The acceleration force can be expressed as in terms of a pressure gradient via Gauss'heorem or a uniform flow acceleration U
(Reference
- 8).
=
py dS
=gVp4,dV S
(4)
=pV Um A
Where VA is the acceleration volume which is equal to the sum of the structure volume plus the classical hvdrodynam' mass (Reference
- 10) divided by the fluid density.
For a cylinder of diameter D, length L with U normal to the cylinder axis, V
8 D L (Reference 11).
The unsteady drag force FD can be expressed in the same mathematical form as the drag for steady flow by defining an unsteady drag coefficient CA F
=, (p
- Py )dS D
W
= gpc~xlu (5) where A.is the projected area and U is the steady flow velocity.
A somewhat limited conclusion, can be drawn from the comparison of Fq.
(5) with experimental data (Reference
- 12).
For cylinders or spheres in unidirectional flow acceleration where the Reynolds Numbe range g'ves approximately constant values for the steady flow drag coefficient C
, the unsteady drag coefficient C
is less A
than twice the steady value i.e.,
CA/CD
<2.
For finite plates the ratio C /C may be as large as six.
A
Thus, the force on a submerged structure (neglecting forces due to shear stress) is the sum of an acceleration force due to a pressure gradient in the flow field plus an unsteady drag force F=F+F A
D (6)
Under certain conditions the pressure gradient is of sufficient magnitude so that the submerged structure force is essentially the acceleration drag force and can be determined f om the action of that force times a correction factor which is approximately unity.
To show this Eq.
(6) is written in the following form F
D F
=F l+
A F
A, A
l + SrC
('7)
U~
D where Sr = ~<
is the Strcuhal number (Reference 13) and g=
C A,
D D
4 9-VA is a geometric factor of order C /< ~ Thus, where the pressure D
gradient is large such that (g/Sr)
(CA/CD) is small, the total force on a submerged structure is properly given by the integration of p> over the structure surface:
A 1 +
Sr CD p> dS (8)
For a cylinder, which is the most common geometrical shape for a SSES submerged structure, g
(CA/CD)
< 2.
Therefore, to ignore the drag force compared to the acceleration forceI the Strouhal Number should be of the order of ten or greater.
For SSES, the Strouhal Number is greater than 30, and in some cases much highe
, and negligible error will be incurred by ignoring the drag force.
Thus, for calculating 0he SSES submerged structure drag forces, Eq.
(8) reduces to F
pC,dS
'Phe pressure a
as a function of time and position is calculated e
ressure by the DREGS/MARS acoustic model of the SSES suppression pool.
Thus p
is calculated in an analogous manner as the symmetric wall loads (see DAB Subsection 9.5.3.4.1),
except that the pressures are calculated at the submerged structure locations
'instead of the containment boundary.
For each structure being analyzed (i.e.,
downcomer),
a pressure time history (PTH) is calculated for every 60o increment circumferential around the structure at each elevation corresponding to a nodal point of the structural model (see DAR Figure. 7-10 for downcomer structural model).
Thus, for each node point elevation, six pressure time histories are calculated.
This is repeated for each source.
These sets of PTHs, calculated for each source, are then integrated across the structure's surface to give force time histories for structual analysis.
~pa e
6
REFERENCES:
~ o 1.
L. Prandtl, "Uber Flussigkeitsbewgung bei sehr Kleiner Reibung,"
Proc. Third Int. Math. Congress, Heidelberg, 1904.
Macmillan Press Ltd., London,
- 1968, p.
709 ff 3.
S. Eskinazi, ~Princi les of plaid 'Mechanics, Second Edition, dllya and Bacon, Inc., Boston,
- 1966,
- p. 436.
4.
H. Schlichting, Boundar -La er Theo
, Sixth Edition, HcGraw-Hill Book Company, New York, 1968, p.
104 ff.
5.
Tbid.
p.
22.
6.
L.D. Landau and E.H. Lifshitz, Fluid Mechanics, Pergamon
- Press, London, 1959, p.'69.
7.
Tbid.
p.
3, Eq. (2.3);
p.
19, Eq. (9.2).
8.
F.J.
Moody, "Forces of Submerged Structures in Unsteady Flow,"
Proceeding of the ANS Topical Meeting on Thermal Reactor Safety July 31-August 4,
- 1977, Sun Valley, Idaho,
- p. 3-516 ff.
9.
K.T. Patton, "Tables of Hydrodynamic Mass Factors for Translation Motion," ASHE Paper No.
65-WA/UNT"2, 1965.
10.
L.)f. Eilne-Thomson, oo. cit., pp.
246-247.
11.
P.J.
))oody, ~o. cit., p. 3-592.
12.
Tbid., p. 3-591.
13.
B. Schlichting,
~o. cir, p. 32.
The following write-up will be included in Subsection 4.2.1.7 of Revision 6 to the SSES Design Assessment Report.
During the drywell air purge phase of a LOCA an expanding bubble is created at the downcomer exits.
These rapidly expanding bubbles eventually coalesce into a "blanket" of air which leads to the pool swell phenomena.
The bubble charging process creates fluid motion in the suppression pool which causes drag loads on the submerged structures.
The submerged structure drag loads due to air clearing, prior to pool swell, are calculated in the same manner as the drag loads due to CO and chugging presented in DAR Subsection
- 4. 2. 2. 5.
- However, the chugging and CO sources are replaced with a source representing the bubble growth prior to pool swell.
This source will be derived from the original 4T data.
All sources are assumed in-phase (87 sources).