ML19326B392
| ML19326B392 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 11/13/1974 |
| From: | US ATOMIC ENERGY COMMISSION (AEC) |
| To: | |
| Shared Package | |
| ML19326B385 | List: |
| References | |
| NUDOCS 8004150761 | |
| Download: ML19326B392 (19) | |
Text
a r
O SUPPLEMENT 1 TO THE STATUS REPORT BY THE DIRECTORATE OF LICENSING IN THE MATTER OF BABCOCK AND WILCOX ECCS EVALUATION MODEL CONFORMANCE TO 10CFR50. APPPENDIX K gov 131974 1
1 d
800415o7//
TABLE OF CONTENTS Page A.
INTRODUCTION.............................................
11 Be A DDITION AL D ISCUSSI0FS...................................
1 1.2 Sc o pe o f Re v ie w...............................
1 2.1 3 Com pliance with 0riteria......................
1 4.1 5 Metal-Water Reaction..........................
2 4.2 Swelling and Rupture of the Cladding and Fuel Rod The rm al Paramet ers......................
4 4 3 1.1 B r e a k S pe c t r um................................ 5 4313 End o f Bl o wd o wn............................... 6 433 Momentum Equation.............................
7 437 Core Flow Distribution During Blowdown........ 8 4.4.?
Containment Pressure..........................
8 4.4 3 Calculation of Reflood Rate..................
8 4.4.4 Steam Interaction with Emergency Core Cooling Water in Pressurized Water Reactors........
14 4.4 5 Refill and Re flood Heat Trans fer.............
15 50 sm al l Br e a ks...............................'..
15 6.0 Doc umen ta t ion................................
16 C.
ERRATA..................................................
17 i
1 B.
ADDITIONAL INFORMATION The following discussions relate to the open items presented throughout the staff Status Report:
1.2 Scoce of Review As indicated on page 1-6 of the Status Feport, it was believed that the similarity in design of the Babcock and Wilcox plant types in categories other than Oconee Class (Category 1) may permit application of conclusions of the present staff review.
Babcock and Wilcox has agreed to address the applicability of the present generic model conformance to Appendix K on a case basis.
Such an approach will confirm that the generic mooel and the technical bases for the generic model remain entirely appropriate for these other plant types.
2.1 3 Comoliance with criteria (a) As stated on page 2-10 of the Status Report, Babcock and Wilcox has presented the results of analyses to demonstrate successful initial operation of the ECCS by showing that the entire core has been quenched. Their analyses conservatively assumed that the peak linear heat rate was at the 10 foot elevation. For a large break (8.55 square foot double enied break in.one cold leg), their analysis shows that the cladding temperature at the 10 foot elevation is reduced to essentially the saturated temperature of the ECCS coolant in the core within 400 seconds of the postulated LOCA.
3 can be shown that the temperature response for each axial elevation in the hot channel envelopes the temperature response of other elevations in the core at the same linear heat generation rate, and that the use of a core power distribution which yielded the highest peak cladding temperature envelopes the use of all other core power distributions. Babcock and Wilcox has recently proposed to modify their model to include a calculation of the highest temperature response over a range of linear heat generation rates and axial elevations. The highest metal-water reactions for each linear heat generation rate is then integrated over the entire core using a core power distribution.
The power distribution used for calculating metal-water reaction is that which yields the highest core wide metal-water reaction.
The staff finds this method adequate.
The Babcock and Wilcox method for calculating the total energy release has been modified to reduce the cladding oxide thickness upon rupture. The oxide is thinned on both reacting surfaces consisten~. with the cladding wall thinning which occurs at rupture.
This modification is acceptable to the staff.
i
5 pressure used in the sensitivity study exceeded the maximum predicted by TAFY. The initial oxide thicknesses were based upon measured values plotted as a function of exposure.
This is an acceptable assessment of the combined effects of pin pressure and oxide thickness. The resultant calculated peak cladding temperatures yielded the conclusion that the value of the pressure combined with the value of the oxide thickness used by Babcock
~
and Wilcox in their current model conservatively predict the highest peak cladding temperature.
Also, Babcock and Wilcox has recently proposed to incorporate a calculation for plastic swelling prior to rupture in both their CRAFT 2 (BAW-10092) and THETA 1B (BAW-10094) codes. A limited amount of information has been submitted to the staff and a more thorough followup with this material is expected. On a preliminary review, it appears that the Babcock and Wilcox calculation adopts the same formulation used by the staff in RELAP-4EM (WREM, October 1974). On this basis, the modification would comply with this aspect of Appendix K.
4 3 1.1 Break seectrum Page 4-20 indicates that the staff requested further information i
2 on the 0 5 ft break. This material was provided by Babcock and Wilcox. The staff considers the range of break sizes analyzed to be acceptable.
7 433 Momentum Ecuation As indicated on page 4-29, the momentum equation used by Babcock and Wilcox was found to be acceptable except for the following terms which needed additional support:
Term 2--Momentum Convection Term 3--Area Change Momentum Flux Term 6--Pressure Loss Resulting from Area Change The staff has reviewed the data submitted by Babcock and Wilcox on pressure loss resulting from area change (Term 6 above).
The pressure loss resulting from area change is based on both an analitical model as described by Babcock and Wilcox and measured plant data.
The total system resistance is within 5% of the measured values and vessel pressure drop is within 1/25 of the measured values. We find this analysis acceptable.
For momentum convection and area change momentum flux terms (Terms 2, 3 above), no additional information has been received in support of their application in the Babcock and Wilcox model.
However, the staff has performed sensitivity studies on various representations of the momentum equation. Although the form of the Babcock and Wilcox equation is not exactly that derived by the staff, we believe that the sensitivity studies performed by the. staff encompass the Babcock and Wilcox model. On the basis of these sensitivity studies, we conclude that the Babcock and Wilcox treatment of these terms results in an acceptable momentum equation.
I
m 9
A.
Noding In general, the small number of nodes is appropriate for the relatively slow moving reflood phenomenon. Any points of concern, such as the node representing the upper annulus of the downcomer, are discussed below and found satisfactory.
B..
The Conservation Eauations The assumption with regard to integration of the momentum equation is acceptable in regions where the density variation is small relative to the area variation or regions of small pressure drop. During reflood, these conditions can be considered to apply and the assumption is acceptable.
With regard to the containment backpressure assumption, a constant pressure is not used to calculate leak flow.
A variable containment pressure is calculated by the CONTEMPT code (see Section 4.4.2 in the Sts,us Report).
Babcock and Wilcox has indicated that vent valve resistance was measured over a range of Reynold's numbers which is applicable to anticipated flows, and that the flows are never at the critical value.
The remaining assumptions in this section are also acceptable.
11 is predicted to occur. The following effects, which were not originally taken into account, were addressed later by Babcock and Wilcox during correspondence with the staff:
(1)
The nitrogen is added to-the vapor which is venting from the upper plenum. This increases the gas velocity around the upper portion of the downcomer and the potential for entrainment.
Babcock and Wilcox has considered the resulting velocity and its effect on entrainment and the reflood rate. When the downcomer is filled to the break elevation, the maximum velocity reaches 200 ft-per-sec. and the entrainment becomes significant.
However, the ECC water is dumping to containment at this point, so there is no resulting reduction in the reflood driving head.
(2)
Dissolved nitrogen will separate out of solution with time as the ECC water passes down the downcomer and up through the core.
In the absence of definitive experimental results, several experts on this subject have been consulted and the staff requested that Babcock and Wilcox consider the following assumptions:
(a) initially, the accumulator water is to contain an equilibrium concentration of nitrogen.
(b) at every point in the course of its travel to the injection point, into the reactor pressure vessel, and down the 1
13 I.
Flow Path of the ECC Water All assumptions used in this section are appropriate.
J.
Sensitivity Analysis It has been established that the purpose of the study in which the quantity of primary metal was reduced to half its reference value was to demonstrate the effect on the peak clad temperature of the primary metal heat release rate.
This accounts for the conclusion referred to in the part J(a) of Section 4.4 3 2 in the Status Report that the use of nucleate boiling for the primary metal heat model is appropriately conservative.
K.
Jncut into the REFLOOD Code As discussed in part B of this section, the containment pressure is not a constant input value; it is calculated by the CONTEMPT code.
No other concerns remain in this area, but to point out that the input must all be justifiable (see Section 4.4.2 of this Supplement).
15 Hot walls, as it has been demonstrated experimentally, can temporarily hold up water flow. Once the conditions for penetration exist, the accumulated water is free to flow down to the lower plenum. This phenomena would result in an increased lower plenum refill time only if the storage capacity of the system is not sufficiant to hold the accumulated ECC water.
Otherwise, there is no effect on refill or on peak cladding temperature.
(c) The staff stated in the Status Report (page 4-54) that the Babcock and Wilcox vent valve flow resistance factor of K=3 9 was under review. Babcock and Wilcox has been supplied information contained in "Overall Calculational Methods For, Two Phase Flow," by Peter Griffith. Two-phase flow resistance factors as calculated from equations I-17, I-18, and I-19 of this publication, with a value of C=15 for gate valves, are acceptable to the staff. Due to the pressure dependence of of "C", Babcock and Wilcox is permitted to make appropriate correction for pressure effects.
4.4 5 Refill and Reflood Heat Transfer (a)
Page 4-55 imposes a 0.8h for 17 x 17 geometry cores until additional information is available. Babcock and Wilcox has indicated that they will adopt this limitation until they are able to establish that a different value is proper.
a
- ~ ~
~
r 17 C.
ERRATA 1.
TABLE OF CONTENTS Correct the following section page numbers in the Table of Contents:
4.1 7................................. 4-13 4.2................................... 4-14 4
3................................... 4-19 4 3 1................................. 4-19
~
4 3 1.1............................... 4-19 i
4 3 1.2............................... 4-20 l
i 4 3 1 3............................... 4-21 4 3 1.4............................... 4-24 4 3 2................................. 4-27 4 3 3................................. 4-28 4 3 4................................. 4-29 4 3 5................................. 4-30 4 3 6................................. 4-31 4 3 7................................. 4-33 6.0................................... 6-1 2.
Page 1-6, last paragraph, first sentence.
Change "eight" to "seven".
3 Page 2-1, first paragraph.
Revise the sentence, "The CONTEMPT code is used to...for the small break analysis" to read, "The CONTEMPT code-is used to initialize REFLOOD with a conservative backpressure calculation and the code FOAM is utilized to calculate swollen water level for the small break analysis."
_.