ML19329A271
| ML19329A271 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 09/23/1977 |
| From: | Seyfrit K NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| To: | Long F NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| References | |
| NUDOCS 8001020845 | |
| Download: ML19329A271 (4) | |
Text
5
/
g/-
l.N v'
's SEP 2 31977 l
)y [
i Docket No. 50-269 t
14DiOPMDUM FOR:
F. J. Long, Chief, Reactor Operations and Huclear Support Branch, RII K. V. Seyfrit. Assistant Director. Technical FROH:
Programs, ROI, IE HONC0hSERVATIVE FLOW COASTDOWN VALVES USED FOR
SUBJECT:
OCONEE(AITSF2147Hl) t De licensee's report that the assumed flow coastdown vaitts, used l
in the Oconee thenna1 hydraulic analysis, was higher than the by as much as 31, was evaluated as you requested.
raeasured wait %
The flow coastdown valas are used in establishing the flux / flow trip setpoint in the reactor protection system. Consequently.
the adequacy of this limiting safety systes setting at Oconee is in question. This concern also applies to the other B&W plants, since the vendor makes the calculations necessary to establish i
the correct setpoints.
This matter was discussed with the NRR Project Nanager for Oconee.
l lie said this problem is under evaluation by the Reactor Safety Branch in connection with the review of the Oconee reload applications.
l As stated in the Oconee occurrence report (RO-269/77-9), B&W is now l
using the measured coastdown flow in the design analysis for new j
This was confirmed by the latest reload report for Rancho reloads.
Secon(dated June 1977), in which the measured coastdown flow is used.
l F5rethese2reasonsi,wasbh11evenhst:this:; concern ttsbiti&gsadequathlyy addressed.;in the review of new reloads. There reraains the question of the acceptability of the flux / flow trip for current and previous fuel cycles.at all but the recent B&W plants.
The flux / flow trip is. based on a power to flow' ratio which would ace:omodate the loss of coolant flow accident from power.
(B&W Sta.,dard Tech. Specs.). The key parameters for this accident, in addition to flow coastdown ctracteristics. are flow rate, Doppler coefficient, moderator coefficient and the peaking factors. As
- 'g shown below, there exists a margin of conservatisa in each of 6
i l
i enNTarTe D_ r Eirknatrick
~ L_
j 49-28183,
77220o197 l
{
^ p I
'$ URN AM E M
'~
),
h$
ku' NRC FORM 31, (9 76) NRCM 0240
- 001020$9.S
&\\
~
i
.y F. J. Long 2
SEP 2 31977 i
I these parameters relative to the vcives used in the original design calculations. These parameters wart reviewed for Arkansas Nuclear 1
One-1. Rancho Seco,~and Three Mile Island as well as the three Oconee Plants.
i Normal Flow Rate - The measured flow ratas at all of the planu reviewed were higher than the original design rates.
These warfod from 111.5% of design at Oconee 2 to 104.9%
at at Rancho Seco. The values for each plant are shown in Table 1.
Credit has been taken for part of this excess ceaseM r
flow in the reload calcusations for some of the fuel cycles now in operation. However, these assumed values are still conservative by about 21 or more compared to the measured flow.
I An additional 55 penalty in the assumed flow wer odginally: count s
required to account for the possibility of a vent valve sticking open. NRR provided for removal of this flow penalty, by letter dated March 10. 1976, to all BW plant licensees. Since calcu-lations previous to this time were done using a lower flow value than is now permitted, there exists an additional 5% margin in the assumed flow relative to the measured flow.
Therefore.
l there is a> 7% or larger, total margin in the flow values used in the calculations for the fuel cycles presently in operation.
Doppler and Moderator Temperature Coefficient'r-The significance-of these parameters to the loss of flow accident, is that negative coefficients result in a power reduction between the time of the I
I flow loss and the time that a scram becomes effective. Both ca coefficients become more negative with burnup in BW plants. They I
are also more negative in reload cores due to the presence of partially spent fuel. As shown in Table 1 the Doppler coefficients I'
~
average 20% to 30*. more negative in the reloads, and tlw moderator coefficients siitch frma positive in the new cores to negatike in the reloads.
Both trends provide additional conservatism in the j
f' analyses of the loss of flow accident.
F Peaking Factors. A penalty for possible densification spiking was required to be included in all earlier fuel cycle calculations.
These include all.of the cases calculated using th'e design coast-l down flow instead of measuras flow. This densification spike penalty is no longue required, so it represents an additional censervatism in these previous calculations.
1714T16. the 35 maximum error in flow coastdown is comparable to to the amount of error to be expected in the flow measurement, and
'which must,be accounted _for in the design calculations.
~u i
'orric s >
.u.....>
a 3
~-
A D' ATE >
' '~
j NRC Penaf 518 (9-76)
CX 0240 _ _ __
' ['M u. Oavsmauswv rainna. oresca s.r.-.a e4
s F. J. Long 2
SEP 2 31977 i
Considering the above it is concluded that the amount of conservatism l
included in the important parameters used to analyze the loss of flow accidents for past and existing fuel cycles is sufficient to compensate for the identified error in the flow coastdown values. Since measured coastdown flows are presently being used in calculations for the latest l
fuel cycles, we believe that the problem is adequately resolved. We plan no further action on this matter. This memorandum closes Action Item F2147H1.
Original Jg..d by K. Seyfrit t
i Karl V. Seyfrit. Assistant Director Technical Programs. ROI. IE I
cc:
H. D. Thornburg. IE e
R. C. Lewis. RII J. D. Neighbors. NRR l
I t
i l
k 6 r'e=
- TE:ROIME----. IP4ROL:.IE--
c
.u===>
DCKirknatrick USeyfrit
.9/_ /77 gh
.gj_j77 s
NBC PORat, 318 (9 76) NROf 0243 W us s. eova NhegNT PR4NMNG OPPICta 1 74 = S28434
t SEP 231977 TABLE 1 Rancho Three Oconee 1 Oconee 2 Oconee 3 AN0-1 Seco Mile Island Flow Rates Measured 108.6 111.5 110 109.7 104.9 108 (5 of Design)
Used in 106.5 last 106.5 106.5 107.5 106.5 cycle Doppler FSAR
- 1. 2
- 1.17
-1.17
-1.17
-1.22
-1.17 Coefficieng)
(ak/k'Fx10 Latest
-1.45
-1.48
-1.54
-1.47
-1.43
-1.49 Cycle Moderator FSAR
+5
+5
+5 0
+9
+5 i
Temperature Coefficient latest
-10
-5.7
-10.6
-10.9
-7.5
-10.6 (ak/k'Fx105 Cycle Flux Spike Included Included Included Inclu-Included Included Penalty thru thru thru ded in thru Cycle 3 Cycle 2 Cycle 2 thru Cycle Cycle 2 Cycle 1
2 i
0 L
1 ht+ev-I
DUKE POWER COMPANY OCONEE UNIT 1 R0-269/77-9 l
_R_eport No.:
i March 10, 1977 Report Date:
February 28, 1977 occurrence Date:
Oconee Unit 1, Seneca, South Carolina Facility:
Two-pump coastdown flow essumed in the core Identification of Occuriene_e: thermal hydraulic design analysis found to be slightly non-conservative Description of Occurrence:
d the associated i
while the core design analys s an4 operation of Oconee 1 were On February 28, 1977, technical specification changes for Cycleinformed Duke Power Company of a The change pertains being reviewed, the NSSS vendor (B6W) change in the core t i
to using the measured 2-pump coastdown flowflow previou discovery that the design coastdown flow was slightly non-during the 2-pump coast-i to the measured coastdown flow) for certain t mes down.
Analysis of Occprrence_:
flux / flow trip h
The 2-pump flov c.oastdown values are used to establish h
vent 1.3.
of a loss-of-2-pump incident will not be less than
- setpoint, 2, Cycles 1 and 2; and setpoints for Oconee 1, Cycles 1, 2 and 3; Oconee f the design coast-Oconee 3, Cycles 1 and 2 were established on the basis oT lightly f 3%), and this less than the design coastdown flow (maximum diffe down flow.
However, a review of the fizx/ficw trip setpoints of the current cycall three Oco
/ flow trip set-cycles fo:
100% design RC flow and points were indeed saf e and adequate.
and ?, the thermal hydraulic analyses were based on lty and densifica-included conservative allowances for vent valve flow penaConsiderin 3 and that the vent tion power spike penalty.108.6% for Unit 1, 111.5% for Unit 2, and 110%
e no longer necessary, l
valve flow penalty and densification spike pena ty ar1 flux / flow trip set-l a significant degree of margin is seen in the Cyc ed and the design coast-points even when the dif f erence between the measureFor Cycles 2 an down flows is considered.2 and 3, design analyses were based on 107 conservative allowances for r.he dencification power sph t the difference in the i
valve ficw penalty, and it has been determined t aconservative flux / flow trip two flow coastdown values did not lead to non-The current flux / flow trip setpoints setpoints for these cycles.
1, 3.0% for nit 2, and vide DNBR margins of a7 proximately 5.5% for Unit 4
7 Min
Thus, the sli htly non-conservative nature of the coabt-
' 3.0%
- Unit 3.
h down f.
used in the previous thermal hydraulic analysen did not in any way result in an unsafe cperation of any Oconee unit, and it has been con-cluded that this incident did not affect the health and safety of the public.
Corrective Action:
A review of the core safety related technical specifications has been per-formed to verify that the existing technical specification limits continue to be valid with sufficient safety margins. The core thermal hydraulic design analysis procedure has been modified to utilize the conservative coastdown flow.
f.,,
sj"