ML19295A103
| ML19295A103 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 09/12/1980 |
| From: | Jensen W Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML111090060 | List:
|
| References | |
| FOIA-80-515, FOIA-80-555 NUDOCS 8009290458 | |
| Download: ML19295A103 (14) | |
Text
.
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATCMIC SAFETY AND LICENSING BOARD
~
In the Matter of
)
)
METROPOLITAN EDISON COMPANY
)
Docket No. 50-239
)
(Restart)
(Three Mile Island Nuclear
)
Station, Unit No.1)
)
NRC STAFF TESTIMONY OF WALTON L. JENSEN, JR., RELATIVE TO PRIMARY SYSTEM NATURAL CIRCULATION UCS CONTENTION 1 Q1) Please state your name and position with the NRC.
A)
My name is Walton L. Jensen, Jr.
I am an employee of the U.S. Nuclear Regulatory Commission assigned to the Reactor Systems Branch, Division of Systems Integration, Office of Nuclear Reactor Regulation.
From June through December 1979, I was assigned to the Analysis Group of the Bulletins and Orders Task Force, Office of Nuclear Reactor Regulation.
Q2) Have your prepared a statement of professional qualifications?
A)
Yes. A copy of this statement is attached to Contention ECNP-ld.
Q3) Please state the nature of the responsibilities that you have had with respect to the Three Mile Island Nuclear Station - Unit 1.
A)
The accident at Three Mile Island Unit 2 (TMI-2) on March 28, 1979, involved a feedwater transient coupled with the equivalent of a small 4009.1104C8
break in the reactor coolant system, though the accident's ultimate severity resulted from a number of interacting elements including lark of complete ur.derstanding of system response, misleading instrument readings and inadoquate operator training and procedures.
Because of the rr ulting severity of ensuing events and the potential generic applicability of the accident to other reactors, the NRC staff initiated prompt action to:
(1) assure that other reactor licensees, particularly those plants such as TMI-1 which have a similar design to TMI-2, took the necessary actions to substantially reduce the likelihood of future TMI-2-type events from occurring, and (2) initiate comprehensive investigations into the potential generic implications of this accident on other operating plants.
To accomplish some of the work, the Bulletins and Orders Task Force (B&OTF) was established within the Office of Nuclear Reactor Regulation (NRR) in early May 1979.
The B&OTF was responsible for reviewing and directing the THI-2-related staff activities associated with loss of feedwater transient and small break less-of-coolant accidents (LCCAs) for all operating plants to assure their continued safe operation.
I was assigned to the Task Force in June 1979.
I participated in the preparation of NUREG-0565, " Generic Evaluation of Small Break Loss-of-Coolant Accident Behavior in Sabcock & Wilcox Designed 177-FA Operating Plants."
Following my assignment to the Reactor Systems Branch, I participated in the evaluation of potential feedwater transients at operating B&W plants 2
and participated in the final preparation of the staff Safety Evaluation on the Three Mile Island 1 restart.
Q4) Please state the purpose of this testimony.
A)
The purpose of this testimony is to respond to UCS Contention 1 whicn reads:
"The accident at Three Mile Island 2 demonstrated that reliance on natural circulation to remove decay heat is inadequate.
During the accident, it was necessary to operate at least one reactor coolant pump to provide forced cooling of the fuel. However, neither the short nor long term measures would provide a reliable method for forced cooling of the reactor in the event of a small loss-of-coolant accident ("LOCA").
This is a threat to health and safety and a violation of both General Design Criterion ("GDC") 34 and 35 of 10 CFR Part 50, Appendix A."
QS) Define what is meant by natural circulation.
A)
Natural circulation is the phenomenon by which circulation of reactor coolant is maintained through the coolant loops under conditions when the primary coolant pumps are not available to provide forced circulation.
The flow through the loops is produced by the unequal fluid densities in that section of the primary coolant loop that is heated by the core and in that section which is cooled by the steam generators.
The unequal densities produce an unbalanced force which produces circulation of the reactor coolant as predicted by Newton's Law of Motion.
The rate of natural circulation flow is a function of the core decay heat rate since the greater the core heat rate the less will be the density of the water in the core and hot legs and the greater will be the unbalanced force.
The cooling rate of the steam generators also acts to increase natural circulation by increasing the density of the water in the steam generator and cold legs.
3
Q6) How is the reactor heat removed from the primary system by natural circulation?
A)
The process of heating the circulating primary coolant water in the core and of cooling that water in the steam generators transfers heat from the core to the secondary system water in the steam generators.
The heat addition to the steam generator secondary coolant water causes it to boil and turn to steam.
This steam will ficw into the condenser, if it is available, or through relief valves to the atmosphere.
Make-up water to the steam generators is provided by the main feedwater system, if it is available, or by the emergency feedwater system if main feedwater is not available.
In the absence of a leak, there would be no net loss of water from the primary system.
However, in case of a small break LOCA, the water lost from the break is made up by the High Pressure Injection system.
Q7) Will power operation be permitted when the reactor coolant pumos are not operating?
A)
No, the core safety rods will be automatically inserted (tripped) so that the reactor will be shut down at any time when the reactor conlant pumps stop operation.
For this reason, natural circulation is required to remove only decay heat.
Decay heat is about 7% of full power when the reactor is first shut down.
The rate of heat generation rapidly decreases so that at I hour after shutdown the rate of heat generation is only 1.4%
of full power.
Q8) How does the natural circulation heat removal requirement for loss of coolant accidents compare to that required during a transient causing reactor coolant pump trip.
4
A)
In the case of a transient, all decay heat would be removed by natural circulation whereas in the case of a LOCA, some or all of the decay heat depending on the size of tha break, would be removed from the break by being carried out with the lost coolant.
For break sizes greater than approximately 0.01 sq. ft., sufficient energy would be discharged through the break so that the system could be depressurized without reliance on the reactor coolant (RC) pumps or natural circulation.
For break sizes smaller than approximately 0.01 sq. ft., only a portion of the decay heat would be removed through the break and natural circulation would remove the remainder of the decay heat.
By these processes, the primary system would be cooled and depressurized.
Q9) What evidence exists that natural circulation is effective in removing decay heat following a reactor transient for which the reactor coolant pumps are tripped.
A)
Data collected at operating B&W plants has shown that natural circulation is an effective means of removing decay heat following a reactor trip.
This information is contained in Appendix 1 to the B&W report, " Evaluation of Transient Behavior and Small Reactor Coolant System Breaks in the 177 Fuel Assembly Plant," May 7, 1979.
Q10) Certain plant transients and accidents including LOCA might produce steam in the reactor coolant loops. What evidence exists that natural circula-tion is effective in the presence of steam voids?
A)
The plant data discussed in the response to Question 9 was for conditions during which there were no steam voids in the primary coolant loops.
Steam voids would effect the natural circulation flow rate by increasing 5
\\
the unbalanced force if the voids were on the side of the loop containing the core and by decreasing the unbalanced force if the voids were on the side of the loop containing the steam generators. Analyses by B&W have 2
indicated that for break sizes of.01 ft and smaller, sufficient steam could be formed in the hot leg region above the steam generators to temporarily block natural circulation.
These analyses are described in the B&W report, " Evaluation of Transient Behavior and Small Reactor Coolant Assembly Breaks in the 177 Fuel Assembly Plant," May 7, 1979.
Q11) Would this temporary loss of natural circulation be significant?
A)
The temporary loss of natural circulation would not be significant.
The core would continue to be adequately cooled as long as it was covered by liquid.
As additional water was lost from the primary system through the break, the region of voiding was shown by the analyses to dip into the steam generators to the level of the emergency feedwater inlet sparger. At this point, two-phase natural circulation was shcwn to be established by the condensation of steam produced by the core on the inside of the steam generator tube surfaces.
Q12) What evidence exists that two phase natural circulation will be adequate to remove core decay heat for those break sizes of 0.01 square feet and smaller for which natural circulation may be required.
A)
The LOFT L3-2 small break test was designed to operate with two phase natural circulation.
In this test, two phase natural circulation did occur and was adequate to remove decay heat.
6
Q13) What conditions are important to assurance that natural circulation established following a SBLOCA?
A)
An important factor in the establishment of natural circulation is that a steam-condensing surface will exist in the steam generators before the core could begin to uncover.
For lowered loop plants like TMI-1, this occurs because the EFW enters the steam generator from the top.
For all B&W lowered loop plants, the Small Break Emergency Procedures require that the levels in the secondary side of the steam generators be raised to 95% on the operating range level indicators if the RCPs are not running.
Emergency feedwater is automatically fed to the steam generators wnen the level reaches the low level limits if the RCPs are running (30 inches on the startup range indication) and to 50% of the operating range if the primary coolant pumps are not running. Analyses by B&W show that emergency feedwater will be initiated before the vessel water level drops below the top of the core.
Assurance that emergency feedwater will be available is provided by improvements made to the reliability of the Emergency Feedwater System.
See pages Cl-1 to Cl-12, C2-6 to C2-9, and C8-34 to C8-40 of the NRC Safety Evaluation for THI-1 restart NUREG-680.
Q14) If natural circulation was not effective at TMI-2 in removing the decay heat, why would it be effective at TMI-1?
A)
Natural circulation did not occur at TMI-2 when the reactor coolant pumps were initially tripped because of insufficient coolant inventory.
This was precipitated by operator action in which the HPI was prematurely terminated while coolant was still being lost th,ough the PORV.
In latar 7
stages of TMI-2 recovery (up to the present day), after an adequate coolant inventory was restored, the core was successfully cooled by natural circulation in spite of the severe flow blockage expected to the damaged core.
Q15) What conditions are important to maintenance of natural c.'culation once it is established?
A)
In order for natural circulation to be maintained, a sufficient amount of coolant must be retained in the primary system during the course of the accident to keep the water level at least as high as the bottcm of the cold leg vessel inlet nozzle.
If this condition is met, steam formed by boiling in the reactor vessel will be transferred to the steam generators and condensed to flow back to the reactor vessel as liquid producing a continuous process.
It less water is in the primary system than that required to reach the cold leg inlet nozzle, water lost from the reactor vessel by boiling will not be sufficiently replaced so that eventually the core would beccme uncovered. The staff believes that this condition occurred at TMI-2.
Although the core could be uncovered for a brief period without exceeding the core damage limits of 10 CFR 50.46, during an extended period of ur.covery the decay heat would raise the core temperature so that these limits would be exceeded.
Increased assurance of a sufficient coolant inventory for natural circu-lation at THI-1 has been achieved by the improved termination criteria for the high pressure injection (HPI) system described in the NRC staff Safety Evaluation Report (Pages C2-4 and C2-5).
These criteria are provided in the control room in the form of emergency procedures which 8
instruct the operator against prematurely terminating HPI in the event of a SBLOCA.
Q16) If natural circulation cannot be maintained (for example, if both main feedwater and the backup emergency feedwater are lost), is there another backup means available to cool the core at TMI-1?
A)
If it is postulated that natural circulation cannot be establisned (because of loss of both main and emergency feedwater), a capability exists in TMI-1 to cool the core in the feed-and-bleed mode. Water is injected into the primary system (cold legs) by the high pressure injec-tion system, and the core decay heat is removed through the pressurizer pilot operated relief valve (PORV) and possibly the safety valves.
The cutoff head of the high pressure injection pumos is above the pressure settings of the safety valves (2500 psig) so that injection can continue even though the system is at high pressure.
This cooling mode would be maintained until the primary system can be depressurized.
Q17) What evidence exists that the feed and bleed method of cooling the reactor will work?
A)
An analysis by B&W for the case of a small break LOCA for which feedwater was unavailable was reported in " Evaluation of Transient Behavior and Small Reactor Coolant System Breaks in a 177 Fuel Assemoly Plant," May 7, 1979.
This analysis indicated that even if feed and bleed were delayed for 20 minutes, no core uncovery would occur if 2 HPI pumps were available.
Additional analyses were presented by B&W in a letter from J. Taylor, B&W, to R. Mattson, NRC, May 15, 1979.
These analyses indicated that in the absence of a LOCA, feed and bleed wou' ' cool the core even if HPI was not initiated for 30 minutes.
9
Q18) Site an Example:
A)
The event at Crystal River Unit-3 of February 26, 1980, was essentially a feed and bleed event.
The HPI system injected water into the reactor system for a period of 28 minutes.
The excess water was relieved from the primary system first by the PORV until it was isolated and then later by the safety valves. Main feedwater was lost initially in the event and emergency feedwater was not established for 9 minutes. This 9 minute period was the tima for the operator to sanually override the failure of the Integrated Centrol System which prevented the i= mediate actuation of emergency feedwater. There was no evidence of core overheating.
Q19) A large amount of hydrogen gas was generated at TMI-2 which could not be condensed. What would be the effect of noncondensible gas on natural circulation?
A)
The effect of noncondensible gas on natural circulation is addressed in NUREG-0565, " Generic Evaluation of Small Break Loss-of-Coolant Accident Behavior in Babcock & Wilcox Designed 177-FA Operating Plants," January 1980.
It was concluded that the expected amounts of noncondensible gases that could accumulate in the primary system during a small break LOCA are small ccmpared to the amounts needed to:
(1) block natural circulation flow; (2) significantly degrade steam generator condensation heat transfer; and (3) invalidate single-fluid (steam-water) analysis models.
As an additional precaution, a safety grade vent system for the removal of noncondensible gases is being installed at TMI-l at the reactor system high points (reactor vessel head, pressurizer, and hot legs).
These vents are discussed in the NRC staff Safety Evaluation Report for TMI-1 restart.
See pages CS-60 to C8-63.
10
Q20) Do you believe that natural circulation will be adequate following a small break LOCA?
A)
Yes.
The TMI-1 reactor core can be adequately cooled by natural circula-tion following a small break LOCA or system transient.
As discussed above, the plant conditions required for natural circulation to be effective have been determined and improvements in the plant operating procedures and in the emergency feedwater system have been accomplished which will ensure that the required conditions for natural circulation will be present whenever natural circulation is required.
Q21) Is it therefore your expert opinion that forced circulation is nct necessary to adequately cool the core following a small break LOCA?
A)
Yes.
Q22) What are the requirements of General Design Criterion 34 of Aopendix A to 10 CFR 50?
A)
The requirements of GDC 34 are as follows:
"A system to remove residual heat shall be provided.
The system safety functions shall be to transfer fission product decay heat and other residual heat from the reactor core at a rate such that specified accep-table fuel design limits and the design conditions of the reactor coolant pressure boundary are not exceeded.
Suitable redundancy in components and features, and suitable intercon-nections, leak detection, and isolation capabilities shall be provided to 11
assure that for onsite electric power system operation (assuming offsite power is not available) and for offsite electric power system operation (assuming onsite power is not available) the system safety function can be accomplished, assuming a single failure."
Q23) Does the IMI-1 design satisfy GDC 34?
A)
Criterien 34 deals with the capability to remove decay heat during a normal shutdown and, as such, is beyond the secpe of this hearing.
Q24) What are the requirements of General Design Criterion 35 of Appendix A to 10 CFR 60?
A)
The requirements of GDC 35 are as follows:
"A system to provide abundant emergency core cooling shall be provided.
The system safety function shall be to transfer heat from the reactor core following any loss of reactor coolant at a rate such that (1) fuel and clad damage that could interfere with continued effective core cooling is prevented and (2) clat. metal-water reaction is limited to negligible amounts.
" Suitable redundancy in components and features, and suitable intercon-nections, leak detection, isolation, and containment capabilities shall be provided to assure that the ansite electric power system operation (assuming offsite power is not available) and for offsite electric power system operation (assuming onsite power is not available) the system safety function can be accomplished, assuming a single failure."
12
Q25) Does the TMI design satisfy GDC 35?
A)
Yes.
The systems which prc/ide emergency coolant to the reactor are the High Pressure Injection System (HPI), the Low Pressure Injection System (LPI) and the Core Flooding Tanks (CFT).
The HPI provides emergency coolant for reactor system pressures above the safety valve setpoint to atmospheric pressure and is designed primarily to deal with small break loss-of-coolant accidents.
The LPI provides emergency coolant for pressures below 200 psla, and the CFT's provide a pass./e source of flooding water for reactor system pressures below 615 psia. These systems are provided with redundant valves, pumps, power sources, and piping and the HPI and LPI may be operated from the emergency diesel generators. These systems were reviewed by the staff as reported in our Safety Evaluation of July 11, 1973, and found to be in conformance with the General Design Criteria.
The above described requirements in CDC 35 have not changed since the issuance of our safety evaluation of July 11, 1973.
Q26) How is abundant core cooling determined?
A)
Abundant core cooling is that which is sufficient, in the eve.1t of a LOCA, so that the limits stated in 10 CFR 50.46 for peak cladding temperature, maximum cladding oxidation, maximum hydrogen generation and coolable geometry are not exceeded. The analytical methods used for this deter-mination must be in conformance with Appendix K to 10 CFR 50.
Q27) Will abundant core cooling be provided at TMI-1 in the event of a LOCA?
A)
Analyses of small and large break loss-of-coolant accidents which are applicable to TMI-1 were performed by Babcock and Wilcox.
These analyses 13
were reviewed by the NRC and determined to be in conformance with the requirements of Appendix K and were within the limits of 10 CFR 50.46.
For large and small breaks, the initial approval was published January 8, 1976.
For small breaks, a revised model was approved September 5, 1978.
Since the accident at TMI Unit 2, further small break analyses were performed by Babcock and Wilcox for additional break sizes smaller than previously analyzed, including a stuck-open PORV.
The results of these analyses were also within the core temperature and core damage limits imposed by 10 CFR 50.46.
14
_