ML19309D567
| ML19309D567 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 11/05/1979 |
| From: | Mattson R Office of Nuclear Reactor Regulation |
| To: | Fraley R Advisory Committee on Reactor Safeguards |
| Shared Package | |
| ML19309D561 | List: |
| References | |
| ACRS-1684, NUDOCS 8004100530 | |
| Download: ML19309D567 (7) | |
Text
-- y UNITED STATES l'
.8 NUCLEAR REGULATORY COMMISSION B
p$
wasumaron.o. c.2assa RECEWED k*****,o yng tot 5 m 4 @
gggyggsc TO ACRS MEMBERS 9
MEMORANDUM FOR:
R. F. Fra'4ey, Executive Director u.s.HutLE
- R R{,I;Cp[pg Advisory Committee on Reactor SafeguagSh0R 5 f; GUARD 5 FROM:
R. J. Mattson, Director Division of Systems Safety, NRR
SUBJECT:
ACRS PWR QUESTION REGARDING EFFECT OF PRESSURIZER HEATER UNC0VERY ON PRESSURIZER PRESSURE BOUNDARY INTEGRITY During the October 3,1979 meeting of the ACRS subcommittee on TMI-2 Implications a question was raised regarding the potential for pressurizer pressure boundary failure if the pressurizer heaters should become uncovered and overheat. At the meeting I had comitted to providing a response within one month.
Please find attached our response to the question.
)
1
\\TA la R. J. Mattso Director 7i Division of Systems Safety Office of Nuclear Reactor Regulation cc w/att:
F. Cherny, DSS D. Eisenhut, 00R E. G. Case, NRR D. B. Vassallo, DPM L. Porse, SD S. Hanauer DSS R. Bosnak, DSS J. P. Knight, DSS R. Satterfield, DSS T. Novak, DSS D. Ross, DPM A. Thadani, DSS F. Schroeder, DSS T. Speis ACRS (21) H-1016
.zs
~w
t 0,
RESPONSE TO
~
AGRS QUESTION RE:
RCPB INTEGRITY IF PRESSURIZER HEATERS FAIL i
Pressurizer heaters are of the immersion type i.e. the heater wire is inserted in a tubular sheathing that is generally made of stainless steel or Inconel. The heater wire is surrounded by a powdered and compacted insulating material such as magnesium oxide through which the heat from the element is conducted to the sheath and hence to the reactor coolant.
One end of the tube sheath is capped and the other end has a cap with small openings that permit the heater wire extensions to pass to the outside connections.
Generally the terminal connectors are designed to withstand the same internal pressure as for the reactor coolant system to p:: vent leakage to the outside in case of sheath tube failure.
The thickness of the tube sheath generally is within the range of 0.108 -
0.125 in, with one design as heavy as 0.180 in. The heater's are inserted through nozzles or penetrations located either in the spherical bottom head of the pressurizer (Westinghouse and Combustion Engineering) or through the vessel wall (Babcock and Wilcox). All of the designs feature a seal weld of the heater tubing to its pressurizer penetration nozzle. Additionally one design provides for threading of the heaters into the vessel penetration, after which a seal weld is applied. Generally speaking, the seal weld is located about 6-12 inches away from the location of the active heating element.
- 79 N
em
. _ _ gw w am de N
1 Pressurizer heaters can fail in one or both of the following modes:
1.
Melting of the heater wire and thus creating an open circuit.
2.
Shorteircaiting to the tubing sheath, creating an electric arc and burning a hole in the sheath.
Coolant could then enter the interior of the heater.
Heater failure in the first mode normally would result only in loss of heater function, with consequent economic penalty for the utility.
However, if a heater were to fail in the second mode resulting in a hole in the heater sheath, a concern can be raised as to,whether reactor coolant could escape through the hole, out the heater wire penetrations, and into the containment thus resulting in a small loss of coolant accident. This does not appear to be a likely scenario.
For some heaters there would be no leakage since the wire penetrations are designed and tested to prevent J
1eakage. For other heaters there would be a seal area about 6 inches long.
A leak through this area would result in a very small amount of steam leakage to the containment with too small a leakage rate to be classified as a small LOCA.
One other possible leakage path would be if the heater to vessel penetration seal weld were to fail from exposure to a temperature much higher than its design temperature with the uncovered overheating heater as the heat source.
l
_ __. v
3-The heaters are not designed to operate uncovered in a steam environment.
They are protected from operating uncovered by either a single or double non-safety grade low level sensing device such that they are de-energized when the liquid level is low. Non-safety grade equipment is used because historically the NSSS vendors have viewed heater loss as strictly an The vendors have estimated that the sheath temperature ecemomic problem.
could rise to as high as 1500-1600 F, if uncovered, before a heater would fail.
i There seems to be a lack of any precise information available as to exactly Those vendors how long a heater could operate uncovered before failing.
that the staff has talked with recently that were willing to estimate heater failure time under these conditions felt that failure would occur within a They further judged that this time interval would be too short few minutes.
to transmit sufficient energy to the penetration to heater seal weld to change the weld materials sufficiently such that weld integrity would be of concern.
On the other hand, one vendor noted they had not performed any tests to determine heater failure time, if uncovered, and did not offer a judgement as to what that time might be.
Instead this vendor referred to some problems that had been experienced with uncovered pressurizer heaters during preoperational startup testing of the D. C. Cook Unit 2 plant.
Information obtained by the DSS Staff from the Office of Inspection and Enforcement regarding the D. C. Cook 2 preoperational problem seems to confirm that heater to penetration seal weld integrity would not be affected if the
---=-
t-
. heaters were to function uncovered for a relatively long period of time.
On the other hand, the events at D. C. Cook seem to throw into question the judgement of some vendors that heaters could only function for a few minutes, uncovered.
In late May,1977, D. C. Cook Unit 2, a Westinghouse Pressurized Water Reactor, was undergoing filling of the primary system for the " cold" hydrostatic pressure integrity test. Before the pressurizer was filled, 54 of a total of 78 heaters were inadvertently energized for a period of about one hour.
Apparently the non-safety grade low level sensor had been bypassed thus permitting the heaters to be energized.
For an hour the energized heaters were exposed to an air environment.
After the heaters were de-energized an inspection was performed.
It was found that eight heaters were open or shorted out i.e. failure mode 2 discussed above. Three heaters were bowed as much as 6" between support plates, but were still operational. Nineteen had experienced some lesser distortion.
The remaining 24 experienced no visible distortion but were covered with a grey oxide coating.
j Prior to the performance of the cold hydrostatic test the 8 open or shorted out heaters and the three severelydeformed ones were replaced. The balance of the 43 heaters that had been inadvertently energized were si.111 operational and were replaced after the primary system hot functional test. This means that they were exposed to test pressures of just over 3,000 psi at a system temperature slightly above containment ambient and a pressure of around 2,250 psi at normal plant operating temperature.
=
... -..- - - - -- -~
5-Based on the appearance of the heaters after they were energized in the air environment at D. C. Cook, it seems highly likely that many of them could have remained functioned for considerably._ longer than one hour before failing..
Since there is convincing evidence that heaters can remain uncovered and functional for a long period of time there is one other scenario that we thought should at least be mentioned, which hypothetically, could result in degradation to the pressurizer vessel itself. Our general under-standing of PWR plant behavior is that if during plant operation an initiating event which causes both pressure and level to decrease, the. pressurizer heaters could become uncovered and remain energized if the low level control channel fails.
If the pressure decreases to below the scram pressure setpoint, the reactor will trip.
If the pressure continues to decrease, Termination the High Pressure Safety Injection System (HPSI)could be actuated.
of the HPSI would depend on the amount of subcooling and the pressurizer water level measured from redundant level channels.
If on the other hand the pressure remains low but above the HPSI setpoint, the pressurizer heater indications on the control board could be used by the operator to recognize that the heaters were energized. Also, it is our understanding that most if not.all, pressurizers have thermocouples in the steam space which could be used by the operator to determine that the heaters are energized and possibly uncovered.
However, if the heaters were to operate uncovered for some length of time, the steam above the heaters could become superheated.
If the superheating process centinued long enough, vessel integrity could be compromised by high temperature tensile or creep failure, f
-se e e q e
-m
- N eN.
- ee me w e es aw N * **** weems nasump
.m.
,.e e
'6-In order to assess the potential for such vessel degradation the Bulletin's and Orders Task Force plans to require the Utilities Owners Groups to address the potential for a small LOCA from this type of scenari.o in.the small break LOCA evaluations now being performed by the groups.
As of this writing, the DSS staff has not completed a detailed review of operating plant events, other than that D. C. Cook, which might provide addittor..J sight into the behavior of pressurizer heaters that have operated uncovered and the ultimate effect, if any, on reactor coolant pressure boundary integrity.
The Office of Inspection and Enforcement has informed us of at least one other case where heaters operated uncovered for a period of time. We will investigate this incident and also try to determine if there have been any others that might provide further insight into the effect on primary system integrity. We believe we will be in a position to report the results of these findings in about one month,.
Based on the design information available to the staff and the information gained from the D. C. Cook event, our preliminary conclusion is that the most probable scenarios for operating pressurizer heaters uncovered for some length of time do not appear to have an adverse effect on reactor coolant pressure boundary integrity.
i y
.