ML20115H511

Operating Procedure OP-402 FPC Crystal River Unit 3 Makeup & Purification Sys, Rev 75

ML20115H511
Person / Time
Site:	Crystal River
Issue date:	05/23/1994
From:	FLORIDA POWER CORP.
To:
Shared Package
ML20115H481	List: ML20115H511 ML20115H514 ML20115H522
References
PROC-940523, NUDOCS 9607230165
Download: ML20115H511 (13)

Text

_ ___ __ . . . _ . _ _ - . _ . _ _ . . _ _ _ _

i hev. 73

! rFe=t1. est. s2Is/w l

t 1

(

1 i

)

i 4 OP D AT!M PROCG W E I OP-401 i

FLORIM POUR CORPORATIM

! CRTSTAL RIVR WIT 3 i

i

!'i IW3UP Am Pa1FICATICII SYSTM -

I TNIS PROCIMAE m SAFETT RELATB m i THIS PROCGWE A00RESSES Elit!ROIBWifALLY ORALIFIS (ES) CONOMES i

l i

l l

i s

1 i

i APPROVED BY: ten Cantact

' a~

) ,

(5 LElun IEORFILE)

DATE: d11/8FV ,

^

IE ERPRETATION CO EACT: Superviser, nuclear Operations Adelaistrative Shift 9607230165 960710 PDR ADOCK 05000302 O PDR

a J .

I

' S W E BLEER 4.4 i

ACTIONS OETAILS l

I NBN: Changes in purificaties lies ups uhtle en Decay Neststen j offest vessel level and the aparatten of the Decay Nest

System. -

i 1

4.4.1 Select *8LEED MINIE' e ,RCST-3A se)ector switch to desired e ,RCET 38

! - RC Bleed Tank e _ EST 3C

! E poll Mendle Up j /

i latttal/Date 4.4.1 Select M 112 Centrol

switch to BLEED /

i . Initial /Date 1

l

? -

4.4.3 m 1817 decreases to les

1evel alars, '

i 3 desired letdeun is j cespleted, j

select M 112 to .

! la,tt1al/Date 4

l

}

i 4.4.4 place 'tLEED feDE' selector switch ta the pushed ta l

, posities /

i 1sitsal/Date 4

i '"

l i '

i I

1 -

1 op-est new. 75 Page 35 i '

i i

h 1

4.s EmL5m v

i ACTIMS DETAILS NBfts Changes la puriflestica lies ops dile en Besty Nest een j effett vessel level end the operetten of the Besat Ilest j System. g 1.

i ,

4.5.1 Deterates recetred essent e. Refer to OP-304

• j of Feed '

a

' , /

j '

lettsal/D. I s

t j

4.5.2 Align Batch Centre 11er 1. _ Select ' Start / Step' evitch u STOP

! Repress 'Cleer' puehbetten 1 Adjust ' Batch Stas' i

thebertsch to destred flew

_ Ataust "Preshutdeus j thusbeuttet to desired

! 2. ,,,,,preshutdemo setestetSelect ' Sta 4

START l

! / l l

?

lattial/Date 4

4.5.3 5eler'c FSID MO0t* selector i switch te desired feed -

leurce A g Pull Handle up /

j latttal/Date i

4.5.4 E selected feed seerte is o. ,, OPS CAV 57  !

3 a 8457, e .,,,,STA N CAP-1A l

j 2 B Open discharge valve enT start en CAP-a e _ START CAP-18 1 /

j ,

letttal/Date i

4 i 4.5.5 Place both postft1ters is e Open postfilter tselatten 4

service valves 1 lWy.93 _ 1g3y.91

,,,,, Illy 97 _ IEN.96

/

letttal/Date OP 401 Rev. 75 Pope 34 s

1 i

i .

1 1

}

i -

J i

4.5 M WN ACTIONS i

1 1 Open m and l 4.5.5 Portficatten Feed POV 1 .

l 4,g,y Adjust flew nte estag

, VALVE LOADINE CONTROL en I the estch Centre 11er to i

destred fles 1

1 2

4.5.0 legg m imot mactes 3

- 7ess red level.

M CLOSE ItN 103 l

4.5.9 57 uns esened.

CLOSE CAV 57 l STOP rematag CAP e

I l

~

j 1

4.5.10 Piece the 'FIED NDOE' I se)ector sutsch ta the pushed ta pasttica ,

4.5.11 Destore postfilter lineup as desired l

l OP-402 i

i

1 I i i l

l 4.19 8 '""" ~ " " * " l i ACT!tull [

l j  !

f 4.19.1 M trod Q addNien is 80 TO Step 4.19.8 saattmus with next step usat salt to MIDT ,

j .

f I

4.19.2 Essere se other drainig er weattee operettens are ta progress to the Waste Gas system j  ;

t 4.19.3 Select WT 1A OR desired Waste las Decay Tant as

- directed by Chem. Dept.

l perfore Valve Alignment 4.19.4 for venttes part l

l -

I 4.19.5 West fulf 1

I .

4.19.6 'dB appreziastely 2 eTastes have elapsed.

3 STOP TOP 1A(IS)

=

4.13 IRK _VINffME AS RAS AIBIT115 (Cent'd)

Actions DETAILS 4.19.7 Remove Waste Sas Decar e F1 Iglector switch te Trat selected in Im StaP 4.19.3 from service .e ace eestred tank ta service l ste '

F

..i.it.'i.'f.f'"re"s"la.l:

c Pegar

!- i.1 1a,'  :

i 4.19.8 Estabitsa N, pressere ta 1. , tefer to Cerve 8 of OP 1038 l M if sestres, othenrtse for mastems saR

{ 1. ME. NCB Centre)

3. IER is at desired the following:

I Asistal/pata 4.19.9 1[ add 1tien with the 1. ,,,,,,, Detemtas assissa IER Nors*tNN$1ewteg 5E WA 1. ,N Y)

Locaj open IRN.4gt,

3. _ b I N NCS

- e. ,1 re : r,, m<ne itett ts not

5. Close 18N 143

6. ."."'.". Locally close IEN.4

/

Anttsal/pate 4.19.10 , saare is 1. _ Deterstne ansteum IRR

' Norsthe e [

$!HvA w

1:=_@7 4.

Ffti1" Nil' estre while ensu. - t of ring IRR store Itatt is not

5. Close IRN 341

6. , Locally close IEN

/

Insttal/Date OP-401 Rev. 75 Page 78

1 i

To8 DFIELDS l Fren: GRALHOW -

i

• subject: Journal l Date: 08-09-94 Time: 2:00p 4

To: . JOURNAL ,

j ,

There have been some questions regarding the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> rule. It is still

alive and well and takes the DNPO approval to exceed 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. Emmeeding i 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> will be sensidered if no viable alternative can be found. I instructed clint to attempt to schedule such that we do not exoned 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> i for a 7 day period. If in doing this, it will require scheduling a person j

to work hours that he/she would normally be sleeping (is the fatigue factor would be high) to attempt an overtime deviation.authorisation. The point i

j is it is better to keep someone working.a set sabedule one extra day or a few extra beurs than to take someone who would neraally be sleeping on that day and ask him/her to work. When attempting to mahadMe around the 72 i

hours, all attempts will be made to get the eeneurrence of the personnel involved. It is not the point to limit the total number of hours, since it will probably wash out as either premium time or overtime anyway. The k

j point is to not ask our personnel to work more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> in a seven day '

period because that is where we set the limit on where we are willing to

push the human body. Admittingly, some people can handle the 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and i

more and soon get more fatigued. Rather than try and label our personnel.  !

as to their physical / mental agility after 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, we set the standard.

! We will be making a change in the clerk's shifts. Andrew will rotate onto days and Marlene onto mies. Tttis change will be effective this weekend.

{.

' The reasons behind the serdlule change are several but suffice it to say it will provide some cross trairaag w r .anhities, help put some emphasis on department trends, and help to alleviate some issues that will work out for the best for operations as a whole. Andrew will start learning the schedule which Marlene has built into an excellent tool for eur people. -

l Marlene will perform the duties of the mid shift and all that is man.==sary j to maintain department re ords and plant status. she will also help me 4 partora some detailed trending of key indicators in ops. I hope ovaryone can make this transition a smooth one and give both your best support.

l There is an ongoing discussion on the issue of R2 pressure in the NUT. I have talked to Mark v. and understand the concerns. I discussed this with Bruce and . Terry campbell and will be working.to resolve the questions. It is important, and truce has asked, if anyone has a sonoern, please write it down and send it to me. Even if it is not now, I need to get all

perspectives of this issue so we can address the right areas. It appears 1

j what has been addressed has not satisfied the concerns to date, so I need to be absolutely clear on the questions.

j ............................................................................

4 ..........

nob Murray has volunteered to look, in detail, at the deleted computer alarm summary. The purpose is to* do away with unraman an=*y alarms so that the real ones are not diluted beyond recognition. No will need same support in being I asked him to get at least a second opinion before he goes to barry Baumgardner to get the points deleted. I also asked'that I be kept informed, not as an approval, just so I can keep aprised of the

progress. Its a good project and I commend those who questioned the need

and are taking action to correct it.

A concerted effort will be undertakan to do away with " operator work-arounds".

4. _ . .

I need your input. Please send me any and all comments on

.! -4 . . - 6... a ww4

. .a44 4 I

4 l

}

I

\

( l

! GHH_

j e ,'

Subject:

c.ammary of actions required to implement the new j make-up tank hydrogen cover gas pressure limits (Refa Cale. 1-90-0024) such that a 25cc/kg hydrogen concentration can be maintained in the RCS.

NorE: There are two separate issues concerning the potential.*

l damage to make-up pumps resulting irom an increase in make-up '

oever gas pressure. These concerns are as follows:

l 1. During normal operations, will raising the make-up tank oever gas pressure to the limit specified in omiculation I !

0024, cause hydrogsn gas to some out of solution in the make-up

! peep suction piping and onllect in the high point of the static

pertions of thLs piping. The aaaamm is, if a buildup of gas i occurs in the BNST piping common to the normally running make-4 up pump, upon an NPI actuation this gas would be transferred te j either pump in the common section header and potentially cause j pump damage (Ref IN 88-023 including supplements 1 and 3).

2. At some point during a racA will the BNST head be overcome by make-up tank head pressure. This tion could

potentially occur as the BNST and make-up tank inventories are 4

depleted to a point eventually allowing make-up tank cover gas

! to anter the NPI pump suction piping common to the make-up taak j (Ref: IN 88-023 Supplement i 2).

i Actions required for item 1 above:

)

I a. Per the Nuclear Operations Engineering response to NRc IN l 88-23, including Supplements 1 and 3, the potential for j hydrogen gas to cellect in the static piping of the non-j running make-up pump may exist at CR3.

4 j Per an NPSE walk-down of the make-up pump suction piping, i high points do exists in the BNST piping that can trap l gas. NPSE will view the high point sections of piping with j an intrared scope to determine if any gas is currently

! present. If this method of gas detection is deemed i

! reliable, it will be used to monitor these high points sa i a routine bases following the implementation of the new l make-up tank cover gas pressure limits. If this method is j not considered reliable, than samples will be taken from-high point vents. chemistry and NPSE will develop a matbed to quantifying the trapped gas. Routine sampling will then l l be performed until it is determined that a problem exista.

If a gas buildup is occurring, routine venting of the high point vents will be required.

NPSE will perform the infrared analysis on 10/27/92. ,"

l i .

l Actior:s required fue iten 2 above:

l a. cceplete the review of GC calculations I-90-0024. The i ihmlear Operations Technical Advisors are performing a

! final review of the subject calculation. per conversation

with Bill stephenson, this review will be completed 'by j October 23, 1992. o ,

! b. Following the above review and subsequent acceptance by

! the NOTA's, a revision to 0P-1035 -(ourve GS) is required.

curve f 8 is the maximum make-up tank pressure vs. indicated water level ourve.

c. calculation I-92-0024 rwa that only one of the two make-up pumps normally lined up to the make-up tank, i.e.

IETP-1A or it, he operatimi during a Iach when the piggyback mode most be estabinshed. When this analysis was performed, all three pumps sould be operating at this switch over point however, recent changes to the Engineered safeguards Actuation procedure (AP-380 now requires that only two of the three pumps are run)ning.

operations should review AP-380 considering t%e.make-up tank pressure concerns for any additions t changes required.

d. calculation I-92-0024 provides the data required to produce a new make-up tank level vs oever gas pressure curve. The curve will represents aooeptable normal plant '

operations that do not endanger make-up (EFI) pump operation in the event of a IacA. This serve does not address the hydrogen oever gas pressure required to i maintain the 25 cc/kg RCS concentration. A caloulations  :

currently being performed by Rocky Thompson will determias  !

the actual make-up tank hydrogen oever gas pressure required to reach 25 oc/kg. Dale Mc collough is working with Rocky on specific plant data needed for this calculation. Based on the results of this omiculation, a MAR will be required to raise the make-up tank high pressure alarm set point. Per conversation with operations, raising thus set point is mooeptable and i should not produos any operating sonoorns. The new alarm  !

set point will likely be approximately 3 pSIG higher than a conservative value for required cover gas pressure.

Because the make-up tank is part of a dynamic system, there may be short periods of time when the operating pressure will exosed the-tank pressure vs level ourve and times when sufficient hydrogen is not present. At this point in the evaluation phase, we believe this should be acceptable.

An REA will be written to request a MAR to change the high pressure alarm set point: however, the cover gas pressure requirement must first be determined. Also, operations

}

must by into the proposed set point change. The plan is to write the Rt.A by october 28, 1992.

e. A revision to OP-402 vill be required to incorporate the proposed new make-up tank set point and any details required by operations to incorporate the new higher operating tank pressure.

f. A revision to AR-403 will be required to incorporate the new higher make-up tank set point.

/o-/5- m A

4 a

i k

e M

l M

. , INTEROFFICE CORRESPONDENCE

,bt%4 kr;wdw$G.h0M l l Ihselaar Plant Technical Sussert osatt

{ .

........... 148. nan j erame suc i.i

! bVic40hIL:.ic j

SUBJECT:

Make Up Tank Hydrogen Overpressure g .

. g. ,

! T0: B.J. Nickle DATE: ~5eptember 2, 1994 '

j NPT5 94-0429 j

' ' On August 5th,1994, a meeting with Operations, Chemistry, SIES. WEIand Irf5 was held to discuss current Make Up tank overpressure condittens. The purpose of the i

meeting was to determine what was required to increase overpressure se desired RCS hydrogen concentration could be maintained with minimal Operator burden. The options

' discussed are provided below. It should be noted that none of these options will recommend changes to OP-1035, Curve #8. Engineering believes this curve is accurata l ' and reasonably conservative to protect the Nigh pressure injection pumps fren hydrogen gas intrusion in the worst case Larne_treak1GCA. In addition, corrective action #8 of PR 94-149 is currently is progress to provide technical bases for the i BWST swap over point. During this analysis Make-Up tank overpressure per Curve 88 will be re-evaluated. This action is scheduled to be completed by September 30, 1 1994. '

j Backeround; -

1 There are two conditions which limit the amount of hydrogen over

i. maintained in the Make-Up tank. These condittens are as follows: pressure being I 1.

I The first Ilmiting condition is based on the LOCA analysis. In this analysis, the differential pressure between the Make-Up tank and W5T was evaluated during maximum ECCS flows at various Make Up tank overpressures. The object was to i

i determine the maximum allowable tank overpressure such that pas from the Make Up tank would not enter the Make-Up pump section header. This condition must exist down to the point where ECCS pump section is transferred from the 9WST to the RS sump due to a low BWST level. NPI pump operation after the transfer would require an LPI piggyback line-up. .In the piggyback mode, gas intrusion into the Make-Up pump 4

1 suction header is not a concern due to LPI discharge pressure.

i This analysis resulted in a revision to the Make-Up tank level vs. pressure curve, (Curve #8 of OP-1038). The curve is based en the expansion of a known volume of j hydrogen for an increasing or decreasing Make-Up tank level. Therefore, for the LOCA analysis it is acceptable to raise the hydrogen everpressure to the curve at any

indicated Make-Up tank level. .

2) The sec'ond limiting condition f5 hydrogen overpressure is based on Section.

Ill.J of 30CFR$0 Appendix R and the Appendix R fire study.Section III.J states that operator ac' 6 minst occur for eight hours to achieve safe shut down unless 4

emergency ',@ing is provided for the required action. The study postulates a fire in the Auxiliary building which causes 18N-143 (Make Up tank hydrogen supply valve) to fail open. In addition, a plant trip is postulated to occur. CR3 cannot take 4

credit for operator action to manually isolate the Make-Up tank hydrogen supply l because there is no emerpancy lighting to the isolation valves (IEN-493 or 4g4) on the 119 ft Auxiliary But ; ding. Also, the fire may prevent access to these valves.

] Ther. fore, the current setpoint of PRN-491 (hydrogen supply regulator) is limited '-

{ the ight hour period. Following the trip, nake up to the RCS is requtred J

10 1984 l Iris94-042s i

{

for inventory control during plant cooldown. With M 143 failed open hydrogen ovoipressure on the Make Up tank will remain at the setpoint of fRN 4gl while the ,

l Make Up tank level decreases. The analysis evaluates Make-Up tank level with respect i

to bd5T level during cooldsun while asintaining Make Up tant overpressure at the i regulator setpoint. The regulator setpoint ensures that hydrogen gas will not enter the Make Up pump suction header within the eight hour liett. Operater action would i

be required to isolate hydrogen to the tank after eight hours. In this Appendix R i scenario Curve #8 of Op-1038 will be violated. However, because Curve 88 is for LOCA considerations and an Appendix R fire is not assumed to occur concurrent with a l LOCA, the violation of Curve #8 is acceptable.

\ .

entians to 1 w = e 'c_2. -- __ ._ m =er '

Option 1: Based on a review of the current M-4gl setpoint and the analysis performed to establish this setpoint, same of the conservative facters used in the analysis can be relaxed. The current to pais setpoint would allow apprestaately 14.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> before operator action would be required to prevent pas intrusten.

If the time frase is lowered to the sintaus 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period, the setesint can be l

increased by approximately 1.2 psig. In addition ' the setting of IRN 4gl per the '

Refuel g MAR functional test can be revised to allow the regulator setting to be based on indicated tank pressure. The Refuel g MAR functional test set the regulator at a conservative 1g.5 psig at a Make-Up tank level of 70 inches. Neuever. the actual tank pressure was 17.7 psis. This difference is due to lesses in the hydrogen supply 1tne between.the tank and the regulater and due to a drastic reduction of gas flow to the tank as pressure approaches setpoint. If both the aheve changes are made, the new regulator setpoint would be apprezimately 23 psig. This mould provide !

the ability to increase tank pressure from the Centrol Room to the curve at approximately 74 inches without manipulating Make-Up tank level. At present, pressure can be increased to the curve at approximately ES inches but then the Make-Up tank level must be increased to increase pressure to a:hieve 25 cc/kg.

Although this option is relatively easy to implement and it will improve the ability I to add hydrogen to the Make-up such that the 25 cc/k be maintained, the overall taprovement is marginal. g RCS hydrogen concentration Option 2: The hydrogen addition regulator setpoint could be raised if a redundant isolation valve was added to the supply lies. This valve could be operated from the Control Room providin IRN-143 to fail open.gTra its power additionisof

. rested independent a manual valve withof the fire zone enervency which lighting cause could also. be added and . accessed free outside the fire zone. If the setpoint of IEN-4p1 is 1 raised to 24 psig, the redundant valve would have to be isolated within approximately one hour. The best location for a redundant valve appears to be on the 11g ft elevatten of the Turbine letidinp. The hydrogen supply line is routed approximately 10 feet above the 119,ft 'urbine Building floor.

This option will also improve the ability to add hydrogen to the Make-Up tank and maintain the 25 cc/kp concentration. However like tien 1. the improvement is marginal. Also, the epistics of installing a new va ve in the hydrogen supply Itne at power could ks cap ex. Temporary hydrogen suppites to both the generator and Make-Up tank would be required.

Opt *in 3: The restoration of !WV 64 with a chain operator would allow the Make Up tam ta he :.- W 1y i nista: in *f e otat of a fire. Because e ergency lighting is tira ,ey anii tia t ta **eaet av.68 the trei t'ea t revle ecrer -4+h4a *he **ah* haa'

= . . - .

page 3 september 1. Ih N WT594-0429 stedy assmes a fire on the !!p ft of the Auxiliary Building could cause M 143 to fall ven. In this scenario, the 95 ft elevation of the Auxiliary Sullding remains acce sible. M H is on the 95 ft elevation. If the fire exists on the 95 ft Aexiliary Butiding then NN.143 would be unaffected.

Based on preliminary NDE analysis, M-H would have to be iselsted within approximately 15 minutes of a fire if the following conditions estats: A fire on the lit ft Auxiliary Building, a Reactor trip, and either the less of Nake-Up' tank i ' indiations .or a high tank pressure alars occurs due to M-143 falling open. The isola. tion 6f M-H in the 15 minute period is based en the Nake-Up tank being able to supply an average of 140 gpa for 15 minutes even if the tank level is initially i

55 inches (Iow level). Once the pump section header is aligned to the Bust and M-64 is isolated, the gas binding concern no longer exists. Also, the hydrogen regulator setpoint (M-491) would no longer be an issue.

The isolation of IEN-H could be restored to the control Ross and a protected centrol board switch could be installed. However, the restoration of power and air would be complex and very costly. The valve would have to be protected from spurious closure in the event of a fire.

In addition to installing the chain operator, a NAA would be required to raise the setpoint of M-431 to ensure operations can maintain a minimum ef.25 cc/kg hydrogen concentration in the RCs without Operator work around. The installation of the chain operator and setpoint change would fall into the category of a minor M .

Canclusiant l Based on preliminary'evaluattens. Engineering prefers tasta11 tag a chain operator per option three. The chain would be locked to the wall to prevent unauthorized valve operation. This would require a revision to $P-381 (Locked /5ealed Valve Check List procedure). Also Ap-800 (Fire Protection procedure) would have to be revised to incorporate the isolation of M H in the event the conditions described above occur during a fire.

fkfY: -

J.P. Hinman, Nuclear Project Engineer Nuclear plant Technical support -

.V. Campbell, Manager Nuclear plant Technical Support -

jph sc: 5. K. Balliet C. W. Bergstrom E. H. Nainen J. R. Nasada

p. E. Saltsman K. C. Vogel P.ecord: P.anagement 4

s