Reducing the effects of Loss Of Off-site Power?

Come here if you are new to nuclear power and are looking for answers.

Feel free to post questions.
Forum rules
  • All questions allowed... provided they are relevant to the topics at hand, that is to say nuclear power, power generation, environmental science and so forth.

Re: Reducing the effects of Loss Of Off-site Power?

Postby Michael » 10 May 2011, 00:47

Well, as a complete layman I'd say:

1) Arrange for ports where you can hook up mobile power units from the outside.

2) Set up pressure relief filters, like the Swedish FILTRA used at Barsebäck. They are dirt cheap and very effective.

http://www.oecd-nea.org/nsd/docs/1988/csni88-156.pdf

/Michael
Michael
 
Posts: 117
Joined: 24 Nov 2008, 05:07

Re: Reducing the effects of Loss Of Off-site Power?

Postby Johan » 10 May 2011, 18:30

Hi Ross,

I know very little about electrical systems and the balance of plant. My thing is the core since I am a reactor physicist. But I can still speculate a bit :D

When something unexpected happens one usually want to be able to isolate the containment from the environment. One shuts valves to all the main steam lines, feedwater pipes etc. If one wants to hook up an alternative alternator it would probably be better to have a dedicated steam line to it with everything housed within the containment so it can work when the containment is isolated.

BWR kind of have this with the RCIC system, check the brief description given here (http://nuclearpoweryesplease.org/blog/2 ... re-cooled/). The steam flow due to decay heat is enough to drive it so there is no need to run at low power. Running at low power can create all kinds of headaches and instabilities so I am not sure it would be an easy thing to do for any extended time. Especially not going from a scram and then up to a partial power while suffering a blackout.

Now that you mention it I actually do wonder why the RCIC isn't connected to a small alternator that would be sufficient to at least provide charge for the batteries needed to operate the valves etc of the system.

One possible retrofit would be to somehow modify the wetwell so that they can get rid of heat in some passive way, that would make the RCIC system into a long term cooling system that could supply cooling for quite some time.

I think the solution in AP1000(PWR) is quite elegant. They have a big pool of water outside the vessel, in case the containment is isolated and the plant suffers a blackout then water can circulate passively from the vessel into a heat exchanger in the pool. After about an hour the pool water starts to boil, the steam created from the boiling condenses on the containment walls and is drained back into the pool. The containment walls in turn are afaik passively air cooled. Retrofitting existing PWR's with such a system would probably not be a small task though.
Johan
 
Posts: 26
Joined: 24 Nov 2008, 14:38

Re: Reducing the effects of Loss Of Off-site Power?

Postby Matte » 10 May 2011, 22:55

@ Michael
Both those facilities exist today, all power plants around the world are capable to be hooked up to a mobile generator. The filter system exists in certain parts of the world and will probably be implemented as standard in the future due to the events in Japan.


@ Ross
You raise some interesting points with your question, please forgive me for starting off with shaking a finger at you;
There is quite a difference between a PWR and a BWR when it comes to reactor control and what you can do with the steam, this will have an impact on what you suggest.
Also, you will be surprised to learn how many reactor scrams occur every year at every power plant, if my memory serves me correct the best performing plants have ~0.5 scrams/year and the worst above 10, I believe the world average is around 1.5-2.5 scrams / year per reactor. You never hear about it due to it being routine and depending on the fault the scrammed reactor will be back online within a few hours, xenon poisoning takes about 20 hours to set in and become a serious problem.

Now, basically what you are suggesting is “house turbine operation”, the plant operates to supply it self with electricity to operate systems. In theory all 2nd generation (pretty sure there are no 1st generation left) and later are capable of running in this mode, in theory. The difference here is of course that the normal turbine-generator-set(s) is providing the power not a designated backup turbine/alternator. This requires a very good reactor operator as the automatic control systems are not capable of running the plant in this mode, the change in reactivity due to xenon and feedback mechanisms will trip out the automatic control system at this power level and quite possibly the reactor as well. Operating at ~5% power is not a normal mode in a system designed to operate at full whack. This depends slightly on whether you are starting up the reactor with no core poisoning (easier) or if you are being disconnected from the grid due to external event and have to drop power from 100% to ~5% (difficult). Modern reactors require the automatic reactor control system to be disconnected below a certain power percentage (Gen III and GEN III+).

I have seen two pretty much identical units (located side by side and built ~3 years apart) being disconnected from the grid simultaneously where one unit managed to stay in house turbine for about 3 seconds before scramming where the other unit managed about 12 hours before tripping out, for example.


What you suggest is having a separate turbine/alternator unit to drive the plant during station black out (SOB) conditions using steam produced by the nuclear heating in the reactor (either PWR or BWR). This is not a bad idea in theory however when you consider the implications of plant operations, I hope I can convey the knowledge in order for you to understand why it would be a bad idea. Please bear in mind I am a PWR buff and know very little of BWR operations, somebody else would have to fill in the blanks or correct me if I am wrong there.

Remember, in a scram situation, regardless of reactor type, the core is taken subcritical within a few seconds, turbine shutoff valves will close and dump valves will open or steam generator pressure relief valves (PWR). If you have off site power grid failure, the emergency power system will start up in order to provide power to pumps and systems required to take the plant to a safe state. In the event of emergency power failing there is always a minimum of 30 minutes where the operator does not need to do anything in order to affect the plant. What happens next is plant specific but for PWRs you can always operate the steam driven auxiliary feed water pumps to pump water into the steam generators in order to cool the reactor circuit, if the auxiliary power fails to start. I don’t know if BWRs’ have a similar arrangement. But what is common for all reactors is that emergency power needs to be restored within a limited timeframe, this includes the AP1000. Because the spent fuel pools will always require cooling sooner or later (depends on the fuel being stored there).

Regardless of which plant type we are looking at, you are suggesting we start up the plant “back wards”. Normally when you start up a power plant you make sure you have all the cooling chains in operation before you reset the reactor trip breakers and start taking the core critical and into power operation mode. If you think about it I am sure you will realise why. What you suggest is starting the reactor up with all systems either offline (or coasting down after a scram). From a regulator point of view this is a big no-no! You must have your cooling systems in operation before you can consider operating the reactor. Let’s consider your seawater pump to cool the condenser does not start when you try and take the reactor active after a scram, suddenly you can’t cool the steam and the turbine will trip out. You are now in a worse state as the reactor is producing heat that you can’t remove. The amount of dangerous situations that become worse if you start up the reactor in a SOB situation are just too many for the method to be considered.

Now you may suggest using batteries to run up a cooling chain capable of starting the plant up again after a trip, however this would mean you drain precious battery power powering systems required for safe plant restart which may fail, battery power that may have been more important to keep cooling the plant while onsite emergency power is restored via a generator or by other means. Reactor coolant pumps (PWR) or main circulating pumps (BWR) need to be operational in order to safely operate the reactor, these pumps are usually quite large (MWs’ per pump range)…

In a PWR you can usually dump excess steam “over the roof” of the plant as the steam does not (usually) contain radioactive substances. In a BWR you have to condense the steam and recover it as it contains radioactive substances that you can’t just spread around, this will require an operational cooling chain.

Hope my aimless garble has not put you off the subject, if there is anything I can clarify please let me know. And please feel free to ask more questions I am sure somebody around here will happily answer it.

//Matte
Matte
 
Posts: 17
Joined: 09 Mar 2011, 18:56

Re: Reducing the effects of Loss Of Off-site Power?

Postby zl2wrw » 12 May 2011, 11:06

Hi,

Johan & Matte, can you please explain why it is hard to run a reactor at a low thermal output power? (does reactor criticality have an increasingly non-linear response to control rod/blade insertion as the neutron flux and or thermal power level is reduced? - this could make it hard to maintain the reactor on the point of criticality at the desired thermal power output?)

Yes, I am talking about house power, but with the house alternator driven from the main turbo-alternator shaft, instead of from a separate turbine (but other than the economics, I don't see why you couldn't have a completely separate house power turbo-alternator). Considering that when grid power is available there will be almost no electrical load on the house turbine, so relative to the main alternator, the torque required to keep the house alternator spinning at synchronous speed will be negligible (it will only load in the event of a LOOP).
I was not suggesting that anyone try to start a SCRAMed reactor without a fully functional cooling system. I was trying to ask why SCRAM the reactor in the first place if the house alternator can meet on site electrical demand (including keeping the cooling systems functioning), when a fault occurs "off-site" (eg a loss of connection to the grid, loss of synchronism between the main alternator and the grid, main alternator over-current relay trip, etc)
I was thinking that the house alternator should be a separate machine to the main alternator so that house power is still available even if the voltage across the main alternator collapses because of a grid fault or the main alternator trips offline (sustained overcurrent, overheating, etc)

Reading the following report got me thinking about this:
Nuclear Generating Stations and Transmission Grid Reliability
http://www.consultkirby.com/files/NAPS_Sept_07.pdf


Regards
Ross Whenmouth
zl2wrw
 
Posts: 5
Joined: 09 May 2011, 11:00

Re: Reducing the effects of Loss Of Off-site Power?

Postby Matte » 12 May 2011, 22:39

Ross,

sorry for confusing the issue. But to achieve house load operations you don't need a separate alternator, the main one will do just fine. As you suggest, you could hang a seperate alternator on the turbine shaft that feeds back to the stations internal equipment only, however I am pretty sure it would not work. During an occurence where the main turbine trips, you have to reduce reactor power pretty darn quickly, often the operator and the automatic reactor protection systems will scram the reactor simultaneously, often these sort of events happen unanounced when the operator is not ready for it.

I know you did not suggest to start a scrammed reactor, unfortunatly that is what you suggest as in an event where a sudden loss of offsite power (disconnection from the main grid) will most certainly lead to a plant tripp.

As to relying on coast down to power systems...was that not why Chernobyl blew up?

The reason why a reactor is difficult to control at low power (after running at full power with xenon-equilibrium) this is Johans department, however I will try and explain;
The plant is designed to run at full whack, the water bled off the reactor coolant is calculated at full power and is replaced with unborated water to reduce the chemical shim in conjunction with the reactivity decrease in the fuel due to burn up. This system is sized to run at 50%-100% power operation.

In PWR operation you control short term reactivity with control rods and long term reactivity with boron poisioning of the coolant. Running at low power would require you to run the reactor with all rods almost at the shutoff point due to decay heat, as decay heat falls off you need to throttle up (withdraw rods) to keep the power level constant. While this happens you build up Xenon-135 which will reduce reactivity of the core requireing you to throttle up some more. At this point small variations of temperature will also feedback and change reactivity of the core (due to density change of the coolant).
As the reactor is settling into the new power level you have several parameters (coolant temperature, xenon-135, boron concentration and core recativity worth, control rod movement) affecting the power output of the core. Power up slightly too far and you burn off more xenon-135 increasing the reactivity of the core, power back too far to compensate Xe can build up further and slow you down, etc. Feed and bleed operations that reduce boron concentration has to be operated intermittenly and on top of it all you have the reactor protection system that will tripp out the reactor if things change too rappidly.

In BWR you control the reactivity with the circulation pump-RPM and control rods and have better fine control, however since the reactor is directly coupled to the turbine you have the added complication of turbine protection systems that can tripp you out.
Matte
 
Posts: 17
Joined: 09 Mar 2011, 18:56

Re: Answering my own question

Postby zl2wrw » 26 May 2011, 11:00

Hi,

Under "Other Design Improvements" on page 14 of http://www.nuclear.gov/np2010/pdfs/esbwrOverview.pdf (that document is dated 2006)
It says "100% Steam Bypass > Island Mode of Operation", so it looks like yes, you can run an ESBWR without outside grid power by running the reactor at full rated power and bypassing most of the steam direct to the condenser, although this may only considered to be safe because General Electric claim that the ESBWR "passive safety systems" work without electric power for 72 hours (page 9)?
Page 11 says that the ESBWR only needs two on-site diesel generators which do not have to be safety rated compared to the three safety rated generators required for an ABWR reactor.

Am I correct in thinking that to keep the passive cooling system working beyond the inital 72 hours, all that has to be done is to keep adding water to the PCC and IC pools so that they do not boil dry, (pages 16 - 17, 23 - 27) as the water in these tanks is evaporated to atmosphere in order to remove shutdown decay heat from the core & RCV?



Cheers
Ross Whenmouth
zl2wrw
 
Posts: 5
Joined: 09 May 2011, 11:00

Re: Reducing the effects of Loss Of Off-site Power?

Postby Matte » 09 Jun 2011, 15:18

100% steam bypass still implies that you have the pumps for the main condenser running, otherwise you will run out of water in the reactor pretty darn quickly. With 100% steam bypass you are not producing any power from the plant and are relying on diesel generators or offsite power to run your equipment. ESBWR may have passive capability, for how long I can't tell you as I don't know.
Matte
 
Posts: 17
Joined: 09 Mar 2011, 18:56

Re: Reducing the effects of Loss Of Off-site Power?

Postby lordsigma » 31 Aug 2011, 01:22

The ESBWR and I think the AP1000 too are designed to support a main generator island mode of operation. I know the Turbine bypass system in the ESBWR is designed to prevent turbine (and reactor) trip on up to 100% load rejection and also to prevent a reactor scram after a turbine trip. I don't know if this works when all offsite power is cutoff or not.
lordsigma
 
Posts: 2
Joined: 31 Aug 2011, 00:51

Re: Reducing the effects of Loss Of Off-site Power?

Postby lordsigma » 31 Aug 2011, 05:06

Actually all the new variants can handle 100% load rejection as well as turbine trip without a reactor trip. It uses a combination of the turbine bypass and rapidly inserting control rods to quickly reduce reactor power to low levels. Only the ESBWR and AP1000 support the "island" mode we're talking about where the plant is totally disconnected and powering house loads only. To get there with a full grid disconnect (and with the reserve offsite power source also down) the rapid power down has to succeed without the turbine tripping or power to all the needed pumps will cease and the reactor will trip and the voltage output obviously has to stay within acceptable levels for the pumps while the power down is going on. The EPR and APWR can handle a grid disconnect without a trip but do not support the island mode of operation. In the APWR it's physically impossible as both power supplies to the plant loads are offsite only and have no connection to the ouput of the unit main generator.
lordsigma
 
Posts: 2
Joined: 31 Aug 2011, 00:51


Return to The Study Hall

Who is online

Users browsing this forum: No registered users and 2 guests

cron