Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Safety Injection during Main Steam Line Break

  1. Mar 25, 2014 #1
    Why is Safety Injection used during Main Steam Line Break?

    All that I can find is that it brings water back to the cold leg to compensate for the loss during the break.
  2. jcsd
  3. Mar 26, 2014 #2


    User Avatar
    Science Advisor
    Gold Member

    Rapid cooldown causes RCS volume to decrease due water density increasing, therefore you need immediate water injection to make up the volume as the volume control system can't handle the change that rapidly. You also risk draining the pressurizer and losing pressure control of the RCS.
  4. Mar 26, 2014 #3
    Thanks that helps a lot! So to clarify.. the steam line break cools the primary system down, which in turn causes primary side to think it needs to generate more power. By doing so, the pressure decreases because it will drain more water from the pressurizer to fill the new steam generator demand.. And so the safety injection system introduces more water so that the pressurizer level can remain stable?

    In terms of steam loss, would the same result occur if extraction steam is lost to the feedwater heaters (maybe not necessarily due to a steam line break)? Under normal conditions, the extraction steam is the main source of heat to the feedwater heaters. So I imagine the feedwater heaters would not be able to heat the feedwater enough before it goes back into the steam generator.. But how would colder water in the steam generator affect the power level/pressurizer/etc.?
    I'm not so good with the thermo to know if more or less steam would be generated..
    Last edited: Mar 26, 2014
  5. Mar 26, 2014 #4


    User Avatar
    Science Advisor
    Gold Member

    The pressure decreases because the water volume shrinks due to cooling down and increasing density. The reducing water volume is what causes pressurizer level to decrease. Otherwise yes.

    Main steam line break supposes the steam generator secondary side instantly reducing to atmospheric pressure. Problems occurring with the feedwater heaters are not a big deal and can be handled online by reducing power until the problem is corrected.
  6. Mar 26, 2014 #5

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    oops a good answer preceded mine...
    Well, here 's mine anyway.


    Back up a little way in your thinking.

    It's all about energy.

    What does a steam line break do?
    It removes a lot of BTU's from the primary.
    That causes the primary to col down, and the water in it to shrink.

    What does a primary side break do ? It removes a lot of mass from the primary , and the BTU's carried by that mass.

    So the initial symptoms look very much alike - pressurizer level decreases, so do pressure and temperature.... and if the steamline break is inside containment pressure there will rise,, so the can two look a LOT alike to the unlucky operator on shift..

    For either event the reactor is tripped immediately.
    SI will inject boron should pressure get below SI pump pressure, which prevents a return to criticality on cooldown. Return to criticality could put more energy into containment than it can handle, so SI is appropriate for steamline break.
    SI provides direct cooling water flow to help cool the core for a primary side break .
    So SI is appropriate for both events, secondary as well as primary side break.


    Extraction steam is not a really large fraction of total steam flow. Losing it results in cooler feedwater which means you can't boil as much of it with the fixed power available from the reactor. You're making~1200 BTU/lb steam from ~430 BTU/lb feedwater when you have extraction steam, or from around 150 BTU/lb feedwater when you don't have extraction. It's unusual to lose all extraction steam, usually it happens just to one feedwater heater and the effect is only a degree or two at entrance to steam generator..
    Last edited: Mar 26, 2014
  7. Mar 26, 2014 #6
    Edit: This now goes for either of you.

    Well for the second part, lets say it is NOT because of a steam line break (So the scenario is, for some unknown reason, there is no extraction heat from the turbine for the feedwater heaters). I guess what would be a direct result in terms of the pressure/power/etc., assuming that it isnt just handled online? aka, the turbine governor valves dont reposition, or anything else, what happens to pressure/power?

    I figure that the amount of feedwater turning to steam in the steam generator (S/G) would decrease if the feedwater is colder (because of the now bigger delta T between the feedwater and the coolant from the hot leg). Since less steam is created in the S/G, less steam goes to the turbines. If less steam is going to the turbine, and the turbine governor valve doesnt reposition, would the power increase or decrease?
    If the steam demand stays the same, then I imagine the power would increase to generate more steam to meet it..?

    Btw, my reason for asking: I'm currently taking a nuclear engineering course which intends to teach us the plant integration. I have NO background in nuclear or mechanical. I am trying to get a grasp of basic concepts so I can speak at least somewhat coherently on the topics, so I have a decent amount of questions. Would you be at all inclined to assist in others? Logically talking through these helps a lot.
    Last edited: Mar 26, 2014
  8. Mar 26, 2014 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Cooldown of feedwater is an event covered in the accident analysis chapter of pwr FSAR's.

    Initially it would cool down the reactor cold leg, which would cause the reactor power to increase.
    It would also shrink steam generator level causing feedwater regulating valves to open, making things worse. .
    That'd be the natural tendency of the plant.

    So the instrument system has something called "overpower runback" which senses reactor power as difference between hot and cold leg temperatures. Temperature rise is proportional to power.

    When cold leg starts to plummet it looks like a reactor power increase is imminent, so an "overpower runback" is initiated.
    That runback drives the turbine throttle valves in closed direction, reducing steam demand to match what the reactor can deliver. Ours would pulse the throttle valves until temperature rise across reactor returned to normal.

    Once again, it's a balancing act with energy.
    Keep energy in back of your mind when visualizing these systems - the plant is just moving energy from the fuel to the grid via first BTU's in hot water, next steam, next mechanical torqueXrpm, and finally electrical kilowatts.
    That''s why a plant requires just a handful of Nuclear engineers but scores of mechanicals and electricals. The reactor runs fine for it was developed by geniuses in WW2. The rest of the apparatus that moves the energy out to real world
    dates back to late 1800's. Even Titanic had a steam turbine.

    if you don't mind my unprofessional qualitative talk-throughs, i'm honored to help. Mentors please advise if i'm not up to PF standards here. My math has got rusty and my Latex was never good - but think i can help beginners with these basic concepts.

    old jim

    ps - see also http://www.westinghousenuclear.com/Products_&_Services/docs/flysheets/NS-ES-0030.pdf [Broken]

    old jim
    Last edited by a moderator: May 6, 2017
  9. Mar 26, 2014 #8


    User Avatar
    Science Advisor
    Gold Member

    Feedwater temperature would decrease, thus slightly increasing the reactor power. The turbine power would remain the same. Thus the net effect is lower plant efficiency (which is the purpose of feedwater heaters).

    In a PWR, the reactor is a slave to the turbine. While the turbine governor valves remain constant, the steam demand will be constant. If you reduce the temperature of the feedwater, it will extract more heat from the primary side in order to keep the same steam pressure. Removing more heat from the primary side cools it down, which thus increases reactor power due to the negative moderator temperature coefficient.
  10. Mar 26, 2014 #9
    Okay so my answer was correct, but for not exactly a correct reason. Thank you both!
    I just have a lot of questions dealing with the steam/turbine/feedwater portion of the plant (as I feel like its the main "big picture" concept to grasp). For example, things we discussed are generator tube ruptures, main steam header (which I can find hardly anything online about), and :

    1.) What happens during a steam generator tube rupture? How would this get resolved (either by the operator or the plant systems itself)?

    My knowledge: A S/G tube rupture means the tubes inside the S/G which contain the radioactive coolant. If this gets mixed in with the steam in the generator its a huge problem. Therefore the first step would be to cut off the steam to the turbine by closing the Main Steam Iso Valves (MSIVs). No steam is going to the turbine and the reactor is shut down. What I dont know is what steps would be taken to startup again, or resolve it without shutting down the plant..

    2.) How does main steam header pressure affect reactivity?

    My knowledge: Not too sure what a steam header even does other than serve as a middle man between the S/G's and the turbine, and between the S/G's and the condenser. I would think that if the header experiences higher pressure, that means more steam is coming from the S/G than the turbine needs. Therefore, more steam would be sent to the condenser. Not sure how reactivity would be affected in this case.
    Same but opposite goes if the main header pressure is low. I would believe that the there isnt sufficient steam supply to the header based on turbine demand. I However I think that this would cause the reactor to generate more steam, which results in an increase in reactivity.
  11. Mar 26, 2014 #10


    User Avatar
    Science Advisor
    Gold Member

    The defective tubes are identified using eddy current testing, and are then plugged. There is a limit as to how many tubes may be plugged without significantly affecting the total heat transfer. If too many need to be plugged, the plant must reduce its maximum power or replace the tube sheet (expensive).

    The main steam header is just the point where the main steam lines for each steam generator meet and combine. Remember, the reactor is slave to the turbine. Changes to steam pressure will affect the steam generator heat transfer rate and thus reactor power. The turbine gets exactly the steam it demands (unless the steam generators dry out - bad). Increasing turbine power lowers steam pressure, which increases steam generator boiling until pressure reaches new equilibrium. If turbine power is lowered, steam pressure increases, reducing boiling until new equilibrium is reached.
  12. Mar 26, 2014 #11
    For your response to #1, could you elaborate more? I guess in PWR's there are more than 1 steam generator, so they could shut that S/G down for maintenance, but I imagine there are a large number of steps that have to be taken if something like this happens, such as certain valves are shut (I mentioned the MSIV above), different processes are stopped (like the turbine), different processes are started, etc.

    For your response to #2, I guess it wasnt clear enough to make me sure that I'm correct in my logic. Can you work through it but using a BWR instead (since in BWRs, turbine is slave to the reactor)?

    If you havent noticed, I'm really trying to UNDERSTAND the flow of processes and their consequences by stepping everything out. Like for #1, in a PWR, if a generator tube ruptures, the operator has to do certain tasks, and those tasks will have an effect on other things. For #2, the main steam header in a BWR is experiencing high (or low) pressure, this means the S/G has to react a certain way, which affects reactivity a certain way.
  13. Mar 26, 2014 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    This link has links to the NRC training materials for the introductory course they give to their inspectors, i think.

    I tried to download them but am on a weak server here and it crashed .

    http://www.law.cornell.edu/wayne/projects/entergy/callcenter_website/orient_train/ont_nuclear_power_for_energy.htm [Broken]

    aha this one loaded, it's a pretty good bird's eye introduction.

    QPion's answers are good, and i admire his direct answers. Makes me feel my thinking is muddied by age. Maybe too many late nights in the plant.....

    Anyhow - big picture on SG tube leaks is this -

    The leaks usually start out small. Your first indication is count rate at the condenser air ejector starts to climb, and this is long before radioactive nuclides become noticeable in secondary water samples .
    So you start getting ready for a shutdown.
    When water samples can tell you which generator is leaking you'll shut down reactor and begin a cooldown, reducing reactor pressure as soon as practical to minimize leakage.
    In old days we could isolate affected generator's secondary and pressurize it during cooldown so leakage goes other way - clean water into reactor instead of contaminated water into secondary systems. I dont know if that's done anymore.

    Once plant is cooled down you drain primary and locate the leaking tube(s), plug both ends , and probably do a lot of testing to see if any other tubes are thinned or near failure.

    It's a big operation, a week minimum and often a month.

    I'm not aware of any leaks that were ever big enough to affect reactivity. One gpm would be a big one where i came from.

    That training material should help you with mental pictures of the parts and terminology.

    Ask your questions, but be aware it'd take a book to describe the systems and that book is already written.

    PWR Steam Header Pressure affects reactivity only through temperature, look at saturation curve for water substance (In my days Civils took Thermo, and Electricals took Fluid Flow - i think so that we'd be still able to communicate after graduation.) It's back to that energy flow - heat transfer through tubes is quite good, pressure variation of secondary translates directly to temperature (saturated liquid), and delta-temperature across tubes defines heat flow out of primary....
    both sides obey the properties of water in the steam tables. Secondary is saturated, primary is well sub-cooled.
    As we said in the control room: " It's only water. Keep it that way."

    In a BWR header pressure affects reactivity via the tiny steam bubbles in the boiling water - more steam bubbles shut down the reaction, fewer steam bubbles enhance it. That's why the control system for a BWR is pressure based vs PWR's temperature based.

    If you haven't had Thermo, suggest to professor he spend one day on navigating the steam tables .

    Enjoy your course. You know, nuke plants need civil engineers too.... If you enjoy this course, interview your local utility...

    happy to address your questions. Maybe Qpion and i can complement one another's answers.

    old jim
    Last edited by a moderator: May 6, 2017
  14. Mar 26, 2014 #13
    In a BWR, the turbine is slaved to the reactor, as you said. BWRs are analyzed such that they must be above about 750 PSIG in order to exceed 25% power. They are not analyzed to be producing power at lower pressures. Additionally, BWRs have a main steamline break protection, such that if the reactor senses steamline pressure low, the MSIVs will automatically fast close, generating a reactor scram if the reactor mode switch is in RUN (and even if you are in STARTUP mode, you'll likely trip out on high flux). In general, BWRs are operated between 970-1050 PSIG, depending on the specific plant, the specific core design, etc.

    That said, BWRs have positive pressure feedback when they are steaming (which is almost all the time, except during cold startup). BWR operations in the US do not allow direct reactivity control using steam header pressure. I've done it before in my BWR's simulator, and you do see a response, but the moderation of pressure to affect reactivity is explicitly not allowed as it is not an analyzed reactivity control method for BWRs and can lead to reactivity transients. (There may be a plant out there that does have permission to raise pressure at end of cycle conditions to help boost reactivity, but it is not the norm).

    Generally, pressure in a BWR is set constant. At my plant, we hold about 920 PSIG at the main steam equalizing header. This equates to about 1020 PSIG in the reactor. As power is raised or lowered, the change in steamflow causes the pressure in the main steam equalizing header to raise or lower. The turbine will respond by automatically opening or closing control valves to maintain pressure in the header at its set point, which in turn, maintains reactor pressure around its setpoint, and ensures that pressure is not controlling reactivity.

    During heatup, you do see effects on reactivity. If you do not have adequate steam removal, you'll have pressure come up, power come up, pressure come up faster, power come up faster, and you'll either swell your water inventory causing a high level trip or steam line closure, trip out on high flux, or even worse, violate your tech spec heatup rate.

    When you are at power, any sudden change in throttle rate (how far open or closed your control valves are) will have large effects on both reactivity, actual reactor water level, and indicated reactor water level. For example, if a bypass or control valve were to suddenly jump full open, you will have an increase in steam flow and a decrease in pressure. In the reactor, your water level indications will rapidly rise, and power will decrease. The other control and bypass valves will automatically reposition to maintain steam header pressure, which halts the transient. If they didnt reposition, the water level swell would continue and would trip your main turbine which would then result in a reactor trip. If you had a bypass or control valve suddenly go full closed, you would see water level decrease, power increase, and again your control/bypass valves would respond to the transient. If they did not respond, the reactor pressure would increase and you would trip out on high pressure. If the high pressure scram failed, you would trip out on high flux.

    This brings another big point, any time you lift safety valves, or perform any non-smooth pressure control action, you will have sudden swings in actual and indicated level. This is very challenging to deal with, as you tend to hit your high and low water level trips and you are dealing with major equipment going offline as a consequence. It takes a bit of training to get used to to these responses.
  15. Mar 27, 2014 #14
    First, thanks so much for the response, it is really informative.

    I have 2 questions about this part you said..

    1.) I'm confused on how an increase in water level causes the power to decrease? Wouldnt the BWR reactor "want" to produce more steam if the water levels were high, in which case would increase power? (lets assume the turbine is malfunctioning which is causing the pressure fluctuations).

    2.) The relationship between power and reactivity is that if power goes up, reactivity goes up...and power goes down, reactivity goes down. And neither one is especially bad if it means trying to get it back to criticality. Correct?
    Last edited: Mar 27, 2014
  16. Mar 27, 2014 #15
    For 1:

    If you were to add water to a reactor, you get a slight improvement in subcooling. This results in a slight increase in power, (but also results in a net decrease in plant efficiency). If you were to decrease water in a reactor, you get a slight decrease in power, until you uncover the feedwater spargers, at which point you get a large decrease in power as the incoming water is not adequately subcooled.

    What I was talking about though. Is when you have a pressure change, the sudden pressure change will result in a change in saturation pressure/temperature. The overall effect is a decrease in pressure will cause MORE voiding, which causes your water density to DECREASE. A decrease in density results in an increase in level. Additionally the pressure change will affect your reactor water level sensing lines causing indicated level to doubly increase. On a pressure increase the opposite happens. In this case, the water level change is due to a change in density. Power is affected due to void content change.

    For #2:

    Reactivity is like "Acceleration" in traditional kinematics. A critical reactor has 0 reactivity. If you have positive reactivity, you are super critical and neutron flux is increasing. Negative reactivity, flux is decreasing. The rate of change is exponential. Eventually, as flux increases, power increases, you produce more heat, and you boil more water (generate more voids). The increase in void content reduces moderation, which removes reactivity until it is back to 0. If you add negative reactivity, power will decrease exponentially, and the reactor is subcritical. The reduction in flux then reduces power, reduces heat output, and ultimately reduces void content. This improves moderator density, brings reactivity back to 0, and the reactor goes to steady state at a lower power level. The exception to this is if you add a LOT of negative reactivity at once (like in a SCRAM), the reduction in void content and other effects in the core wont be able to make up for the large reactivity reduction, leaving to the reactor shutting down. If you had a large reactivity increase, the scram system will detect high flux and automatically actuate to stop the increase, bringing the core subcritical.

    Hope this helps.
  17. Mar 27, 2014 #16
    Okay that makes sense, thank you! Out of curiosity, what is your experience? Operator? Professor? I may just have to contact you again in the future with more questions as they pop up, since you seem to have a wealth of knowledge. I like the whole "if this happens, then this happens" step-through approach, as it really helps understand the concepts and theory that general manuals and literature online seem to lack.
  18. Mar 27, 2014 #17
    Nuclear engineer for several years. Currently getting a senior reactor operator license at a BWR. Feel free to ask questions. These are the fun questions to answer.
  19. Mar 27, 2014 #18


    User Avatar
    Science Advisor
    Gold Member

    If power goes up, then reactivity goes down. But you need to insert positive reactivity to make power go up initially. If reactivity went up when power went up, you'd have uncontrollable positive feedback.
  20. Mar 27, 2014 #19
    BWRs have a positive pressure coefficient. Normally, the turbine and/or steam bypass valves will automatically modulate to control pressure. If pressure control is lost, you will have a pressure change which will affect flux, and you will have a run-away effect, where pressure causes power to go up, which causes pressure to go up.

    The worst case scenario for a typical BWR is a fast closure of all main steam line isolation valves. The pressure spike will result in a prompt peak of around 210% flux, which is arrested by the lifting of safety valves and the reactor scram. If the scram fails, the lifting of safety valves will reduce pressure and your ATWS (Anticipated Transient without scram) logic will trip your recirculation pumps to drop power down to about 40%.

    In the event pressure control is lost, operators are supposed to manually scram the reactor. It's not a requirement, but you will likely end up either in an unanalyzed condition or, more likely, with an automatic reactor scram.
  21. Mar 27, 2014 #20
    That's cool Hiddencamper, and good luck on the license.
    Thank you Quantum.

    Okay, I imagine this question is a simple thermo one, I just dont know thermo. In a heat exchanger, what would happen to the heat transfer rate if there are particulates/solids in the tube side of the heat exchanger?

    This is part of a much bigger picture I'm trying to paint, which involves the main condenser and particulates in the tube side, the resulting efficiency, etc. But I'd probably figure that out on my own if I knew whether the steam would condense less or more because of particulates in the tube side.

    I imagine the heat transfer rate decreases, because the particulates take up space that the cooler water would normally occupy. That means the heat exchanger wont function as properly (less steam condensed => hotwell temps increase => eliminates vacuum => plant efficiency decreases)..
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook