ML19257B593
| ML19257B593 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 08/22/1974 |
| From: | US ATOMIC ENERGY COMMISSION (AEC) |
| To: | |
| References | |
| TASK-TF, TASK-TMR NRC-593, NUDOCS 8001170690 | |
| Download: ML19257B593 (12) | |
Text
.
,7 NRC #593 REPORT ONfSPECIAL EVENT NO. 74-13 YEAR:
1974 DATE:
22 AUGUST TRIP TG-1/ REACTOR TRIP /SE [ SAFETY INJECTION] TRIP MAY29ISN Translated by:
Thomas IV. Appich Jr.
i916167 8001170[70 pl:
Distributor List Dir. Aemmer/Elmiger 1
Bossh/Lampr/ Imper.
1 Kueffer 1
Ba/It+Zirkul. FTT 1
Mh'/Nef 1
Samue1% feller 1
Tra/Sg 1
Schlosse rJB/Rb 1
II:/lig 1
Shift supervisor KKB I 1
Shift supervisor KKB II 1
Sedlacek 1
Ptm/Wn/Mi 1
W 1
BBC 1
Sec. Baden 1
Sec. Be:nau 1
Oper. office Original ASK 3
(Incl. BV-5403)
}h
}hh ii I
TABLE OF CONTENTS Page 1.
B r i e f S um a ry - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
1 2.
Chronological Description of the Event-----------------
2 3.
Analysis of the Cause of the Event--------
4 4.
Effects and Consequences-------------------------------
5 5.
Conclusions--------------------------------------------
7 6.
Requests for Possible Changes--------------------------
7 7
L i s t s o f App en d i c e s - - --- - --- - - -- - - -- -- - - - -- -- -- - -- --- -
8
[916 169 iii
i Brief Summary d August 20'at 11.20lthere occurred altrip.of TG-1 causes by high
~
bearing vibrations (bearing 6:60 p).
At the time of the trip, the blind load output of generator 2 was
,approximately 140 MVar.
Due to.a failure of the Atmospheric Exhaust!
1 Stationi (AES) which caused the! blow-off valves not to open, the mean reactor coolantatemperature_ rose very quickly; on the one hand this caused a rapid rise in the secondary steam pressure and on the other.it caused an equallyasteep rise in.the pressurizer le. vel anj pressure.
16_0_ _bar_ primary pressure, thelpressurizer_ relief valves PCV_-455 At z
C/456; opened properly, and this very quicklyireduced the pressure in _
the primary system. Approximately 10 seconds after the order to open, the pressure had dropped so_ far that the relief, valves were again ordered to close.
Due to a malfunction in valve PCV-456, it! remained open so that the pressure in the reactor coolant loop The: reactor dropped below 100 bar after approximately one minute.
wast tripped by the signal Ppressure low", 126.5 bar.
Due to the open pressurizer relief valve, the pressure in the reactor coolant group dropped to approximately 70 bar; this corresponds to
,a saturation pressure of approximately 284 C.
This caused the
' generation of-steam in the hot leg of the primary loop, and this
- caused the refilling of the pressurizer.
The observation that the flow rate in the' two loops was varying
( widely.and the elevated vibrators of the RHP. [ expansion uncertain, probably " reactor main pump"] confirm.d the assumption that the steam was being generated in the loop.
Approximately 2-3 minutes after the reactor trip, the shift personnel became taware of the
- malfunction 3in the relief valve and isolated 'it with motor-actuated valve 531. 'Then the water level in the pressurizer began to drop, and 11 minutes after.the reactor trip SI was automatically initiated by " pressure and 1cvel low in the pressurizer". The start-up of the I
SI systems proceeded properly, and initially 40 1 s of water was _.
inj ected by each of the four SI spray lines.
Due to the inj ection, the pressure in the primary system rose to 110 bar and the level rose to 70%.
Then the SI. pumps _ were shut dwn and the motor-actuated valves in the spray lines were closed.
From then on it was possibic to r 4 ~ att the level in the pressurizer with the charge pumps and the S "
then the cool-down of the primary system was started.
Approximately;3 minutes afte2 the reaetor trip, the alarm " containment:
} pressure high" occurred, and approximately 1 minute later there.
( followed.the alarm " containment activity high".
The-pressure peak i di6 170
in the containment was approximately 100 mbar above the normal value. To counter this, the shift personnelistarted all four
! circulation fans in the containment. Since various clarms of the
.pressuriter relief tank had also been actuated, it was very soon
, possible to assume, wi,th a high degree of confidence, that the irupture diska had burst,or that theLrelief line must have been -
. defective.
Af ter the trip of TG-1, the secondary-side steam pressure rose to 66 bar due to the failure of the AES.
Due to a load load transient, the turbotrol [ sic, " turbine control"] of TG-2 shifted to emergency operation immediately after the trip of TG-1.
This rendered TG-2 uncontrolled, and the position of the inlet valves remained constant during the subsequent pressure transients of the live steam. The output of TG-2 thus rose to approximately 214 MWe because of the higher steam pressure (a rise from approxi-mately 52 bar to 66 bar),
After the trip of TG-2, the live steam pressure rose to more than 70 bar due to the reactor trip; then the safety valves opened and reduced the pressure to approximately 65 bar.
2.
Chronological Description of the Event
! 20 August 1974 2.1.
Time Sequence of the Reactor Trip by " Sequence of Events" First event at 1120' 12" Generator rain switch of TG-1 off Pressurizer-pressure low, reactor trip 39.7" later Trip switch open 39.8" Generator main switch TG-2 off 40.3" SI trip (pressuri:cr pressure and Icvel low) 11'55.9" 2.3.
Summary of Alarm Printer After Trip of TG-1 Time 11.15 Blind output of generator 1 high 13.5 MVar 11.20 Safety oil pressure TG-1 too low 11.20 Pressurizer pressure high 158.2 bar 11.20 Pressuri:cr pressure high 159.9 bar (Reactor trip 1I916.17i 2
Time 11.21 T coolant loop-A high 302.2 *C g
11.21 Steam pressure in front of HA [ expansion unknown], TG-1 high 66.3 bar 11.21 T coolant loop-A high 305.2 *C M
11.21 SG-A steam pressure high 67.3 bar 11.21 SG-B steam pressure high 67.2 bar 11.21 Steam pressure in front of HA [ expansion unknown], TG-1 77.6 bar 11.21 SG-A steam pressure high 73.3 bar 11.21 SG-A steam pressure high 65.4 bar 11.22 Safety oil pressure TG-2 too low 11.22 T coolant loop-A 285.2 *C g
11.23 Steam pressure of HA, TG-2 68.1 bar
~'
11.23 'lTemperatsre of pressurizer relief tank!
- thigh, 62.8 *C 11.24 GRelief tank level 79%
11.24.. Relief tank level 88%
11.24 ) Pressure of containment high 1.1 bar abs.
11.24 ' Pressurizer relief' tank level low 20.2%
11.24 Pressurizer relief tankipressure high.
0.59 bar 11.25 Pressuri:er relief tank pressure 0.15 bar 11.25 SI-A+B steam pressure normal 63.7 bar 11.25 R Activity of containment high 17.3 mr/hr 11.26 Coolant loop B low 88%
11.27 E Containment, air temperature high; 53.4*C 11.32 ; Pressurizer level low..
6.8%
11.32 Pressurizer level normal 18%
11.33
. Temperature.in surge line too low 271.2*C 11.34 Pressurizer level high 58%
2.3.
Excerpt of ZFM [ expansion unknown] -Concerning Pressurizer and Pressurizer Relief Tank 1120 11.1 Pressuri:er pressure above control range 111.9 Pressurizer relief valves unblocked-22.8 Pr :ssurizer relief tank pressure high Y23.0 Pressuriv.er relief valves blocked 23.0 Pressuri:cr pressure in control range 23.1 Pressurizer relief tank level high 24.2 Pressuri:er level high 33.0 Pressuri:er relief tank pressure too high 35.0 Pressurizer pressure under control range 21 00.4 Presstirizer pressure low, reactor trip 01.2 Pressuri:er pressure low, SI systems unblocked 05.1 Pressuri:er relief tank 1cvel high 13.5 Pressuri:er pressure low, SI systems unblocked 3
.l9.1.6.172
23 27.6 Pressurizer level high, I channel trip 43.3 Pressurizer relief tank pressure too high 43.5 Containment pressure too high 47.1 Pressurizer relief tank level low 24 29.4 Pressurizer relief tank pressure normal 51.2 Containment temperature high
-25 17.8 Containment activity high For a complete ZFM printout of the primary, see Appendix 1.
For a complete ZFM printout of the secondary, see Appendix 2.
For the most important printer rolls, see Appendix 3.
2.4.
Start-Up of the Plant This occurred on 8/25/74 without particular problem.
3.
Analysis of the Cause of the Event The cause for the trip of turbine 1 was the high bearing vibrations, especially in bearing 6.
The vibrations were excited by a blind load surge from the network. Previously it had been noted that unit I reacted particularly sensitively to blind load surges.
At the time of the accident, the generator was excited for the maximum possible blind load output, so that the additional blind load surge was abic to cause the vibrations.
The subsequent malfunctions in the primary section of the plant were definitely caused by failure of the AES.
The cause of the failure is discussed separately in Appendix 5.
A single-machine trip is by no means a novelty and has no significant effect on the primary system when the AES is operating normally (see for instance Special Event No. 73-03).
A tour of the containment after_ the primary _ system was cooled off showed that the' support. structure between the. valve body of PCV-456
- and the air motor [ sic] had failed, very. probably due to being
' struck by_ the line when the valve opened. i This. prevented the valve from closing, and this failure caused the rapid drop in pressure
,in the primary system.
In addition, it was determined that the Lanchoring of the relief valve was torn loose.
The' rupture disk of the pressurizer relief tank burst due to the
- prolonged blowdown.of primary water into this tank.
The excerpt of the ZFM in section 2.3 shows that the' rupture disk burst only after mthe relief _ valves had already. received the order to close.
4 1916 17'3 1
Further details on the cause of the defect both in the valve and in the rupture disk are given in Appendix 7.
, Below is a balance sheet of the water which accumulated in the
~' containment. sump:
Reg. water tank A 38'6 + 100%
= 9.8 m Reg. water tank B 16%
- 36%
= 3.2 m Total quantity of water accumulated =13.0 m Relief tank 80%
- 19%
=11.2 m Water from the system 1.8 m Since no other damages or leaks were found in the containment, it can be assumed that these 1.8 m3 of water were blown off through the defective relief valve.
The behavior of the operating personnel during the event is described iin Appendix 4.
4 Effects and Consequences 4.1.
Pouer Failure Power failure relative to 100*6 rated output (364 Mie)
Failure time:
8/20/74, 1120 to 8/25/74, 1055 Duration of failure:
119 hours0.00138 days <br />0.0331 hours <br />1.967593e-4 weeks <br />4.52795e-5 months <br />, 35 minutes Nominal production of TG 3/4 from 8/20 to 8/25 52'416 MWh Effective production 12 ' 266 Mih Production failure relative to 100% (364 MWe) 40'150 bnih Failure time relative to 100% (364 bnie) 110.3 hr 110 hr 18 min 4.2.
Thermal Stressing of the Primary Loop In addition to the rapid rise in water temperature by approximately 6*C after the trip of TG-1, followed by an equally rapid rise in the primary pressure from 154 to 160 bar, a major temperature transient occurred when the SI system was tripped, especially in the area of the inj ection no::les.
Since the main reactor pumps were in operation the whole time and thus ensured rapid mixing of the cold injection water with the hot primary water, it can be assumed that the other.
}916174 5
components were not subjected to a large temperature jump.
The temperature and pressure transients are within the design values.
The stress to the inj ection no::les is discussed in Appendix 6.
4.3.
Damage to the Relief Syste i On the tour of the containment after the cool-down of the primary system, the following damages were noted in the pressurizer relief
, system:
-- Relief valve PCV-456:
- 7. drive _ yoke broken on both sides and spindle bent slightly.
--. One anchor point of the relief valve was torn loose.behind the valve.
-- The rupture disk of the relief tank burst.
No other mechanical damage was initially noted in the containment.
In this connection the only other item to note is the fact that the
. Erelief tank is not designed to absorb steam from the pressurizer over a longer period of time. The bursting of the rupture disk is thus directly attributable to the failure of the relief valve.
4.4.
Turbines TG-1 The cause of the high bearing vibrations was very likely a blind load surge from the network. The actuation of the AP alarm of the seal oil system was due to the bearing vibrations and had already been observable beforehand.
Damage to the packing rings or bearings is therefore extremely unlikely so that inspection was not necessary.
TG-2 The load transient from initially 172 bMe to approximately 110 MWe and then to 215 MWe initially gave rise to fears that the expansion bolts of the high pressure cylinder had been overstressed (high wheel case pressure) and lost a portion of their pre-stressing. However, it was decided not to remeasure these bolts (remove their insulation) since inadequate prestressing would reveal itself immediately upon start-up due to a Icak in the high pressure cylinder flange.
Because of the high turning moment during the load transient to 215 MWe, the couplings of the turbines and generator were checked for distortion. No damage was noted.
f916 i75 6
It was decided not to inspect the multi-collar thurst bearing since the shaft monitor had not indicated any displacement of the shaft.
5.
Conclusions If the entire event is viewed in retrospect, we come to the conclusion that despite the failure of two systems, namely initially the AE system and then the pressurizer relief system,Lthe plant never entered
- an uncontrolled. state and no after-damage occurred.
Even during the entire event, there was never any uncontrolled release of activity either in gaseous or liquid form from the containment. All of the.
? safety systems performed their functions completely.
The safety valves on the live steam side, which were intended as safety mechanisms in case the blow-off valves should fail, maintained the steam pressure within the permissible limits, and the. SI system brought the primacy system back to a safe pressure so that subsequently it was possible to cool the system down normally.
6.
Requests for Possible Changes 6.1.
Voltage Regulation of Generator 1 As had already been noted previously, generator 1 reacts very sensitively to blind load surges with bearing vibrations.
A study should be made of whether the voltage regulator can be modified in such a way that rapid blind load changes can be suppressed.
6.2.
Turbine A check must be made as to whether when the turbines are running without the first row of vanes the pressure limiter must be reset in order to limit the maximum output of the turbine to approximately 190 FMe.
A study should also be made of whether it is possible to set the night run mechanism [ sic, possible misprint, possibly " follower mechanism"] of the turbotrol in such a way that after the turbotrol is shifted to emergency operation the turbine output is always smaller than before the transfer.
6.3.
AES a)
Before the start-up of the plant after a prolonged shutdown or after revision and calibration work is done on the blow-off control, a serviceability check of the AES must be carried out in a last test, from the actuating transducer to the opening of the blow-off valves (function check with the plant shutdown).
7 1916 176 7
b) A study should be elaborated in order to investigate wheth'er it is possible to carry out a periodic check of the blow-off function during power operation.
In this process the safety of the plant must not be endangered (e.g., accidental opening of a blow-off valve).
c)
For the blow-off station a printer should be studied (such as the turbine printer) which would begin operation when the oil pumps start up at high speed and would record the valve openings of the blow-off valves and the bypass valves.
6.4.
Pressurizer Relief System As an initial measure, a defective valve must be repaired, the defective pipeline anchor point must be remounted and the rupture disk of the pressurizer relief tank must be replaced.
With these initial measures it will be possible to start the plant up again.
Subsequently a study must be made of how the relief line can be -
- better anchored in order to prevent the line from being struck when the relief valves open.
As a third measure a study should be made of whether and:how the existing cast structure of the support between the valve and the drive can be replaced by a better steel welded structure.
6.S.
Emergency Feedwater System During this event it was noted that the' emergency feedwater pumps
_ failed due to overflow because of= the low steam pressure in the
- steam generators (rapid cool-down of the_ primary systems). This caused one operator to be occupied for a long time in closing 'the manual slide valves in the emergency feedwater lines.
Therefore a check should be made of whether it would be possible to:
install a regulating valve, with the control in the control room, in the main lee of the emergency feedwater lines behind the pumps.
Upon the. -rival of the signal " reactor trip, SG 1evel low" the valve would i. ave to be automatically opened completely until the valve can be deblocked by the operator and then controlled with the regulator.
7.
List of Appendices Appendix 1:
ZFM Strip Primary Appendix 2:
ZFM Strip Secondary Appendix 3:
Printer Rolls Appendix 4:
Behavior of the Operating Personnel Appendix 5:
Study on the Cause of the Failure of the AES 8
1916 177
~
4 Appendix 6:
Effects of the Safety Injection in KKB I on the Reactor Pressure Vessel Appendix 7: Damage to the Relief System
)h ]f b 9 I i i}}