ML20056B371

Safety Evaluation Supporting Amend 42 to License NPF-39

ML20056B371
Person / Time
Site:	Limerick
Issue date:	08/16/1990
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20056B370	List: ML20056B369 ML20056B371
References
NUDOCS 9008280193
Download: ML20056B371 (10)

Text

f*,"%

4 Io UNITED STATES (g

NUCLE AR REGULATORY COMMISSION y,

(

g r, e '

/-

W ASHING TON, D. C. 20$55 b..Y.h'al e,

• SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SUPPORTING AMENDMENT NO. 42 TO FACILITY OPERATING LICENSE NO. NPF-39 PHILADELPHIA ELECTRIC COMPANY LIMERICK GENERATING STATION, UNIT 1 DOCKET NO. 50-352 1.0 IllTR000CTION By letter dated June 14, 1990, Philadelphia Electri: Compan* i -

licensee) reauested an amendment to Facility Operating Lice 3 % y J'

for the Limerick Generating Station,(TSs) to add new isolatic.

W.

Unit 1.

The proposed ai m e - c.N revise'the Technical Specifications each common Control Rod Drive header to the table of primary c o u 4 isolation valves that must be operable and to delete the existing individual Hydraulic Control Unit isolation valves from the TSs.

Limerick, Unit 1, is scheduled to shutdown for the third refueling outage on September 8, 1990. During the refueling outage, the licensee plans to install eight new check valves (four pairs of valves) on the control rod drive (CRD) supply headers to the hydraulic control units (HCUs). These new valves will constitute a new isolation boundary for the Integrated Leak Rate Test (ILRT), replacing the existing HCU isolation boundary valves. The proposed changes to the TS are to include the new valves in Table 3.6.3-1, "Part A - Primary Containment Isolation Valves" i

and to remove the current numbers for the HCU boundaries.

2.0 DISCUSSION i

The Integrated Leak Rate Test (ILRT) is a pressurization of primary containment and measurement of total leakage from all isolation boundaries. The current method of testing the isolation boundaries of the CRD system is to collect leakage through the HCUs at the vent valves on each of four supply headers during an ILRT.

If the total leakage exceeds specified limits, approximately 1300 individual check valves or solenoid valves must be examined to find and repair leak paths.

To minimize critical path outage time, new check valves will be installed in the CRD supply headers in four locations, effectively extending the isolation boundar These eight new valves (4 pairs) y from the HCUs to these new valves.

will reduce the number of testable CRD penetrations from approximately 1300 to four.

The licensee included a sketch of the proposed arrangement which is enclosed (Figure 1). As shown, the licensee proposes to install two lift check valves, a manually controlled block valve and two test connections on each of the four CRD headers to the HCUs (drive, cooling, charging and exhaust). The general process diagram for the CRD hydraulic system is shown in figure 4.6.7 in the Limerick Final Safety Analysis Report (FSAR) which is also enclosed. The new valves will be installed between the main control station and the existing vent valve shown in Figure 1.

9008280193 900816 PDR ADOCK 05000352 P

PNU m

i i

i l \\

In a BWR 4 such as Limerick, Units 1 and 2, there are 185 control rods, each of which has a HCU. As shown on Figure 4.6.8 in the FSAR (enclosure 3), and in Figure 1, there are 3 check valves and 4 directional control valves on each HCU for a total of 1295 valves. The Integrated Leak Rate lest (ILRT) is ? pressurization of primary containment and measurement of total leakage from all isolation boundaries. With the present arrangement, the 1295 valves constitute the isolation boundary. The current method of i

testing the isolation boundary of the CRD system is to place a bucket under the vent valves shown in Figure 1 and to collect. leakage through the HCUs at the vent valves on each of the four supply headers during an ILRT.

If the total leakage exceeds specified limits, approximately 1300 individual check valves or solenoid valves must be examined to find and repair leak paths.

Experience has shown that the total leakage is due to a drop or two of water leakbge from many of the ball check valves rather than failure of one or two valves.

To avoid these problems, the licensee proposed to relocate the isolation boundary on the headers to and from the HCUs rather than on the HCOs.

2 Although a TS change is not required prior to installation of the new i

valves, a TS change is required to take credit for these new isolation boundaries, and also to remove the current valve numbers for the HCU boundaries from the TS. Therefore, the licensee is proposing that TS Table 3.3-1, "Part A-Primary Containment Isolation Valves," be revised to remove the existing HCU isolation boundary valves and replace them with the newly installed isolation boundary valves. Note 12 of that table also has to be revised to reflect the addition of the new valves. Also since the affected CRD lines are water filled and would remain water filled for a minimum of thirty days after a loss of Coolant Accident (LOCA), Note 22 applies to these isolation valves.

3.0 EVALUATION 1

The proposed TS change will take credit for the new valves installed in each of the CRD headers to the HCUs (drive, cooling, charging, exhaust) between the main control station and the vent valve. These valves l

L constitute a new isolation boundary.

Each check valve station consists l

of two check valves, a block valve and two test connections. This enables each d.eck valve to be tested individually instead of during the critical path ILRT.

Each check valve station is accessible from an existing platform near the existing vent valves. The charce will move the isolation boundary out on the CRD headers.

The licensee has provided analyses to demonstrate that the new design does not change the design criteria previously approved in the staff's Safety Evaluation Report (SER), NUriG-0991, Section 6.2.4.1.

The present method of leakage monitoriag was accepted by the staff in Section 6.2.6.3 of the SER.

i

6

.. The piping to be included within the new isolation boundary complies with the same standards and specifications as the original boundary. The number of active components making up the boundary will be reduced from approximctely 1300 to four. The current CRD isolation boundary includes the insert and withdraw lines, the scram discharge volume and the HCus.

The relocation of the boundary will add some of the supply header pip Pg but will not affect the existing equinment. The added piping is small diameter (2" or less) comparable to tie previously analyzed scram discharge drain line. The consequences of a pipe failure inside the isolation boundary remain within the envelope analv na in NUREG 0803.

An analysis has been performed on the piping oeing upgraded for inclusion in the extended I!.RT boundary. The pipirg and related pipe supports are designed to meet the criteria of Seismic Category I and ASME Code Section III, Class 2 or 3 as appropriate. Analysis has shown that the existing piping and the modified piping is within the ASME Code allowables.

Piping supports have been evaluated and modified as necessary to accommodate the newly analyzed loads.

The licensee has evaluated the hydraulic effect of the new valves. The additional pressure drop will not introduce significant line loss and is well within the CRD pump capacity. The calculations show that there will be adequate flow in each of the headers to meet required design flows. The performance of the CRD system is well within the system capability for normal operation, and control rod scrtm performance is unaffected.

The testing of the new valves uses techniques and criteria accepted for i

other similar applications as documented in the Limerick Final Safety Analysis Report (FSAR) Table 6.2-25. " Containment Penetrations - Compliance with-10 CFR, Part 50, Appendix J," Note 14.

The staff has evaluated the potential if the lift check valves fail closed, blocking one of the headers. The new check valves will be similar to the present check valves shown in Figure 1 on the cooling, charging and drive headers, but located farther upstream. The licensee plans to " blue" the seats for a polished surface.

Experience has demonstrated that unless the lift mechanism is distorted by temperature, becomes severely corroded or is coated by gunny organics in the line, the valve is highly unlikely to fail closed.

In any case, even if one of the check valves were to fail closed, it would not affect the ability to scram the reactor.

As discussed in the F3AR, at system pressure above 600 psig, reactor pressure provides adequate energy to insert the control rods without the assistance of the accummulators. At low or zero reactor pressure, the accumulators provide the energy to scram the rods.

A typical curve of scram time vs reactor pressure is shown in enclosure 4.

This was verified as p(art of' the Limerick Unit 2 startup test program as shown in enclosure 5 letter from G.M. Leitch to HRC dated A transmitting Limerick Unit 2 startup test report.) pril 2, 1990 If a valve failed a

p s

l e

i

-4 dosed on'the charging water header, the pressure in one or more accumulators might bleed down if one could postulate a leakage path.

If the pressure decreased to 955 psig, the alarm setpoint, this is i

annunciated in the control room. When a HCU accumulator alarm condition occurs, the Main Control Roun (MCR) rean:or operators receive a flashing accumulator trouble alarm indicathn on the Full Core Display panel in the hCR for the specific HCU (pane? *00600). The reactor operator must examine the Full Core Display to identify the specific HCU accumulator that is in alarm, Any alarmed accuou utor trouble alarm on the Full Core Display will flash until the reactor operator acknowledges the alarm on a i

specific eccumulator trouble alarm ackr.wledge button on the reactor console (panel *00603). This alarm condition is accom 1

b and flashing annunciator alarm, " Accumulator Trouble "panied by an audible O

in-the MCR. The Reactor Operator must acknowledge the alarm on a gene,ral annunciator I

acknowledge button to silence the alarm noise and stop the flashing alarm window.

If a second HCU accumulator alarm is received after the first alarm is acknowledged, the annunciator re-alarms and the operator must again acknowledge the alarm to silence the alarm neise and stop the flashing alarm window. The second HCU accumulator trouble alarm also flashes on the Full Core Disphy. This sequence is the samc for multiple HCU ac;umulator alarms. Therefore, adequate MCR indication exists for

'the operator to be alerted to multiple HCU accumulator trouble alarms.

If more than one accumulator is inoperable and if reactor pressure is less than 900 psig, the TSs require that the reactor mode switch be 4

placed in the shutdown position. Thus, if a check valve on the charging water line failed closed and this somehow allowed the pressure in one or more HCU accumulators to bleed down-the operators would have ample i

warning of the condition and the plant would be shutdown while there w N1d be sufficient reactor pressure to scram the rods.

If a check valve failed closed on the cooling water header, the lack of flow would be alarmed in the control room, followed by high temperature alarms. The high temperatures in the drives would require plant shutdown but would not have any effect on screm capability.

The drive and exhaust headers are only used to move control rods.

During a scram, the water is discharged to the scram discharge header, not the CRD exhaust line.

• f either of these lines were blocked by a check valve failing closec, ie operators would not be able to perform the weekly exercise test.

sry Sunday, the operators move each control rod lu and out one notch (6-) to demonstrate operability. This would not effect the ability to scram the rods.

In summary, if a check valve on one of the CRD headers failed closed, it would not effect the ability to scram the rods.

The staff has reviewed the licensee's analyses and found them sati: factory.

The proposed TS changes to relocate the isolation boundary are acceptable.

t

h l

p i

4.0 ENVIRONMENTAL CONSIDERATION

This amendment involves a change to a requirement with respect to Liie i

installation or use of a facility component located withir, the restricted area as defined in 10 CFR Part 20. The staff has determined that the amendment involves no significant increase in the amounts, and no i

significant rHnge in the types, of any effluents that may be released

-(

offsite and that there is no significant increase in individual or cumulative occupational radiation exposure. The Commisslon has previously issued a proposed finding that this amendment involves no significant i

hazards consideration and there has been no public coment on such finding. Accordingly, this amenament meets the eli ibility criteria for

'3 categorical exclusion set forth in 10 CFR 51.72.(c) 9).

Pursuant to 10 CFR 51.22(b), no environmental impact statement nor environmental assessment need be prepared in connection with the issuance of this amendment.

5.0 CONCLUSION

The Comission made a proposed determination that the amendment involves:

no significant hazerds consideration which was published in the Federal Register (55 FR 26290) on June 27, 1990 and consulte( with the State of Pennsylvania.

Po public comente 1ere received ani *he Commonwealth of Pennsylvania did not have any comNnts.

The staff hat. concluded, based on the consideration discussed above, that: (1) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, and (2) such activities will be conducted in compliance with the Comission's regulations and the issuance of this amendment will not be inimical to the comon defense and the security nor to the healthy and safety of the public.

Principal Contributor: Dick Clark Dated:

August 16, 1990 i

I

' ' 'f (

~

m

'y

.h

}

E a

3,.

j g,

r.

T 21 Q

1-

_v g

of a4 id a

j wy R

=

='%&

wm k __

b n-o

sf 3

}.___-

l g

a4Iw-

. d '"~ ' "

w

<2 o

' y, 4 4

U

• y J.

5 p<

.S g

{-

gW-

~~

+

, 4 -=

tO*

b O dt p

a s=

a N

3

~ 32 u)

=

4 g

e o

}.

a)

]

h

-h o

y e

o 2

h 1

~

u w

3

(

w e

E W

e g

b SI k

u a4 i

L,:

,, i K

.J p

=

z

_ g.___ w _.

.. ~.t E

.u

_ -.mmm L

h

> s k

O v

g m

.&re4ma /

e

= _ _ _ _ _ _ _ _. -....

-i!.i!

n l

I!I e

p1,le;>!i1190 l l.l g. 5

iiii, it '

. pin.nnivy i

n.,,

,1 a i

frII!E I I3 III

• )

te 8

i' "I llI gNB

~

i Ijlit 2 jgf l

g jl!Tlllild!!!!!!!!]j

{-

l i

!!!!S!!!!,

!pu'"ll !ii;;="!r="=!;; iirm;ll!!;m! I!'Il8E M

l m.

iiii g

. n zl!!!i s!!! 11;

,I 2!

, m p

ari sii i!!illi

-I u

3 i-si

..q.

u..

, a,,

i g kr

f

'r Yl' I

h a

Jf=

g y

e l

'i i

.I l

I l

a

(

n b

. ~,,..,

p e

l n

cq T@-

N ; *1!

I 1

~~~L!!!.

S l

, q l,ij!hl.

t 7..

y W

egi g,

i I

.a i

I 5

e l

c-3 h

g I

4 ll--

J NDJ.

l k

kD t

d?

i i

(fjp w,li s : : : :.h..!:, :

n se 6 88:

I l

a[ s s A,

(,y @

e G : l i-r v

j E"j

ade, p nui sli:b.

i, a

,j iu;!

19 p.

5 g

' j igy$..

_3 gjjj---m i.

h.!!Il Inr -

b c e^

I h:

e l

1 ISOLATION VA LVE SCRAM VALVE PlLOT AIR CHARGlNG WATER RISER l

t EXMAUST WATER RI ISOLATION VALVE

[-

g WITHORAWAL RISER

t. -

- ISOLATION VALVE

. ISOLATION VALVE E M L}

N 5 ORIVE WATER RISER INSE RT RISER ISOLATION VALVE -

m SCKAM PILOT VALVE ASSEMSLY.

COOL'40 WATER RISER

_ O' f

,/

/

OUTLET SCR AM VALVE

/

ISOLATION VALVE SCR AM OISCHARGE RISER l

r

. INLET SCRAM VALVE f

Y a

h DIRECTIONAL CONTROL VALVE (WITHORAW)

OIRECTIONAL CONTROL VALVE N

be

'.,' es S E P fl

\\

OIRECTIONAL CONTROL VALVE (INSERT)

MAP.sFOLO - -

oha:.L vat <<,

" #' W OIRECTION AL CONTRO'. <ALVE #

p (WITHDR AW ANO SE 'TLE) -

SHUTO7FVALVE WATER ACCUMULATOR ORAIN SCRAM ACCUMULATOR N CYLINDER 2

SCRAM WATER ACCURiLAroR #

i-

'l b

\\'" #/)

l-ACCUMULATOR N i

2 PRESSURE l

FAAME INDICATOR

=== CARTRIDGE VALVE ACCUMULATOR N, CHARGING ACCUMULATOR INSTRUMENTATION I

t ASSEM8LY k

LIMERICK GENERATING STATION UNITS 1 AND 2 l

h,

• lNAL SAFETY ANALYSl4 REPORT Sd*

l.

i CONTROL ROO DRtVE HYDRAULIC CONTROL UNIT FIGURE 4M istchse/rn 0

.4 64 =

U

(

l' J

l-l l

~'

Ao e

44 c*

.t VegethPARWWRtONbV

=

IAccWesukAf0M DISCMAA458)

. h.

4 44 t

~

m'

$J

,f l h, g ACCutsVLATOA PAE8EJRt ONLY %

NESBSLPOR75 PLVOCEDI icNAA444G waftM VALVED OUT)

%,6

=-

e i

h 9* m %

s gg m,

fvMCALSMtVE 34

=

=

{s 1

I

-,.. L I

_L I

e m

.=

'48 nio eme im l-naActoa vassai. mamune emf 88 -

L V

Figure 9 w

E 4 se f

-1 e

i 2STP~5.4, Scram Testing of Selected Rods From the results of previous CRD testing, the four rods with the slowest scram times to position 05 or with unusual operating characteristics-were selected for further teisting.

This subtest was performed at the following test conditions:

at zero reactor pressure with accumulator presnure just above the low pressure alarm point, at 600 psig reactor pressure with normal

=

accumulator psessure, and at 800 psig reactor pressure with normal accumulator pressure.

Each control rod was scrammed three times at

.-=

every test condition.

There were no acceptance criteria verified in this subtest, but aach selected rod scram time was verified against Technical Specifications (TS) when performing this subtest during Test Condition Heatup.

The scram times of selected control rods are as follows:

Measured Time to Position 05 (see)

O psig 600 psig 800 psig TS Limit Selected Rod (TC OV)

(TC HU)

(TC HU)

(sec) 22-11 1.78 2.33 2.47

<7 1.55 2.56 2.71 77

~

1.56 2.32 2.58 57 26-11 1.65

< 7 1.69 77 1.66 57 18-39 1,71

<7 l.69 57 1.71

<7 26-47 1.65

<7 1.69 77 1.69 57 26-23 2.65 2.64

<7 2.52 2.79 E7 2.53 2.75 57 30-15 2.36 2.51

<7 2.34 2.32 E7 2.38 2.39 57 34-07 2.32 2.32

<7 2.30 2.39 77 2.28 2.33 57 a

=

.4 4-74 a

sm-------i-mi- -

--ie iiii ismi i

i

"