ML17187A866

From kanterella
Jump to navigation Jump to search
Summary of 970121 Meeting W/Comed Re Emergency TS Change Requesting Use of Containment Over Pressure to Compensate for Net Positive Suction Head Deficiency for Emergency Core Cooling Pumps.List of Attendees & Presentation Encl
ML17187A866
Person / Time
Site: Dresden  
Issue date: 03/07/1997
From: Stang J
NRC (Affiliation Not Assigned)
To:
NRC (Affiliation Not Assigned)
References
NUDOCS 9703210071
Download: ML17187A866 (21)
v • d • e


Text

March 7, 1997.

LICENSEE:

Commonwealth Edison Company (ComEd)

FACILITIES:

Dresden Station, Units 2 and 3

SUBJECT:

SUMMARY

OF THE MEETING CONCERNING THE EMERGENCY TECHNICAL SPECIFICATION CHANGE REQUESTING THE USE OF CONTAINMENT OVER PRESSURE TO COMPENSATE FOR A NET POSITIVE SUCTION HEAD DEFICIENCY FOR THE EMERGENCY CORE COOLING PUMPS

  • On January 21, 1997, the staff met with ComEd to discuss an emergency Technical Specification (TS) change involving the use of containment over pressure.to compensate for t deficiency in net positive suction head (NPSH) of the Emergency Core Cooling Pumps (ECCS).

A list of attendees is provided as fnc 1 osure* 1.

The objective of the meeting was to resolve open issues from a previous m~eting held ~ith ComEd on January 16, 1997:

The topics disc~ssed were the NPSH calcOlations and the ECCS pum~ performance during pump cavitation. A copy of the 1 i.censee' s presentat.i on is included as Enclosure 2.

The lic~nsee provided specific details and clarificaticin on hriw the new NPSH calculations had been performed, including all input par~meters and conservatisms.. The licensee *also provided the :de.tails and s*pecific justifications for the EtCS pwnp flow under cavitatfo*g co*nditions.

/sf John F. ~t~~d~* Senidr Project Manager

-~.Project Director.ate.. IIJ-2

~

  • }Dfvision of. Reactor Pr'oj~c:ts - Ill/IV

. Office, *of. ~ucl ~ar

  • React6r: Regulation Docket Nos.

50~237 and 50-249

Enclosures:

  • 1.

List of Attendees

2.

Licensee,.s Pres en tat ion I

i./

~-

cc w/encls:

see ne~t page DISTRIBUTION w/encl. 1 & 2:

Docket PUBLIC PD3-2 r/f.

OGC 015-B18 E-Mail w/encl 1:

S. Collins/F. Miraglia R. ZilTITierman J. Roe(JWR)

R. Capra(RAC1).

E. Adensam CEGA1)

C. Moore(ACM)

D. RossCSAM)

B. McCabe CBCM)

K. Kavanagh CKAK)

P. Hi land CPLH)

H. Dawson CHFD)

K. Dempsey CKCD)' 'J, Kudrick CJAK1)

J. Lyons CJEL) 200069 I'

COPY tfO\\

ACRS T2-E26 J. Stang CJFS2)

G. Hal!ITier CCGH)

R. Wessman CRHW)

G:\\CMNTJR\\DRESDEN\\012197ET.SUM To receive a copy of this docunent, indicate in the box- "C" = Copy without enclosures 0 E0 = Copy with enclosures 11N11 = No copy OFC

<2

~~p (L...--

D:PD3-2 G:.

DATE 9703210071 970307 PDR ADOCK*OS000237 F, *

".*** PDR

'1FIGIAL RECORD COPY I I RCAPRA (>.,/

03/ 7 /97

' y v

cc:.

Ms. I. Johnson Acting Manager, Nuclear Regulatory Services Commonwealth Edison Company Executive Towers West III 1400 Opus Place, Suite 500 Downers Grove, Illinois 60515 Michael I. Miller; Esquire Sidley and Austin One First National Plaza Chicago, Illinois 60603 Site Vice President Dresden Nuclear Power Station 6500 North Dresden Road Morris, Illinois 60450-9765 ;

Station Manager Dresden Nuclear Power Stati6n 6500 North Dresden Road Morris, Illinois 60450-9765 U.S. Nuclear Regulatory Commission Resident Inspectors Office Dresden Station 6500 North Dresden Road Morris, Illinois 60450-9766 Regional Administrator U.S. NRC, Region III SOI-Warrenville Road Lisle, Illinois 60532-4351 Illinois Department of Nuclear Safety Office of Nuclear Facility Safety 1035 Outer Park Drive Dresden Nuclear Power Station Unit Nos. 2 and 3

Spr-i-ngf-ield-, I-l-l-ino-i-s-62-104-------------------------

Chai rman Grundy County Board Administration Building 1320 Union Street Morris, Illinois 60450 Document Control Desk-Licensing Commonwealth Edison Company 1400 Opus Place, Suite 400 Downers Grove, Illinois 60515

NRC

  • Richard Wessman Ken Dempsey Jack Kudrick
  • Jim Lyons Commonwealth Edison Bob Rybak Ross Freem~n Linda Weir
    • Harry Pal as Kevin *Ramsden Frank Spangenberg Pedro** *Wong STS., Inc.

Bill Cross I

LIST OF MEETING ATTENDEES JANUARY 21, 1997

  • I. *-
  • *. COMED Meeting withNRC Emergency T:echnical Specification-Amendment
  • ECCS Suction _Strainer Head Loss e

--- -_ January 21, 1997

.* Question 1...

Discuss Addition.al Conservatisms in the Methodology used to Calculate Peak Cladding._

Temperature r:

Additional Information Regarding.Conservatism in PCT Evaluations The' purpose of this response is to provide additional information regarding the evaluation and assessment of PCT penalties with respect to the runout flow conditions predicted in SIL 151 scenarios.

This is being provided to respond to NRC staff questions occurring during review of the Dresden amendment submitted for review on January 17, 1997. Specifically, the staff has requested that additional information be provided to facilitate a qualitative assessment of PCT margins inherent in *the methodology applied in the amendment.

Original Basis of SIL 151 This SIL was primarily focused on the potential for loss of long-term contairiment cooling due to the potential for damage to. the LPCI pumps under single failure assumptions that would.cause LPCI pump injection* to a broken recirculation piping discharge loop. The concern was that operation in cavitation conditions could cause

  • loss of the LPCI pumps and subsequent loss of the containment heat removal function.

The evaluations performed in response to this concern were reviewed and accepted by NRC staff in an SER dated January 4, 1977.. This SER eoncluded that for recirculation discharge line breaks with failure of the loop select logic causing multiple pump injection -

to the faulted loop, that "... your facilities' design provides sufficient safety margin to preclude LPCI pump damage following a LOCA due to either pump cavitation or pump.

motor overload."

It is important to note that none of the evaluations at this time were*

concerned with core spray pump operation, other than as an input to the over~ll flows used in LPCI pump runout Net Positive Suction Head (NPSH) evaluations.

I Current Assessments In the current assessments, the principal concern being addressed is the potential for the high LPCI flow rates to affect the total Core $pray pump flow.

This concern

  • surfaced as a result of review questions and investigations conducted during the recent Independent Safety Inspection. The design basis LOCA analysis for Dresden is the recirculation suction break with assumed single failure of the LPCI injection valve.* This results in core recovery and reflood based on two Core Spray pumps injecting. The
  • original calculations employed runout flows at depressurized vessel conditions of 5650 gpm per pump. The most recently reported assessments (November 6, 1996, 50.46 response) were also based on 5650 gpm per pump flow rates. Based on hydraulic characterizations of the LPCI and CS runout flows under bounding assumptions for this**

SIL 151 (recirculation discharge line break) case, a CS flow rate of greater than 5300 gpm per pump is expected.

A value of 5276 gpm per pump was utilized in an evaluation performed by the vendor, Siemens Power Corporation (SPC) for the limiting recirculation suction break and shown to result in a PCT of 2163 F.

Margin in LOCA PCT Approach The approach described above contains significant conservatisms, beyond those applied in the generation of CS pump flows under cavitation conditions. The most significant of these is.that the PCT evaluations are being performed ori the basis of a recirculation s*uction piping break. As noted above, the only break location of concern to this runout flow condition is a break of the recfrculation ctischarge piping. Discharge piping breaks.are less limiting than the suction side breaks due to the more restrictive blowdown flowpath. New break spectrum studies currently being performed indicate that a PCT difference of approximately 100 F is anticipated between these break locations, with the recirculation suction piping location bounding. Therefore, the use of recirculation suction break models to assess PCT penalties for this scenario is clearly conservative and results in additional margin in the overall assessment.

  • Question 2 Discuss the Applicability of NPSH Curves in the UFSAR with the Quad Cities SER Approval r

i Applicability of NPSH Curves in the UFSAR with the Quad Cities SER The purpose of this response is to provide additional documentation for the proposed use of 2 psi over pressure (16.7. psia) as an input for ECCS pump NPSH calculations during short term runout conditions in the initial 10 minutes following a design basis LOCA. As indicated in a previous response, the Quad Cities SER states that a few psi of containment over pressure will be needed to ensure adequate ECCS pump NPSH for a period of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> following a OBA LOCA. A comparison of key containment parameters for Dresden and Quad Cities has been provided, demonstrating that post-LOCA

  • conta.inment pressure. response can be expected to be virtually identical for these units! particularly in the short term behavior.

Additional questions have addressed the long term containment pressure NPSH curves in the Quad Cities UFSAR and the applicability of these curves to the original SER stated "few psi".

Long Term Response The Quad Cities and Dresden long term containment response curves, UFSAR Figures 6.3-41, 6.3.42, and Figure 6.3-80 respectively, generated to support ECCS pump NPSH during the post-LOCA suppression pool* heatup transient have been reviewed.

These response curves indicate very little long term.overpressure exists, based on a number of conservative assumptions, Specifically, the UFSAR discussi.on supporti.ng

. these curves indicates* that they are based on minimum* initial levels. of rion-condensibles as well as containment leakages of 5%/day (10 x the maximu*m allowable). Probably the most significant assumption applied in the generation of *these.

  • curves was the assumption that the.drywell temperature is calculated as being equarto the containment spray temperature, which is. an implicit assumption of.zero mixing of

. the break discharge fluid to the drywell. While these assumptions ~rtainly do minimize the over pressure *that would exist, especially in

  • the IC?ng term analyses, the mechanisms for minimizing the pressure would not be active in the short term cases, with the exception of the minimum hon-condensible assumptions.

ComEd believes that these very conservative assumptions underly.the SER wording of *

"several psi of overpressure", since the actual pressure would be anticipated to be.

. several psi above that predicted in this manner.

  • This has been observed in recent reanalysis previously mentioned, where long term pressures of approximately 3 psi are predicted for the same containment temperatures as originally calculated. In contrast, the UFSAR containment over p*ressure calculation Figures 6.2-19 and 6.2-16 for Dresden ahd Quad Cities respectively, d~monstrate that long term pressures of

. approximately 8 psig would exist.

These pressures are based on a model that determines drywell temperature by adding 5 F to the temperature based on assuming the break fluid mixes completely with the drywell spray flow. Discussion with GE with respect to the break flow mixing assumptions used in containment analysis have indicated that best estimate values of mixing appear to be near 40%, and that 20%

mixing is typically assumed in design applications where minimal mixing is desired.

Short Term Containment Pressure Response The short term pressure response for both Dresden and Quad Cities are attached. As can be seen, the initial response is identical up to the end of blowdown. The Dresden pressure then decays more rapidly than the Quad Cities curves: The basis of this difference is at present unknown, since the original analysis has not been recoverable in either case.

What is notable with respect to these curves is that a significant containment pressure is predicted throughout the initial 10 minutes, prior t0 initiation of containment s*pray.

While these curves are intended to* predict. the maximum containment pressure to demonstrate compliance with containment pressure limits, it is clear that over pressure conditions will exist throughout this interval.

Physically, signifieant overpressure *should continue until the quench of the drywell steam occurs

. -due to* spray initiation. Prior to spray initiation, significant fractions of the initial non-condensibles are stored in the suppression pool airspace. Based on these curves, the minimum overpressure for the interval is between 1 O and 20 psig. What has been proposed is the use of 2 psig over pressure for the ECCS pump NPSH calcul~tions being performed in this time interval, at 200 se*conds and _at 600 seconds. *The 200 second calculation point is the most significant with respect to PCT performance, since this is the time period when quench/r~flood occurs, and at which the most Core spray flow. is desired. At the 200 second point, both Dresden and Quad Cities containment anaiyses support containment *pressures in excess *of 20 psig.

New containment studies currently in progress demonstrate that with the most conservative break flow mixing, minimization of _non~condensibles in th~ drywe!I airspace, _as well as heat sink modeling consistent with CSB 6-1 guidance, that the pressure in this interval remains above* 5.5. psig. Therefore the use of minimal amounts of over pressure in the short term is conservative; particularly. in the case of the 200 second data point important to the ECCS performance evaluation:*

Additional factors that would ensure that the overpressure would not be less than 2 psig include the* current tech spec requirement of maintaining the drywell at 1 psi greater

. pressure than the suppression pool. This requirement, used to provide mitigation of suppression pool loads, would cause a higher non-condensible mass than the original calculations assumed. In addition, it should be noted that subatmospheric conditions

_ are precluded by the presence of the reactor building to suppression pool vacuum

  • breakers.

These factors as well as the fact *th~t the guillotine failure of a recirc discharge line, with assumed single failure of the LPCI injection valve causing all LPCI flow to be directed to the faulted loop, is the only scenario leading to the severe runout conditions bein*g* evaluated, ensure that a minimum assumed overpressure for the short

  • term evaluation of 2 psig is a conservatively low estimate of the actual conditions.

anticipated.

y

.!! l -

w a:

~

fl) fl) w a:

A.

.... z w

E z c rz 0

(,) '

40 240 220 INITIAL POOL TEMPERATURE 1:190°F 30 200 180 Ci:'

0 -

20

. 160 "'*

a: I

~*

c 140 c::

CONTAINMENT PRESSURE w

A.

E 15

~

120 a:

w m*

E c 100 %

(.).

CONT Al NM ENT z

PRESSURE

  • REQUIRED

~

10 FOR ADEQUATE NPSH 80 0 w a:

A.

. A.

60

~

CD 5

40 20

.O""-~~~~~~~--~--~--::--~~~~~~.......,i.-..~~~~~~~~--JC 103 10 4 105 10&

TIME AFTER ACCIDENT (sec)

QUAD CITIES STATION UNITS 1 & 2 CONT All'.'MENT PRESSURE REQUIRED AND CONT AINME.'l.'T PRESSURE AVAILABLE FOR ADEQUATE NPSH AT THE LPCI AND CORE SPRAY PUMPS FOLLOWING A DESIGN BASIS LOCA WITH AN INITIAL POOL TEMPERA nIRE OF 90° F AGUf~ C 6.3-41 i

i I

~ -

  • 1,
i l -

25 10 5

INITIAL POOL TEMPERATURE: 95°F 1

SUPPRESSION POOL TEMPERATURE (ILPCI, IS~, IHX)

~~

. MINIMUM CONTAINMENT PRESSURE ',

(WITH CONT. SPRAY)

--~----

'~

~

~~-

  • ~.

"""'---------~~ ---

CONTAINMENrPRESSURE REQUIRED FOR ADEQUATE NPSH LPCI (RHR) PUMP CORE SPfU\\ Y PUMP 200 160 Ii:'

0 -

w IC 120 j:

c IC w

80 40 t

~

OL-~~J...---1~.L.-'-'...i..u...L-~--i...,--1---i__.1._,_.L.L..L.L~~-'-~.;..L.-L-'-...1...1u...i"-'-.;...J 1a' 104 1rr 10 6

TIME AFTER ACCIDENT (sec)

QUAD cmES STATION UNITS 1 &2.

CONTAINMENT PRESSURE REQUIRED AND CONTAINMENT PRESSURE AV All.ABLE FOR ADEQUATE NPSH AT THE LPCI AND CORE SPRAY PUMPS FOLLOWING A DESIGN BASIS LOCA WITH AN INITIAL POOL TEMPERATURE OF 95° F FIGURE 6.3-42

I

. ~ -,

. ___ :::*TPIE._... llAaE7-

-....,_ -~...;.-- ------

  • CGlllPIAY--

DRESDEN STATION UNITS 2 & J MINIMUM CONTAINMENT PRESSURE AVAILARLE ANO CONTAINMENT PRESSURE REQUIRED FOR PUMP NPSH

  • FIGURE 6.3-80

. ;*.*';/

80 60 6

~

w 40 a:

~

w

. a:

A.._

z

~ 30 z

~

8 20

. 10 0

  • J
  • case A - Operation of both RHR containment coollng loops
  • Operation of one complete containment coollng

. loop

  • with containment epray Case D - Operation of one.RHR containment coollng loop with partial pumping

I

  • -------------------------~------......... -r--

II 10*

1. ONE CONT. COOllNG LOOP* J CORE SPRAYS
b. BOTH CONT' COOLING LOOPS* I CORE SPRAY
c. ON£ CONT COOllNG LOOP* I CORE SPRAY
d. ~CONT COOLING LOOP* I CORE SPRAY INITIATION OF CONTAINMlNT SPRAY 10J 10>

10*

TIME AFTER ACCIDENT CSECONDSl DRESDEN STATION UNITS i &. J PRESSURE RESPONSE TO LOSS-OF-COOLANT ACCIDENT FIGURE 6.2-19

~esponse to Question *..

The cavitation pump tests that were performed by the vendor involved setting the flow to a desired value and. reducing the pump suction pressure (per Hydraulic Institute Stan.dards -

Attachment.1 ). As the suction pressure is redUced, at some point the pump begins to cavitate, resulting in a degradation of total developed head. If the test system were to remain unchanged (no change in valve position), th.e pump flow 'would also degrade due to cavitation. To understand why this occurs, a review of the pump and system interaction is warranted.

A system is defined by the.piping and fittings that comprise the system, as well as any elevation

. changes that occur in the system. For a fixed set of valve positions, the system is also fixed and has a given system resistance curve which is quadratic in nature. Therefore, for any given flow rate through the system, the System resistance is proportional to the flow ~uared (Loss = K x Flow2). A typical system curve is provided !>Clow:

  • System Curve Flow A pump that is in a given system will always operate at the intersection of its pump curve and the system curve. This is the point at which the pump and system reach 8n equilibrium;. that is,' *at a certain flow rate the pump develops exactly enough head to match the system resistance. In order
  • to vary the flow in a given ~stem, a valve in the system can be throttled, which will result in a..
  • c~ge to the system and the 'system curve as shown below:

Varying FIQW by Throttling'

" Flow 1

I Jr the available NPSH i&u~ below the required NPSH, the.p will cavitate, resulting in a degradation ~ the developed head, i.e. the pump performanee curve changes. In other words, the pump will operate at the maximum flow that it can deliver with the available suction head and at a total head below. that which it can develop with adequate NPSH. Cavitation tests can be performed to determine how the performance curve of a particular model *and size pump changes when cavitation occurs. A typical degraded pump curve is shown below:

Degraded Pump Curve Due To cavitation Flow Since a pump will always operate at the intersection of the pump and system curves, for a given system (e.g. the test system in the cavitation report), the flow at which the cavitating pump will *

  • operate can be illustrated as:

Operation of a Cavitating Pump In a. Fixed System Flow In* can be seen from the. figure above that the degraded pump curve will intersect (reach

  • equilibrium with) the system curve at a reduced flow and head as Compared with the original pump curve. Essentially, the pump will deliver as much flow as it can given the available NPSH. It is the equilibrium point between the pump curve and the system curve that determines the reduced head at which t~e pump*will operate.

2

Since the intent of the !talion test is to maintain a fixed flow, th'lstem throttle valve must be throttled further open, re5ulting in a lower system resistance, thus returning the pump/system combination to the original higher flow:

Throttling of System to Maintain Fixed Flow of Csvitating Pump Flow

. Since the LPCl/Core Spray systems post-accident cannot be throttled in the short-term (< 600 second~). their system curves remain fixed. Therefore, when the pump cavitates, the pump curve will degrade and intersect the system curve at a reduced head and flow as described in the discussion above. It must be emphasized that pump will operate. at this point of intersection

  • .. because it is here that the pump and system reach equilibrium. It is at this reduced flow that the reduced head produced by the pump equals the system resistance.
  • As the NPSH available to the pump is reduced further, the pump curve degrades further and will intersect the fixed system curve at an even lower flow(~ Fig. 30 below). It is the purp9se of the proposed NPSH calculation to determine at what ~ow the *pump will operate given a cer:taffi available NPSH. For the purpose of this calculation, the pump curve degradation is assumed to be perfectly vertical* starting at ihe point of initial head collapse. Since the pump degradation is not.

perfectly vertical, this assumption results in the absolute miniriium flow that the pump is expected

~to deliver. Since this NPSH calculation uses this conservative assumption, there is no further flow penalty to be assessed in determining the anticipated flow reduction due to cavitation. It should be rioted *that this methodology was disaJssed in detail with the pump vendor and with independent pump consultants; All parties involved agreed with the methodology presented in the proposed NPSH calculation.

250 A 9

- 200 8

d 7 t:

c 6.i

!I! t50

..J 5 CL z

§ 100 4

3 50 2

O 20 40 IO IO 100 GPM FIC. 30 Test of a 1.5-ia (S.81-cm) single-stage pump at

$470 rpm, on water, 700F (21°C) (ft X 0.3048

  • m; gpm X 0.06309
  • l/si (~. 12).

3

~

llTillU llUyUI JIJlllP

  • ~ f

,~

  • . * ~

test sranda*

  • Fie. 49A CLOSED LOOP SET*UP HORIZONTAL PUMP ever, when.the suction condition is specified as 0 ft.

NPSH at the suction datum elevation, the test must be performed by reducing the distance between the elevation at the impeller eye entrance and the datum elevation through reduction of discharge column

. length, or removal of pump series stages. The test results must then reference the difference between-.

test datum and application datum elevation.

Vertical pumps for free surface applications can, if practical, be tested in a deep sump in which the liquid level cari be varied to establish the desired suction lift or N PSH requirements. A more desir~ble method is to locate the bowl assembly (first stage only for multi-stage pumps) in a suitable suction can or tank in which the pressure can be regulated and reduced to the desired level to meet the test cri*

teria. See Fig. 498 for caVitation teSt set.up of ver* * *.

tical pump.

For large pumps cavitation testing may, for prac-tical reasons, be performed on models. Reference is

.made to th~ section on model testing at the end of the Test Standard.

. Determination of Limiting Suction Requirements The suction requirements to be met by a pump a~e defined by the cavitation coeffidents, sigma (o) as determined by the specified field conditions. Sigma

' iS defined as:

  • .NPSHA O*

H*

  • where NPSHA = Net positive suction head available, as defined on page 71 in feet H
  • Total pump head per stage in feet 74 VACUUM PUMP v

v Fie. 498 CLOSED LOOP SET *UP VERTICAL PUMP The cavitation characteristics of a pump can be determined by one of the following procedures:

  • Using one of the test arrangements shown, the pump may be run at cpnstant capacity and speed.

with the suCtion condition var_ied to produce cavi*

tation. Plots of head, efficiency, and power input shall be made as shown in Fig. 50. For the higher.

values of sigma (o), the values of head (H), effi~

ciency (fJ), and horsepower (bhp) should remain

~substantially constant. As sigma (CJ) is reduced, a point is reached.where the curves break away from this trend, indicating a con*dition under which the performance of the pl.imp may be impaired, the degree of which will depend upon the specific speed, size and service of the pump, and the char-acteristics of the liquid. Fig. 51 shows results typical of tests for sigma (o) at capacities both greater and less than normal.

One alternate technique for determining the cavitation characteristics is to hold the *speed and sUction pressure (pJ constant, and to vary the ca-pacity. For any given suction pressure, the pump head may be plotted against capacity. A series of

  • such tests will result in a family of curves, as shown in Fig. 52. Where the curve for any suction pressure (pJ breaks away from the envelope, cavi-tation occurs. Sigma Col may be calculated at the break-away points.

~:

r ii --

fla. 50.

PLOTS Of HEADS,

. ~iCIENCY AND POWER INPUT

. VS SIGMA

  • . P'je. ~

. --Q1 r_

oa r

Qa Fla. 51 SIGMA AT VARIOUS CAPACmES

. \\

\\

\\

I I

I 0

I I

I I

I I

I I

I Pal 0

I I

I P.z Pal I

I P.t Pd fla. 52 ILLUSTRATION OF

.HEAD BREAKDOWN UNDER CAVITATION

  • Another alternate technique js tD hold the speed

( constant and run head-capacity tests at severai different suction throttle valve 5ettings, bracket*

ing the suction condition point. A series of such.

tests will result in a family of curves similar to those obtained above from which sigma (o) may be calculated at the break-away points.

  • Accurate determination of the start of cavitation, i.e., the cavitation point,.requires very careful con-trol of all factors which* influence the* operation of the pump. A nu~ber of test points bracketins the point of change must be taken, ~nd the data plotted to determine when the performance ~ks away fri>m normal. Any chanie in performance, either a drOp in head, power or efficiency at a given capacity, or a chanse in sound or vibration may be an indica*

tion of cavitation, but because of the difficulty in determinins just when the change starts, a drop in head of 3 per cent is usually accepted as evidence.

that cavitation is present.

_When testins with water, an accurate temperature measur~ment usually is sufficient to establish the vapc)r pressure, but the degree of aeration of the

. water may have a considerable influence on perfor-

.. mance. Consistent results are more readily obtained when the water is effectively de'aerated. When test-

. ing with other fluids such as hydrocarbons, a vapor

. pressure bulb in the silction line not far from the

I ATIACHMENT2

  • Some of the conservatisms Used in the NPSH cal~tions perfonned for Dresden are listed below. In order to quantify these value5, the amount of conservatism is expressed in terms of a "Flow Penalty" that was applied to the estimated Core Spray pump flow. For example, the use of a conservative suppreSsion pool temperature resulted in a Core Spray pump flow reduction of about 100 gpm.
  • PARAMETER CONSERVArtSM CS FLOW PENALTY Temperature Recent GE ~alysis indicates pool temperature at 200 100 gpm seconds is about 10°F lower than Quad Cities pool
  • tern erature that was used in the calculation 118°F v. 129°F Strafuer plugging Sensitivity analyses were perfonned that detennined the 60 gpm
  • worst-case strainer to be blocked. (Flow penalty based on com arison of worst to best blocked strainer Strainer plugging One fully blocked strainer was assumed (three clean 260 gpm.

strainers); a more realistic assumption would have beeit to

  • uall tu all four *strainers 25%.

Reduced NPSHR The NPSH resulting in initial head collapse was selected froin 50 gpm the cavitation test report as opposed to the NPSH that results.

in.total head colla se.

  • Pump Flows LPCI runout pump flows. based on calculations of system ' _
  • 40 gpm
  • resistance' that a5sume clean commercial steel pipe (no account for i e
  • n

\\...

Retrieved from "https://kanterella.com/w/index.php?title=ML17187A866&oldid=5131742"