Why did Plymouth Pilgrim power plant shut down


by Naty1
Tags: pilgrim, plant, plymouth, power, shut
Naty1
Naty1 is offline
#1
Feb9-13, 11:22 AM
P: 5,634
The Plymouth Pilgrim [Mass] nuclear power plant apparently shut down and stayed down during the nasty northeaster of last night:

The Pilgrim Nuclear Power Plant in Plymouth shut down after losing off-site power. Authorities say there’s no threat to public safety.
Since when is a nuclear POWER PLANT subject to 'offsite power' losses?? And even if it shut
down temporarily, can't a nuclear power plant be operated via on site power or emergency diesel power??


http://boston.cbslocal.com/2013/02/0...r-power-plant/
Phys.Org News Partner Science news on Phys.org
Lemurs match scent of a friend to sound of her voice
Repeated self-healing now possible in composite materials
'Heartbleed' fix may slow Web performance
russ_watters
russ_watters is online now
#2
Feb9-13, 12:38 PM
Mentor
P: 21,994
That's par for the course. It is a safety requirement: yes, it could be powered from on-site* or diesel power, but there are very stringent redundancy requirements and loss of external power calls for shutdown.
If there is a loss of off-site power, reactors automatically shut down as a precaution and the emergency backup diesel generators begin operating to provide electrical power to plant safety systems.
http://www.nei.org/resourcesandstats...cted-against-h

[edit]*Actually, I'm not sure if they can power their own systems, with no grid connection. The sizing mismatch between generator/turbine size and load would be problematic. Remember: the turbine is part of the cooling of the reactor. If there is little electrical load on the generator, the plant will overheat.
Naty1
Naty1 is offline
#3
Feb9-13, 01:37 PM
P: 5,634
Maybe they anticipated loss of the power distribution network....many thousands did lose power in Mass.

That the operational safety of a plant depends on 'off site power' sure seems odd without knowing any details. Sounds like maybe a combination of bureaucracy and maybe environmental over zealousness has intruded here.

gmax137
gmax137 is offline
#4
Feb9-13, 02:36 PM
P: 819

Why did Plymouth Pilgrim power plant shut down


Big power units in general do shut down on a loss of the grid. Not just nukes, but the big coal fired units as well. This is not unexpected since when the grid goes away there is nowhere for the power being produced to go, so the plants are turned off. The normal design has the large "house loads" energized from the grid. At nuclear plants these include the pumps normally circulating water through the reactor and the pumps circulating water through the condenser. Neither of these is needed for plant safety with the reactor shut down. The pumps and other systems that are needed under these conditions are powered by the diesel generators and/or by very small safety-grade steam turbines that run off the decay heat produced in the shut-down reactor.
Astronuc
Astronuc is offline
#5
Feb9-13, 04:22 PM
Admin
Astronuc's Avatar
P: 21,625
Quote Quote by Naty1 View Post
The Plymouth Pilgrim [Mass] nuclear power plant apparently shut down and stayed down during the nasty northeaster of last night:

Since when is a nuclear POWER PLANT subject to 'offsite power' losses?? And even if it shut
down temporarily, can't a nuclear power plant be operated via on site power or emergency diesel power??

http://boston.cbslocal.com/2013/02/0...r-power-plant/
From the article, " A massive snowstorm packing hurricane-force winds has knocked out power to hundreds of thousands of . . . ", and it appears the plant lost connection with the grid.

Nuclear plants along the Atlantic coast routinely power down, or shutdown, when a hurricane or severe storm approaches, just in case the grid goes down. When the grid is down, there is nowhere to send the power. At startup, turbines are synched with the grid around 15% of full power, give or take. At low power, the steam would have lower quality, i.e., would be rather wet coming out of the turbines. Water droplets tend to cause erosion, which is undesirable.

When the plant shutdown, decay heat from the reactor (about 7-8% at shutdown) has to be removed. The decay heats 'decays' rapidly in time as the short-lived radionuclides decay in seconds, minutes, and hours.

http://www.nrc.gov/reading-rm/doc-co...3/13-003.i.pdf
jim hardy
jim hardy is offline
#6
Feb9-13, 05:12 PM
Sci Advisor
jim hardy's Avatar
P: 3,134
In that circumstance one shuts down his nuke plant for the same reason he puts his car in the garage and shuts its engine off.

As others have said, the plant consumes about 5% of its own capacity for internal pumps and stuff.
No need of running that equipment when you can't make any power to sell because the grid is down.
It's prudent to take the plant most of the way cold beforehand so that, should grid go down, you don't have to make that cooldown maneuver "in the dark".

We had about 36,000 horsepower of huge pumps,
two @ 7,000hp for boiler feed, 3 @ 6,000hp for reactor, 4@~1000hp for condenser water ,
used during normal operation but not at shutdown..
When shutdown you run on much smaller pumps that are hundreds of horsepower not thousands.
It takes a few hours to get the plant cooled down and valves aligned for those smaller pumps and heat exchangers.
One prefers to do that before it becomes urgent.

If you've shut down and the grid stays together, that's fine, you start back up after the storm.
If you've already shut down and the grid disappears, no big deal you're already aligned and running on the smaller shutdown equipment.

That's prudent operation.
Don't try to drive through a hurricane or blizzard.

old jim
Naty1
Naty1 is offline
#7
Feb9-13, 05:20 PM
P: 5,634
Big power units in general do shut down on a loss of the grid.
a lot better description than that in the news report....
Hiddencamper
Hiddencamper is offline
#8
Feb10-13, 08:55 AM
P: 164
The more technical response:

First, as many are saying, the loss of power causes you to lose house loads. This means all plant systems were unpowered for a short amount of time. All of the non-safety systems, like the condenser, the condenser cooling water, the turbine, feedwater, instrument air, reactor recirculation and cooling pumps all went offline and did not come back. Non-safety systems are on a separate bus. Some plants have connections to manually restore these systems using the generators, but not all.

The safety/class 1E systems were all restored by the diesel generators. This includes ECCS, spent fuel pool cooling, residual heat removal, battery chargers for the DC power distribution system, incore monitoring. The diesel generators will autostart in emergency mode on a loss of power signal, and within 10-15 seconds of the loss of power they will start restoring safety-grade power to the plant.

Now what actually caused the scram on loss of power, and what will answer OPs question:

First, based on what I've seen, the loss of power clearly took out the main power transformer output. You need to think of the plant as a big continuous system. The reactor makes steam, steam goes through the turbine in the form of mechanical energy, the turbine transfers mechanical energy to the generator which converts it into electrical energy which then goes out to the grid. A loss of power means the generator has nowhere to send power to, and could also mean an electrical fault which can damage the generator. The generator immediately locks out (shuts itself off). Without anyplace for the turbine to transfer its mechanical energy to, it would overspeed and catastrophically destroy itself, so the turbine will automatically initiate a turbine trip, which closes all turbine steam stop valves and turbine steam control valves. Now we go back one more step, to the reactor and main steam lines. Without any place for the reactor to transfer its steam to, you are going to get a very large pressure buildup. In fact, when the turbine valves close, the steam becomes a shock wave which bounces off the valves and travels back into the reactor, causing a large pressure change, followed by a large neutron flux spike (in a BWR, an increase in pressure causes an increase in the boiling point of water, which causes steam to 'collapse' back into the liquid phase, which increases moderation, and causes a prompt neutron flux spike. Peak spike is over 180% of rated reactor flux/power). This is clearly undesirable, so reactor protection system (the scram system), is designed to sense when the turbine valves close, and will automatically scram the reactor the moment that occurs. This is most likely the first scram signal which actually shut the core down.

With the loss of power, the loss of condenser, the reactor still needs to remove steam. Under loss of power and isolation conditions, which is what the pilgrim plant was in, pilgrim will start up their RCIC/HPCI steam driven water injection pumps. Steam will be vented to the suppression pool either through some combination of the safety-relief valves and the steam discharge from the RCIC/HPCI turbines to slowly reduce reactor pressure until the reactor is below the shutdown cooling interlock (~100 PSI) at which point the shutdown cooling system will be placed into service to cool the core down to 90~120 degrees F.

There are a few more details than this, but its the overall picture that counts.

Here's a very terrible mspaint diagram I made of the main steam system:
http://imgur.com/wVbKBZM

Note that the bypass valves in that picture are non-functional due to the loss of power.

I'm an engineer in BWRs, so if there are any detailed-technical type questions let me know.
jim hardy
jim hardy is offline
#9
Feb10-13, 10:41 AM
Sci Advisor
jim hardy's Avatar
P: 3,134
Some early PWR's were built so that it would be physically possible for them to "island" and carry their own in-house load following a loss of grid. To that end, all internal equipment is powered during operation from an auxilliary transformer that taps power from main generator leads between the generator and the stepup transformer to grid. A different transformer powers internal equipment from the grid for starting up.

See pages 3 and 8 here:
http://www.pdhonline.org/courses/e184/e184content.pdf

Administrative limitations now pretty well preclude 'island' operation.

But 'way back when' the capability was designed in.
Hiddencamper
Hiddencamper is offline
#10
Feb10-13, 12:01 PM
P: 164
Quote Quote by jim hardy View Post
Some early PWR's were built so that it would be physically possible for them to "island" and carry their own in-house load following a loss of grid. To that end, all internal equipment is powered during operation from an auxilliary transformer that taps power from main generator leads between the generator and the stepup transformer to grid. A different transformer powers internal equipment from the grid for starting up.

See pages 3 and 8 here:
http://www.pdhonline.org/courses/e184/e184content.pdf

Administrative limitations now pretty well preclude 'island' operation.

But 'way back when' the capability was designed in.
BWRs are capable of being operated this way as well (if they designed it into their plant). I don't think any BWR in the US has the ability to start in island mode. Island mode was an option that GE offered to BWRs, it would allow them to operate with low loads and start in black-start conditions. However, even if you had island mode, the plant would still scram during a load reject. US BWRs are not designed to stay online with load rejects above about 30% power. (In other words, the core could not back off steam supply fast enough to deal with the sudden drop from 800+ MWe down to 30 MWe, and the scram would still be required to prevent a prompt neutron flux spike in the core).

I know the BWR/6 plants were offered an option called "REVABS", or "RElief Valve Augmented Bypass System", which would allow them to survive a load reject. No US plant has this feature though.

How REVABS works, (or would work if it was ever installed), is that during a load reject, a select number of control rods would individually scram and the recirculation pumps would trip to low speed (similar to a power cutback in a PWR). Then, a select number of safety relief valves would lift in power operated mode for up to 10 seconds, using the suppression pool as a temporary heatsink. After 10 seconds, reactor power would be low enough to keep the core critical using just the main condenser bypass valves and the safety relief valves would reseat. If power was not low enough after 10 seconds, the core would do a full scram similar to any other load reject scenario.

That said, it also depends on how the grid fault occurred. If the grid fault only caused the main power output breakers to open, then yes, if you had both REVABS and Island Mode, you might be able to keep the core critical running its own house loads. However, if the fault caused any form of lockout on the main power system, then the generator and turbine would have tripped and even if you had REVABS, the loss of power would cause a loss of condenser vacuum, followed by a group 1 isolation (Main steam isolation valve closure) and an MSIV closure scram.
mheslep
mheslep is online now
#11
Feb11-13, 05:14 PM
PF Gold
P: 3,021
Quote Quote by Astronuc View Post
...When the plant shutdown, decay heat from the reactor (about 7-8% at shutdown) has to be removed. The decay heats 'decays' rapidly in time as the short-lived radionuclides decay in seconds, minutes, and hours.

http://www.nrc.gov/reading-rm/doc-co...3/13-003.i.pdf
And for days, weeks, and months, with a megawatt of decay heat still lingering six months later.
mfb
mfb is offline
#12
Feb11-13, 05:18 PM
Mentor
P: 10,766
Compared to ~2GW during regular operation, ~10MW after a day is a rapid decay.

And years, decades, centuries, and millenia, but again with a power decreased by some orders of magnitude.
mheslep
mheslep is online now
#13
Feb11-13, 05:18 PM
PF Gold
P: 3,021
Quote Quote by jim hardy View Post
Some early PWR's were built so that it would be physically possible for them to "island" and carry their own in-house load following a loss of grid. To that end, all internal equipment is powered during operation from an auxilliary transformer that taps power from main generator leads between the generator and the stepup transformer to grid. A different transformer powers internal equipment from the grid for starting up.

See pages 3 and 8 here:
http://www.pdhonline.org/courses/e184/e184content.pdf

Administrative limitations now pretty well preclude 'island' operation.

But 'way back when' the capability was designed in.
In "island" mode was it possible to generate enough steam from decay heat to generate power to run the pumps and remove decay heat, possibly intermittently?
mheslep
mheslep is online now
#14
Feb11-13, 05:22 PM
PF Gold
P: 3,021
Quote Quote by mfb View Post
Compared to ~2GW during regular operation, ~10MW after a day is a rapid decay.

And years, decades, centuries, and millenia, but again with a power decreased by some orders of magnitude.
Of course, but as Fukushima demonstrated, not decreased enough as the couple megawatts were to sufficient to cause heat related destruction of an otherwise shutdown reactor and subsequent release to the environment.
Hiddencamper
Hiddencamper is offline
#15
Feb11-13, 05:31 PM
P: 164
Quote Quote by mheslep View Post
In "island" mode was it possible to generate enough steam from decay heat to generate power to run the pumps and remove decay heat, possibly intermittently?
Nope. The main generator will automatically shut down anywhere from 10-20 minutes post scram due to not enough steam load. One of the post scram checklist requirements is to verify the generator did indeed trip itself, otherwise the grid turns the generator into a motor and reverse powers it (not good). Operators are supposed to manually trip the turbine/generator set either right before or right after this occurs (I've seen procedures both ways at plants).

You should have enough steam to run the RCIC pump on decay heat for over 72 hours though. RCIC is a small flow controlled 600 gpm steam driven pump. We know this because Fukushima Daiichi unit 2 ran for 70 hours on its RCIC, and I personally know that Clinton power station did this several times in the '90s for days. It's part of the reason they got shut down for 2+ years.

On a side note, RCIC only injects water to the reactor. It does not remove decay heat from the containment system. (Again, we saw this at Fukushima Daiichi unit 2)
jim hardy
jim hardy is offline
#16
Feb11-13, 06:25 PM
Sci Advisor
jim hardy's Avatar
P: 3,134
Quote Quote by mheslep View Post
In "island" mode was it possible to generate enough steam from decay heat to generate power to run the pumps and remove decay heat, possibly intermittently?
Island mode is exactly that, reactor power low enough to make just enough steam for turbine to carry the plant's internal equipment. Turbine bypass valves dump excess steam to condenser if need be. In theory you could island forever.

The biggest problem in PWR is with controlling water level in the steam generators (boilers). The boilers are tall skinny affairs and level swings up and down quite a bit with pressure changes. That's because of the bubbles in the steam water mix, just like in a BWR. At such low power the automatic controls and their valves aren't operating because you are using smaller manual valves.
One usually experiences a reactor trip (scram in BWR terminology) from inability to control steam generator level precisely.
We contemplated automating those smaller valves but never did it. R&D projects just aren't justifiable on an operating plant.


These days there is something called "Degraded Grid Protection" which is an automatic subsystem intended to disconnect the plant from the grid should the grid voltage get low . Utilities had some leeway in just how to implement that and i believe many plant designers chose to just abandon a troubled grid.
However it was done, a plant will automatically place itself in a safe condition should the grid disappear. That could be "islanded" if designers chose to try and hang on.

Of course it's moot if equipment necessary to operate, like feedwater or condenser cooling pumps, is powered from the grid instead of auxilliary transformer.

Assuming the grid goes down and you don't island therefore lose the big pumps, there are steam driven pumps to feed the boilers for cooldown. When you have cooled down enough you can no longer make steam, you no longer need those pumps. By then the plant is in a condition the diesels can power everything you need.

Since you're interested, you might enjoy exploring this site - it is intended for interested folks.
http://www.nucleartourist.com/

old jim
rmattila
rmattila is offline
#17
Feb12-13, 04:46 AM
P: 242
In some European BWRs, the strategy in load rejection is to perform a partial scram (about 10 rods are scrammed, another 10 screwed in within a couple of minutes) to reduce power to around 30 %, switch the turbine controller from pressure to RPM mode, and let the steam bypass valves take over the pressure regulation duty. The plant thus takes itself to islanding mode. In practice, success rate to avoid full scram in load rejection has been around 50 %.


Register to reply

Related Discussions
generator from a power plant delivering electric power Introductory Physics Homework 1
Power output of a coal burning power plant Introductory Physics Homework 2
How to design a nuclear power plant reactor power control system ? Nuclear Engineering 25
Power plant Introductory Physics Homework 5
A power plant that uses more power than it generates General Engineering 30