ML20137S117
| ML20137S117 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 12/23/1985 |
| From: | Ball S, Harrington R, Moses D OAK RIDGE NATIONAL LABORATORY |
| To: | NRC |
| Shared Package | |
| ML20137S001 | List: |
| References | |
| CON-FIN-A-9478 TAC-59787, NUDOCS 8602130580 | |
| Download: ML20137S117 (15) | |
Text
Enclos'ye 7 l
~
l TECHNICAL EVALUATION REPORT FORT ST. VRAIN NUCLEAR GENERATING STATION DOCKET 50-267 LICENSEE:
PUBLIC SERVICE CD. OF COLORADO t
EVALUATION OF CONFIRMATORY ACTION ITEMS TO SUPPORT 35% POWER OPERATION PREPARED BY:
S. J. Ball R. M. Harrington D. L. Moses Oak Ridge National Laboratory Oak Ridge, Tenn. 37831 December 23, 1985 NRC Lead Engineer:
K. L. Heitner - NRR Project:
Selected Operating Reactors !ssues, FIN No. A-9470 0602130500 060207
'^
POR ADOCK 05000267 P
PDH
i.
r l
l
- e,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
NOTICE This report was prepared as an account of work sponsored by an' agency of the United States Government.
Neither the United States Government nor any agency thereof, or any of their employees, makes any warranty, expressed or implied, or assumed any legal liability or responsibility for any third party's use, or the results of such use, of any information, apparatus, product or process disclosed in this report or represents that its use by such third party would not infringe privately owned rights.
- wson**s,sewee,e,,,,,,,,,,,,,,,,,,,,,,
e
_.m
?
Dec. 23, 1925 PAGE 1 L
l I
I EVALUATION OF CONFIRMATORY ACTION ITEMS TO SUPPORT FORT ST. VRAIN r
l NUCLEAR BENERATING STATION 35% POWER DPERATION THE ENVIRONMENTAL i
DUALIFICATION SCHEDULE EXTENSION PERIOD l
INTRODUCTIDN Oak Ridge National Laboratory (DRNL) was requested by NRC to i
provide technical evaluations of selected issues relating to the
[
acceptability the Fort St. Vrain (FSV) environmental qualification i
schedule extension.
The extension request by the licensee was made I
by the letter from Public Service Co. of Colorado (PSC) to NRC dated November 22, 1985 (P-85432).
The NRC Policy Issue (Affirmation) letter on this matter, dated November 22, 1985, is from W.
J.
Dircks j
to the Commissioners.
The purpose of the DRNL technical evaluations j
is to provide technical support for the NRC staff review of the 4
licensee's submittal (R. F. Walker to H. N.
Berkow, December 10, 1985, P-B5460) of the 14 confirmatory action items.
Preliminary I
evaluations of design basis accident scenarios by ORNL (S. J.
Ball to j
T. L. King letter of October 14, 1985 - Attachment 1) had shown that i
f or a postulated permanent loss of f orced circulation (LDFC) accident j
at 35% power, where the reactor cooldown utilizes only the liner j
cooling system (LCS) as a heat sink, the predicted total fuel failure y
is insignificant.
The DRNL analyses make use of the DRECA (ORNL
[
Reactor Emergency Cooling Analysis) code as applied especially to i
postulated FSV severe accidents. (Ref. ORNL/TM-5159).
This Technical l
l Evaluation Report addresses only action item statements 1, 2, 4, 5, L
j 9,
10, and 11 as described'in the Policy Issue.
1 t
i 1.
PCRV LCS COOLDOWN AFTER HIGH-ENERGY LINE BREAKS (HELBs):
i
!~
ACTION ITEM STATEMENT:
" Complete an evaluation which confirms that the prostressed concrete reactor vessel (PCRV) liner cooling system using firewater can be utilized to prevent significant damage to any j
of the fission product barriers, including fuel particle coatings, in the event of a high energy line break at power levels up to the 35%
r j
power restriction."
i j
A sensitivity study was run using the severe accident version of
}
the DRNL DRECA code f or an FSV initial power level of 35%.
The i
)
object was to determine if reasonable variations in either the
{
operational assumptions or the model/ parameter assumptions could affect the results such that significant damage to the fission product barriers would be predicted. In previous correspondence (Ref.
i S.J.
Ball to T.L. King letter of 10/14/85 - Attachment 1) it was j
noted that for base case permanent LDFC assumptions, the reactor l
could cool down on LCS(one loop)-only operation without significant fuel failure.
Subsequently it was determined that the LCS normal cooling system might be assumed to fail due to the HELB.
Other I
variations were postulated relating to the possible periods of time 5
af ter the LOFC/ scram / loss-of-LCS flow before the PCRV could be i
i i
4 j
Dec. 23, 1985 PAGE 2 depressurized and before the LCS flow could be restarted, wither by using the normal cooling water supply or by connecting the firewater system to the LCS.
These actions require operator access to the reactor building if the normally available remote controls are disabled by the HELB.
Estimated access-delay times are in the order of 4-5 hours and depend on a variety of assumptions not addressed here.
Earlier studies suggested that NOT depressurizing the reactor 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the start of the LOFC (per the Tech Specs - LCO 4.2.18),
and perhaps not at all, may result in an even more favorable cooldown.
FSV has the option (by law) during an accident situation to decide to disregard the tech specs and choose not to depressurize, although they have opted to attempt depressurization if f easible.
We agree with their decision to attempt depressurization.
In our reference case, with the 7-hour depressurization beginning 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the LOFC and 1 of 2 LCS trains operational, I
the maximum core temperature reached 2894 F (1590 C) after 3.5 days.
Using the Goodin (GA Technologies) time-at-temperature f uel failure i
model, the (5 day) fuel failure was 0.25%, and only 2 of the 37 coverplates above the top of the core reached the (conservative) assumed failure temperature of 1500 F (at t=3.3 days).
In variations in which the LCS is assumed to f ail initially,
)
there is a question about how hot the PCRV liner may get and still permit a satisfactory restart of the LCS.
NRC has apparently accepted 1500 F as this maximum liner temperature (per FSAR).
RECA l
code (GA) studies show that the inlet plenum liner temperature for an l
inoperative LCS reaches 1500 F shortly (2 hr) after the assumed i
failure of the upper " cover plate" (assumed to be one average plate) at 1500 F.
In the ORNL ORECA studies, where the coverplates are modeled individually (but the top LCS is modeled as a single average tube), there is typically localized coverplate f ailure predicted before the times RECA predicts the gross failure.
To be on the conservative side, one may assume that a liner hot spot, possibly precluding LCS restart, may occur shortly following loss of the first coverplates.
On the other hand, recent coverplate/LCS failure experiments at KFA (Ref. J.
Walters, personal communication 12/12/85) indicate that for similar coverplates, essentially no distortion or failure was seen for coverplate temperatures of up to 1000 C (1832 F), the FRG's limiting temperature.
Another possible consideration, (from the standpoint of structural or thermal stress problems) for choosing a conservative limiting liner temperature would be to assume failure at the nominal maximum operating temperatures for conventional boilers of the same material.
A rule of thumb value for this tem erature is about 500 C (930 F).
In our reference case pressurized run, the maximum (average) liner temperature reached 930 F shortly after the first coverplate failures occurred (at 18 hr after the LOFC).
The ability to restore (l i quid) water flow to the LCS given an j
initially hot liner may also be in question, besides the structural integrity questions that may arise.
These concerns are addressed in our response to action item 11, where it is estimated that, in the worst case, the minimum required restart time would be 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
D c.
23,.1985 PAGE 3 Hence the structural con si derati ons, as noted above, would be limiting, and we would conclude that restarting the LCS within 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> of the HELB/LOFC would assure a safe shutdown.
Other scenarios studied included various delays in.
depressurization and various initial periods for which the LCS is disabled.
These also resulted in satisfactory LCS restarts, but with somewhat higher maximum fuel temperatures and slightly higher estimated fuel failure fractions.
Sensitivity studies included a model variation for the pressurized runs in which values of the assumed coverplate failure temperatures were much higher than the reference failure limit of 1500 F.
This gave " good" results in that the coverplates remained intact, but " bad" results in that the ' core remained much better insulated from its LCS ultimate heat sink.
In the limit, however, the results of " worst-case" assumptions of no coverplate failure would at worst approach the depressurized results, which have been shown to be satisfactory.
Cases corresponding to those presented by PSC (P-85460, Att. 1) were run, and the results compared and presented in Table 1.
Case 1 assumes depressurization begins at 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and LCS flow is resumed at 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />.
Case 2 assumes that depressurization begins at 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> and LCS restart at 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />.
Case 3 assumes the PCRV remains pressurized and LCS cooling is restarted at 29.4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, the time at which PSC predicts massive failure of the upper plenum coverplates (at 1500 F).
Case 4 is like Case 3 except that the coverplates are assumed not to fail.
In general, the results are in good agreement, and the ORECA results support PSC's conclusions.
The worst cases from the standpoint of fuel failure were for the depressurization runs.
In those cases the maximum fuel temperature approached 2900 F (1600 C) after 3.6 days, and the total (5 day) fuel failure was 0.3%.
- Again, it should be noted both that the Goodin fuel failure model is not claimed to be very accurate at these threshhold failure temperatures, and that due to holdup of the fission products in the graphite lattice, the actual release fractions for most nuclides would be less than the fuel, particle f ailure fractions.
2.
PCRV STRUCTURAL INTEGRITY:
ACTION ITEM STATEMENT:
" Evaluate the leak tightness and structural integrity of the PCRV during the heatup which.would occur following and extended LOFC resulting from an HELB f rom 35% of rated power.
Consider the cold reheat helium interspace leak, PCRV penetrations and seals, and other portions of the PCRV where leakage may be a concern.
Actual PCRV leakage experience should be considered (e.g.,
the recent LER on the PCRV penetration cold reheat helium leak)."
A review has been performed which examined and evaluated PCRV leakage claims and limits recorded in the FSAR, the Technical Specifications, the Reference Design Book, and recent submittals by the licensee.
Based on this review, the PCRV leakage assumption made by the licensee in their most recent licensing submittal (P-85460) e
. Dec. 23, 1985 PAGE 4 appears to be very conservative or at least made that way by the design basis assumption of primary coolant radioactivity (30,000 Curies) available for release early in the core heatup transient.
In effect, as an upper bound the licensee could have simply compounded the DBA No. 1 and DBA No. 2 doses at the LPZ and still have been well below 10CFR100 limits (compare Table 14.11-1, Updated.FSAR to Table 1 in Attachment 5 to P-85460).
As indicated above, the licensee has recently submitted a Jrevised analysis of the DBA No. 1 assuming an initial leakage rate of 1267 lb/ hour with no depressuriation (see attachments 1 and 5 to P-85460, December 10, 1965).
The total initial leakage is derived 1
from assuming that no circulator shafts are initially sealed (1250 lb/ hour) plus a continuing leak of 401 lb/ day which continues after the circulator shafts are sealed. The source of the esM AF;w of circulator seal leakage could not be found in either Seti.un 4 of the FSAR or. System 21 of the Reference Design Book.
The 401 lb/ day derives from a 400 lbm/ day leak across secondary closures at full i
pressure (688 psi per LCO 4.2.9) plus a continuing leak of 1 lb/ day into cold reheat.
The 1 lb/ day was reportedly derived based on conservative assumptions.
Under these assumptions, approximately one fifth of the primary coolant is released during the first hour of the accident.
Technical Specification LCO 4.2.9 limits total leakage of purified helium from the penetration through the primary closure and into the primary system to less than 400 lb/ day at 10 psi differential pressure.
This is conservatively assumed to be equivalent to the leakage of primary system radioactive coolant out at 1145 lb/ hour at 688 psi.
As referred to above, LCO 4.2.9 also limits the total leakage of purified helium through secondary closures to 400 lb/ day at 688 psi.
Although not described in the basis for LCO 4.2.9 or SR 5.2.16, there is a flow alarm function (FAH-11263) on the main purification train pressurizing line, which feeds all penetrations.
This alarm has a nominal set point of 400 lb/ day.
The actual alarm setpoint is below the nominal set point.
The alarm is located in the control room.
In addition there are separate high flow (275 lb/ hour) alarms and isolation switches on both lines to the steam generator penetration, each line to a circulator penetration, a the common feed line to the other penetrations.
Furthermore, the licensee controls the 700 lbm/ day limi,t on the steam generator penetrations having a reheat steamline leakage path by monitoring the differential pressure between the penetration and the reheat steam line.
Penetration pressure is controlled to slightly above that of reheat steam pressure by using the penetration inlet line helium pressure control valve.
In so doing, the leakage rate for purified helium is apparently kept to 1.3 lb/hr at 2 psi, based on reported surveillance results.
This value linearly extrapolates (laminar flow assumption) to 450 1b/ hour at 688 psi, which is within the 700 lb/ day limit in LCO 4.2.9, and also less than the 1145 lb/ hour limits on the primary closure leakage into the depressurized penetration at 688 psi, i.e.
the existing leak path appears to be less severe than the primary closure limLt.
A major
=
Dsc. 23, 1985 PAGE 5 1eak developing into the reheat steam lines would not only be detected by the pressurizing line high flow alarm (16.7 lb/ hour) but would also cause the loss of condenser vacuum due to non-condensibles who'e the (helium), as occurred in the event reported in RO 81-068, r
purified helium leak rate apparently reached 1312 lb/ day.
In addition, LCO 4.2.9 limits primary coolant.1eakage indirectly by limiting the allowable radioactivity leak rate through the condenser air ejector to less than 1.4 Curies / day.
Such leakage would be through the. reheat steamline leak path.
Also, other alarmed monitors are provided along the reheat leak path.
The sensitivity of these monitors is stated to be much lower than the allowable leak rate with an additional LCO requirement to check primary closure integrity if the monitored radiation levels along the leak path increase by 25%.
i If we assume a 1.4 Curie / day leak rate at an expected equilibrium primary system radioactivity level of between 300 Curies (full power) and 105 Curies (35% power), then the primary closure leak rates could be projected (assuming no path losses) at between O.47%/ day and 0.95%/ day at 2 psi differential pressure.
On page 11 i
of Attachment 1 to P-85460, the licensee estimated only 0.6 lb/ day because they normalized to the design value rather than the best estimate level of primary coolant radioactivity, although they did a
assume 50 psi for the differential pressure.
This assumption may or may not have been conservative with regard to the size of potential leak rates, but is apparently made conservative by assuming a circulating radioactivity level of 30,000 Curies.
At the depressurized condition (5 psi differential) assumed in Appendix D of the FSAR, the upper limit on the reheat path leak rate j
would be between 1.2%/ day (or l'ess) and 2.4%/ day, which is much higher than the assumption in Appendix D.
However, if we examine the 400 lb/ day (5.4%/ day) limit at 10 psi for primary closure leakage, we get 3.8%/ day at 5 psi using conservative downscaling (turbulent flow I
assumption).
This result is higher than that from the radioactivity release through the condenser air ejector.
Therefore, both of the LCO 4.2.9 limits equate to higher depressurized leak rates than the l
Appendix D assumptions, but the licensee's most recent analysis (Attachment 5 to P-85460) is much more conservative by assuming that':
(1) the reactor remains pressurized, (2) the circulator shafts are delayed in being sealed, and (3) the primary coolant radioactivity is at design level (30,000 Curies).
4.
MAXIMUM TIME BEYOND WHICH DEPRESSURIZATION SHOULD NOT BE ATTEMPTED 4
ACTION ITEM STATEMENT:
" Provide an evaluation of the estimated time and primary coolant temperature beyond which the PCRV should NOT be depressurized during the remainder of PCRV liner cooldown following a postulated high energy line break f rom the 35% power level. "
i i
For 35% operation, it has been established that depressurization following an LOFC is not necessary, assuming a satisfactory and
D c.
23, 1985 PAGE 6 timely restart of the LCS.
However, there is concern about the ability of the High Temperature Filter Adsorber (HTFA) cooler to handle the sensible heat load from unusually hot letdown gas during the delayed depressurization.
The FSAR states that depressurization should not be attempted if the upper plenum gas temperature exceeds 1350 F, and PSC calculates that this temperature is reached 12 hcurs into a pressurized LOFC accident with LCS failure.
The ORECA calculation confirms this (1344 F).
However, we were also concerned that the gas letdown temperature may increase substantially.during the later stages of the depressurization and overtax the HTFA cooler.
In a very conservative calculation in which all of the letdown gas was assumed to flow through the hot core, the ORECA calculation predicted a maximum letdown temperature of 1400 F.
Independent performance calculations of the HTFA. cooler showed that cooling of the letdown gas, in the worst case, gave a maximum inlet gas temperature to the HFTA of 700 F (compared to a 1000 F design limit per the FSAR Sect. 9.4), and a subsequent maximum outlet temperature i
from the helium purification cooler of 200 F, compared to its normal temperature of 120 F (no design limit is specified).
Hence we conclude that the PSC estimate of 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> as a maximum permissibic del'ay prior to depressurization is quite conservative.
5.
MAXIMUM OPERATING POWER FOR PRESSURIZED PCRV COOLDOWN ON THE LCS ACTION ITEM STATEMENT:
" Provide an estimate of the maximum reactor power level at which it would be safe to perform a pressurized PCRV liner cooldown while protecting the integrity of the PCRV liner and liner cooling system."
The Tech Specs (LCO 4.2.18) require depressurization within about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> of an extended LOFC accident due to the possibility of overloading the upper plenum LCS.
With the reduced loads expected from a 35% power LDFC, however, it was established that depressurization was not required.
To address concerns about the margin'available, LOFCs from higher power levels were run, assuming only one of the two LCS trains to be operational and with no flow redistribution or cooling system cover pressure.
The results indicated that for power levels up to 60%, no boiling was predicted.
This result confirms the PSC estimate of
">50%".
9.
PCRV DAMAGE DURING COOLDOWNS ACTION ITEM STATEMENT:
" Document the extent of PCRV damage expected during a liner cooldown from 35% power following an HELB."
Estimates of overheated PCRV concrete temperatures were noted in item 1.
The worst case shown, with the >400 F penetration to 15 in.,
would not present a safety problem, especially if depressurization is accomplished.
There is some discrepancy in the PSC and ORNL estimates of upper plenum coverplate damage for the scenarios noted t
l 1
l Dxc. 23, 1985 PAGE 7 g
in item 1.
Assuming a 1500 F failure temperature, the ORNL' calculation predicts failure of a small number of coverplates and its j
accompanying insulation (above the center of the core) for the' scenarios in which the primary system is depressurized.- (Even these few failures enhance the core cooldown process.)
The PSC calculation apparently uses a single coverplate model for the upper plenum i
ceiling, and as a result its average temperature does not reach 1500 F or fail (and consequently the peak core temperature is higher i
than that predicted by ORNL) for the depressurized cases.
As noted in item 1, however, the 1500 F limit is probably very conservative, so the depressurized cases may not lead to individual cover plata failures (Ref. the personal communication with J.
Walters of KFA).
j In the pressurized case scenarios, coverplate failure is predicted at 1500 F by both PSC and ORNL.
The first coverplate failure (ORNL) is at 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />, while PSC predicts (total) upper plenum coverplate failure at 29.4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.
For the case where the coverplates are l
assumed not to fail (case 4), the maximum oredicted coverplate i
temperatures are 1780 F (ORNL) and ~1600 ~ L'SC).
Again, much of this j
discrepancy may be due to the differences in the coverplate modeling schemes.
Conservative calculations of possible LCS liner failure for the worst-case scenarios require postulated delay times for LCS restart which range from the ORNL estimate of 18-20 hours (i. e., 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> for the first coverplate failure plus 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> for the liner behind the coverplate to reach 1500 F), to 30-32 hours (PSC).
10.
PCRV HOTSPOTS ACTION ITEM STATEMENT:
" Evaluate the effect of a PCRV liner cooldown from the 35% power level on the previously analyzed hot spots."
PSC submittal P-78037 provides an evaluation of seven hat spots at various locations throughout the PCRV.
These locations are dubbed
" hot spots" because data from individual LCS tubes or from thermocouples embedded in the concrete showed during low power operations through 1978 that local concrete temperatures might exceed 250 F if the plant were operated at 100% power.
P-78037 gives predictions of maximum local concrete temperatures, LCS tube delta-Ts (outlet minus inlet temperature), and stress effects on the affected structures, and concludes that the temperatures and stresses would be acceptable for full power operation.
We have performed a very crude extrapolation of the calculations reported by Table 1-1 of P-78037 from full power to the case of an LOFC from 35% power, with the assumptions of continued LCS coolant flow and no PCRV depressurization.
The method of extrapolation is to ratio the heat transfer rates and temperatures in accordance with the appropriate environmental temperatures predicted by ORECA-FSV for the subject LOFC accident.
The conservative assumption is made that the heat transfer coefficients are the same under the accident conditions:
i 4
wa---
--v3, cyv.,
m,--,v...,.,,,,.,,m..me.-.,~,w,.-,m,-ww.,,..,
,w.
,--,,,,,..m,
,,-----..,.-.ve
I i
I Dec. 23, 1985 PAGE 8 LCS cooling LCS cooling Local HOT SPOT
- tube deltaT tube deltaT Maximum (F) w/o re-(F) w/ re-Concrete.
distribution distribution Temp (F)
Top head pen-65 74 311 l
atrations Core outlet thermocouple penetrations 45 14.5 587 Core Barrel seal area 72 347(boiling) 448**
Crosover i
pipe 6
7 531
- there are three hot spots not mentioned here because they are at the i
bottom of the PCRV and thus out of the picture.
- this prediction not valid for the case of boiling.
These results predict concrete concrete temperatures that are acceptable from the standpoint that the outcome of the LOFC at;cident is not significantly influenced one way or the other.
The LCS coolant delta-T results are also acceptable from the same point of view.
The prediction of boiling for the Core Barrel Seal hotspot tube's occurrs only if the LCS flow is redistributed.
ORECA-FSV predicts that the bottom of the core barrel would be heated to about 1100 F after 5 days,.so no major change of the concrete geometry would be expected.
No discussion is provided here of the stress levels expected in affected strutural components.
The temperature levels predicted do not seem to indicate catastrophic structural failure although some of the components might sustain some damage.
This is acceptable f or an infrequent severe accident such as permanent LOFC.
11.
LCS RESTARTS:
ACTION ITEM STATEMENT:
'" Evaluate the impacts on the integrity of the PCRV liner cooling system of re-establishing liner cooling after prolonged periods of core heating without liner cooling or forced circulation cooling.
Verify that the impacts of re-establishing liner cooling fs11owing a postulated HELB from 35% power with a pressurized liner cooldown would be no worse than those for the
-depressurized liner cooldown after 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> from full power previously evaluated in the FSV FSAR.
Compare the associated heat fluxes to the maximum acceptable heat flux to the PCRV liner cooling system assuming single loop operation."
If steam is produced upon flow resumption, it mighf create a
Duc. 23, 1985 PAGE 9 back pressure that could not be overcome by the normal or f'irewater pump coolant supply.
A variation of the LCS flow model was incorporated into ORECA that allowed parametric studies of restart with boiling.
Variables that affect the eventual cooling flow include minimum LCS (saturation) pressure for full (li quid) flows and the ratio of inlet-to-outlet flow resistances.
With. inlet throttling, the sensitivity of flow to additional pressure drop from flashing would be reduced.
For what we considered to be "best estimate" parameters, flow for the base (pressurized) case could be re-established for the " normal" flows at any time during the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after scram and for the " enhanced" (planned) firewater flows at any time during the first 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />.
Hence the limiting conditions for LCS restart are more likely to be due to structural / thermal stress considerations (noted in item 1), which (conservatively) require restart in 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />.
The LCS is capable of removing, without boiling, much more heat than the core can supply.
This is especially true af ter the LCS flow is redistributed to maximize flows to the " top head / upper barrel" regions that are directly above or across from the reactor core.
For example, assume for illustrative purposes a cooling water inlet temperature of 85 F and an outlet temperature of 240 F (saturation temperature at the estimated 10 psig minimum LCS tube outlet pressure); assume also that a coolant flow of 850 gpm is available to the top head / upper barrel region after redistribution.
A simple flow
- delta-h calculation then yields the maximum non-boiling heat removal rates of 4.7 MW from the upper head region, and 14.5 MW from the sidewall region opposite the core.
These heat removals are greatly in excess of the total decay heat production rate by the core--about 1.5 MW at 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after shutdown from 35% power.
For this reason, when the restart of LCS flow is delayed long enough f or there to be boiling, the boiling typically' lasts only for a short time, on the order of minutes.
When the LCS flow is delayed too long for the boiling to be overcome and flow to be reestablished, it is not because of excessive heat fluxes from the core, but rather because the steam binding effect cannot be overcome by the outlet pressure of the firewater pumps supplying water to the LCS.
e
,_,r.
TABLE 1.
Cases Corresponding to PSC Scenarios (Attachsent 1 to P-85460)
PSC FUEL FAILURE 1500'F COVER MAX. LINER MAX DEPTH CASE T-PEAK FUEL PSC:
ORNL:
PLATE FAILURES TEMP T-CONCRETE >400'F
%>2900'F
% FAIL (Goodin)
_ORNL _
1 3140*F 2898'F 5%
0.3%
NONE 1 of 3,?
340'F
'4480'F 0
0 0 4.5d 63.6d 05h 05h 2
3160*F 2890*F 5%
0.3%
NONE 1 of 37 570*F E635'F 4 in.
- 3 in.
95.2d 93.6d 014h
@l4h 3
2150'F' 2485'F 0
0 ALL ALL 900'F 1480'F 14'in.
15 in.
029.4h 036h 029.4h 024h 029.4h 030h 4
2350'F 261&*F 0
0 (assume no fail) 570'F N635'F 4 in.
~3 in.
96.3d
@4d 014h 014h t
I f
.-s Enclesuf.13-Revision l PUBLIC SERVICE COMPANY OF COLORADO COPMITMENTS RELATED TO THE OPERATION OF THE l
FORT ST. VRAIN NUCLEAR GENERATING STATION l
(FROM NOVEM8ER 19, 1985 CAL)
\\
Item Cossnitment Description / Requirement (Reference)
Schedule 1.
Implement a program to maintain CRDM temperature within acceptable limits.
(P-185199 and P-85242) complete (see Item 8) 2.
Provide an improved CRDM Surveillance and preventative maintenance p ogram (P-85199 and P-85242),
6 months which includes consideration of the new bearing performance (P-85201.
after restart 3.
Implement a procedure to prevent overdriving of the CRDM.
(P-85040) (see 50P 12-01,. Issue 15) complete 4.
Implement a proc'edure to require a reactor shutdown under conditions where.CRDM purge flow is lost-complete or when high levels of moisture exists in the coolant.
(P-85040)
(see Item 8) 5.
Monitor CRDM temperatures.
(P-85199 and P-85242) 6.
Provide status' reports / progress reports on the Nuclear Performances Enhancement Program on a approximately quarterly basis.
(P-85217) every 3 months 7.
Implement the Technical Specification (TS) Upgrade Program (P-85098) 11/30/85 l
4 a.
Provide a final draft proposal of the Upgraded TSs for NRC comment.
(P-85243) b.
Submit a license amendment to incorporate the Upgraded TSs.
(P-85243) 90 days after 4
NRC coments on Item 7a c.
Implement the Upgraded TSs.
(P-85243)
Approximately the 4th Refueling Outage 1
(License Admend-ment will define) 8.
Improve control rod and reserve shutdown reliability.
a.
Propose Technical Specification changes..(P-85242) complete b.
Implement proposed requirements through interim procedures.
(P-85242) complete c.
Incorporate the proposed Technical Specifications in the Upgrade Program.
(see Item 7) i
.4 -
"4 i
Revision'1 '
Page 2 PUBLIC SERVICE COMPANY OF COLORADO COMMITMENTS RELAltu TO THE OPERATION OF THE.
FORT ST. VRAIN NUCLEAR GENERATING STATION 7
I I
Item Commitment Description / Requirement (Reference)
Schedule i
9.
Develop a plan to implement approved modifications to control moisture ingress and submit annual.
Annually reports on progress.
(P-85022 and P-85082)
Report-i Progress l
10.
Control moisture in CRDM purge system.
(P-85032) 4th Refueling Outage l
11.
Perform environmental'requalification testing of a CRDM assembly (P-85032) including the 12/30/85 l
temperature sensor epoxy.
(P-85195) i 12.
Refine the " watt-meter" test for verification of control rod full insertion or develop an 6 months alternative test.
(P-85040 and P-85199) after restart
~
(see Item 2) 13.
Investigate a design change to provide a positive stop'to prevent CRDM overtravel; report results 1/1/86 to the NRC.
(P-85003) 14.
Conduct an integrated systems study to resolve control rod position indication, maintenance and 6/30/86 operability questions.
(P-85003) i 15.
Implement QA Procedure Review Program.
(P-85028) 7/1/86 4
16.
Liquid Effluent Releases from the Reactor Building Sump
- a. Inglement procedures to perform all effluent releases from the reactor building sump in
. complete i
batch mode.
(P-85212)
- b. Investigate installi.g in-line, beta sensitive, effluent monitors and report on progress.
in progress j
(P-85212) 4 i
.s
- 1C Revision 1 Page 3
-PUBLIC SERVICE COMPANY OF COLORADO COMMITMENTS' a
RELATED TO THE OPERATION OF THE FORT ST. VRAIN NUCLEAR GENERATING STATION Item Commitment Description / Requirement (Reference)
Schedule 17.
PCRV Tendon Surveillance Requirements
- a. Incorporate the revised Tendon Surveillance Program into the Technical Specifications.
(see Item 7)
(P-85199)
- b. Implement the revised program through the use of interim procedures.
(P-85199) complete
- c. Provide the results of the revised program to the NRC Approximately Every 6 months 18.
Propose modifications in the Emergency Diesel Generator circuit breakers' control circuitry to 9/15/85 resolve the independent problem.
(P-85208) 19.
Fire Protection-Interim Requirements
- a. Implement the special (and interim) repair and operating procedures.
(P-85113 and P-85245) complete
- b. Implement a fire watch program.
(P-85245) complete 20.
Complete the CRDM bearing performance testing and incorporate results Report on into the improved CRDM surveillance and preventative maintenance program progress identified in Item 2 above.
(P-85201) quarterly i
-