ML20148Q186

Forwards Addl Documentation Supporting Environ Qualification of Bunker Ramo Containment Penetrations,Per Indication in Author to Recipient.Evaluation Shows That Midland II LOCA Test Accurately Represents Unit Parameters

ML20148Q186
Person / Time
Site:	Braidwood
Issue date:	04/07/1988
From:	Hunsader S COMMONWEALTH EDISON CO.
To:	Murley T Office of Nuclear Reactor Regulation
References
4432K, NUDOCS 8804120351
Download: ML20148Q186 (17)

Text

)

N 1 ) Commonwealth Edloon One Fast Natonal Plaza. CNcago, innois G A33ress Repfy to: Post Of5c) Box 167 i

• ~.y CNea00. Ill'00's 60690 0767 April 7, 1988 l

I I

Mr. T. E. Murley  !

Office of Nuclear Reactor Regulation  !'

U.S. Nuclear Regulatory Commission Washington, DC. 20555 l

Attn: Document Control Desk  !

Subject:

Braidwood Unit Rnvironmental Qualit2c,+.2vi. j Bunker Ramo penetration NRC bocket No. 50-452 Reference (a): March 23, 1988 S.C. Hunsader letter to T.E. Murley i

i

Dear Mr. Murley:

Reference (a) provided the NRC staff with additional documentation to i provide support for the environmental qualificatior., under 10 CFR50.49, of a  :

Bunker Ramo penetration used at Braidwood Station Unit 2. i As indicated in reference (a) additional supportive documentation addressing the application for post-accident monitoring would be provided for  ;

the NRC staff's review. Attached to this letter you will find Attachment A,  !

entitled "summary or Calculations. Thermal Analysis of Bunker-Ramo Containment penetrations" which includes this information. l

)

i Attachment A provides an expansion of the answer to NRC questions included in pagea 3 through 6 of the Supplement to Appendix B, attached to reference (a). It provides a further description of the heat tran',fer model i for the penetration and includes the thermal analysis model representative of the post-accident monitoring (pAM) condition. Also included in Attachment A is a description of the subsequent review performed of the Midland II environmental qualification test time /temperatute strip charts. This review shows that temperatures seen by the penetration module during the first two days that insulation resistant values were measured, on average, were 228oF and 230or, respectively. These values envelope the maximum temperature (224.loF) expected to be seen by the Bunker Ramo penetration at Braidwood Unit 2 during the PAM condition.

(

kb 9804120351 880407 ~T I PDR ADOCK 05000457 ,

P PM \\ \

l

i i

)

s

. 4  ;

l i

e t i

Overall, the evaluation shows that the Midland II LOCA test accurately represents the Braidwood Station parameters to support the  ;

j environmental qualification of the Bunker Ramo penetrations.  :

i Please address any questions concerning this matter to this office.

4 Very truly yours, 4

i C'

S. C. Hunsader Nuclear Licensing Administrator ,

/klj l cc: NRC Region III Braidwood Resident f j S. Sands [

4432K i f

I I I

\ i

)  ;

j  !

i i j  !

y 1 {

J 1

i .

5 I .

+

4

, i k

4 1

i' i i

i t

l $

i i 1

i

}

i i

?

I

ATTACHMENT A

SUMMARY

OF CALCULATIONS l

l Thermal Analysis of Bunker-Ramo Containment Penetrations Calculations have been performed to determine the effects of the Containment Building pressure-temperature transient on the ,

Bunkor-Ramo instrumentation penetration assemblies. The time dependent temperature profile in the vicinity of the penetration l

feed through module is calculated for comparison to the avail- ,

able qualification data. This calculation documents that the tamperatures at the penetration feed through modules will not exceed 157 F prior to initiation of the necessary instrument signals to trip the reactor and initiate safety injection.

This calculation also documents that the temperature at the feed through module was similar to the temperature in the Mid-land Il test during the remainder of the accident profile.

l The calculation documents that the temperatures at which the IR values were taken in the Midland II test envelope the antici-pated feed through module temperatures during the period follow-ing the accident when the instrumentation circuit, passing through the penetrations are required, j

To evaluate the effect of the accident pressure-temperature transient on the feed through module, a computer model of the containment and the penetration was prepared. The first node in the model is the area between the center support plate and

[

l the closure flange of the penetration. The feed through module l is partially exposed to the environment in this node. The sec-ond node in the model is the portion of the penetration assembly between the inboard support plate and the center support plate.

The third node in this model is the containment volume. The l

nodes communicate with each other by means of the openings in the support plates.

1

-- 1

?

i The initial analysis performed for this study utilized the KITTY Computer Code and took into account the heat transfer into the penetration sleeve and the cuerounding concrete. This analysis was sufficient to demonstrate that the temperature of the steam-air mixture in Node 1 was below 1570P at the time when the reac-tor trip and safety injection signals would be initiated. The i

results of this analysis was provided to the NRC as a part of the Supplement to Attachment B to the March 23, 1988 letter from S. Hunsader to T. Murley. The model utilized was given in Exhibit B-1 to that letter and the temperature curve was given in Exhibit B-2.

i l

In order to evaluate the long term temperature profile at the feed through module location a new computer model was prepared.

This computer model, utilizing the COMPARE /MODT Code, took into consideration several heat transfer processes. These heat trans-

fer processes are as follows:

1) Isentropic comp assion of the gas in the assembly due to pressurization of containment.

i l

2) Convection and conduction to the penetration steel and surrounding concrete and to the Auxiliary Building air.

3

3) Influx of mass and heat into the penetration due to i

condensation.

4) Condensation heat transfer to the penetration steel surfaces. I l

l 5) Natural circulation flows around the cable support plates and the resultant mass transfer into the penetration assembly. j 4

I

__ .. . . .. . ._ \

?

t

, i l This new model simultaneously incorporated the above processes for the three nodea and also modeled the penetration sleeve, the containment concrete, and the Auxiliary Building enviror.nent (Se9 Exhibit B-3 for a description of this model).

The COMPARE /MODT analysis accounted for the flow f rom Node 3 te Node 2 and from Node 2 to Node i due to the pressurization of the containment. These flow paths are shown respectively as mass flow paths 2 and 1 on Exhibit B-3. The analysis modeled the natural circulation flows between the adjacent nodes. Upper i bounding time dependent flow rates based on the density differ-ences between nodes were input. These mass flow paths are shown i

as paths lA, 1B, 2A and 2B on Exhibit B-3. The analysis modeled axial conduction along the penetration sleeve and radial conduc-tion into the sleeve and through the sleeve into the adjacent concrete. The model accounted for this two-dimensional affect '

j in the one-dimensional COMPARE /MODT code by using air nodes and dummy heat sinks as shown on Exhibit B-3. The capacitances l and conductances of the air nodes ar.d sleeve and the heat sinks were modeled to represent the sleeve and the adjoining concrete.

i Node 4 on Exhibit B-3 represents the portion of the penetration l 1 sleeve exposed to the auxiliary building environment. The COM-

! PARE /MODT model used natural convection to transfer energy f rom l the outer surface of this node to the auxiliary building envi-j ronment. Axial conduction of thermal energy along the penetra-tion sleeve was included in the model as represented by the heat transfer path from Nodes 8 to 6 to 5 to 4. In accordance with NUREG-0588, the larger of natural convection or condensa- i tion heat transfer was applied at the inner surface of the sleeve. l The COMPARE /MODT Code also accounts for the influx of mass and heat into the penetration due to condensation effects.

't

(

4 L_________-_______________-_____________-___.

The accident that was chosen for this calculation was the Main Steam Line Break (MSLB). This accident was considered limiting for three reasons:

1. A comparison of the pressure-temperature curves in the FSAR Chapter 6 shows that the peak containment temperature of 318 F for the main Jteam line break significantly exceeds the peak temperature of 267 0F for the LOCA (FSAR Table 6.2-1) .

2. A comparison of the time to actuate for the protective func-tions for LOCA and MSLB shows that the protective function actuates in 10 seconds for MSLB (FSAR p. 15.1-19) and in one second for LOCA (FSAR p. 15.6-30).

3. As shown on Exhibit B-4, the Main Steam Line Break (MSLB) envelopes the Double Ended Hot Leg (DEHL) LOCA except for two relatively insignificant periods. The first period is in the first ten to twenty seconds of the transient.

This period is after actuation of the safety injection and reactor trip instruments for the LOCA (FSAR p.15.6-30) but well before the PAMS would be required to operate.

The second period is from 150 - 300 seconds into the events.

During this period, the differencas in temperature between two curves is not significant when compared to the earlier peak of MSLB curve.

The computer analyses utilized the Westinghouse mass-energy release data and the containment pressure - temperature curves from FSAR Figures 6.2-13 and 6.2-14 to establish the time depen-dent conditions in the containment node (Node 3) . Using this input, the time history of the pressure-temperature conditions in Nodes 1 and 2 was calculated. Exhibit B-4 shows the temper-ature on Node 1 (adjacent to the feed through module) plotted on the containment temperature curve for the Main Steam Line Break / Double Ended Hot Leg LOCA (FSAR Figures 6.2-11 and 6. 2-14) .

It should be noted that the maximum temperature at the feed through module prior to trip initiation (10 seconds) is 157 F.

This temperature is well below the temperature at which the Midland IR value was measured.

In order to more accurately determine the temperature at which the penatration IR value was recorded during the Midland II LOCA test, a more detailed review of the time / temperature strip charts from the test was performed. Exhibit B-5 is a sample of a portion of the Midland II LOCA test strip chart. The first IR measurement was made between 3:12 a.m. and 7:00 a.m. on Octo-ber 25, 1978. The second IR measurement was made at 10:00 a.m.

on October 27, 1978. These times are documented in ANCO Engi-neers, Inc. March 21, 1988 letter. The time / temperature strip charts for these time periods record temperature data from sev-eral thermocouples in the test chamber. Because neither the test report nor the test log document the installed location for each thermocouple, the average temperature of all thermo-couples during the periods when the IR measurements were made will best represent the temperatare of the penetration feed through module. This methodology is consistent with that used in other test applications in which several redundant instru-r-v'.s are used to reduce calibration error.

ANCO Engineers, Inc. reassessed the time / temperature strip chart data recorded between 3:00 a.m. and 7:10 a.m. on October 25, 1978. Exhibit B-6 documents the results of this review. ANCO counted the number of temperature data points recorded in each 10 F increment between 190 F and '50 F. By averaging the data provided by ANCO, we can establish an average temperature of 228 F (Standard deviation 12.3 F) at the time when the IR mea-surements were made on October 25, 178. Additionally, we can

l establish that 81% of the temperature data points recorded, when the IR measurement was made, are above 220 F. A similar l evaluation was performed for the time period between 9:42 a.m.

and 10:42 a.m. on October 27, 1978. The results of this review are documented on Exhibit B-7. By averaging the data provided  ;

by ANCO, we also can establish an average temperature of 230 F O

(Standard deviation ll F) at the time when the IR measurement was made on October 27, 1978. We can establish that 70% of the temperature data points recorded at this time are above 225 F.

Exhibit B-4 shows the calculated temperature in Node 1 of our i 1

model during the MSLB transient. As can be seen on the Exhibit, i the Node 1 temperature initially peaks at 157 F after 20 seconds and gradually drops off. This early peak is due to the initial pressurization of the containmen't forcing the steaia/ air mixture from containment into the penetration. The Node 1 temperature cools somewhat when the heat sink provided by the containment penetration steel and the containment concrete begin to absorb '

heat. From about 25 seconds into the transient onward, the other modeled phenomena (especially natural convection) become predominant. The calculated Node 1 temperature crosse: 210 F approximately 335 seconds after initiation of the tranaient.

It approaches the MSLB containment temperature curve and peaks at 224.1 F approximately 15 minutes into the transient. The calculated Node 1 temperature parallels the MSLB containment temperature for the duration of the transient returning below 210 F approximately 48 minutes into the transient. From this point onward, the calculated penetration temperature decreases i

slowly as does the containment temperature. 1

.. .~ - . -_ - - . -

A comparison of Lne time / temperature strip chart data for the Midland II LOCA test and the calculated Node 1 temperature for the COMPARE /MODT model of the Braidwood Unit 2 containment in-strumentation penetration assemblies, provides additional justi-fication for our use of the Midland II LOCA test to evaluate the Braidwood Unit 2 instrumentation penetration assemblies. <

The calculated temperature is well below the test temperature durin. the time frame when the Safety Injection and Reactor Trip instruments are required to activate. During the later stages of the accidenc, instruments passing through the. subject penetration are only used for the Post Accident Monitoring Sys-tem (PAMS) . However, even during this time frame, the tempera-tures predicted by a conservative model of the penetration assem-bly during the transient are less than the average temperatures at which Bunker-Ramo measured IR values during the Midland II LOCA test. Therefore, we believe that the Midland II LOCA test also establishes a basis that supports the qualification and demonstrates the acceptability of the Bunker-Ramo instrumenta-tion penetration assemblies during the time frame PAMS is re-quired, e p

It should be noted that the use of the Midland II LOCA test is conservative for several other reasons as discussed previ-ously. The chemical spray used in the Midland II LOCA test ,

is more electrically conductive than the chemical spray u.1ed at Braidwood Unit 2. Independent test data has shown that the -

penetration IR measurements are af fected much more signif f c:antly by chemical spray initiation than by temperature. This 'fas j included in "Supplement to Appendix B" attached to S. C. Hunsader's letter to T. E. Murley, dated March 23, 1988. However, during j j

the LOCA test, the synergistic effects on IR measurements from.  !

all parameters (i.e., pressure, humidity, chemical spray) must ,

be considered. We have shown in this evaluation that the Mid-land II LOCA test accurately represents the Braidwood Station parameters for acceptance of these penetrations. In addi-tica, the Midland II LOCA test specimens did not contain envi- f I

1 l

. , - - . - , , ,- - -,7 - - - - . - . -

. . .c -

. , . . . . - , . , , , . - .....e , ~ ~ _ . . - . . ,7

i ronmentally sealed connections. The installed Braidwood Unit 2 penetrations utilize environmentally sealed connections that have been qualified separately for the LOCA environment.

In summary, we believe that the documentation submitted with out March 23, 1988, letter and supplemented by the above evalu-ation, provides additional bases to support the environmental qualification and demonstrates the acceptability of the Braidwood Unit 2 Bunker-Ramo Instrumentation Penetrations.

I f

I i

l i

1

)

1 B

t t t i nn b ee i mm nn h iaai x t a

t a

t t nn oo E e hht h h e cc _

t _

sf o n a n a e y

• o p o p d r

a it r c

it r u es d uf e cef e lch n e

• ds vs inc n u d *h t nn nn o *a oar oa r 3 7i _

B t' P Ct Ct e _

d s AQg y o l u e NP _

x

• n objf4

% 7 3

c5

- r e ev 3 6 gg tuee l l

e _

1 an i d

Q 0s N _ -

6 od "5 3 Mil b u

1 mo 1 aM ,

7.

_ 4 o 2 r llg[ lllI 2

j R " _

i 1 4 5 -' -

1 A 1 2 -

r T 1 -I- - m" re ef "

1 h h.___ -

• fA 0 5

• _

ks 1

~ -

5 l' 3

1 nn T AI - - .

r ua r h hn e _

" - - - o BrT 0 1 -

6 E*x 2 m"

ir A

6 4

1 c_

T I 2t  : ~ - _

- h___

t a "

0 1

- 2 7

• Mh - 7 g 5 1

ie  : - _ b_ ^

p L nH l lgll l U

^

g1

- 3 -

"7 I n t - -

- 3 2g5S oU -

do oi 6' t n

ota ,~1 7 r e m

wr n g 8 r 3i 8-3 a t 3 t

1 n n 0 o

de 1 n e 1 C M

/

2 a

ren i P 4 0

0 5

9 5

BP 1 C -

' ,* l ^ j

Braidwood Unit 2 Bunker-Ramo Penetration Short-Term Heatup Curve Exhibit B-2 1

Temperature ( F) 350 y S team

• 300 i

s

/

200 Penetration feed

! j[ A , through module j/ ]x% l .

Water

• 50 lb 10' 10 2 10 10 4 -

Time (sec)

• Source: FSAR Figure 6.2 - 14, "0.9 4 2 f2 t split rupture at

! C1595.008 3-23-88 30% power with steamline stop valve f ailure" 1

4

Exhibit B-3 Braidwood Unit 2 Bunker-Ramo Penetration Model COMPARE /MODT 1

7" 20" 15" 13.5"

+ 5 4 0 t # 0 i I I 11- -- ---

i  !

6" 2 1 Pen 3r3g _ _ _ _

1 6 5 4  :

8 - ..-- . - - - - -- --

HS 9 HS 8 3 3' I' l' 7 Containment --d--------1------ Auxiliary Building

. A ,, A .

M g

NodM m Node 2 Node 1 K'Y:

3 (Air) > (Air)

(Catet.) a 2 h w

] Boundary node

~

if if if if if Active l HS 12 ll HS 4 l l HS 3 l l HS 2 l lHS1 l -

node if if

_ 1P if HS Heat sink gs Node 6 g5 Node 5 Node 4

~

3h (Mode Sleevel +a 7 -> (Sleeve / -> 6 (33**'*/ 4 HS

$ IiSleeve)

Ccacrete) -

Concrete) -

Concrete) + Heat transfer path 1r 1r 17

_ A Mass flow HS 9 HS 8 gg l HS to l Path (Concrete) (Concrete) -b gg 1f ode

• Node 3 includes containment 7 plus 7 inches of containment 4

i C1595.043M 03-88 (Aux. waH.

Bldg.)

Braidwood Unit 2 Bunker-Ramo Penetration Short-Term Heatup Curve exhibit B-4  :

i

.i Temperature ( F) 350 y MSLB*

)'

300 '

DEHL*

["

^

250 b r o

200 /

/ = Node 1 Temp Profile COMPARE / MODT 150 /

100

• Sources: FSAR Figure 6.2 - 14, " 0.9 4 2 f2 t split rupture at 30% power with steamline stop valve failure" FSAR Figure 6.2-11, Containment Temperature 50 Response for Double Ended Hot Leg

]

0 '

2 3 4 10 10' 10 10 10 Time (sec)

C1595.009 4-4-881 4

i '

EXIIIBIT B-5 Bunker-Ramo Midland II LOCA Test Example Strip Chart Recordings of 10-25-78 (Data from 3:00 a.m. to 7:10 a.m.)

T "#

6 AM l  ;

I  !

i 3

. , . k to eccacts FAHRENHEIT  ;

Wq i i  :  ; .  : l 0 5 10 15 2 0 " ., .g'5 q 3,0 3,5 40 45 i i i

1

! 1 l

i l.

M 10 !AM j.

.l.l- l-l i

'i Ii t '*T t ,

, i,

' e l I il l g

ll 1 l I l  ; l!

.j i

,i t

,i 4

! 1 t

P .

j i l!

, }. , !

! g gg i  ;  ! I' 9fAM j. , II l r'. .I i

l l~  !

a

!i ' '!! , I!;, ,l,j i l l- ,

l , g ,

.j,4 4

i

1 ..  !

!. l

. l. r.M.

l, 8 AM  !

1

l

^'

'l.

], r kioctcars FAHatxHrit

. j;l. , I 0i 5 10 l'5 2d , w25

. 30 , ; :35 1 40 45  !  ;

f,r4k

, i - -

I i

!  ; l . - -

! 4 AM ! ,

..- . i 7 'AM  ;

I i  !: i l

!: 10 '%

I;

=

I l l  ;

i

. 'l I

..i ; l i .

1 i  ?

l 1

. $ ' [t.

- i

-1:

i l l

i i

'O i.A i i j 'c!

i i

l j. 6 AM ! ,

l I

!~  ! l l , I ~ lh I  ! i 41 , j j i i M9..' -

i.

e i

• 7 l. ,'.! 4 I i

3 AM l 5,AM  !

j:  ; l ,I Note: Tnc above Midland II strip chart represents a sample portion of the data points recorded on October 25, 1978, during the period from 3:00 a.m. to 7:10 a.m. The time increments noted on the chart do not reflect the actual test time periods shown above. The test data was recorded at a scale of one (1) nch =

two (2) minutes not one (1) inch = one half (1/2) hour. The ambient temperature during the test was 75 F. I The temperature  !

recorded is in degrees Fahrenheit and must be multiplied by 10 degrces to obtain actual recorded temperature. j l

l l

l l

EXHIBIT B-6 Bunker-Ramo Midland II LOCA Test Strip Chart Recordings of 10-25-78 (Data from 3:00 a.m. to 7:10 a.m.)

(A) (B) (Bf C) = (D)

Temperature Midpoint Data (DxA)

Range ( F) Temperature ( F) Points  %  %

190 - 200 195 50 .019 3.844 200 - 210 205 200 .079 16.167 210 - 220 215 271 .1068 22.975 220 - 230 225 778 .307 69.026 230 - 240 235 872 .344 80.804 240 - 250 245 365 .1439 35.262 250 > 0 g # #

Total (C) 2536 0.9997 228.08

a====== ========= =======

(A) Average Temperature = 228'F 2557 (B) Total Percentage @ 210*F and above = 91%

2807 6

(C) Total Percentage @ 220*F and above 28

= 81%

l l

i 1

EXHIBIT B-7 Bunker-Ramo Mid;and II LOCA Test Strip Chart Recordings of 10-27-78 (Data from 9:42 a.m. to 10:42 a.m.)

(A) (B) (B C) =(D)

Temperature Midpoint Data (DxA)

Range ( F) Temperature ( F) Points  %  %

205 - 210 207.5 56.4 .10 20.75 215 - 220 217.5 112.8 .20 43.50 225 - 250 237.5 394.8 .70 166.25 Total (C) 564 1.00 230.5

======= ========= =====e

(A) Average Temperature = 230.5 F (B) Total Percentage @ 215*F and above = = 90%

4 (C) Total Percentage @ 225*F and above =

564

= 70%

l l

l I

l l

l l

1