ML20024B264

Suppls 771101 Ltr Re Input to Teco Rept Concerning 770924 Depressurization Event.Evaluation of Reactor Coolant Components Encl

ML20024B264
Person / Time
Site:	Davis Besse, Three Mile Island
Issue date:	11/07/1977
From:	Lauer J BABCOCK & WILCOX CO.
To:	Domeck C TOLEDO EDISON CO.
References
TASK-*, TASK-03, TASK-3, TASK-GB BWT-1561, GPU-2400, NUDOCS 8307080132
Download: ML20024B264 (3)

Text

-

r _ . _ . . _ - . . . . . . . . _ - . _ . - _ . . . _ . . _ . . . . . . . _ . . . . . . _ . .. _.. _ _ .

,' Y $**

G f d 2 91 p p nor, ,, ,. , t . n t r e _ - . .

.w..w.....wvis TELEC0PY Power cenera'tien croup P.O. Box 1200. Lynch:urg. Va. 245c5

?!ovember 7,1977 menmuso4) 3e4.si t t bht-1551 File: T1.2/12B bec: F. R. Faist AHL/DBT JAL Hr. C. R. Domeck Records Center T1.2 Nuclear Project Engineer R. C. Luken Toledo Edisen Ccepany. W. H. Spangler Power Engineering & Ccnstruction A. H. licBride 300 liadison Avenue G. A. !!cyer .

Toledo, Ohio 43652* C. W. Bruny, lit.V.

J. P. Jones

Subject:

Toledo Edison Comcany R. B. Davis REPO?,7 ON DEPRESSURIRATION E. Kane Davis-Besse Unit 1 -

D. C. Rice B&W Reference !!SS-14 3. Berchin

Dear !!r. Demeck:

Our ' letter BWT-1529 dated flovember 1,1977 fon:arded input for a TECO report to !!RC regarding the-depressuri:stien event on September 24. To supplement that letter, you will find attached another writeup which evaluates the reactor coolcnt components.

It may be inserted into the previous report as Attachment B.

Very truly yours, A. H.. Lazar Senior Project '-tanager JAL/hj

, CW J A. Lauer Attachment ojectfianager cc: J. D. Lenardson w/a J. C. Lewis .

D. J. DeLacroix P. P. Anas/4c w/a E. C. iiovak/lc w/a G

B18741

. g.

,os l} !

j .,

Inc D1Nocir & WdCos CemNny / Estab! .fied I$67 *'

8307080132 771107 PDR ADOCK 05000289

- S E --

.;_ 1 2. . . _ __ . .__ ._. _. _

4

,y b ..

s

. s -

C M ochC.".::cox ATTACHMENT B REACTOR CCOLA'iT CO::?0: TENTS

/

.ET.! has cc:pleted their evaluation of the Septenber 24 incident at Davis-Besse and found no harmful effects were incurred in the reactor vessel steam generator pres-surizer and pricary piping pressure boundary.

During this rapid depressurization event the reactor coolant pressure dropped from 2300 psig to 930 psig in 7-1/2 minutes and gradually recovered to 1800 psig in 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

During tne firgt 7-1/2 ginutes the reactor outlet temoerature dropped at varying rates from 553 F to 533 F. For this evaluation it is assumed that the total temperature drop occurred at the in.tial rate. This results in a 49" te perature drop over a 6 minute period. Approximately 30 minutes after the jnitial terperature drop a second slower and smaller temperature drop from 540 F to 505 F occurred over a 21 minute period. Following this second temperature ramo, the temperature gradually increased over a 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> period to 5280F. The reactor inlet temperature ramps and durations were the sar.2 as for the reactor outlet temperature. -

The secondary side pressure in steaa generator tio.1 reached a maximum of 1050 psig and decreased to 860 psig within.15 minutes and remained at that level. The seconday side pressure in steam generator tio. 2 reached a maximum pressure of 980 psig decrecstd to a minicum of 610 psig in 14 minutes, and returned to 850 psig in 2 minutes. 3.%nty minutes later the pressure again dropped to 610 psig and recovered gradually over a '

2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> period.

The Design Specification for the Davis-Besse I plant required evaluation of 40 cycles of a rapid depressurization event which included a drop in the reactor coolant pressure from 2200 pgi to 803 csi, a drop in the reactor coolant system average temperature from 553 F to 500 F in 15 minutes, and a drop of secondary pressure from 1050 psi to 640 psi. -

The major difference between the actual transient and the design transient is the rate of the temperature drop in the reactor coolant system. ThE actual rate of te=;:crature drop was twice the rate of the design transient but the total temperature change was cnly 78". of that of the design transient. The net result is that the fatigue usage of this one rapid depressurization is about the same as that predicted for one cycle of the design transient.

As a more direct comparison, the transient event 1.dentified was analyzed for the reactor vessel shell and compared to the design transient. The results were that the range in ther=al radial gradient stress for the actual transient was 5400 psi and the ranga of thercal radial gradient stress for the design transient was 6500 psi. This comparisen would be representative of other thicknesses throughout the pressure boundary.

The conclusions,of the analysis are:

1. Stresses in the pmssuroboundary did not exceed those already calculated on a design basis. This is verified by the actual pressure not exceeding 2500 psi and the themal transient being less severe than a combination of design transients for a rapid d,: pressurization and a reactor trip.

, B18742 .

. I (V *

? *l- '\

i 1

- - ,. l

.-p<awp a,7m- -my w- - y w----+v-a Y

+ ,-g. y 4 .-#9-- 3

e

.j, .

s

- /

... ,j B-2 n...........r....

2. Fatigue life of the reactor coolant co monents is not affected if one cycle of the reactor trip design transient and tao cycles of a rapid depressurization design transient are considered to be used for this transient. Two cycles of the rapid depressurization transient cre necessary because the hPI systcm was actuated twice during the event and two cycles are necessary to reflect the thermal transient ,

in the high; pressure injection no::le.

The effect of the entirc event on the fatigue life of the steam generators can be accounted for by using one cycle of the design transient for rapid depressurization and one cycle of the cosign transient for loss of feedsater to one generator.

3. The effect of the change in water level on the pressurizer has a very minor effect

- on the pressurizer shell stresses. The pressurizer has been previously analyzed for the thermal effect of water-steam interface and the change of level does not affect that analysis.

4. No significant thermal shock should occur to the heaters since the heaters were de-activated due to a low water level sensor and not reactivated "till the level re-covered. . ,

5. No dynamic effects were caused by the rapid pressure decrease. No specific analysis was done but a dynamic response of the shells would require a large pressure drop in the order of milliseconds and the actual change was on the scale of minutes.

The reduced feedwater " flow to steam generator No. 2 was not sufficient to maintain a water level during the first five minutes of the event and'this steam generator boiled dry. The primary concern with a dry generator is the tube to shell temperature difference. In this event a water level was established before the system cooldown was started and acceptable tube to snell temperature differences were maintained. This condition is similar to the loss of feedwater design transient followed by restart of a dry pressurized generator using the auxiliary feedwater system.

The burst rupture disc of the pressurizer quench tank resulted in a stream of steam and water impinging on steam generator No. 2. This stream removed a section of insulation 10' high and 20' wide from the lower shell of the generator and imoinged directly gn the generator shell. The temperature of the impinging water was assumed to be 212 F. A conservative evaluation of the rapid te perature change in this local region of the vessel shell was perfomed. The results of this evaluation indicate that this one event used less than 15 of the total fatigue life of the vessel. The predicted fatigue usage factor for the 40 year design life of the vessel in this area was less than 0.10. This jet impingement did not significantly reduce the fatigue life of the generator. - .

e B1S?a3, -

t

,l p ..

.