ML20010F255
| ML20010F255 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 05/31/1981 |
| From: | BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20010F250 | List: |
| References | |
| TRG-81-3, TRG-81-3-V01, TRG-81-3-V1, NUDOCS 8109090535 | |
| Download: ML20010F255 (95) | |
Text
{{#Wiki_filter:- - - - - TRG.81-3 May 1981 lI ti
'l -Study Guide-il
!I lI
- I i OPERATOR TRAINING-DEGRADED 15 ll CORE RECOGNITION AND
!I MITIGATION l lI 4 l 'I Phase 1 l ll l 1 l ll i lI !! Volume 1 Babcock &Wilcox
~
l 810 9 0 95L5 ,;
i TRG-81-3 D May 1981
- STUDY CUIDE -
OPERATOR TRAINING-DEGRADED CORE RECOGNITION AND MITIGATION Phase 1 Volume 1 4 BABCOCK & WILC0K Nuclear Power Group NuclearP.Power Generation Division O. Box 1260 Lynchburg, Virginia 24505 Babcock & Wilcox
lI il CONTENTS il 4W Page Lesson 1 - CORE COOLING MECHANICS . . .. . . . . .. .. . . . ... 1-1 lE u Lesson a 1AS/ STEAM BINDING . . . .. ... . .. . . . . . .. . 2-1 il i Lesson 3 - BORON PRECI"ITATION CLNCERNS FOLLOWING A LOCA Lesson 4 - EQUIPMENT FAILURE SEQUENCES THAT COULD LEAD TO A
.. . .. . 3-1 DEGRADED CORE . . .. . . . . .. . . .. . . .. .. . . 4-1 -
l5 Lesson 5 - AVOIDING DEGRADED CORE CONDITIONS .. . . . . . . . . . . 5-1 ,!I 'l !I
!I lI I
I I I
- 111 -
~.
1 I I i I Lesson 1 - CORE COOLING MECHANICS I Introduction
- 1. 1ecturer -
- 2. Purpose - To describe the bas *c principles of heat transfer from the core, the effects of loss of this heat transfer, and the reasons L e disruption of normal c'rculation patterns.
Objectives The following subjects will be covered during this lesson:
- 1. Basic heat transfer from the core to the ultimate heat sink.
- 2. The factors affecting this heat transfer and the operator's con-trol over them.
- 3. Heat transfer characteristics of the various states of water.
- 4. N'atural circulation theory.
- 5. High-pressure injection (once-through) method of core cooling.
Key points to be retained are as follows:
- 1. The various methods available for cooling the core.
- 2. How to interpret primary system status from saturation curves.
- 3. Factors that affect natural circulation.
- 4. How the operator can effect heat transfer from the core to the ultimete heat sink.
- I ,
i I 1-1 Babcock & Wilcox
I I I Lesson 1 - CORE COOLING MECilANICS I 1. Heat Transfer 1.1. Fundamentals of Heat Transfer Heat generated in the core (and by the reactor coolant pumps) is transferred to the reactor coolant, which in turn transfers the. heat to the steam genera- .g 'W tors. The steam generators then reject heat to either the atmosphere or the condenser. The exchange of heat takes place between a source and a sink. For example, the reactor core is the heat source for the coolant, which is the heat sink for the core. If the plant is tripped and the reactor coolant (RC) pumps are running, they become a significant heat smtree. The coolant is the heat source for the steam generator fluid; however, if the reactor coolant is I colder than the sterun generator fluid for any reason, then the steam generator becomes the heat source and the coolant the heat sink. 1.2. Overcooling and Undercooling Events if steam generator heat removal matches the heat production in the primary system, the reactor coolant temperature will remain constant. If the steam generators remove more heat than the primary system is producing, the coolant I temperature will drop. Normal cooldown is an example of a controlled cool-down; however, if the condition is abnormal or not controlled, it is classi-fled as an overcooling event and corrective actions are necessary to bring it under control. On the other hand, if the steam generators remove less heat than the primary system is generating, then reactor coolar". temperature will increase. Normal heatup (from 0 to 15%) is an example of a controlled heat-up. If the condition is abnormal or not controlled, it is classified as an I overheating event and corrective actions are necessary to bring it under con-trol. Equations can be used to describe the heat transfer path from the core to the .I steam generntors. When the heat transfer is balanced, 1-1 Babcock & Wilcox
I 4 core =6 rc (1-1) 3 for the heat transfer path from the core to the reactor coolant, and Q core =Q sg (1-2) for the heat transfer path from the reactor coolant to the steam and water in the secondary side of the steam generators. 6istheheatrateinBtu/ hour. When heat transfer is balanced all the way from the core to the steam genera-tor, equation 1-1 equals equation 1-2. But when heat transfer becomes unbal- g anced, they will not necessarily be equal. The heat transfer path can be B interrupted when the reactor coolant is not a good heat sink for the core (k core tc) r when the steam generator fluid is not a good heat sink for the reactor coolant (k ore f g)
- The unbalanced condition of concern for core heat transfer to the reactor coolant is when there is not enough heat transfer from the core to the reac-tor coolant. This can happen when the core is covered partly by water and partlybysteamorcoveredcompletelybysteam;then6 core + m . When this happens, not enough nuclear heat can be passed from the core to the reactor coolant, and the core will heat up. The stored heat of the fuel cladding will increase.
When the steam generator heat flow path becomes unbalanced, the steam genera-I tor fluid will remove too much or too little heat from the reactor coolant, and it will be an overcooling or overheating condition. When this happens duringatransient,6 will increase or decrease depending on the heat re-moval by the secondary side. The reactor coolant temperatures will change in order that temperature (thermal) equilibrium can be re-established between tae primary and secondary side fluids. To show these effects, equetions 1-1 and 1-2 can be written to add temperatura terms: Equation 1-1 can be written as Q core =$ rc Cprc(Th -T)c (1-la) where
$ = RC system mass flow rate, lbm/h, I'
Cp = specific heat capacity of reactor coolant, Etu/lbm *F, T =cr inlet temperature, F, h T = core inlet temperature, F. Babcock & Wilcox Il l-2 E,
I 'I Equation 1-2 can be expanded as follows: 6, = UAAT (1-2a) I I U = overall heat transfer coefficient, A = total area of heat transfer surface, AT = temperature differential across the heat transfer boundary. The overall heat transfer coefficient is dependent on many factors, including the fluid conditions (primarily dens.ity and flow rate) on both sides of the boundary and the properties of the boundary (primarily the thickness and ther-cal conductivity of the barrier and oxide layers). For this discussion we can assume that the properties of the boundary (steam generator tube walls) re-main constant and thus can be ignored. The secondary side of the steam generator has three different regions alor.g the tube bundle during power operation: nucleate boiling, film boiling, and superheat. Each region has a different coefficient (U), surface area (A), and ::emperature differential across the tube wall (AT) . The nucleate boiling region has the highest U of the three and accounts for approximately 70 to , I 85% of the total heat transfer into the steam generator over the power range. The heat transfer coefficient decreases by a factor of 3 to 10 in the film boiling region and again by another factor of 3 to 10 in the superheat region. The heat transfer surface areas and AT values involved for each of the three regions vary over the power range with the two boiling regions accounting for an increasingly higher percentage of the total heat transfer with increasing power levels. Thus, to determine the effects of transients on secondary heat removal during power operation, the effects in each of the three regions along the tube bundle must be studied. However, we are concerned mainly with control of heat removal by the steam generators after a reactor trip. After trip the steam generators are at sat-uratica conditions with two basic regions - water and saturated steam. Almost all of the heat transfer occurs in the water region, and most of this occurs ,I in the nucleate boiling portion below the steam / water interface. Saturated water is absorbing the latent heat of vaporization, and the nucleate boiling region provides a much higher heat transfer coefficient (U). Below this level the water is subcooled, with a considerably lower U, although this U value is still much higher than that in the steam space. 1-3 Babcock & Wilcox
I, 1 I Very little heat transfer occurs in the steam space (primary side temperature I' can be considered equal to T throughout the steam space). Even though the g area is large, the heat transfer coefficient is small due to low steam flow 5 rates and low density with respect to the water region. During forced circu-lation the AT across the tube walls in the steam space is also very small since T g is close to T of the steam. The AT is larger between T and T,,g during natural circulation, but the U value is even smaller due to the lower primary flow rates. The major factors affecting heat transfer in the water region are surface area and the AT between the primary and secondary sides. Surface area is increased by increasing feedwater flow to raise the water level. The primary increase in area takes place in the subcooled water region. Even though most of the heat transfer occurs in the nucleate boiling region, overall heat transfer is increased becat. e the area of the steam space (with a very small U) is de- g creased and replaced by area in the subcooled water region (with a relatively much larger U). The major method of effecting primary-to-secondary AT is on the secondary side by varying steam pressure. When steam pressure is decreased (e.g. , by opening the turbine bypass valves), saturation temperature also decreased, which in-creases the AT across the tube wall. The higher AT causes heat transfer (Qsg) to increase, thus cooling the primary side. Heat transfer can be increased significantly by injecting feedwater (main or emergency) through the upper nozzles. The increase in heat transfer is due to two factors. First and most significant, the spray of feedwater into the steam space reduces steam pressure in much the same way as the action of pres-surizer spray. This reduces the saturation temperature, w'ich increases heat transfer as described previously. Second, where water contacts the tube sur-faces in the steam space, the heat transfer coefficient is increased, essen-tially replacing steam area with water area as in the case of raising the steam generator level. Emergency feedwater will have a greater cooling ef-feet than main feedwater through the upper nozzles (for the same flow rate) because of its colder temperature. Assuming that a minimum adequate level is maintained in the steam generators, variations in steam pressure will have a greater ef fect on heat transfer than I 1-4 Babcock & Wilcox E
g/:
??
~/
variations in level. The best method of decreasing heat transfer is to close the turbine bypass valves and allow the steam generator pressure to increase.
/ Allowing the steam generator level to decrease will have no appreciable ef fect on heat transfer until the level becomes inadequate (too low for maintaining natural circulation or virtually dry with forced circulation).
In summary, the operator can control primary-to-secondary heat transfer (and therefore, RCS overcooling and undercooling) af ter reactor trip by controlling two major parameters on the secondary .:ide (assuming that the capabila / of the reactor coolant to' transport core heat to the steam generators remains in-tact). The operator can increase heat transfer by reducing steam pressure or by raising steam generator level. He can decrease heat transfer by allowing the steam generator pressure to increase. Note: Equations 1-la and 1-2a have been simplified to show the I general heat transfer process. To be complete, addition-al heat transfer terms would have to be included. All of the water that flows through the RCS loops does not I flow through the core and get all the way to the steam generators. Some flow is let down to the makeup system, some goes to the pressurizer sprays, and there is some
" leakage" through spaces in the internals. This small
}I amount of flow 1.as been ignored for these equations. Also, all the heat of the core does not go to the steam {g generators; some is lost through the " skin" of the pip-g ing to the reactor building or through the letdown water. But this amount of heat is small compared to the total amount and has been neglected. Heat is also added by }l 'W the RC pumps (as in plant heatup to power operation), but this is small compared to core heat when the reactor is at power (but the RC pumps are a large heat source after trip or at low power). j Control of heat transfer requires control of all the parameters in the two equations. Some are fixed by design or properties of fluids; the remainder can be influenced by the operator. The general methods of heat transfer con-trol are discussed next. I I I i B 1-5 Babcock & Wilcox
I
- 2. Control of llent Transfer 2.1. Overview The preferred way to remove heat from the core is to transfer the heat to the reactor coolant and then transfer the coolr.nt heat to the secondary fluid in the steam generators. Steam generator heat removal is controlled by adjust-int steam pressure and feedwater. To keep the core-to-steam generator heat transfer in balance, the heat removal rate from the steam generators must be W l equal to the heat generation rate of the core. In order to balance the heat removal, two very basic conditions must be satisfied (1) there must be enough l liquid reactor coolant in the vessel and piping to transfer the heat to the steam generators, and (2) the steam generator pressure and level (feedwater flow rate) must be balanced at the correct heat removal rate. Figure 1-1 11-lustrates five fundamental methods of heat transfer control that are used to g support these two basic conditions.
When an abnormal transient occurs, one or more of five fundamental methods will he out of control. It is the operator's job to determine which methods these are and to make corrections to restore the right heat transfer balance so the core heat can be removed by the steam generators.
- 1. Reactivity Control - Reactivity control is usually taken care of g automatically by ICS rod control or reactor trip. Reactor trip 5
lowers the core heat output to the decay beat level.
- 2. Reactor Inventory Control - The link between the core and the steam E generator is the reactor coolant; it is the fluid which transports W the heat. To do its job best the coolant should be in a liquid state, that is, subcooled.
- 3. Reactor Pressure Control - The reactor coolant system is pressurized to keep the reactor coolant in a liquid state.
- 4. Steam Generator Inventory Cont rol - The reactor coolant transfers its heat t t, the water and the steam in the secondary side of the steam generator. The water-steam inventory is the heat transfer g fluid that removes the heat from the reactor coolant. In order for 5 it to remove heat at t he correct rate, the amount of fluid and its rate of change must he controlled.
- 5. Steam Generator Pressure Control - The reactor coolant temperature is controlled best by controlling the pressure of the steam genera-tor. In combination with reactor presaure control, st eam generator g pressure control will maintain the reactor coolant in a subcooled g liauid state.
I 1-6 Babcock s. Wilcox E
I ' I Each of these control methods is discussed individually as it relates to heat transfer. lI 2.2. Reactivity Control Reactivity control is usually taken care of automatically by ICS rod control or by reactor trip. Reactor trip lowers the core heat output to the decay heat level. The operator usually has nothing to do except to verify rod in-sertien. After the trip, no more heat transfer control can be achieved by use If one or more rods are stuck I of the rods unless they did not fully insert. out after trip, the operator should try to manually trip them in so that the heat source is as low as possible. An attempt may also be made to drive the stuck rod in; however, care must be taken to prevent rapid depressurization of the steam generator when the reactor trip is reset. Since the rods also help to control reactivity, reactivity imbalances because of stuck rods must be adjusted with boron addition. 2.3. Reactor Coolant Inventory Control Reactor coolant heat transfer can be affecced by changes in the amount of the fluid mass in the RCS or by changes in the density of the reactor coolant. Several ways exist to vary the mass of reactor coolant: LOCA or small break l and changes in HPI or makeup, RC pump seal injection, seal return, and let-down. There are also several ways to vary the density of the coolant. Changes of the rate of heat transfer from the reactor coolant to the steam generator i can cause the coolant to cool down when the steam generators remove too much l heat (low steam pressure, too much ferdwater), or the reactor coolant can heat up when the steam generators do not remove enough heat (not enough feedwater). These effects cause density changes in the coolant; it contracts or expands accordingly. Regardless of the cause, changes in the inventory in the reactor coolant sys-tem have two effects- .
- 1. A loss of mass can affect the ability of the reactor coolant to transport heat from the core to the steam generators. If the RC pumps are not run-ning, steam can collect in the hot legs and block natural circulation.
When circulation stops and heat transport stops, then the steam generator temperrture will not " set" the temperature of the reactor coolant, and T will n t change when T changes. Md s e sg 1-7 Babcock & Wilcox l
I If the RCS mass continues to decrease and the core is mostly covered by steam, it will not provide a sufficient heat sink and the core will retain g the heat and heat up. Fuel failures can result if the RC pumps are not 9 running.
- 2. A change of mass or density can affect the ability of the pressurizer to provide pressure control over the RCS (this is discussed in section 2.4).
Operator control of reactor coolant inventory requires the ability to balance mass increoses or decreases by adding water with makeup or ECCS systems or re-moving mass with the letdown. Control of reactor coolant density changes re-quires control of the steam generator pressure and inventory. The RCS inventory cannot be measured directly. However, the operator has two indications to determine whether the inventory is sufficient for core cooling. Pressurizer level is an accurate measure of the inventory when the reactor coolant is subcooled (except for a rare possibility when free hydrogen gas may exist in the loops; this condition will probably only exist after fuel fail-ures caused by uncovering of the core). The other measure is the incore ther-mocouples; if these read subcooled or saturation temperature, then enough mass exists in the reactor vessel to cover and cool the core. However, the incore W thermocouples will not show if the loops are full. 2.4. Reactor Coolant Pressure Control Reactor coolant pressut* control is required to keep the reactor coolant sub-cooled so that it is in the best state to transfer the heat from the core to W the steam generators. For all cases of reactor operation except LOCAs, RCS pressure control is provided by the pressurizer. The use of oressurizer heat-ers and sprays is the usual way to increase and decrease RCS pressure when a steam / water interf ace exists in the pressurizer. The heaters maintain the reacter coolant in a subcooled condition; the sprays prevent pressure increases to li tt operation of the pressurizer relief and safety valves. Neither the heate s or the sprays have enough capacity to prevent large abrupt pressure changes, but they can moderate small changes. As a backup, the pilot-operated relief valve (PORV) can be used to reduce pressure, but it is not as desirable 3 to use as the sprays. I l-8 Babcock & Wilcox I
I I RCS pressure control by the pressurizer can be lost in two ways:
- 1. The steam / water interface in the prescurizer can be lost by I draining the pressurizer or if the pressurizer fills solid with water.
- 2. The heaters and sprays can fail.
Each of these is discussed below. Draining the Pressurizer - If the pressurizer drains, the heaters cannot pro-vide pressure control because no water is available for them to boil and create steam. The pressurizer heatere have a low-level cutout, but even if it were bypassed, the heaters would not increase pressure when the pressurizer water level is low. When the pressurizer drains the RCS, pressure will immediatcly drop to the saturation pressure of the reactor coolant in the hot leg. The coolant will be saturated at that pressure. Pressure control will then be controlled in- ! directly by the steam generator (the steam generator sets the reactor coolant j loop hot and cold leg temperatures). Since the hot leg is at the highest tem-I , perature, the reactor coolant in the hot leg will flash to steam. In effect, the hot leg will become a "new" pressurizer. 1 Filling the Pressurizer - Sprays depressurize the RCS by condensing the steam in the pressurizer. If the prr.srurizer fills with water, the sprays cannot be effective for depressuruins because the steam space is lost. When the pressurizer fills, the RCS may or may not lose subcooling and become saturated, depending on what caused it to fill. If the filling was caused by HPI or makeup and the steam generator is still removing heat, then the RCS will i remain subcooled because the makeup (HPI) pumps will cause the pressure to stay l at the PORV setpoint and the steam generator will keep the temperature con-trolled. If the filling was caused by heatup and swell because the steam gen-lg erators were not removing enough heat , then the system may become saturated l5 because the heat from the core will only go into the reactor coolant and not 1 out the steam generators. W When the pressurizer fills, because of either heating the reactor coolant or too much HPI, the water will be lost through the pressurizer valves. This loss is considered to be a LOCA, even if the action was done deliberately. I I 1-9 Babcock & Wilcox
I' Failure of Heaters and Sprays - Failure of the sprays and heaters in the pres-surizer control system can also cause a loss of pressure control. If the sprays fail and cannot be turned off, the system will depressurize. Depres-surization may also occur if the heaters fail in the "off" mode. The reverse is not true; failure of the sprays in the "off" mode will only limit the abil-ity to depressurize. Unless something else happens to the plant, no pressure increases or decreases will occur. If the heaters fail "on," no large pres-sure increase will occur because the sprays will operate to provide a balance. However, if the sprays are not working because the RC pump may be stopped, for g example, then the heaters can cause the system to pressurize to 2500 psi and "" cause coolant (steam) to be lost f rom the pressurizer valves; subcooling will not be lost as long as water covers the heaters. When only steam covers the heaters, they will no longer raise pressure and subcooling can gradually drop. If the heaters fail "on" v'.en they are uncovered, no water exists to cool them and they will burn out. 2.5. Steam Gene'ator . rn ojv Control Heat from the reactor coolar.t is transferred to both the steam and the feed-water. When changes in feedwater flow or steam pressure occur, the volumes occupied by the steam or water will change and the heat transfer may change. For example, when the volume of water increases, it occupies space formerly occupied by steam, so the mass of steam has to decrease. This changes the relative amount of OTSG tube surfase area covered by water and steam. Because water has a greater heat capacity (MC ) than steam, it is a better heat sink u for heat transfer from the reactor coolant than steam. Simply stated, there are more pounds of water in a cubic foot to absorb heat than there are of steam. If the water inventory increases, then the generator will become a better heat sink for the reactor coolant, but if the water inventory decreaser or is lost, the generator will lose is ability to absorb heat from the reactor coolant. For example, after trip when the core heat is nearly constant, if the water level in the steam generator is raised rapidly without changing steam pressure, the reactor coolant temperature will drop and remain low until the feedwater addition reaches a new level and that Icvel is held. Once the new level is fixed, the reactor coolant will reheat and temperatures will return to their g former values. 5 I 1-10 Babcock & Wilcox E
I I This cooling ef fect of feedwater is caused by the inlet feedwater temperature, which is colder than the general temperature of the bulk of the fluid in the I steam generator. The inlet feedwater temperature allows a colder heat sink to be established in the steam generator. The steam generator level can, however, be increased slowly after trip, without a large drop in reactor coolant temperature, by controlling the rate of feed-water addition. Too much inventory can also be the result of overfeeding with the emergency I feedwater system. Even though its flow rate is lower, emergency feedwater will have a proportionally larger cooling effect on reactor coolant than main feedwater for three reasons:
- 1. It comes on when the reactor is tripped and core heat is lowest.
- 2. It is colder (T gg fy is less).
3. I It has a steam pressure reduction effect that main feedwater does not have. On the other hand, if the steam generator inventory is too low, insufficient feedwater or loss of feedwater can lower the water level, and the reduced heat sink will not allow the reactor coolant to transfer all of its heat to the steam generator. When the generator's heat sink is reduced, the coolant must retain more of the core heat and it will heat up. For example, if all feedwater is lost, the water in the generator will boil and only steam will remain to remove heat. But because the steam does not have I enough heat capacity, the reactor coolant must retain the core heat; reactor coolant temperatures will increase. When all feedwater is lost, the reactor coolant pressure will increase to the PORV setpoint and the coolant will even-tually become saturated as the core continues to add heat. The steam remain-ing in the generator will flow out through the steam lines and steam pressure will drop; loss of the steam eliminates the heat sink of the generators alto-Fether. Finally, another part of steam generator inventory control is feedwater tem-l perature. The heat sink of the generators will be affected by an abnormally low feedwater temperature. A reduction of feedwater heating steam or loss of a feedwater heater will cause the reactor coolant temperature to decrease. ICS I 1-11 Babcock & Wilcox
I operation will usually stabilize the plant, but the decreased feed temperature fill cause a change in the heat sink and increased heat transfer from the re-actor coolant. The operator should control the rate of feedwater addition to control the steam generator inventory. Level measurements in the steam gen-erator downcomer give a good indication of the steam generator inventory for control. 2.6. Steam Generator Pressure Control Heat transfer from the reactor coolant to the steam generators goes to both the steam and the water in the generator, After reactor trip the steam and feedwater in the generator are saturated, and changes in steam pressure will cause direct changes in the saturation temperatures of the steam and the feed-water. A review of the saturated water and steam sections of the ASME Steam Tables will show how much the steam and water temperatures will be changed by E increasing and decreasing steam pressure. In the usual situction the operator 5 must handle, he would manually decrease steam pressure using the turbine bypass valves or the atmospheric dump valves. When the steam pressure is lowered, the heat transfer from the reactor coolant goes up because the steam and feedwater in the generator become a colder heat sink, causing more heat to flow away from the reactor coolant. Two reasons combine to create the colder heat sink: first, the temperature of the steam and feedwater is reduced and the rate of boiloff I W goes up; the increased boiloff takes away more heat. Second, the increased boiloff requires more feedwater flow to be added to maintain level. The inlet feedwater is colder than the water already in the generator, so its addition contributes to the colder sink. Because a colder secondary sink exists, the primary side temperature will drop as heat is transferred. Steam pressure can be lowered in two ways:
- 1. By opening the steam line and releasing steam (turbine bypass, steam line break, atmospheric dump valves, steam to emergency feedwater pump turbine driver) .
- 2. By spraying cold emergency feedwater into the steam and condensing it. This is similar to the way pressurizer pressure is reduced by the pressurizer sprays.
Steam pressure can also increase, but normally it will only increase from the operating condition to the reactor trip condition, where it will be limited by the steam safeties or the turbine bypass valves so the effect on reactor cool-ant temperattire is small. However, if steam pressure is low because of a 1-12 Babcock & Wilcox I
'I I failure - for example, a steam line break - the change in reactor coolant tem- ! perature could be much larger. When the steam break is isolated the reactor coolant adds heat to the generator and causes the steam pressure to increase. 4
- The operator can limit the increase in reactor coolant temperature under these j conditions by lowering the turbine bypass valve setpoint and keeping steam
! pressure low. 'I ll lI lI 1 lI il 4 I i I I I 1-13 Babcock & Wilcox
I'
- 3. States of Water in Primary and Secondary Systems The state (solid, liquid, or gas phase) of the water in the RCS or the steam system is determined by the existing pressure and temperature conditions. The terms subcooled, saturated, and supertaated are normally used in operating pro-cedures. These terms are defined below.
Subcooled Water can exist only in the liquid phase. Saturated If heat is added to subcooled water, a temperature for the exist-ing presstr_ce will be reached at which the water can exist as l cither a liquid or a gas (steam). At this point, the liquid is up called saturated water and the gas is called saturated steam. The liquid and steam phases exist at the same temperature aN g pressure. Heat mus'. be added to saturated water to change it to saturated steam. Heat can also be removed from saturated 5 steam to change it to saturated water. The heat required to make the change is called the " latent heat of vaporization." Superheated Water can exist only in a gaseous or steam phase. This phase can be distinguished from saturated conditions because the tem-perature will be higher than the saturation temperature for the E 3 existing pressure. The state of the steam generator outlet fluid is superheated 6tring power op-erat ior. and sat urated af ter t rip. The state of the reactor coolant can be de-termined by watching the RCS pressure and temperature on a pressure-temperature diagram (see below). P-T conditions that are to the left of the saturation line are in the subcooled region, and those to the right of the saturation line E are in the superheated region. 5 SUDC00 LED
^
SATURATION LINE I P SUPERHEATED T r I 1 - 1 t. Babcock & Wilcox I
I 3.1. se'qcooli"E Subcooled conditions are maintained during normal operation. During a reactor transient it is desirable to maintain the reactor coolant subcooled. When sub-cooled, the following conditions exist:
- 1. The primary loops are solid water, and a water level is present within the pressurizer.
- 2. The pressurizer water level is a true masurement of RCS inventory.
(Note: A very special case can exist when the reactor coolant is sub-I cooled and a water level is in the pressurizer, but the loops are not full. In thot case, pressurizer level is not a true measurement of inventory. That condition is when there is a large amount of free gases in the loop. The gases w!11 be mostly H2 that has been created I after a large amount of fuel failure. Since this would be an un-common event, reliance on pressurizer level is u.ually acceptable when the reactor coolant is subcooled.)
- 3. The reactor coolant is liquid and is ideal for heat removal from the core and heat transport to the steam generator by either forced or natural circulation.
- 4. RC oressure can be maintained by the pressurizer and can be regulated by using normal procedures and equipmtnt (sprays.. heaters, and regu-lati an of pressurizer level by the MU and/cr HP1 system).
- 5. RC temperature can be controlled by the secondary system (with feed-
,g water available) and can be regalate.d by adjusting f eedwater flow g and steam pressure. i Subcooling should be checked in all parts of the loop, especially when natural circulation is removing heat. The operator should check T g and T eold in bo@ l loops. 3.2. Saturation l Subcooling can be lost when the pressurizer drains or is filled solid (if the pressurizer is solid because of HPI and cooling is by the steam generators, l then the reactor coolant can remain subcooled). A loss of subcooling can be caused by an overheating or overcooling transient or a loss of reactor coolant. Saturated conditions can exist in isolated peckets of the loop (i.e , within one or both hot leg pipes and not in cold leg pipes) or within the .ystem as a whole, as would be the case during a major LOCA. Therefore, temperatures I should be checked in the hot and cold legs of both loops. When the RCS is saturated: l I-S Babcock & Wilcox 1
l I i
- 1. Neither the reactor coolant temperature nor pressure will show whether the saturated fluid is liquid or gas (steam).
- 2. Voids (steam bubbles or pockets) can exist within the primary system.
! 3- The pressurizer water level indication is not a true measurement of l reactor coolant inventory.
- 4. If the RC pumps are off, a loss of natural circulation may occur be-cause steam voids can form at the top of the hot lag and block water f'ow.
- 5. Normal pressure control by the pressurizer has been lost. The RCS hot Icg loops, which have a steam bubble at the top, now work as a pressurizer. RC pressure will be controlled by the amount of steam in the loops. The amount of steam can change because of steam con-j densation by the steam generators, by the addition of cold HPI water, g
! or by the loss of steam generator heat removal. E l l'nder ideal conditions, subcooling should exist in all parts of the reactor E coolant loop to be able to transport heat f rom the core to the steam genera- 5 l tors; botever, the steam generators can remove heat when the reactor coolaat is saturated. For all events except a LOCA or a total loss of secondary fluid, naturated conditions should be a temporary effect. For example, if steam gen-erator overcc oling causes the pressurizer to drain, saturation will occur, but HF1 will start and restore the reactor coolant to a subcooled state.
- 3. 3. Superheating Superheated reactor coolant conditions are to the right of the saturation line of the P-T diagram. Superheated steam results when the core is uncovered.
Heat from the core is passed to the steam and its temperature rises above sat-uration. k' hen the reactor coolant is superheated, the core is cooled by steam. Steam canno? remove enough heat to prevent the core and cladcing frc heating up, and f e' failure may result. Superated steam indicates inadequate core ecoling (ICC) The only accurate measure of temperature is the incore thermocouples, which should be used along with ho* 1*.g pressuri to determine the amount of super-heating. More will be said in Lesson 3 about recognition and criteria for operators n. der these condit ions. I 1-16 Babcock s. Wilcox I I
l
/ 4 Natural Circular ton _
4_ . l . Overview When the RC pumps are tripped, forced circolation is method of removing core decay heat lost and an alternate must he found. The preferred method is to t ransport this beat net or coolant. to the steam generators by natural circulation of t he r e-Natural circulation is possible as long na a few simple require-ments are mets (1) n beat source is available to proo. warm (Iow density) water, (2) a heat a flow path (loop) sink is available to produce cold (high density) water, (3) is nvailable connecting the warm and cold water, and (4) the cold water is above the warm water. Requirements 1, 2, and 3 are simple to understand. Decay heat in the core is t he heat side of the source, water on the secondary steam penerators pravides a heat connect the t wo . sink, and the hot and cold legs reality, bent Req ui rement 4 involves a concept enlled thermal center.In core and again in ant ransferred continuously as the wat er moves up rough it th t he moven down t hrough t he st eam generator. ter is t he point The t hermal cen-at average temperature. in the core or the st eam generator where the primary water s i It can be used to represent the in ita " average" cond i t lona. entire column of water Thermni Cent er Definit ton 1. Core thermal center: t hat auy be considered to go from elevat T ion "in t he core at which t he coolant eold hot' 2. Steam generator t hermal center: that tor at which t he coo lant may be considered elevation in the steam genera-to go from T Requirement g to T cold' 4 for natural circulation can be met if the themal center of the st eam generator is above t he t hermal cent er of t he core . This will put the everage cold water above the average hot will nink, the water, the cold water (more dense) hot water (lens denne) will rise, The and there will be circulation. rat e of natural circulat ion (gpm) depends on the following f actorn: 1. The friction (renistance to flow) the primary loops: this of t he piping and component s around built, and t he operator has noin determined when the plantis designed and cont rol over it.
- 2. Ti c wt rengt h of t he heat heat, which in source:
a f unct ion of past thin depends on the available decay teactor t rip. It will, of course, power history and t ime since the The operator han no decrease with time af ter t rip. the reactor in shut down cent rol of this af ter trip except t o criture t hat no t hat the ,nly heat input is decay heat. 1-17 Babcock a Wilcox
j I E l strength on natural cir-I Figure 1-2 shows the effects of decay heat culation flow rate. the colder the heat sink, the more it The strength of the heat sink:
- 3. passing through the steam a will be able to cool the primary coolantThis will make the water 5 The operator can make the heat more den genera or.
(opening the culation flowbyrate(1) will increase. lowering secondary steam pressure ondary satura-sink colder turbine bypass or ADVs more); this will lower t e secf r across the tubes; h g e from main to tion or (2) temperature, which will increase heatacross heat transfer trans elower feedwat emergency feedwater); this will increase the d ry LT. the tubes by providing a larger primary-to-secon a h team
- 4. Difference in height between the core thermal center and t e sthe generator thermal center:between the high cold water and the lower hot balance will exist ll result. The core water, and the more natural circulation flow withe operator transfer can control g the ste thermal center is fixed, but (1) Most of the heat g erator thermal center by two methods: boiling area just below the established secon occurs in the violent Tharefore, the operator can raise the thermal (Figure 1-3 gives ary side water level. center by raising the steam generator water h= level.
d feedwater high in the typical water level requirements.)the(thermal top ofcenter) the generator; of gthis wil If EFW is generator and thereby raise the average heightThis he only works g while heat removal. thermal center will move back down to just below t stopped, the water ~ cel. ing the h In summary, the natural circulation flow rate can be changed by c ang and cold water or changing the dif ference in AT (density) between the hot height between the core thermal center and steam generator thermal center. This can be expressed in equation form as follows:
~
LP gy =hgg(O c h) g
"" LP =
P driving head, the driving head available for natural 5 dh circulation,
^
= distance between core thermal center and steam generator .
b eff thermal center. p
= density of cold water at steam generator thermal center, c r h = density of hot water at core thermal center.
1-9. This is shown graphically in Figure
]
Babcock & Wilcox l l-18
~ - - - - - - _ _ _ _ ---%.-
I I 4.2. Natural Circulation DurL.g Normal Operation When the RC pumps are tripped, the operator should check two things to ensure that natural circulation is being initiated properly. First he should make sure the reactor coolant remains subcooled. I If it does not, he should make every ef fort to restore subcooling. Second, he should ensure that the thermal center is being raised in both steam generators. Automatic equipment will normally start EFW and increase the level to 50% on the operating range of each steam generator when the RC pumps trip. The operator should monitor this process while keeping the following in mind:
- 1. As long as EFW is flowing into the top of the generator, it is not I necessary to get a 1cvel in the generator to have natural circula-tion. If the heat source (decay heat) is high enough, the EFW may come in and boil right off and go out as steam. This is acceptable; the thermal center is high and natural circulation will develop.
I Figure 1-5 illustrates the insensitivity of natural circulation flow rate to water level while EFW is flowing. 1-4 again determines the driving head. If EFW is stopped, Figure
- 2. With two EFW pumps running or with low decay heat levels, it is likely that the reactor coolant will be overcooled and could drain the pressurizer.
I If the pressurizer drains, subcooling will ba lost. This will not happen if the rate of EFW flow is limited. The opera-tor can do this by throttling EFW flow. After initiation of EFW the operator should watch steam pressure, pressurizer level, and cold leg temperatures. If necessary EFW should be throttled.
- 3. If no EFW is available, natural circulation can be initiated using
'g main feedwater. Again, the level should be raised to 50% on the lg operating range. This method is not preferred because it takes longer (than EFW) to raise the thermal center and establish natural circula-tien. By the time the level is high enough to get the flow going. ,l there may be so much colder main feedwater in the generator that over- 'a cooling cannot be prevented. I Ideally, main feedwater is best used for natural circulation if (1) the w-quired steam generator level is established before the pumps are tripped or (2) the steam generator level is established first by EFW. As natural circulation is delivered in the RCS, the cold leg temperatures (Tcold) will be about equal to the saturation temperature in the steam genera-I tors. The hot leg temperatures will increase as necessary to develop the driv-ing head required for flow (by developing a density change between T and T ). h The best measu u to use to determine whether natural circulation has started 'I 1-19 Babcock & Wilcox I
I 1 is the coupling between T and the steam generator temperature and the tem-perature difference between T nd T . When both T and T are subcooled, h h they should follow steam generator T, when it changes; the temperature dif-ference between hT and Tc should not exceed 50r. If only T is subcooled and c T is saturated, natural circulation characteristics should be the same as if b th-y are both subcooled. Once natural circulation is established and the higher steam generator levels are reached, the operator's job is to ensure that feed-water is availabic to replace the steam generator water being boiled of f re- W moving decay heat, and to keep the RCS subcooled. Natural circulation flow will regulate itself. That is, as the heat source (decay heat) dies down, the LT (T - T ) will decrease and there will be less h driving head available; therefore, flow will decrease. 4.3. Natural Circulation During Abnormal Operation Thus f ar, the discussion has concerned expected or normal natural circulation conditions. That is, the RCS is subcooled, the level in both steam generators is 50% on the operate range, and both steam generators are being steamed. This 5 section examines of f-normal conditions: natural circulation with one OTSG and natural circulation with a saturated RCS. There may be times when an operator does not want to steam a generator (OTSG tube leak) or cannot steam it (steam line break and isolated generator is dry), if the system is also in natural circulation, the operator can expect the fol-lowing: T hat n both loops will k about equal; T eold n Perat nE Fenera-tor will be equal to T in the operating steam generator. Teold '" E * '" - lated generator will not be equal to T in the isolated generator; it will g probably be much colder since it is influenced by the temperature of the seal W injection water coming into the idle pumps. (T h -T)c n the operating steam generator should not be more than 50F (but the Icvel may have to be raised above 50% to keep the AT below 50F). Steady-state operation under these con-ditions is simple and safe. Plant cooldown, however, is complicated because the loop with the isolated generator will lag behind the steaming generator. If there is water in the isolated generator, it will become a heat source in-stead of a heat sink. In fact the isolated generator may add enough heat to cause the reactor coolant in its hot leg to flash to steam. If this happens, that hot leg will act as a pressurizer and slow the depressurization during a cooldown. This will also slow the cooldown rate. The operator must watch 1-20 Babcock & Wilcox I I
carefully in both loops under these conditions and ensure that adequate margin is maintained by regulating the rate of cooldown with steam pressure control of the operating Fenerator. A subcooled RCS is the desired state; however, natural circulation will remove core heat when the RCS is saturatad. As long as the four requirements for natural circulation are met, heat will be removed from the core aad transferred to the steam generator. The problem with saturated natural circulation is that the operator does not know how much of the reactor coolant is steam and how much is water. If the RCS is losing inventory, steam will form in the hot legs and eventually stop natural circulation flow (this is a violation for the re-quirement that a flow path exists and connects the hot and cold water). Another form of natural circulation could still exist under these conditions -- reflux boiling (boiling in the cor- and condensing in the stcam generator); however, it requires a higher steam generator level (95% on operate range). This could also be violated by a large collection of non-condensible gases in top of the hot legs, but such a collection could only exist following core un-covering. The point to remember is that primary inventory (mass) is nknown under saturated conditions, and thus, every effort should be made to keep the RCS subcooled. I'I Babcock s. Wilcox
I
\
Ei
- 5. HPI Cooling lf primary-to-secondary heat removal is lost bec9use of the loss of feedwater, the core can be cooled by the HPI system until feedwater is restored. The core I energy is removed by the reactor coolant (HPI) and released to the reactor building: which serves as the heat sink instead of the steam generator. The core is kept covered and cooled by HPl. A total loss of all feedwater is 11-lustrated in Figure 1-6 and is discussed below.
- 1. With the plant at power, a loss of main feedwater and a f ailure of emer-gency feedwater would result in a reactor trip (anticipatory trip on loss of MFW or RPS actuation on high RCS pressure). A loss of all FW could also occur during hot shutdown or plant heatup/cooldown.
j 2. The secondary side of the steam generator will boil dry and the RCS will f then heat up due to decay heat. i I 3. Subcooling will decrease and the reactor coolant will expand into the pressurizer. Note: The rate at which this occurs will depend on the ini-l tial inventory in the steam generators and the core l decay heat level. For example, the RCS heatup rate 3 l may be as high as 4F/ minute with high decay heat or as low as IF/ minute with low decay heat follow-ing boiloff of the steam generator inventory. ! 4. RC pressure will increase as the steam space in the pressurizer is com-I pressed due to the insurge of reactor coolant and steam / water will be re-1 lieved out of the electromatic relief or pressurizer safety valves.
- 5. The reactor coolant will eventually saturate and water will boil through-l out the core.
- 6. Without corrective action, the reactor coolant will slowly be vaporized to steam and relieved to the containment, causing core damage.
To avoid these consequences the operator should make every attempt to regai1 feedwater to at 1 cast one steam generator. This includes main feedwater, emer- ; gency feedwater, condensate pump (if steam generator pressure is low enough), l or service water. If feedwater cannot be regained before primary-to-secondary heat t ransf er is lost , he should manually start two HPI pumps and open and j leave open the electromatic relief valve. HPI flow should be balanced to give I the maximum flow possible. :
' 1-22 Babcock & Wilcox E
T The event can be recognized by observation of generator level and equipment status checks or through use of the P-T curve. Figure 1-7 illustrates the P-T i response of the RCS af ter a loss of main feedwater f rom 100% power with no EW and appropriate operator action. When HPI is started, the RCS will eventually go to a water-solid condition (subcooled) with RC pressure controlled by a com-bination of the HPI pump head rise and the relief capability of the electromatic relief valve (ERV). The number of running RC pumps should be reduced to one to reduce the heat load. The time to become water-solid depends on the core de-cay heat level and *.he number of HPI pumps. If core decay heat is high, the reactor coolant will saturate and its pressure will rise to the pressurizer l i sa4ety valve setpoint. The runnlag RC pump should be tripped when subcooling margin is lost (or the low pressure ESFAS setpoint is reached). The water in-l ventory in the RCS will drop because of core boiling and relief out the pres-surzier relief valves until the amcunt of water added by HPI can take away all the decay heat of the core. Until that time, water from the RCS and HPI water together are needed to take out core heat. When the amount of HPI water can take out all the decay heat, HPI is said to " match" decay heat. When HPI matches decay heat, the liquid inventory in the RCS will start to increase, e steam will be condensed, and the system will slowly go to a water-solid (sub-cooled) state. If the core decay heat level is low at the time feedwater is lost, HPI may be enough to take out all the heat and the RCS may remain sub- ) i cooled until feedwater is restored. When all feedwater is lost it is very important that two HPI pumps be run until subcooling exists. One HPI pump is adeouate for core heat removal, however, because decay heat drops very slowly, the time required for one HPI pump to l match decay heat is much longer than for two pumps. For example, the times i to match decay heat with one and two HPI pumps are approximately 70 and 8 min-l utes, respectively. When subcooled conditions (based on incore thermocouples only) are reached, l the operator should start one RC pump and throttle HPI flow to maintain the i l reactor coolant subcooled but within equipment design limits (NDT/RV brittle fracture) (see Figure 1-8). Figure 1-8 is an emergency curve and should only I < be used if the normal NDT limit curve cannot be used. If an RC pump cannot be restarted, the operator should still throttle HPI to maintain an adequate subcooling margin and avoid NDT problems with the reactor vessel. The intent l I l-23 Babcock & Wilcox i
I is to run an RC pump to improve thermal mixing in the core. However, if the pump is not ready in all respects to be run, it should not be started. Core g protection is ensured by the llPI cooling. 5 In summary,llP1 cooling will maintain core cooling if secondary heat removal is lost. It is not a normal operating mode for many reasons; three examples as are as follows:
- 1. liigh RC pressure will occur, causing the pressurizer relief and safet, valves to be cycled with water and two-phase flow discharge.
This increases the potential for a LOCA that cannot be isolated g) (failure of code relief valve). gl l
- 2. Long-term operation (subcooled solid-water operation) is extremely l sensitive and must be closely monitored to prevent exceeding equip- ,
ment design limits (NDT and RV brittle fracture).
- 3. The degraded containment environment may cause failure or bad read- g ings of instrumentation. 3 Consequently, secondary cooling should be restored as quickly as possible so g that normal primary-to-secondary heat transfer can be resumed. 5 I;
I I I I I I I I l-24 Babcock & Wilcox
i i
- 6. Lesson Simarv I' Briefly review the following:
l. How to detect abnoraalities affect the flow of heat.
- 2. Actions necessary to control heat transfer.
- 3. The rationale for guidelines presented for IIPI operation.
s 1-25 Babcock 8. Wilcox
l I l l Figure 1-1. Fundamental Methods of Heat Trenefer Control SPEAY I I f e al s
'" REACTOR PRESSURE g
e [ ' STEAM ~ REACTIVITY
)
p CONTROL ( j HEATERS l STER 4 GENERATOR r E [ ',
' d t
PRESSURE b CONTROL 6
~
N I /
/
/ \
FEE 0 WATER
\ >
STEAM GENERATOR M INVENTORY CONTROL p LETDOWN (HPI) REACTOR INVENTORY m I I I I I I I Babcock & Wilcox 1-26 I
Figure 1-2. Calculated Natural Circulation Flow Rates Vs Decay Heat Power - All Document Cases to Date on B&W 177-Fuel Assembly Plants 7 , , , , I RC PUbPS POT FULLY STOPPED IN~THIS CASE T k I o e4 x 6 - X - 5 - h l m p jI ! E 4 - - i d 5 # !l
' P 3 -
IE d im M l C 2 - x g - 0 i g x MINIKJM ACCEPTABLE l @ 1 - x NATURAL CIRCULATION _ FLOW RATE I O , t , , 0 25 50 75 100 125 l DECAY HEAT POWER-EGAWATTS I 1 F L F L 1-27 Babcock 8. Wilcox t
I Figure 1-3. Comparison of Lovered- and Raised-Loop Pesigns for B&W 177-Fuel Assembl*/ Plants m , ( h ' I I
;(
b I f \ l l l ? I I JUR - r a , I
'r 1 S ,
-35 " LEVEL
' =
\
'l-~ j l
505 r y OPERATE ( g ,i f d( [] j L EL ^ I t TOP c- 3-- i , . CORE CORE L__4 _J 4 ..\ w N mss m, W 1 g 3 I4 Babcock & Wilcox ! 1-28 I
_r. Figure 1-4. Parameters Contributing to Natural Circulation Driving Head
/ T
~Z NOT TNERNAL CENTER FOR NEAT REMOVAL COLD FLulo COLUMN
(-) FLUID COLuna - J y~ ~y / - ,. 7 - -. ....
*e ft t Y /
-- ] ----- - -- g TNERNAL CENTER "r / N FOR MEAT ADDITION
/ N W
N 1 O' eriving heas * "eff (P c -pg) 1-29 Babcock 8. Wilcox
Figure 1-5. Duration of Phase II of Natural Circulation Test at Davis Besse I, November 3 and 4, 1978 NATURAL CIRClLATION FLOW M G r a mU ww
~
- a. su o
r 4.0 __ a. m 16 - aU m yn* - 3 u. u_ o WATER LEVEL at w 3.0 m<r 12 a * - a wm x
- a o u.
53
- a. <
o m x m 2.0 - o u. 8 - t.e no x a g Q; - REACTOR POWER i j ww
/
O m 1,0 - J 4 ----__/___.,._____________.f 0 I I i e i I I L m
- w g 0 1 2 3 4 5 6 7 l 8 i X e TIME-HOURS
' :E E E 1 M M M M M M M M M M M M M M M M M M M
Figure 1-6. Backup Cooling by HPI for Loss of All Feedwater, No Operator Action
- n. /- ,
[b ' \
/~ i
,1 Q [K ~
[- b I
\ !
,xNs N. i l , l ,
1
~1 l
-l l - .
_J 1 ,i l l i
, l
)
} Y.. e /
lI l
! l I 1
i ! ! O ' l i l l A ' l ( , ; =d w_ = l ' l\ t l l i ~ l(> l i - b H
} -
@eix 1
w ' -y c l j j ! 1
~O,) 'f' I
ll l lg [ l O
}
W j
}
l {
\
w/ j i
\
[/}~}l-:d) D?E l G c;) l
'_?LLL
- l. 1 toss cf f ee.$.e t er .
j j, k ee c t er tri r on Pi;* $ U, press.sre f or ant ic ipiter,v tr ip i .
- l. E:
- eat s F cue to les s c' se<.cncary %et s tre am' earends into l p ,g gg,j gpe ,
i renso,g, r ie,et o,r resses f ace 1 ress.cse voi t o.i nG
- reac t or tr'p) t *,* wreeses m ouse of 't ado t ton ene 2 G pressere inreases t: tw setpo int. eresssrire stes- is
,;ected ost of 47 res c et e' ree:ter coolant.
! a tetoNery s tee bc.il s cry. L Fr wr., sist er goes -4te- sclid ist111 surcooted) at 2500 ri t , . I I C or t i euco,
- es t ut o' reacto c oC aat causes . ster relie' out of gg o, g, et) .eivn .
5 18. con et s ta rt . l
- f. s' pensee sh<i.s norma t po s t t e t t, reioc.cse t ner s ec renes '**""''# '#"C"*9 '*I' rat ed condi t iers e 2s00 :s s
.. prniwriter level is restoret I
l I is!
,li- (
4 c.__L g.3.. ,a -. . ,o .- _e c._.,- L_,.-s- - s
) * ,\
Io j th t g gg 6g [,,*
- f'o , Peirast tiCl st:cscant SICE
] ,L :\ h !::.! O~ a .,u, y -
- i. ;
!c
- i. . !
l0 Y7./ h f eS Pmait liggi; e - pi sitiv e) .-s ! l t i
. $W
$ [e ,C III'"
+ , )g
- eo L nyo i '
[ } *j [I _A 4 0 **A-
! F ,- *I o, 'e*!
j . l
'* i s
]
[' _j i
*i; e
.o to ; '*
- i ; * * !
l* , l ,! ( L_ e_-
": \e o of
- f. .wy; J; s .o '
s-o
' i ."i._.Ul i a .. s s. . .r ee u e t ca e es,;
y 1. t ee- a vee tes in t or. 9ca ne o' tc o < at ne ao t ** < 3 , p t.e-n .r e rc..m c er .i n be c u t mt w
.<% ne t o, w s.e si me) w c . v e s'esa I'
r.,, , ,Mr s e, sie.ea 4*
.~ m e, 111 i t ~ .ne. severst
,e . e ,,'nr t v e .t u (Continued) e b 1-31 Babcock 8. Wilcox
)
I Figure 3-6. (Cont'd) I, I pm- ,- c-. , v ,s w y.
, v. 1
.,m s . .. ;
tw N1 o N i
,9 3 Q Wy Q Nq ( N_Q eq9 $
iJ1[6. Nx?N> s Ej
.j } x1
,N 4
R1FxN pN sv ' ._ N1,
's %[bx N.
x xs s 3 N3 t v ,, 1 sj
.N! N s:. l xN N; v
;h t, n A_
k_==_ ~~a
~
,N l,k k ;
6.@._ , . . t.,N. Ib
.si ,. , 3 s ' N NNq (' - N ;
q (_1-t1'txxsxs,'.y' e- -? i
' g ' 1v
<N. s is' 3
e - -. h. .rrn,Q . s_ us r, s s s 4 ss .S. l N N,1 7.I 0, 4:x'. N,
; 3:sN N'1 U 4
;4 , \\
s[ o l j
" . _,~\d d. N, (sNNN _%
,- i. , , * */ '
--" t yyy v .
i j e w Q;]~_Jw , %J wy.! rnmmm u.r .. m m c,,3 [nd feedit t9e fSe tota' 10$1 0 #F( mit%ut HPl SCisat107 I nc r.Ga11 t ter f or t et el 101% of Fe we t%gt MP! ar,tvet ier. #iti C PJM Of' te'l) if the twent , ret;tof COOI5et C.8F L ytt e weald (Oeklet ely tC f d litteF Da l , tR Rf. 5 L e r? r, will irr,telf SM C8 A9 SuperhPa t e If IF8FIM I Oh*r-..FP ef
,5 ' D*% 0' ,I OO& 4 94 red Cterd,os c, t.io. to, intl .
I sF ligr a t w
%,ou.1. u,, om , , Co~ i t ie . .. ,d %I t ta* 10 a,, . m *.14. , . t .. . ; ..i Ed *I . I n4 3*Cwe t t (Orf (OC I ' Pg mil l 9 3 ', $1.
I t l l I
- I I
I I I I I 1-32 Babcock s. Wilcox I
Figure 1-7. Backup Cooling by HPI for Loss of All Feedwater, With Operator Action 1800 l PCl? tatP l 3 I'0" " sim000 F - I \ ..
, l 2700 - .
'sd'
; L. - .'. I I E g
c M00 1s00 SusC00L10 its,0 d 2 7 v
!uNantat
'Ist0h I [f. . .
- 1600 - d /
l <C 0 ! : 1400 y [6 II f A 1200 - sitas Pat 11URI 3
~
[ thC P0lht PC11 faar esta 80=L1D Cettutat:06 .i n01 j j tleif EICOL D , A40 (09 ha tbRa. 1000 Cittutaflom e t C0i O '
; U 40Reat Optaatagg P0 int P0ste i O 800 - l j OPIsafl0h afM07*
E " !afussil0m P l Inc P0lti P011 TRIP siin 600 -
/ i
[,j nafusa6 CiDCuiat!Ch >fot0 i '
! - SutC00Lt0 l 440 - sangl= Lint 1 1
i 0 i n i i i i 400 aSO S00 $2 E00 6 50 100 Reactet Coolant ana $tte Outit! Itaptfatuft. I 1 ! Geference Tine Points (Mtnut*S) Gemarks i 12 0-1 Heactor tripped on anticipatory loss of feedwater. i horral post-trip Cooldown and depressurl24 tion in progress. IfW does ny initiate. IW 0 1-2 Stean generators dry. RCS L,* sins to reheat and repressur ite due to loss of secondary cooling. j 3 3- 4 Occrator diaeneses loss of heat transfer, opens I PORV. starts twe H PI pumps and talances HP! flow. { PO4V release rate enceeds HP] capacity initially l and RCS t>egirs to depressurite. Operater trips ' all tut one RC par to reduce heat input. l
; 4 5-0 Subcooled maratn is lost. Operator trips rew ining RC pump.
) 5 6-7 RCS reaches saturettor. 6 7-8 Pressurizer in solid or near solid condttton. HPI J flo. " matches" cecay beat and tiegtns to repressurite ! RCS to subcooled conditions. i 7 F-10 RC5 suticooled margin restored and RCS ts beginning to cool due to HP! flow and PORY release. Operater tnrottles HP! flow to maintain sJacooled conditions I at a pressure lower than the safety valve setpoint and restarts an RC puPT to prodiote thermal mining of HPI to prevent therml shock.
- 1-33 Babcock s.Wilcox
I I
=
~
N N N _ __ __ _ _ -. _ _ _ o g o
. : \ E,
- U E \ Nh E 3
.a 5 \E E x
e i I o o u b
$ DN s
e w w
\
.._ __4(
u g _r s ea Ws o.
~
m _ v - u A $ E EE , .
-4 =
i o
"U e a t E
o -- - = u u o- " 5
" ~"
= 5-O ,w . y -
u - o o .. m 2 Em E - 1 J I T __L_ ___ g
~
m yr ~: w t_ e , . E i a, e e ;
. == t : :
m ! g u
- s
.e o o,
?o s5 . a.
I
- t,' -
o w - u - - - C 90 2
; Vo SE E W
- = 5 l n: : SO l cm
- o. ojo, e =
l as 7 1 o o e o o o o o o e o e o n = = = s = =
~ ~ ~ =_ =_ =_ =_ _
l oisn> .inssa y a I l-34 Babcock & Wilcox I I l
E I Lesson 2 - CAS/ STEAM BINDING Introduction
- 1. Lecturer -
1 I 2. Purpose - To describe situations that could Icad to introduction of large
- quantities of gases into the RCS. To list potential gas sources - conden-j sibic and non-condensible gases. To discuss pot ential gas ef f ects on nat-
) ural circulation and on OTSG heat removal. To give guidance for recog-nizing and dealing with those gase's. Objectives l ]lW The following subjects will be covered during this lesson:
- l. Why gases right enter the RCS.
- 2. What gases might do once insidt t he RCS .
I 3. Ilow an operator might recognize and deal with gas presence and effects. Key points to be retained are as follows:
- 1. There are situations which could lead to gas volumes in the RCS.
! 2. Once there, gases can affect natural circulation and heat removal through the OTSGs.
J
- 3. These effects can be recognized and dealt with by the operator.
l 4. Actions are dependent on availability of certain equipment (e.g., OTSGs. I!PIs , RC pumps) .
- 5 1
i 1 I 2-1
- Babcock t. Wilcox l
I i I I l 1 Lesson Outline I
- 1. Introduction i
l 2. Situations That Might Result in Cas Accumulation in RCS 2.1. Loss-of-Coolant Accidents 2.2. Total Prolonged Loss of OTSC Feedwater 2.3. Severe Overcooling 2.4. Inadvertent Cas Discharge
- 3. Potential Gas Sources f
g
- 4. Potential Detrimental Effects of Cases in RCS W l S. Symptoins of Cases in RCS 4
- 6. Removing or Otherwise Coping With Cases in RCS I
- 7. Lesson Summar,i
- 8. References !
I L r I i P I I,: l l l 2-il Babcock & Wilcox l
4 J i i t ill e l Leeson 2 - GAS / STEAM BINDING 1 i
- l. 1. Introduction Certain situations that may arise during plant operation can result in the ac-jg l W cumulation of gases within the reactor (oolant system (RCS). Should these j gases be introduced into the RCS in sufficient volume, tl.uy can potentially ,
hamper efforts to regain or maintain control of the plant. This lesson will l deal with possible ways gases may enter the RCS, what their harmful effects may be, and how an operator may recogr.1ze their presence and eliminate them. E I 1 lll 4 i 3 t !,il lll 1 l 2-1 Babcock & \Vilcox
- 2. Situations That Might Result in Gas Accumulation in RCS 2.1. Loss-of-Coolant Accidents Loss-of-coolant accident s (LOCAc) can cause gases to be introduced into the RCS in a number of ways. First, LOCAs can depressurize the system to the sat-uration point, causing steam to be produced as the primary coolant liquid flashes. In conjunction with this are a loss of liquid inventory out of the break, and boiling of liquid by core heat. The boiling process can produce steam and other gases.
Fuel cladding failures can result from LOCAs causing gases that were confined within the cladding to escape into the RCS. During a LOCA, the makeup and ECC systems will probably be activated, making it possible for gases in the borated water storaFe tank (BWST) or the makeup systems to reach the primary systen. If the LOCA is severa enough to cause the core flooding tanks (CETs) to discharge, the gas in these tanks could also enter the RCS. If the LOCA is large enouFh to cause a significant loss of primary system lig-uid, the pressurizer could drain causing the gases that were above the pres-surizer liquid to flow into the RCS. Finally, as the system depressurizes, the ability of the RCS liquid to retain dissolved gases would be diminiched, causing any gases that were in solutien to be released. 2.2. Total Prolonged Loss of OTSG Feedwater This is simply a loss of all heat removal from the primary system. This would result in RCS pressurization to the pressurizer relief valve setpoints (and could cause a small LOCA), and eventual saturation and boiling of the RCS lig- ' uid if heat removal is not restored. Gases could then accumulate in the RCS in t he same way as in a LOCA.
- 2. 3. Severe Overcooling The removal by the steam generators of more heat than the core is producing would cause a system contraction. This could be severe enough to cause the pressurizer to drain. Were this to happen, the steam and other gases original-ly in the upper portion of the pressurizer could be released into the rest of the RCS. The makeup system, in trying to maintain pressurizer level, could also inject gases.
I 2-2 Babcock s. Wilcox I
2.4. Inadvertent Cas Discharge I During plant operation or testing, gases could accidentally be injected into the primary system from the makeu;; system, the CFTs, or the BWST. '/or example, this could be caused by malfunction or improper operation of valves and pumps. When considering gases in the RCS it must be remembered that there are basically two types - condensibic and non-condensible. Steam is a condensible gas; it is normally within the cooling capability of the steam generators to suf ficiently cool steam and change it to water. Steam can affect RCS pressure control and circulation; however, because it is condensible, it can be dealt with by means other than removing it f rom the RCS (this will be discussed later). Other gases cannot be cooled suf ficiently by the generators to liquify them. Their volume within the RCS can be significantly reduced only by venting or forcing them back into solution. With present NSS designs, this could poten-tially be very difficult. The RCS geometry, which could allow gas pockets to form, the unavailability of vents in the present system, and the possibility of inoperative equipment in post-accident situations are factors which could restrict the options available
; to operators in these situations.
- I l
2-3 Babcock & Wilcox iI
i
- 3. Potential Gas Sources j A list of non-coi.densible gas sources is given in Table 2-1. Also shown are the types of gases that could be released into the RCS and the maximum gas vol- m i
4 umes each would be expected to contribute under the worst conditions, i.e., if each source released as much gas as pos.ible into the RCS.1,2 Note that for ) fuel fill gases, fission gases, and hydrogen (from cladding Zircaloy-steam / water reaction) the release rate is linearly dependent on how much of the fuel
- cladding has failed or reacted.
Note also that the gas generated by radiolyt'c decomposition of primary coolant j due to boiling in the core is given as 480 standard cubic feet (scf) per hour.2 1 Stated in simple terms, radiolytic decomposition of water results from core radiation causing dissociation of water molecules into various ions (112, 0,2 l 011, e t c . ) . This is an on-going process and normally (with the core covered with water) nearly all of the ions recombine immediately. During situations when part or all of the core is exposed to steam, the molecular density in these voided regions is much lower. This results in a reduced rate of recom-bination of the ions and can ultimately cause a net gas production (ions being produced f aster than they recombine). The resultant excess of gases could ac-cumulate within the RCS in large volumes if the core is partly or completely volded, or core boiling is taking place, for a long period of time. t I i 2' Babcock 8. Wilcox
Il
- 4. Potential Detrimental Effects of Cases in RCS Basically there are two harmiul ef fects which might result from the presence of gases within the RCS. The first is blockage of single-phase natural circu-lation. Figure 2-1 shows a system that is in single-phase natural circulation.
There are four requirements which must be fulfilled before this type of flow can be established and maintained within the RC5.
- 1. A heat source; this role is fulfilled by the core.
- 2. A heat sink; this is provided by the steam generators.
- 3. The heat sink thermal center must be higher than the heat source thermal c.cnter. With the reactor coolant pumps not running, steam generator secondary water level is at or is being raised to a height covering approximately the lower 20 feet of the tubes (50% on the operate range). This, in combination with the added cooling effect of emergency feedwater (EFW) which is being sprayed into the upper I tube region of the generators, fulfills this requirement.
I 4 There must be communication between the heat source and heat sink. This is provided in the case of single (liquid) phase natural cir-culation by the RCS coolant liquid which flows around the loop be-tween the core and the generators. Should enough gas enter the RCS, it is very likely that it would migrate to the highest points of the system - the hot leg 180* elbows, upper hot Icgs, and upper steam generators. Some gas would collect in the RV head, which would have no effect on natura. ion circulation unless the gases filled the upper RV down to the tops of the outlet nozzles. Then these gases could migrate to the tops of the hot legs and cause (or contribute to) flow blockages. When the 180* (U-bend) elbows becor2 gas filled, natural circulation flow would be stopped, interrupting the core to OTSG communication. Approximately 125 ft' of gas are required to completely occupy the hot leg U-bend.1 Once a U-bend I is gas filled, natural circulation in that loop stops (see Figure 2-2). If sufficient gas is present, both U-bends could be filled thereby stopping na6-ural circulation and core-to-0TSC communication altogether. 2-5 Babcock & Wilecx I
l The second potential effect of gases on the RCS could occar if the system is in the two-ohase (steam and water) circulation mode (boiler-condenser or re- ! flux cooling) . If this type of circulation exists, liquid is being boiled in W j the core while steam is being condensed at approximately the same rate in the l steam generator (s) (see Figs 3). If the condensate level is above the { bottom inside surface of the reactor coolant pump outlet pipe (the sp111over point), a sufficient driving head will exist to force condensate through the l pump suction piping, over the spillover point, and into the reactor vessel (RV). ! There is evidence that non-condeneible gases would be attracted to the. inside surf ace of the steam genert. tor tubes where steam condensation is taking place. Once there, some evidence indicates that the gas could act as a barrier to the l heat removal process which causes the steam condensation. If this is true, non-condensible gases in sufficient quantity within the RCS could actually shut off the boiler condenser process (Figure 2-4). I I I I I I I I 2-6 Babcock & Wilcox I
I I 5. Symptoms of Cases in RCS As discussed in the preceeding section, the presence of a large volume of gas within the RCS could halt natural circulation, either single (liquid) or two phase. There are certain associated symptoms which could indicate to plant I operators that 1 css of circulation has occurred. Single-phase natural circulation is characterized by hot and cold leg tempera-W tures (T g and Tcold) approximately 30 to SLr apart and steadily decreasing with time. Also, RCS pressure is slowly decreasing as the core heat output declines. If gases interrupt circulation (Figure 2-2) , RCS heat input to the generators decreases to essentially zero. T he w uld reflect this by becoming constant or increasing toward the saturation temperature for the existing RCS pressure. Steam generator pressures and temperatures could show a decreasing trend; there would be some leakage and ambient heat losses from the OTSG while RCS heat in-put is stopped. Also, if the OTSG steam-production rate is measurable, this might also show a decrease. T w uld decrease with steam generator tem-cold perature. Two-phase natural circulation requires time after the accident occurs before it can be established. For thase accidents which can evolve to this type of circulation, the RCS liquid level must have decreased sufficiently so that when the steam generator liquid level is at or is being raised to the appro-priate level this secondary level will be higher than the primary liquid level. The required level is 95% on the operate range, approximately 380 inches of the tubes covered, for the lowered.-loop plants. For the raised-loop plant, the required secondary level is 96 ine' - . Fulfillment of these conditions is characterized by a decrease in RCS pressure to approximately the steam gen-erator secondary pressure. - Should non-condensible gases accumulate in sufficisnt concentration in the re-I gion where condensation is taking place, they could block condensation alto-gether. If this happens, the operator would have observable symptoms that re-semble those for loss of single-phase natural circulation - decreasing steam generator pressure, temperature and steam production, RCS Tcold, a increasing hot
- I 2-7 Babcock & Wilcox I
I In r.ny case of reduced stean generator h:at removal whether due to gas-induced flow blockage or condensation blockage, an increase in RCS pressure would re-sult. This is a distinctive, observable symptom for the operator, whether the RCS had been in single- or two-phase circulation before the interruption oc-curred. Even if gas pockets within the PCS did not cause such flow interruptions, they j could make it difficult to control system pressures. This would be the case
- if the gas volumes were sufficiently large to overcome efforts to control RCS pressure with the pressarizer. The mechanisms available to the operator for controlling pressurizer pressure (heaters and sprays) could have only negli-j gible offect on gas pockets elsewhere in the RCS.
1 i Increasing RCS pressure or lack of pressure control slong with abnormal trends in primary temperatures, secondary temperatures, secondary pressures, and steam-g ing rates provide information to help the operator determine whether or nor 5
- large gas volumes exist within the RCS. Based on these indications, the opera-tor can then take action te remove the Fases from the RCS. Possible actions are described in the next section.
7
\
I I I I I I 2-8 Babcock & Wilcox I
I i 1
- 6. Removing or Otherwise Coping With Cases in RCS The options available to the operator for controlling the effects of gases within the system or for removing them, are varied and dependent on systems and hardware availability.
Operating plants now have (or will soon have) gas vents installed in the tops of the 180* hot leg c1 bows. In addition, vents will be installed in the RV head of some plants. These can help to relieve gases that might collect in the high points of the RCS. However, analyses have shown that their effective-ness is limited by their small (approximately 1-inch diameter) size. As long as boiling is taking place in the core, gases will probably be produced (steam and radiolytic decomposition products) at a faster rate than can be relieved by these vents. It is only after the core is sufficiently cooled by the ECCS -I or other means and boiling is stopped that these vents will be effective in venting gases and enhancing system refill. The relief valves in the pressurizer, primarily the electrically actuated re-lief valve iPCEV), provide another RCS vent. This would only be an effective outlet for gases that might collect in or be forced into the pressurizer. Tle reactor coolant pumps (RCPs) if operable in post-accident situations, pro-vide a very effective means of dealing with gases. Existing guidelines define the use of the RCPs, either by " bumping" or restarting. The intent of their use is to restore circulation (natural or forced). In doing this, gases in the RCS should either be swept through the steam generators and (in the case of steam) mondensed to liquid, or mixed with RCS liquid to eliminate gas pockets. The makeup and purification system can be used with or without RCPs running, to eliminate gas pockets. With inoperative RCPs this system would be acting to refill the RCS with liquid. With the RCPc operating or with a liquid-filled RCS, it rould simply be replacing liquid containing gases with liquid contain-E ing less gas volume.
- The combinat:*.on of steam generators, RCPs, vento, and system refill by the
'3 makeup /ECC systems, will be um.ful in eliminating gsses from the RCS. How many of tocse are available will determine the effe.. .iveness of operator ac-tion to correct the problem of gas accumulation. g 2-9 Babcock & Wilcox I
E It is possible that gas blockages may interrupt heat transfer to the OTSGs. ! This may happen If the above methods of restoring communication between the core and the OTSGs have failed or cannot be carried out because of (for ex-I ample) inoperative equipment. Should OTSC heat removal be stopped in this way, I the operator must rely on HPI (feed and bleed) coolinF via the pressurizer PORV l to remove core heat. This is the alternative cooling method to OTSG heat re-l moval and must be initiated quickly if OTSG heat removal is lost. It is a j viable, although time consuming, means of achieving cold shutdown. The opera-l tor should employ it until the gas blockage problem is brought under control { and OTSG cooling is restored. E I l I l - I' , I l l \ I I I I I I I I 2-10 Babcock & Wilcox I
l i l ^
- 7. Lesson Sirmnarv Briefly review the following:
- 1. Explain how gases can Fet into RCS.
- 2. Discuss the harmful effects they might cause once there.
i, l 3. Outline ways the operator can recognize Fas presence and gas-related effects. j 4. List possible actions chat are dependent on equipment availability. il i e i ll l l l !I i. il i lI J l I lI i ?E l t 8
'-11 Babcock & Wilcox I . . . . _ _ - . _ . - _ _ _ . . _ _ _ _ _ . _ _ _ _ _ . _ _ , . . . _ _ _
i 4
- 8. Peferences
- W. O. Parker (Duke Power Co.), Letter to H. R. Denton (NRC), October 17, 1979.
2 J . H. Hicks (B&W) to C. D. Morgan (B&W), Memorandue, " Post-Accident Radio-lytic Cas Generation," March 8, 1980. i i i
- I i 5
\ E i I I ! I l 3 1 I 1 i f E 2-12 Babcock & Wilcox
!E l Table 2-1. Approximate Gas potential Scurce introduced quantity, scf ,g Ig
- 1. Dissolved in reactor coolant 112 , N2 560
- 2. Pressurizer l
!g E Steam space H2, N2 140 Liquid space 112, N2 30 l
- 3. Fission gases l 100% failed fuel Kr. Xe 190 1% failed fuel Kr, Xe 1.9
{'
- 4. Fuel red fill gas (100% He, N 2 , 02 1,130 i
failed)
- 5. Zirc-steam / water reaction H2 420,000
; (100%)
- 6. Makeup system H,N2 2 800 i= 7. BWST Air (N 2 , 1,390 CO2, 0) 2
- se
- 8. CFTs (total discharge)
- Gas space h2 26,000 i Liquid space N2 960 j 9. Radiolytic decomposition 112 , 02 480 scf/h i of primary coolant
't 18 I \E I
i iI lI 2-13 l Babcock & Wilcox .I . - . . . _ . . . . , . . _ . .
__ - n + - - - I I I ( h r. ! !. !_ i
""" e1 l ...m x
/'[/(( '
l
- i. s NNN W ANNNNNNN
! i hmwwwwwa s i M E
- w%
l. a 3DDk g <
~
g_ i l I I I 2-14 Babcock & Wilcox
Figure 2-2. Natural Circulation Flow Interrupted by Cas Acetsnulation Occupying Entire flot Leg Elbows
= r
/ O
// 11 c E'"
; ,/
'; F ) PZR
, g g. f.
*/
// LEVEL RISItG i
- g. ..- ,
- / / /
I f(=
?.
fi/ i
/
/ =& ;SS.cruse kU INCEASIfC h
/
y[/ fh:* , '
/ *
/ * * .y i
b kd
/N ,_A, LIQU10 LEVEL
. .. '. INCREASIPC 10 esx o.R.
g / . g / -
... II x
x
$ . /~
w
Figure 2-3. Boiler-Condenser (Reflux) Circulation and Cooling f t lilll M 1 GAS FL0nr a s L PRIMARY STEAM l
- C0f0ENSATION N 5: q to
, cases i i s f ". 2 > 77sx 7 E *.
^
- r
' ses %
A
~ _
./ - * . , x=Jy /y. / /\x -
( , . ./ , n
/ /\N ,,,)/)
Y ' x ,
- x// N f-{ '
eata^av j/ '
- ='"' gg ki!E'41on ="at"""
c z w
S _ E S _ A g 7 g p n i l o C o r o o W
?
,=__E
_m n o i t sx\NN a l u c r 1 1 n//j i 1 C 1 1 o N d e t p e: u r r e t n I N r O e I S T E s A S n S A e m,gW 4 t G d E C l i B n o t e o O Y J O C D M K A C E O T L S D B E
/
/ _
~
4 _ 2 _ r,sj6< e _ r u g i S E F S A G ( _ 7 l ,v fx i
, : i
I I . Lesson 3 - BORON PRECIPITATION CONCERNS FOLLORING A LOCA I introduction
- 1. Lecturer -
- 2. Purposes -
- a. To explain mechanisms for obtaining high boron concentrations in the core following a LOCA.
- b. To discuss the dangers of high boron concentrations relative to long-term, post-LOCA core cocling,
- c. To discuss the interface between natural circulation and operator-induced circulation in preventing boron precipitation.
I d. To outline the long-term flow requirements to prevent boron precipi-tation. Obj ect ives The following material will be presented during this lesson: I 1. How the boron precipitation problem can arise.
- 2. The dangers of boron precipitation and how .hese could lead to a I degraded core situation.
- 3. Natural circulation and subsequent operator-initiated circulation to prevent boron precipitation.
- 4. Specific results of operator actions.
The key poit.ts to be retained are as follows:
- 1. How and why boron precipitation can become a problem following a LOCA.
- 2. How long the operator can wait to take action before the danger point is reached.
- 3. In general, whu actions (including NRC requirements) must be taken to manage the boron precipitation problem on a long-term basis.
- 4. There is nothing to indicate to the operator that boron precipitation is causing a problem. The best approach is to prevent rather than correct.
3-1 Babcock & Wilcox
4 5 I
- I Lesson 3 Outline
- 1. Introduction i 2. Mechanisms That Can Produce High Baron Concentrations l
- 3. Possible Severe Consequences of Boron Precipitation I
( 4. NRC Restrictions on Boron Concentration
- 5. Maintaining Boron Concentration Within Limits
)
- 6. Maintaining Minimum Required Flow Rate
{ 7. Operator Actions to Obtain Required Flow i t i i ! I i l I I l E B B I I 3-11 Babcock & Wilcox I
E I I Lesson 3 - BORON PRECIPITATION CONCERNS FOLLOWING A LOCA
- 1. a troduct ion During plant operatioa, certain upsets can result in high boron concentration I
levels developing in the core coolant fluid. Operators mu ;t have knowledge of the events and mechanisms that can lead to these high boron concentrations and of the dangers associated with such conditions. They cust also be famil- l iar with the required operator actions that must be taken to avoid the problems
'I of high boron levels and possible resultant boron precipitation. Each of these is explained in detail in the following text.
- 2. Mechanisms That Can Produce High Boron Concentrations This section discu.ae,some plant transients that can cause the core coa 1aN to reach a highly borated state. In addition, some of the possible mechanisms that can lead to this development are described.
Basically, these conditions can be caused by any plant transient that results in highly borated fluid being injected into the reactor coolant systerr (RCS) from the makeup system, the borated water storage tank (BWST), the core flood-ing tanks (CPIs), or the reactor building (RB) sump [via the low-pressure in-
-I jection (LPI) system). lossible transients that can cause this effect are less-of-coolant accidents (LOCAs), prolonged loss of steam generator feedwater 8 (LOW), or steam generator overcooling. In these types of accidents, highly borated water can ba introduced into the RCS in an effort to restore pritrary system liquid inventory, to help restore subcooling in the primary system, or both.
LOCAs are the most likely of these accidents to produce severe consequences. Figure 3-1 illustrates a LOCA in a cold leg. Figure 3-2 shows regions of dif ferent densities within the reacter vessel (RV). Alsc shown in Figure 3-2 are a postulated cold leg LOCA and the path of injection of highly borated I 3-1 Babcock & Wilcox I
E water (CFT, BWST, or RB sum,>) into the RCS. Another injection path, not shown in the figure, is froe the makeup and high-pressure injection (HPI) systems into the cold legs through the makeup lines, the HPI lines, or the reactor coolant (RC) pump seals. Along with this input of highly borated liquid, two other conditions must exist before excessive )>ron levels and boron precipitation become problems. One is the boiling of f of RCS coolant liquid by core fission or decay heat. The other, described below, is a flow rate through the core that is insuffi-cient to " flush out" or allow low-borated fluid to enter the core while high-borated fluid is scing swept or drawn eit of the core. n gure 3-3 shows low-density, highly borated fluid in ar< ab m e the core. ~~.e r fluid in the core region is being boiled away by core heat and is exiting the core as steam, while the boron that it brought into the core remains be-g hid. If the fluid above the core is all steam, it will contain no boron at 5 all. In the lower RV ad the downcomer, Figure 3-3 shews ~t accumulation of relatively low horon concentration, high-density liquid. If stagnant or near-ly stagnant flow conditicus exist within the RV, the high-density fluid may not reach the core rapidly enough to flush out the highly borated core fluid. The result could % an increase in the localized boron level in the core. The core coolant could become boron-saturated, and the excess boron could begin to f all out of solution Jn the form of boric acid (H 3 B0 3 ) crystals I W (Figure 3-4).
- 3. Possible Severe Consequences of Boron Precipitation This section explains the dangers associated with allowing the core coolant to become boron-saturated. Basically, the precipitation of boric acid crystals out of solution can lead to degradation of core cooling; there are two possible paths that this degradatior: can take. .
The first ,'ath is illustrated in Figures 3-5 and 3-6. Figure 3-5 is a cross-sectinnal sketch of a fuel pin uith the UO2 fuel pellets enclosed by and sepa-ratel f rom the Zircaloy-4 cladding by a gap which is filled with fill gases and fission gases. Under the normal ennditions illustrated in Figure 3-5, heat is transferred from the fuel through the gap to the cladding. An equal amount of he't is transferred through the cladding and is then removed f rom the cladding w I 3-2 Babcock & Wilcox I
I surface by the RCS coolant fluid. Cladding sarface temperatures are thus kept in equilibrit.m with the surrounding fluid temper :ure. 8 % If conditions were to develop that allowed a large amount of boric acid crys-tals to fall out of solution, this precipitate could potentially form a layer 8 on the outside cladding surface (see Figure 3-6). This layer could retard heat removal from the cladding to the coolant. The resultant heat transfer imbal-ance, more heat reaching the cladding from the fluid than is being removed to the coolant, could result in increased cladding temperatures. If allowed to continue uncorrected, the ultimate resu)'c could be cladding heatup to very dangerous levels. The other potentially dangerous aspect of boron precipitation is coolant flow blockage. The layering of precipitate on the clauding surface, as discussed above, could also significantly reduce or completely block the area available I for coolant flow along the fuel pins. This would, of course, aggravate the degraded heat removal problem discussed previously. Coolant flow reduction would also be possible if the precipitate were to f all to and collect on the lower core support structures or the lower RV flow distributor, as shown in Figure 3-7. In either case, the degraded heat removal could result in elevated cladding temperatures, possibly to the point of cladding tallure. And this could poten-tially happen with no advance warning to the operator. It then becomes impor-tant for plant operators to be knowledgeable about necessary preventive actions to keep this system within safe limits and what these limits are. These are discussed in the next section. They should also be aware of the types of sit-I uations that might require them to take such actions. Examples of these are given in section 2. I 4. NRC Restrictions on Eoron Concentration The question of how much boron concentration is a lowed in the coolant fluid I of the core region has essentially been answered by the NRC, who specified that the maximum allowable concentration is 4 wt % below the solubility limit of the coolant at its boiling point. Analyses nave shown that this translates into a concestration ratio (C/C ) of about 32:1.l* The ratio parameters C and'Cg are d.. lined in Figure 3-8. I 3-3 Babcock & Wilcox I
5
- 5. Maintaining Boron Concentratior Within Limits Basically, there are tw ways to maintain the boron concentration within '
limits. The first is to prevent the RCS coolant 1.iquid from boiling. However, in some situations, partla larly pcst-LOCA, it may be difficult or impossible for operators to puvent saturation (boiling) conditions f rom developing with-in the primary system. Inability to prevent boiling means that sufficient circulation through the core muct be established and maintained. This must be done to replace liquid as it is boiled away by the core (or lost for other reasons such as LOCA), in order to keep the core covered and to maintain long-term cooling. It is also neces-g e sary in order to flush out the highly borated coolant in the core region and to replace it with fluid of relatively lower boron content. This will enst e that the core coolant is being continuously diluted and not allowed to stagnate. Analyses have shown that a liquid flow rate of 40 gpm or more through the core is sufficient to maintain boron IcVels there within limits.l'2 This long-term flow rate must be maintained until the core is removed from the RV.
- 6. Maintaining Minimum Required Flow Rate Establishing and maintaining the necessary 40 gpm flow through the core re-cuires operator action and monitoring. These are discussed in this se-tion, along with the time constraints involved.
Scudies have shown that for a period of approximately 7 days following reactor shutdown, enough core heat remains to maintain natural circulation within the RV. If the reactor has been operating at full power for an extended period W before shutdown, the natural circulation period could last much longer. Figure 3-11 illustrates the results of one such study, conducted for B&W's 205-FA i plants. Results are expected to be similar for the operating 177-FA plants. This natural circulation could be both within the RV and around the loop for certain post-accident conditions, e.g., small LOCA with steam generator feed-water available. Liquid-phase natural circulation, shown in Figures 3-9 and 3-10, would be expected to continue for a long time. For a large LOCA in a hot leg, flow would simply pass from the injection point through the core in the I normal direction and exit through the break. For a large cold leg LOCA, circu-l lation would be primarily inside the RV, as shoen in Figure 3-9. For this ) 3-4 Babcock & Wilcox I'
I case, the flow is down the downcomer, up through the core, and through the vent valves to the break. During this 7-day-plus time frame, core flows will be sufficient to prevent boron buildup. As the core decay heat level decreases, the natural circulation flow inside 8 the RV will eventually decrease below 40 gpm. At that point, operators must take action to initiate forced flow through the core.
- 7. Operator Actions to Obtain Required Flow General actions for providing the required flow are discussed in Chapter 10, tne long-term cooling section, of references 1 and 2. These include the fol-lowing:
- 1. Taking suction from the decay heat drop line in the hot leg with the LPl. This should be done only if both LPI strings I are operable.
- 2. If mode 1 cannot be utilized, initiate gravity feed from the hot led through the decay heat drop line to the RB sump.
I Thie s(11 W work if the elevation of the drop line exceeds that of the bottom of the hot leg nozzles at any point.
- 3. If modes 1 and 2 cannot be used, initiate hot leg injection by either (a) injecting from the LP1 through the decay heat drop line, or (b) injecting through the pressurizer auxil-I inry sprays, through the surge line, and into the hot leg.
Ff gure 3-14 illusttraes taking suction, either gravity or forced, f rom the hot leg, via the decay heat drop line, to the sump. Sufficient liquid driving head exists between the downcomer (the bottom of the cold leg) and the bottom of the hot leg to provide adequate flow through the RV with the drop line in the bottom of the hot leg, as shown. This will be true as long as the down-comer liquid level is maintained by makeup or ECCS injection. Specific oper-I ator actions are, of course, plant-dependent. Current plant licensing agreements between B&W, the owners, and the NRC re-ur quire that the operators initiate forced flow within 24 hours af ter an acci-dent. This is well before the time when decreased natural circulation flow rates would actually require operator action. Figures 3-12 and 3-13 show that, for two different initial core power levels, beginning the operator-induced circulation at 24 hours into the accident will ensure that maximum concentration levels will be kept well below the NRC limits. This is the B 3-5 Babcock & Wilcox lI
E basis for concluding that bcron precipitation will not be a problem if re-quired actions are taken within the specifie,i time.
- 8. Summary I,
- 1. Discuss how boron concentrat Mn in the core can become too g great. 3
- 2. Describe possible consequences if this happens (consequences may not be readily detectable by the operator).
- 3. Show that natural circulation is a short-term preventive mechanism.
- 4. State that eventual operator-induced circulation is required.
- 5. Describe generally how to obtain the 40 gpm required flow.
- 6. Stress that the NRC requires that 40 gpm flow be establisned within 24 hours and that it must be maintained until the l
' core is removed from the RV. W t References 1 ECCS Analysis of B&W's 177-FA Lowered-Loop NSS, BAW-10103, Rev. 3, Babcock i
& Wilcox, July 1977.
2 ECCS Evaluation of B&W's 205-FA NSS, BAW-iO102. Rev. 2 Babcock 6 Wilcox, December 1975. j I I' i l i I II 3-6 Babcock & Wilcox I
. - - - a.- . _ _ _ _ .w,. m . . _ ,
,i t
- I '
't 'G" I 25 I m u '. u 3 y 18 s I .s mmw=m , a I - 1\4 t E g
\ \ \ \ ','s\ \
2 s ,t i a
. . a.. w -~ -
3 s- - j s u a R 5
,]*5 b xxu..u3
- l u i i N 4 -
. N 3
u mmm'm l Ih 1 l L\m&L y lI lg : s I 3-7 Babcock & Wilcox j
~ . . _ _ _ _ _ _ . . . _ _ - - _ _ _ _ _ _ _ _ - _ . _ _ . _ . . _ _ _ . _ . _ . _ _ _ . _ . _ . - _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ . _ . _ - _ . . _ _ . _ . . _ _ _ .
E Figure 3-2. Cold Leg LOCA With ECCS Ir.jection I i
TTTTTs g
. .. g STEAM, , e l LIQUID-STEAM *
) MIXTURE, OR f ,
g! , j LOWER DENSITY y, , Ij ,,
+'
y VENT VALVES h LIQUID %I + 1 -- r j ,i .i;. In . ECCS [h4
- l
. , y..
l
' \
- l . y, .- # BREAK l s .
, g
! AN..
~
+ .-
h; g - , ., (k
+* N .".
j 4 b f g. i , 't HIGHER g
\
, <f DENSITY LIQUID g ! , yjp
' gl' hs '
$ I IlB,K s
s-l- - i I- <- a g
$N a.t . +
N, )K
- l I cvovw<q, L
.r. = 5
/
l / r ws g I I 3-8 Babcock & Wilcox I
il !I Figure 3-3. Areas of Different Boron Concentration i !l '
T T T T T ' " . .
ig M t % is 1 l .ut !g , Uhl!Riisic' ' is i P_-
,r ,
SW* Wl1 . W (NO BORON) lE!
- l P 60 T
5 5, V, \, ' l ,
}\1\%NY(
s HIGH ' RON BORON - ,,. , ,Q CONCENTPATION CONCENTRATION b l5 l N,,i'i;- ,
. 4[ :
Q k % h4 '* 4 ffrY!I - w .e .
~
I I I l 3-9 Babcock & Wilcox l
3 Figure 3-4. Liquid in Core Is Boron-Scurated,
; Boric Acid Crystals Precipitate
! T T T T T ' '. ! l i M r s 3 1 5 I lll
@ h h $!!I l l
- l. ~! '. ! ' '-
l %
._- i j i -.
i [! i p
'* l I pl e STEAM l
l'l l l ! l ,s' l (NO EDRON) li i
~ > s ; ; I
', ;I 5 l )p f Q.' - -=
b b en/n/a trV ,/ '
/ 4; i
't lf .
s - , j :fLafj g 6
? , ~T T d 8 f,,-
,/ /s- /
. 7-J l
,/
El i
%*l, l
8 I, ' 3-10 Babcock & Wilcox -
)
l l
Figure 3-5. Fuel Rod Cross Section and Nomal j licat Transfer ; l fuel to cladding " cladding to coolant 3 i
- , , ~
\ '
0 FUEL \ Q CLAD i TO TO COOLANT i
/ l N
N l I / Zr4 CLADDING
\ 0 U02 N P LLETS
/
,E N ' N COOLANT Flul0
\ \
5 /
- y N '
N l5 N N l i / k '
/ \
N l lI l I >I
- 3-11 Babcock & Wilcox 1
S Figure 3-6. Fuel Rod Cross Section With Boron Precipitate Plating Onto Cladding Surface and Degrading Heat Transfer to the Coolant fuel to cladding cladding to coolant
,,,/ _
0PUEL \ O CLAD TO CLAD TO COOLANT N i - & - H 803 3
\ CRYSTAL PLATING If4 P N \ ,
3 l CLADDING %
\ GAP UO2 \ [
PELLETS
\ \ g
\ N I
, g' N
N . 8; L\ N ; g
- r I
I 3'2 Babcock & Wilcox I
l8 Figure 3-7. HsB03 Crystal Precipitate Fallo to l Lower Core and Reactor Vessel I ' T T T T ' e. Il M % ll - gg f' 'l
- i.i .-~. W M P
(/AV///A>
/ d-la i 4 l3 4
- i , -
M,, 1 ls i ! - \ ! ,! i [ gl 4 l l T , X l ,$ s'l ' .' x, 'i ', 2 SUP ORTS s 5 lg h ' FLOW l DISTRIBUTORS l \ 8 5 3-13 Babcock & Wilcox I._ .. _- . . . . _ _ _ - _ _- -__ . l
8 Figure 3-8. NRC Restrictions on Concer.tration C: O BORON CONCENTRATION IN BWST (APPROX 2270 PPM) l 1' C: TIE DEPENDENT BORON CONCENTRATION IN CORE 3 NRC LIMIT: C:C 0 LESS THAN 32:1 h I THIS IS SAFELY BELOW THE SOLUBILITY LIMIT OF THE REACTOR COOLANT FLUID AT ITS BOILING POINT. 8
- e t,
E 8
- 8
- 8 I
i I 3-14 Babcock & Wilcox I
(;
?
Figure 3-9. Inside the Reactor Vessel - Natural Circulation for Large LOCA
, ' T T T T T ' e.
's s . .
STEAM, -y I 1 A > - VEW WATER-STEAM l ' ' i ' VALVES
.. I ~i 4 MIX , OR LOWER - ', !
--ECCS I
DENSITY WATER- g' i (WITH HIGH --
\ e' ' \
- BREAK N NTRATION) 2,. .
j . !\ A \ N h 1[ g h d-h, 1 LIQUID-LOWER BORON E N' CONCENTRATION N , I F; Q I
- N' '! y N1 d '; hN rc E=,
' E}-VWil5 s_ _m__ rmn_icf- _
h( ) 3-15 Babcock & Wilcox
l l l Figure .-10. Natural Circulation Around the Loop (SBLOCA) LOWER
/ DENSITY - ;
FLUID
<}.
j
/ t
[v t Q / , d Q
/ , -
/ <
~
py L Puff j 1 p
/ J
' i :,
0FF j / t - FF j 91 /
, t 4 '
m f 19
/
/ \
' ' ' ' 'g i
K/g 5/%
'r y j M,' \ / N - ---(([gM//$ N/N #
r x, u 5
, \/\ ,,
a l
,e r \/\ /
L1 1 1 -1 zb %f\ , ff] ,,
\ /
\/N 7 -
N/s
- I' s
, 'N , a I , ss) HIGHER f)j DENSITY FLUID
?
er n-
" " W W # # W W W W W W W W W W es W m
l M M M E O O E M M M M M M M M I Figure 3-11. Duration of Natural Cir:ulation , 1000.0 - 4 i i i t
$ 100.0 -
j N C INFINITE OPERATION w 1 5 20 FULL REACTOR
~ -
POWER OAYS 10.0 - END OF NATURAL lENDOFNATURAL CIRCULATION l CIRCULATION l9.26OAYS l763.8OAYS ! l l l l 4 e g i.0 I I I j h 1 10 100 1000 10.000 e I I Time, nrs. J g
- E i
i
l 100 Figure 3-12. Boron Concentratic.. Vs . time: 40 gpm Dump-to-Sump at 24 Ilours l MUM SOL E llTY LIMIT 32.0 l l O L E g 10.0 - Y ; I ; 3 u 5 I MAXIdVM C/Co = 1.463 20 FULL REACTOR POWER l AT 7.71 DAYS DAYS l a ~ i R - l N l l0 i i i
~
100 Iric] 10.000 x Time, nrs W W W -_ W W W W W W W W ww _ _ __ aea g g
il 0 0
, 0 I 0 0
1 6 3 5 1 -
=
S o Y C A
/ O C
7 M . U 6 M 6 s I 7 r X u A T o M A H ll l 1l I I 4 2 S t a p m u
/ I 0
0 0 0 o 1 t p m N u O D I T m A p R g E P s 0 O r 4 n R
- E e W e m O m i P T 0 i
. T 2 E s 3 I
V - I T N n I o I M F i I N t I a L r Y t T n 0 e I L I 0 0 c n 1 8 1 o U C L n O o S r M o U B M I X 3 A 1 M 3 e r u g i F
- 0 0
1 3 0 0 1 0 1 1 8L 2%x g5ZU8a S FX87 $ax ll
3l I l Figure 3-14 Suggested Long-Term Cooling Method With Minimum h ur l Required Circulation Through Decay Heat Drop Line i
- TTTTT.. ;
# #**[% 1 8
i s!~ ds [s Jy* ~ h 3 dl@ @;h l k: D I I + INJECTION h! i* \f i b i l -_--_ _- 1){(_*=_.- _ _= - -{I i 1l-^ ig
-- ~- -- - -
BREAK A ..2-- o i 4..
,~ -
~~ T h b
.....c$
j .
" i Q ! ;&
TO SUMP P
% \N ' j![ $
'/ ?
! $ N /.I 'sv i
y d i m . iL< e , , , i i f - j/ \ i t,, / I ,,ja%-
- . t g .
/ -
;~
m g TrTrTv m r1 1 1 1
! H H F-- HHW -
5
'J E F P P U ER] 0W
~
# \
A . l - I
%A i A A
l l 8 I 3-20 Babcock & Wilcox I
3
- I lI
'h' Lesson 4 - EQUIPMENT FAILURE SEQUENCES THAT COULD LEAD TO A DECRADED CORE Introduction
- 1. Lecturer -
- 2. Purpose - To describe mechanistic equipment failures that could, if un-corrected, lead to a degraded core condition. Alternate success paths are discussed where applicable when normal components or systems are not available.
Objectives The following material will be presented:
- 1. Event sequences that lead to inadequate core cooling.
I 2. Casual failures of certain event sequences. l3 3. Discussion of the more probable of these highly unlikely g evente. Key points to be retained are as follows: .I ! 1. Inadequate core cooling, while unlikely, can happen. The operator should be convinced of the possibilitv. l
- 2. Several sequences leading to ICC 0 e possible.
- 3 l8
- 3 I
I 4-1 Babcock & Wilcox
8 8 'I Lesson 4 - EQJIPMENT FAILURE SEQUENCES THAT 8 COLLD uEAD TO A DEGRADED CORE ,I
- 1. Funicional Failures
! A degraded core could evolve from several vulnerable plant conditions. Such conditions include anall LOCAs (including a stuck-open pressurizer safety ,E "" "" ""d "' """" "*"' '""'"'*) ' "" ' " '"" """' "*"' ""' "" ' ""*'"
= power. In all cases, additional failures are required to cause inadequate core cooling. Any event path leading to a degraded core has a low probability of occurrence. Before addressing combinations of equipment failure and time frames, it is important to note failures in system functions - for example, a small LOCA with a loss of the HPI function could lead to a degraded core con-dicion. Similarly, loss of the secondary side cooling function with loes of l the HPI function or tripping of RC pumps during a small break while the RC void f raction is 70% or greater could also lead to a t'.egraded core condition.
Loss of any system function can result from combinations of independent equip-ment failures, operator errors, or common-cause failures. Corbinacions of in-dependent equipment failure in different systems, e.g., MFW pumps, auxiliary 1
- feed pumps, and HPI pumps, are a remote possibility. It is the incent of these i lesson plans and ATOG to lessen the probability of operator error that would propogate a degraded core condition (e.g., turning off HPI and auxiliary FW),
leaving common-cause failures as the most probable (but still very small) path to inadequate core cooling conditions. These include hardware ccamon causes and the operator common cause (which is the result of conflicting information being displayed to the operator). Lesson 5 describes recognition and proce-l 8 dures to employ for these conditions. I I I l 4-1 Babcock & Wilcox I
8
- 2. Scenarios 2.1. Natural Catastrophic, Sabotage, and ATWS Scenarios These sequences are beyond the immediate scope of these lesson plans although operator corrective actions can be taken to mitigate their effects. The ini-tiatirg event itself is outside the control of the operator .at depending on the resulting effects of the initiating event, some portions of the ATOG guide-lines are applicable. For example, if the initiating event caused a loss of secondary heat removal, operator actions would coincide with guidelines for loss of all feedwater. It should be noted that ATWS sequences are manageable if secondary heat amoval is available and if peak ATWS pressure does not re-E sult in primary system rupture (if it ruptures, it is a LOCA condition). B&W W plants also have a manual trip and runback capability in the event of failure of the RPS breakers. HPI boration should also be attempted as soon as possi-ble (the initial primary system pressure scike may be too large for the HPI pa..pc., to deliver water, but they should be flowing as pressure drops). !-TPI initiation wi.11 be by manual operator action. To summarize, if the RPS fails to drop the rods, try to run the rods in manually while initiating HPI flow.
g It should be noted that this course of action is consistent with ATOG proce- W dures in that HPI flow should be established an loss of subcooling margin. 2.2. Scenarios Terminated by Procedural Action Most of the corrective actions for the INPO vulnerable plant conditions have been accounted for in the first lesson plan. This section addresses only those items not discussed in Lesson 1. Vulnerable plant conditions identified by INPO include the following:
- 1. Loss of offsite power while one onsite power source is out of ser-vice. Loss of offsite power results in loss of main feedwater. If the diesel supplying the motor-driven emergency feed pump is the one that is out of service, secondary side heat removal is dependent i g;
1 on the turbine-driven pump (unless condensate / fire pumps, etc. can be aligned und are supplied by the good diesel). If the HPI that W , is aligned to the good diesel is out of service (e.g. , down for i maintenance), the HP1 function is lost unless the standby HPI pump I 4-2 Babcock & Wilcox E
I I is aligned (or can be through a swing arrangement *) to tha good diesel, l'nless these mubiple equipment failures (feedwater and I HPI pumps) occur, normal procedural action terminates this event sequence. If the multiple failures do occur, this event would be classified under section 3 (common cause and causal f ailures).
- 2. Extended station blackout. If the emergency turbine feed pump is available, secondary side cooling is initiated and the event is within the scope of the guidelinet. However, see section 3 for a discussion of vulnerability while operating in this mode.
- 3. Stuck-open pressurizer safety valve. This is a small LOCA, and guidelines are in place; however, depressurization to decay heat removal pressure is encouraged before the BWST is depleted.
- 4. BWR event. In its place we will consider a stuck-open steam safety valve or steam line break. The affected steam generator should be isolated by stopping feed flow, and cooldown should be initiated I under the "one good OTSG" guideline.
.E '- ' "" '" '""' """' "*"' ' " "'"* """"' """" "- ' ' ' "" = of the secondary heat removal function, employ HPI cooling under Lesson 1 (part 5).
I 6. Loss of d-c control power to a 4160 V emergency safeguard bus. If this should occur, the diesel supplying the good 4160 V bus should be started and left running until the situation is rectified. If the diesel fails to start or fails to continue to run, a shutdown should begin. A loss of of fsit, nower during this t t te is unlikely, but in case it should occur, a station blackout results as discussed in item 2 above.
- 7. Loss of automatic emergency turbine feedwater control. If the motor-driven emergency feedwater pump (s) are not available, then manual control of the turbine pump is necessary. Two methods of control I __ __
*However, remember that if the third HPI pump is a swing pump, all supporting = subsystems, such as cooling water, should also be aligned to some power ser-vice.
I 4-3 Babcock & Wilcox I
IIi are possible: (1) manually start /stop pump based on steam genera-tor level, or (2) continue to run the pump but regulate flow to the steam generator by varying the amount of recirculation to the con-denser. The second method is preferable (as long as the bearing temperature on the turbine and pump are maintained in the safe re-gion) due to the relatively large probability of failt.re of the cur-bine pump to start on demand (e.g., overspeed trips while coming up to load) and the fact that even small flow to the generators has a quenching (cooling) effect. 2.3. Scenarbs Unaffected by Procedural Action There are scenarios for which r.o guidelines are written. In shart, these are the ones in which nothing operates properly, so the operator has little if any-thing to do concerning the primary system condition. No guidelines address E W Oe actions to be taken in the event of losses in the secondary heat transfer and high-pressure injection functions, or in fact for any combination of losses of functions that would prohibit eventual dep essurization of the primary sys-tem. Under these conditions, the operator could adjust tne containment func-tion if it were operable. The reason guidelines are not written for these "e mrything fails" scenarios is that their probability is remote. The most likely of these improbabic scenarios are the common-cause sequences, such as loss of all a-c poner, and those in which the operator fai:s to act (analogous g to "everything fails") or acta incorrectly due to inadequate feedback of plant J conditions. The next section discusses the loss of all a-c power event se-quence viewed from a common-cause/ causal viewpoint. Some simplified event tree diagrams follow in section 4. 8 I B B I 4-4 Babcock & Wilcox I
I I 3. Common-Cause and Causal Failures A loss of power is a common cause event that affects equipment in more than I one system. There are three sources of high-voltage a-c power: (1) in-house, (2) offsite, and (3) emergency onsite. The pre-TMI setpoint (on PORV and high pressure trip) for operation of B&W plants allowed for a possible runback of station power to house load in the event of load rejection. With the post-TMI setpoints, this capability is no longer available. Therefore, if offsite power is lost, on!y emergency onsite power is available. Most plants have two diesel generators for emergency onsite power. Safety systems, such as high-pressure injection, are loaded on their respective diesels (train A on diesel A, train B on diesel B), minimizing the consequences of failure of one diesel 8 to start. If both diesels fail to start, only the steam-driven auxiliary feed pump is avullable as an active source for heat removal. In the event that flow blockages (e.g. , closed valves, pump f ailure, or inadequate steam supply to drive the turbine pump) exist, then this source of heat removal is also lost. It should be noted that after a loss-of-feedwater event there is a lim-ited amount of steam inventory (even if there are no stuck-open steam or at-mospheric dump valves, which could be caused by loss of control p)wer), so it I is imperative that the auxiliary feed pump be started (if it has failed, for example, on overspeed trip) as soon as possible to supply feedwater not only for cooling purposes but also for steam generation to diive the pump. In the loss of C 1 a-c power scenario, there is a limited amount of time before the d-c power source gives out. In addition, the loss of ventilation / air condi-tioning will eventually result in erroneous signals being displayed for opera-tor information or used in automatic steam generator level control, even if d-c power is still available. This introduces additional event sequence paths that can Icad to a degraded core condition. Other connon-cause initiating events include seismic events *, the effects of which are discussed in more detail in Lesson 5. Before the plant modifications I on the separation of NNI-X and -Y, loss of power to either NNI would have been a
- Note that seismic is not classified as naturH catasttophic (section 2.1) ;
because the more probable seismic events are the less severe or mild-amplitude l occurrences which will not cause massive equipment damaga but would generate l 8 conflictine information to the operator. l 1 I 4-5 Babcock & Wilcox I
- I 4
potentially serious initiating event for degraded core conditions due to the I j loss of indication or conflicting information to the operator. Now, however, j with this separation -- coupled with knowledge of where the instrumentation g
; power is supplied from - the probability of this event sequence is minimized, s
)
i l I l i J 1 5 l 8 i 1 i I l l I I I. I l I I 4-6 Bcbcock & Wilcox I' I ,
I I 4. Event Trees The first event tree is taken from the IREP Crystal River Study (Figure 4-1). An "up" branch on an event tree means system success, and a "down" brat ch means failure of that system. The systems are specified in colums at the top. The "Results" column is gven in terms of S (safe shutdown), L (LOCA condi-tion), or CM (core melt). Note that this event tree is not correct (e . g . , 12, 13, 14 should be L. not CM, as is also the case with the cms in te ATWS branches), but it is a good graphic representation of an event tree. Figure 4-2 is an example of a Eimplified et ant tree. A probability 1s assigned to each branch of the tree, so the outcome of any particular path is multiplication of the individual branches. If, for example, the probability of each failure branch were 0.01, then T*MFW*HPI*AFW = probability of T x 1x 10-' or T times 1 in a million. However, if we investigate common-mode / common-cause failures, such as loss of !I all a-r power (Figure 4-3), we see that these events have much smaller event tree branches (if we lose all a-c power, we have lost the HPI pumps and the motor-driven auxiliary feed pump). B&W plants now have no a-c dependencies associated with the auxiliarf feed pump turbines (which was not the case before the reliability study of late 1979). However, there are still some connect, ions between loss of all a-c pow-er and steam availability for the turbine feed pump which are site-specific. Many other event trees can be drawn - for example, Figure 4-4. However, these involve either multiple equipment or operator failures and are more remote than the common-cause/ causal plant sequences. In summary, three general event scenarios lead to a degraded core cooling con-dition: (1) multiple (random) equipment failures, (2) multiple (random) op-erator errors, and (3) common-mode / common-cause scenarios. The most probable category is common cause scenarios; both (1) and (2) are less probable. Com- ; mon-cause scenarios include both equipment (e.g., from loss of all a-c) and I operator (e.g., from conflicting aforeation displayed) paths. Both of these l l event paths are important. Conflicting information to the operator is normally initiated by an abnormal plant condition (e.g., seismic event, loss of ventila-tion and air-conditioning, etc.). While a degraded core cooling condition is I 4-7 Babcock & Wilcox I
- l. l an improbable event, the operator should be aware that it is possibic, and he l
W
! should be cognizant of the various scenarios that could bring the plant to such conditions.
4 ! s' I
- 5. Lesson Summarv Briefly review the various failures that can lead to a degraded core condi-tion. Review insights on the most probable of these failures.
l i I i ) I i I i l 8 l t l i l .t 1 I I I l l Babcock & Wilcox I' 4-8 I'
I 1 i Figure 4-1. Crystal River IREP Event Tree
,I c.n.m . m. s , m.o . g;'- ,,n, . 1i. ,,i i. ,c,... . .
Me' ill &' N A .f t P0(15ust it'Ilie:TT essellP .3tK 'see et0gief ee9
- enesesCTier71 g gggg ages,gg stLIEf ageg ?
, , , , e P, 0 6 e e
$ e I I I
Det eceot ees L.ros
. -- s i st.e.
e s 2 P.
*G m.---.--an-==== L 3 e.
-I 7t 8 e e p a s
-g .
?
.I -
e
> e W 9 l =D
- 'T a ie
-. -.-.- ==m- m - ===
Sutt t 19
- g gi CN II f .
e DB $6 9 f e stgeg P g I IS e
-- - - - amme -- m= L e4 e
e Cis 07 6 --m -- Fe . ,, r (Is 19 9 S 30 I 'f . O
- .m. -um amm -- a s ex- m.au L 21 i *
& I2
,I g
T2 On 23 e th It I e a a Os 76 7 Os 24 On 11 g(v #se ellet t totgue Os 20 O P 3
- Sa*8 (0#9 fies 2 .,
I L
- LM A Sl '14 tff
.....e,.
e O, = I 4-9 Babcock & Wilcox
.I
Figure 4-2. Simplified Event Tree MFW g3 TRANSIENT HPI o3 MFW AFW gg l HPI ALTERNA7E SECONDARY WATER g3 AFW SOURCES AVAILABLE i O l ASWSA O(CC) IF THIS CONDITION EXISTS FOR LONGER THAN l 20 MINUTES A { DEGRADED CORE CONDITION j y l 7 RESULTS s l W E [_ en an e as eme en m me e em m me m as as es em
Figure 4-3. Simplified Loop Event Trees POWER FROM DIESELS SUPPLIED ,3 TO LOADS DIESELS TRANSFORMER AND CIRCUIT AUX FEED TURBINE g TRANSIENT IS LOOP BREAKER FAULTS PUMP DIESELS AUX FEED TURBINE PUMP T C DIESELS ,3 LOOP AUX FEED PUMP, g ALTERNATE REPRESENTATION DIESELS I AUX FEED PUMP
' (DC) . I I k DIESEL ELECTRICAL SUPPLY FAULTS SYSTEM FAULTS sw
Figure 4-4. Example of Other Trees aS i ( LPI OPERATOR FOLLOWS DEPRESSURIZAiION 3S l . GUIDEuINES HPI LPI , FEEDWATER .DEPRESSURIZATION 3 l AVAILABLE OK RATOR FAILS TO FOLLOW GUIDELINES = OVERCOOLING 33 EVENT OPE'@ TOR FOLbMS , 33 FEEDWATER GUIDELIPP.S I , i UNAVAILABLE o OPERATOR FAILS TO FOLLOW GUIDELINES
?
X O 7:" e a ! x l l E E E E E E E M M W W W W W W W W W g
I I I Lesson 5 - AVOIDING DEGRADED CORE CONDITIONS Introduction
- 1. Lecturer -
- 2. Purpose - To discuss recognition of ups.ets in the decay heat removal process that are significant enough to lead to degraded core conditions and to provide instructions for correcting those upsets.
Objectives The following material will be presented during this lesson:
- 1. Use of incore thermocouples during abnormal transients.
- 2. Recognition of reflux boiling.
- 3. Recovery from loss of natural circulation.
I 4. LOCAs that cause a loss of heat sink.
- 5. Recovery strategy for multiple-failure events.
Key points to be retained are as follows:
- 1. An understanding of plant dynamics under adverse conditions.
- 2. The preferred methods of operation to relieve these conditions.
I I i I I l 8 I 5-i Babcock & Wilcox l I
l l l l l Lesson 5 - AVOIDING DEGRADED CORE CONDITIONS
- 1. Thermocouple Indiestion l
The incore thermocouples can be used for a variety of functions: I 1. They are used to detect core uncovering. They are the most valid indication of core cooling. If the incore thermocouples clearly indicate superheated conditions, then actions to counter inadequate I core cooling should be taken.
- 2. They provide a backup indication of natural circulation. If Thot I and Tcold do not show natural circulation, then the incore thermo-couples can be used co check for this condition. Thot should read within 10F of the incore thermocouples when the plant is subcooled and solid-water natural circulation is occurring.
- 3. They provide an indication of NDT margin when no circulation (forced or natural) exists. The five highest thermocouple readings I displayed should be averaged, and that average should be used to control HPI flow to keep the RCS within NDT (thermal shock) limits.
I 4. The thermocouples are the only valid indication of core outlet con-ditions when no circulation exists. l I I I l I I 5-1 Babcock & Wilcox
I
- 2. Recognitien and Treatment of RCS Heat Removal Problems "
2.1. Loss of Natural Circulation When the RC pumps are of f, heat j a removed from the reactor core by natural circulation. Some accidents can lead to a loss of natural circulation, but methods exist to restore it. This section highL*.ghts the recovery measures and gives an understanding of why certain actions are recommended and when g they are to be taken. W Loss of natural circulation has two symptoms, depending on the cause. First, if the thermal center in the steam generator drops (level too low or even a 3
- hot will g int a continuous increase as decay heat is added to the water. In the second case flow is blocked, by either a steam bubbic or non-con Jensible gases in the top of the hot legs. This time Thot "i11 #***i constant (or slowly increase toward saturated temperature), but T g will drop below T in the steam generator as seal injection water is added to the cold sat legs. In either case, the result is a large AT (more than 50F) and a loss of heat transfer across the steam generator. If there is any doubt in the opera-tor's mind, he sliould check the incore thermocouple readings. An average of g
five thermocouples tracking closely with T hot will verify that there is natu- W ral circulation flow. Corrective actions for function 1 are standard procedures for the restoration of feedwater. Steam void formation (function 2) is more complicated because the reactor coolant system (RCS) can operate differently depending on what has happened. The two principal accident types that lead to steam void formation are overcooling transients and loss-of-coolant accidents (LOCAs). For these transients, voids are formed in the following manner: Overcooling Too r:uch primary-to-secondary heat transfer causes a drop in l RCS temperature, which causes a contraction in iluid inventory,
- a Acrease in reactor coolant pressure, and a loss of pres 7urizer liquid. Some of the stcam in the pressurizer flows into the RC piping and collects in the hot legs. Because the RC pressure g
g drops, some of the coolant may flash and cause void formation in the hot legs. LOCA A LOCA results in a loss of RC inventory and reduced RC prescure. W Voids are formed directly as a result of the loss of RC invento-ry and also because of flashing of the reactor coolant as the g RC pressure drops. The RC temperature drops less than it would for an overcooling event. g I 5-2 Babcock & Wilcox I
I I Figure 5-1 illustrates the buildup of steam voids and the formation of a steam bubble in the upper ho' leg piping. IIPI should be started whenever the RCS is saturated. The size of the steam bubble will depend on the rate of system overcooling or loss of inventory I versus the rate at which IIPI adds water to the RCS to refill it. If llPI is large compared to the other factors, no steam bubble will form at all and
; natural circulation will not be lost.
2.2. Reflux Boiling If a steam bubble does fc,rm, its size has a direct ef fect on primary-to-secondary heat transfer. If the bubble is large enough, such that the hot leg level is at or below the secondary side feedwater level, then steam can be condensed within the steam generator and they can still remove a large amount of decay heat. This boiling mode of natural circulation (reflux) is illustrated in Figure 5-2. Reflux boiling is an expected small break LOCA I condition. For thts condition, it is important to ensure that (1) the SG 1evel is at 95% on the operating range (to allow the condensed reactor coolant to flow over the cold leg pump elevation and into the core), and (2) IF' is on at a high capacity (two pumps). A check of containment pressure and tempera-ture conditions should also be made to confirm that the cause is a 1,0CA. If LOCA conditions are indicated, and immediate plant cooldown at as close to design rates as possible should be initiated. The P-T diagram should be moni-tored to determine whether subcooled natural circulation returns or reflux boiling is lost. 2.3. Syntem Refill by llPI If natural circulation has been lost and steam cannot be concensed in the steam generators (that is, the steam bubble is in the top of the candy cane but not low enough to be in the steam generator), the RCS will repressurize. I This mode of operation will be indicated (see Figure 5-3) by saturated hot leg conditions with reactor coolant pressure higher than the steam generator pressure (SG pressure may be dropping due to lack of primary to secondary heat transfer). Several exampics of accident conditions that could fall into this mode of operation are as follows:
- 1. Small LOCAs where liPI can match the Icak rate (at reduced RCS pres-sures) and refill the RCS.
I 5-3 Babcock & Wilcox I -
I
- 2. A severe overcooling ever.t (e.g. , major steam line break) in com-bination with delayed actuation of HPI.
- 3. A total lors of feedwater, where EFW is restarted after the RCS is in a highly voided condition, and the HPI is refilling the RCS.
As indicated tu Figure 5-3, the same actions identified for reflux boiling apply to this operating mode. In this mode, the HPI is refilling the RCS. During refill, steam in t.e upper region of tha hot leg piping will be com-pressed as the water level in the loops and stenn generator rises. In some cases (i.e., low decay heat with all HPI pumps on) subcooling and natural cir-g culation will occur with minor increases in RC pressure. Under other circum- E stances, it may be difficult to fully compress the steam in the hot leg and restore natural circule. ion. Figure 5-3 shows the actions required to re-start natural circulation if the steam generator can be used as a heat sink and the RC pumps are available for restart. If the RC pumps are available, pump bumps (short run times of 10 seconds' du-ration) are allowed. This momentary use of forced circulation tries to force reactor coolant steam condensation by mixing it with liquid reactor coolant and by moving the steam into the generators where it can condense. Use of the PORV to limit RCS pressure rise and to increase HPI flow is also allowed (sepa-rately or in conjunction with RC pump operat. ion). To be effective, the steam gu.erator must be a heat sink; the selected steam generator saturation temper-ature (507) less than the incore thermocouple temperature) is somewhat arbi-trary; it was chosen to ensure a strong temperature gradient for condensation. When the pumps are bumped and steam is condensed, the RCS pressure will drop considerably (around 500 psi); HPI flow vill increase to help refill the voids. If natural circulation starts, the RC pressure will stay low; if not, it will repressurize and another bump can be used about 15 minutes later. 2.4. LOCA With Equilibrium Pressure Below Steam Generator Pressure For LOCAs of a certain size that depressurize the RCS below steam generator i pressure before it settles to an equilibrium with HPI (HPI starts automatically because this size LOCA will drop pressure below the ESFAS setpoint), the opera-tor should lower the steam generator pressure (using :he TBS or LDVs) until the saturation temperature on the secondary side is 50F below the primary sat-uration temperature. This will ensure that the steam generators are heat sinks. Other actions are the same as those diseu Jed above for refin boiling. 5-4 Babcock & Wilcox I
I I 3. Strategy for Multiple-Equipment Failure Events I 3.1. Seismic Events The most probabic consequences of a seismic evenc are loss of some power loads I and the display of conflicting information to the operator. The effects of a scismic event on instrumentation are complex. The electronic modules them-selves are qualified for seismic environments, but the power source to the modules may be lost (corrective action should be initiated to cicar and reset tripped breakers). Some of the fluid instrumentation lines may rupture (cr develop leaks), which results in conflicting information on plant parameters
- I to the operator and possible inadvertent initiation of automated safety sys-tems. Judgment must be exercised to determine which indications are in error and which are good (those that use fluid instrument lines are level, flow, and pressure). B&W has reviewed the adequacy of the incore thermocouples, includ-ing the guide tubes, to withstand seismic accelerations. The results indicate that they are or can Nsily be made acceptable. Since thermocouples are self-powered, they can be read (at a point before the signals are input to the com-puter) even in the event of loss of all pu.'
3.2. Loss of All a-c The loss of total a-c means that both offsite sources and the diescis are un-available. Both of these sources have a time-dependent probability cf being recovered (e.g. . the longer the time, the greater the chance of recovery). For example, if the diesels did not start automatically, they may be started manually. Once started, loads may have to be sequenced manually. Similarly, l if the diesels cannot be restored, the procedures for res; oration of offsite power should be employed. It is important in either power restoration case to ensure that support systems (e.g., cooling water, lubricating oil) for HPI, auxiliary feedwater, and RC pumps (in addition to cooling and ventilation equipment for instrumentation rooms) are being supplied. Without proper sup-port systems it doesn't matter if a-c power is recovered. In the event of l auxiliary turbine feed pump control problems, the operator (if he has good ! ur indication of steam pressure) should regulate pressure at just under the steam relief valve setpoints through the atmospheric dump valves. This procedure can be followed from the cuna vl room until the air supply for valvcs is de-pleted- After that, it must be done locally at the valves. I 5-5 Babcock & Wilcox I
1 I Il l 3.3. Steam Litie Break-Induced LOCA The most probable LOCA resulting from a steam line break is a steam generator tube rupture. Containment sprays may be actuated if the steam line break is inside the containment. The BWST is being depleted from the flow path through g ruptured tubes and from containment sprays. Procedures to be followed for 5 this accident are identical to the tube rupture guidelines with the added re-quirement to shut off containment sprays as soon as practical (e.g. , if con-tainment coolers are operating, turn off sprays immediately). I I
- 4. Lesson Siumnary
- 1. List the thermodynamic conditions under which tiiermocouple readings are useful.
- 2. Emphasize that temporary loss of RCS heat removal can be recognized and treated to avoid degraded core conditions.
- 3. Review the basic philosophy for treating multiple-equipment failure events.
I I I I I I I 5-6 Babcock & Wilcox I I
2 !I i ll jW Figure 5-1. Loss of Natural Circulation Due to Buildup of steam in Reactor Coolant System I ! r, 7*
~ - c o of.
- o E- 0 C' -
[X ) D% M J\ N e 3 ) NW - i o l 6':D \ -- e NY
- l T l
l ! .
} >,
l l-lhs7h
- i r'O' N;,
l I l
; '. I b i l &-
l ; a W b i f Q $ j l l C fa g. o
* ~
eG t / 20 )
, ; , ,' f- . ,
/> ,
jl l l
, I .' ~
- a '
E _ y 's/,1T ! ll 1 ) ( / !
. W-H l
. 'V i
Q ! v
, fd}
J l 4 } Tl"t i T!*'t 4 ! J
- 1. tier reasin; pressartier level recasse of Icss o' reactor Pres sar tier liosic vcipe is 1 st stea- f ror pressur ire- I j
t coolant h.0; A; or cortracticn of reactor cool er an erter C loors =n ea the P ! deteest it?s. i (O ve rr001 ing !
' &eacto- coolant tressure Araps to a valva atout equal to l l 2. Pea c tor tris on lo. reacter coolaat pressure. stear venerator pre s sure .
3 ISA5 actuation or low reactor coolent pressu re. 3. Stes' fer~s in not les recause of aces Mation of rre s sur t rer stear aN tw.a sse of flestirs of reactor jI i
- '-P a c t uet t or,
- IF* actuation 4.
c or,l a n t . Stea- is ir tne fec* of tv t i e. w ti r ic RC',
?-; nase retur al circalatior. wi 1 occs c - teilie; may occur i n c,re.
l *
.. If sa'ficient staae is created. tt will start tc cellect j 1r ucrer recon or ;oor reasse stea- cea r'se tt a f aster
, .eioo t , t,.ar .a e er.
/ \\
!l &
= Smc w ;
i s .a m ll s_ c. _ 4lh 0 . D LICtht I I -~ ~ Z ; E ,O y e---, o a 1-- 8' Pa t ria t fler $ttC Q 8v s10t
./
- I ,o n._v.
j lc
-]
c 'AII8 h 91TtR !, Mjl w b O tc east oaic 31,,a 4 s m/,/ (.
\v> O situ I
I 1. I (t er s eparates in wr per tercas of luas 2-lhJte est Jeal circsl et t0'1 s t C P' 5tre 08 I Pf- S t ras' PtJble dettadi o*. stvarit y Of ac SeeM HIl f)De rett , ge9 4 ! Startup t ree , 3_7 Babcock & Wilcox
m___._ I Figure 5-2. Reflux Boiling fp-, . %;. ,., .w- ., . .
.% $ C: g ind *6
'3 N Iv,'
- h. Listas y .-
1 ...- .- m g
, f , ,, , ,,, g *y raisas, siet stkasant sist l
+
J,ms'""......g. * .,
,y , ,1, , g ,,,,,
l ', ___. *
- T- r ' '
,,," , . , , ,. %. . t .# 5m ,*.' ' '
j , @ fe; Peast ticulo sigan
...n.; 3 -m~
{ , h i ;! !
+
sitas 1~ ' ~'M ,
! \
y jj., 'd .. . . . . . . . . . . . . . v - t ...u..... l is: 1 :e ! -- -- W ,
.. s. ,,,, s ,
g;;g 8008 % r..., I , t ..- O sul;DM t B i, 2:0c
,. . t c o. l #e- NOTES ON P-T DIAGRAM l :. 4
- iin I .-
; F
.- - + surin iar itsici, 1. RC pressure is slightly l j iseo .
higher than steam generator pressure. W _L i.30 i - 3 1/ 2. Thot is equal to Tsat for
'ic8 'cet e
'ac t existing RC pressure.
i
,.x
'c*c ss entssvat /._ .d .
r I'8 '0'ai Post * '" 'catto
, _ ._ _ _ _ _ _r_
jj cintutstica 3. Tcold is equal to Tsat for existing steam pressure. l
\e a g q ,, 9 %C##A CPillfl.5 r$l.f.Pollt see ..,..u.,
ss "g% [j c""*s' ' 0* eac t- s'.' sv a tin ir. c , ro,i
, c , ,r.,,cast in , ii. ..ru..:
.:n est sco sse saa sso 18:
. . .o . , c. . . .. t i . . . . t . . . . s OPERATOR ACTIDN REQUIRED
- 1. Turn HPI on to highest flow eate.
i l 2. Turn EFW on and raise steam generator level to 95% on Operate Range.
- 3. Start plant cooldown at 100F/tr.
4 Monitor plant conditions for a loss of reflux boiling or a return to norrr.al n3 tural circulation (subcooling). 5-8 I Babcock a.Wilcox I
i l ! I
-I Figure 5-3. Loss of Natural Circulation -- System Refill by High-Pressure Injection s s s s ' !N Nqsa 'I &%
l , f, """
- g 4 li '
.-l@
- taism siot stcocm sier
!l 4 h I __[., ' sit' E satta
.. // 9a !r
' ll',,,, , ,..
O 150 rNAsl L10010 gyggg _c g ,' I I E sitAa
- "" ;" 3.'l'.;"." ! l .u l llU:'5"."* .l"*;' .
%;.1.'; ;,ll.'"" "'
tsee seco { -- -- - rsst to:p 3 I "" to . su8c00tlO I l D [ . ~_~. ]O e 1.
~ NOTES ON P-T DIAGRAM Thot is equal to Tsat I g ines ; [ 3"'y'" for existing RC prassure.
( .s 3 isu ,3 + 2. RC pressure will increase I l; above steam generator P - iscc I" pressure and can go as
- I
"/ high as the pressurizer
" " ss enssuas ussi ' cot: i'
'a' safety valve setpoint
{ { (2500 psig).
;I : ien
,3
__;___3______,.._. i
] ro poimi.ecs gi negt . ,,,, inir eita roncre
- 3. Steam pressure may drop because heat transfer from f s' 3
sn ,8n '[y .,# h noenai oi ssuins roiar-ranareactor coolant is low. oriaatio, ;
'. ,, I w., w cootto '-8s*U'""'
n 8"nc'u'atio'n y ci t
- 4:s au see sse saa su iso I
- .. .o. , c -- e i . . . t.. . . r OPERATOR ACTION REQUIRED
!. Same as reflux boiling.
- 2. Establish the steam generators as a heat sink (Tsat - SG should be
' = about 50f less than the incore thermocouple temperature).
s Open ERV e Bump one RC pump l I 5-9 Babcock & Wilcox}}