Storing nuclear or thermal electricity
disgoftunwells
energy storage technologies, such as pumped storage. Proposals for
solar power systems now often involved storing the solar heat in a
molten salt so that electricity can be produced in the evening.
But it doesn't matter where the heat is stored. So the question is,
could heat storage be used in a nuclear power station, to effectively
decouple the nuclear side from the power side?
Lets say a 3GW (thermal) reactor - of any type, but this works best
with a high temperature reactor - is used to heat and melt salt. I
can't find the latent heat of fusion of NaCl, but assuming 4E5 J/kg,
such a reactor could melt about 600,000 tons, or 300,000m3 per day of
salt. A "cube" 100m on a side could store 10 GW days of heat, or 4GW
days of electricity (at 40%).
By comparison, Britain's biggest pumped storage facility, Dinorwig,
can hold 9 GWhrs of electricity, or 1/10th as much.
Electrical generators would tap the heat when needed. A variety of
baseload generators and peak generators could tap the electricity. If
one side of the system fails for a short while, the other side can
keep going for some time. Even a refuelling could be done without
interrupting the power supply.
Expensive? Perhaps, but seems better value than pumped storage (though
pumped storage is probably still needed for it's slew, or speed of
response - Dinorwig can switch on, from 0 to 1.3 GW power, in 12
seconds.).
Is this being considered anywhere for 4th generation reactors, or are
the developers convinced that they have to supply baseload power?
Rob Dekker
"disgoftunwells" <disgoft...@yahoo.co.uk> wrote in message
news:5c214bfb-6aaf-4042...@o7g2000yqb.googlegroups.com...
Interesting thought.
Power storage normally makes sense if the power source is intermittent (such
as with solar or wind).
In such cases, the generator (turbine) is not used to it's full potential,
and thus there is inefficiency.
Nuclear power plants have a very reliable power source, being a baseload
provider.
So it's turbines will always run at full capacity, and there is little
incentive to change that.
Let me explore this a bit :
If we were to equip a nuclear plant with heat storage capacity, then we
would need to make the generators (turbines) larger than the full capacity
of the plant. Otherwise it could never off-load the extra (stored) power.
The extra cost would thus be the heat storage facility and the larger
turbines.
Also, by adding heat storage and control systems, there will be some losses
involved.
On the 'benefit' side, the nuclear plant can now react to instant changes in
power demand, and thus can take advantage of the higher prices paid at
'peak' power usage.
Overall, it will thus depend on the pricing schemes used on the grid if this
set-up will be economically more beneficial to do heat storage at nuclear
plants, or installing heat storage at (intermittent) solar plants, or
separate pumped storage facilities.
I think the economics on these tradeoffs are not yet established, so I think
you have a very viable concept at this time.
It would be interesting to see more detailed cost/benefit analysis on this.
Thanks for posting an interesting idea !
Rob
disgoftunwells
> "disgoftunwells" <disgoftunwe...@yahoo.co.uk> wrote in message
Yes, probably about 3GW of turbines for a 1.2GW average. For
comparison, natural gas CCGT are about £300-400 million / GW.
So, an extra 1.8GW of generator might cost £500 million, and perhaps
£100 million for the storage facility. In dollar, perhaps $1 billion
to convert a nuclear plant into a fully flexible generator.
> Also, by adding heat storage and control systems, there will be some losses
> involved.
Yes - though the larger the system the lower they are as a percentage.
This might also mean it works better with low temperature reactors,
with a salt that melts at temperatures a PWR can acheive.
> On the 'benefit' side, the nuclear plant can now react to instant changes in
> power demand, and thus can take advantage of the higher prices paid at
> 'peak' power usage.
>
and if one side fails (or is being refuelled) the other side can
continue.
> Overall, it will thus depend on the pricing schemes used on the grid if this
> set-up will be economically more beneficial to do heat storage at nuclear
> plants, or installing heat storage at (intermittent) solar plants, or
> separate pumped storage facilities.
>
> I think the economics on these tradeoffs are not yet established, so I think
> you have a very viable concept at this time.
> It would be interesting to see more detailed cost/benefit analysis on this.
> Thanks for posting an interesting idea !
>
Your points are all spot on. But no one seems to think of the bigger
picture, but their own source:
"These wind turbines will need a stand alone gas generator in case of
no wind ......"
"This solar thermal station will need heat storage for night time
use ....."
As all these systems are connected, where's the best place to put the
energy storage?
One bigger picture view point is the idea of using car batteries to
store surplus electricity. But is it easier and cheaper to store
energy as heat or electricity?
rlbell
> "disgoftunwells" <disgoftunwe...@yahoo.co.uk> wrote in message
Nuclear reactors do have trouble following load swings. A good
storage technology would allow to a nuclear plant to bank energy when
demand is low and release it when demand ramped up. It is pretty much
the death knell of wind and solar as it would reduce the peaking rate
to only a little higher than the baseload rate, so only baseload
powerplants would be economical.
>
> Let me explore this a bit :
> If we were to equip a nuclear plant with heat storage capacity, then we
> would need to make the generators (turbines) larger than the full capacity
> of the plant. Otherwise it could never off-load the extra (stored) power.
> The extra cost would thus be the heat storage facility and the larger
> turbines.
> Also, by adding heat storage and control systems, there will be some losses
> involved.
> On the 'benefit' side, the nuclear plant can now react to instant changes in
> power demand, and thus can take advantage of the higher prices paid at
> 'peak' power usage.
>
> Overall, it will thus depend on the pricing schemes used on the grid if this
> set-up will be economically more beneficial to do heat storage at nuclear
> plants, or installing heat storage at (intermittent) solar plants, or
> separate pumped storage facilities.
The heat storage would cost the same at either type of plant, but the
baseload electricity costs less.
>
> I think the economics on these tradeoffs are not yet established, so I think
> you have a very viable concept at this time.
> It would be interesting to see more detailed cost/benefit analysis on this.
> Thanks for posting an interesting idea !
>
> Rob
As a way of reducing electricity costs, it is great. As a way of
making internittents competitive, it fails-- unless you pass
regulation that forbids baseload plants from using it.
disgoftunwells
What's the price of nuclear electricity, regardless of when it's
produced?
What's the price of wind / tidal / solar thermal, regardless of when
it's produced?
Now attribute the cost of the thermal storage system to the (non
nuclear) renewables, and you'll probably find they're more expensive.
However, we don't know the cost of 3rd generation nuclear, because no
one's finished one yet (and I hope the Finnish example is a one off
lessons-learned). Equally, no one knows the cost of electricity from a
20 GW offshore wind farm, because no one has planned or built one.
In reality, both nuclear and renewables are going to be needed. (If
the economy moves to an electric footing, the UK alone will need over
100GW average power), A nuclear thermal store could help
"intermittents", but normally the two industries are busy slagging
each other off.
rlbell
Nuclear power is only slightly more expensive than coal-fired
generation.
> What's the price of wind / tidal / solar thermal, regardless of when
> it's produced?
>
I would be pleasantly surprised if it was only three times as
expensive as nuclear.
> Now attribute the cost of the thermal storage system to the (non
> nuclear) renewables, and you'll probably find they're more expensive.
>
> However, we don't know the cost of 3rd generation nuclear, because no
> one's finished one yet (and I hope the Finnish example is a one off
> lessons-learned). Equally, no one knows the cost of electricity from a
> 20 GW offshore wind farm, because no one has planned or built one.
Third generation should (on paper) be noticeably cheaper than second
generation, or you could not convince anyone to build one, but actual
costs are currently unknown. High temperature, gas cored reactors
have an easily achievable thermal efficiency of 70%, so they would be
a bit more than half the cost per kilowatt*hour of a PWR.
>
> In reality, both nuclear and renewables are going to be needed. (If
> the economy moves to an electric footing, the UK alone will need over
> 100GW average power), A nuclear thermal store could help
> "intermittents", but normally the two industries are busy slagging
> each other off.
Why would both be needed except for renewables to be a stopgap measure
while nuclear was brought on line?
If the UK needs 100GW of power, the cheapest solution by a longshot
(if thermal storage solves the load following problem) is gas-fired
combined cycle plants (not cheap, but able to be quickly built)
replaced by nuclear as it comes online. If nuclear plants can be
commissioned quickly enough, you build nukes from the start. Wind/
solar/tidal production is a distraction that incurs costs for little
benefit.
@libero.it Romeo Gigli
"rlbell" <rlbell...@gmail.com> ha scritto nel messaggio
news:30c835f8-d1c4-419e...@a37g2000prf.googlegroups.com...
On Jul 8, 10:29 am, disgoftunwells <disgoftunwe...@yahoo.co.uk> wrote:
>Third generation should (on paper) be noticeably cheaper than second
generation, or you could not convince anyone to build one, but actual
>costs are currently unknown. High temperature, gas cored reactors
have an easily achievable thermal efficiency of 70%, so they would be
>a bit more than half the cost per kilowatt*hour of a PWR.
70 % efficiency is physically impossible to achieve in a nuclear reactor,
even in high temp gas *cooled* (not cored) reactors
Rob Dekker
On 8 July, 10:24, "Rob Dekker" <r...@verific.com> wrote:
> "disgoftunwells" <disgoftunwe...@yahoo.co.uk> wrote in message
> > I think the economics on these tradeoffs are not yet established, so I think
> > you have a very viable concept at this time.
> > It would be interesting to see more detailed cost/benefit analysis on this.
> > Thanks for posting an interesting idea !
> >
> Your points are all spot on. But no one seems to think of the bigger
> picture, but their own source:
>
> "These wind turbines will need a stand alone gas generator in case of
> no wind ......"
> "This solar thermal station will need heat storage for night time
> use ....."
>
> As all these systems are connected, where's the best place to put the
> energy storage?
>
And that may be the case, although some plants (such as solar thermal) will want to have some form
of storage for logistical reasons : You just don't want to shut down your plant when some clouds move in for a few hours.
Other than that, assuming the grid can handle the excess loads to/from each plant, it should not matter much WHERE the storage is
done.
That's why I like your idea.
> One bigger picture view point is the idea of using car batteries to
> store surplus electricity. But is it easier and cheaper to store
> energy as heat or electricity?
Typically, when storing energy for no longer than a day or two, it is WAY cheaper to use heat storage (especially high-heat like
with liquid salt) than it is to use batteries or even pumped storage.
But note that battery powered systems are only used for phase stabilisation, and in that sense have a different purpose than
spinning nat.gas. reserves or other heat-driven plants (like your liquid storage at nuclear plants).
Rob
P.S. Here is one link of how liquid salt storage is typically used in a solar thermal plant :
http://www.sandia.gov/Renewable_Energy/solarthermal/NSTTF/salt.htm
Rob Dekker
"rlbell" <rlbell...@gmail.com> wrote in message news:30c835f8-d1c4-419e...@a37g2000prf.googlegroups.com.........
> > What's the price of nuclear electricity, regardless of when it's
> > produced?
>
> Nuclear power is only slightly more expensive than coal-fired
> generation.
Maybe that is true for plants in operation (and investment pretty much paid off),
but it is not certain for plants that would be newly built.
Do we have investment cost (per GW) for currently built plants ?
The US is not building anything right now, but other countries do. China for example, and India is experimenting heavily too...
I have seen some press releases of at least plans for US built nuclear plants, and they seem to be in the $2-$5/watt investment
cost.
That's very significant, and combined with the long lead time (10 years?) to get a plant up and running, it is not
surprising that investors prefer wind (at $2/watt) and solar ($2-$4/watt), even though these have intermittent power sources.
.......
>
> Third generation should (on paper) be noticeably cheaper than second
> generation, or you could not convince anyone to build one, but actual
> costs are currently unknown.
...
> > High temperature, gas cored reactors
> > have an easily achievable thermal efficiency of 70%, so they would be
> > a bit more than half the cost per kilowatt*hour of a PWR.
40-50% is about as good as it gets even for high temp reactors.
The Carnot limit applies, remember ?
>
> > In reality, both nuclear and renewables are going to be needed. (If
> > the economy moves to an electric footing, the UK alone will need over
> > 100GW average power), A nuclear thermal store could help
> > "intermittents", but normally the two industries are busy slagging
> > each other off.
>
> Why would both be needed except for renewables to be a stopgap measure
> while nuclear was brought on line?
> If the UK needs 100GW of power, the cheapest solution by a longshot
> (if thermal storage solves the load following problem) is gas-fired
> combined cycle plants (not cheap, but able to be quickly built)
> replaced by nuclear as it comes online. If nuclear plants can be
> commissioned quickly enough, you build nukes from the start. Wind/
> solar/tidal production is a distraction that incurs costs for little
> benefit.
You mention the answer to this yourself : there is still great uncurtainty about cost and feasibility.
That is not surprising, since the timeline for commisioning the first 3rd gen plants (such as thorium breeders) is still estimated
to be 20-30 years in the future.
Meanwhile, renewables will not sit still, and cost for these is bound to go down significantly with volume production.
But notwithstanding these developmental problems, once the unknowns become better known, and remaining issues are solved, 3rd gen
plants have enormous potential..
Thanks for your post
Rob
Bret Cahill
> > > produced?
>
> > Nuclear power is only slightly more expensive than coal-fired
> > generation.
>
> Maybe that is true for plants in operation (and investment pretty much paid off),
> but it is not certain for plants that would be newly built.
> Do we have investment cost (per GW) for currently built plants ?
It would have to be at least an order of magnitude higher to even
appear in any cost analysis comparing grid powered transportation to
liquid fuel.
It's just not an issue.
Just start building whatever the locals will let you build.
Bret Cahill
z
> > What's the price of nuclear electricity, regardless of when it's
> > produced?
>
> Nuclear power is only slightly more expensive than coal-fired
> generation.
given a pre-existing plant (and subsidized insurance). but if you have
to include construction costs of a new plant, nuclear is hugely
expensive, even amortized over the life of the plant.
>
> > What's the price of wind / tidal / solar thermal, regardless of when
> > it's produced?
>
> I would be pleasantly surprised if it was only three times as
> expensive as nuclear.
they're all in the same ballpark as coal/oil/nuclear (from existing
plants) except for photoelectric, which is very expensive. however,
focusing the sun to produce steam for a generator is competitive.
The DOE estimates wind power delivering electricity for 6-8.5 cents/
kwh, (without government subsidies), compared to 15 cents/kwh for new
nuke plants (in addition to heavy government subsidies). That includes
the cost to tie the power generators into the grid, half a cent per
kwh. http://www1.eere.energy.gov/windandhydro/pdfs/41869.pdf
disgoftunwells
Just a thought, the 100m cube I described above would be for 4GW days
of electricity. If the nuclear power station simply wants to avoid
selling electricity between 11pm and 6am, then only 1/10th the storage
(about 100,000m3) is needed, so a 45m "cube". This is smaller than
most LNG containers (about 200,000m3).
Rob Dekker
Good point.
Indeed, for nuclear plants riding the dayly demand wave, there would be little incentive to store than 7 hours or so.
One note : you need two of these cubes, one hot and one 'cold'.
By the way, what did you assume as the temperatures for hot and cold ?
These temperatures determine the storage capacity and also fix some of the heat cycle parameters of the system.
Rob
disgoftunwells
the latent heat is the store. What salt combination is chosen depends
on the reactor design. High temperature reactors could be used to melt
NaCl (801C), but most reactors would need a lower temperature salt. I
suspect at low temperatures (PWRs) the efficiency hit would be too
great.
The reactor coolant would be pumped through the base of the stack,
with convection heat melting it.
At the top, helium would be heated to drive combined cycle turbines.
There isn't a need for a cold cube, as water would do, as in most
reactors.
The only engineering issue I can think of is salt solidifying around
the helium pipes. There would also need to be redundant heating tubes
as repairs would need a full cooling of the stack. (one month down in
20 years?).
Rob Dekker
Ah ! Sorry I misunderstood !
Now I see : you only use only the heat of fusion of NaCl for storage. So the tank stays at 800 C all the time.
Then you can indeed use a single tank system.
In the dual tank liquid salt storage systems that I'm familiar with, the salt is always liquid, both hot and cold, and the stored
heat is determined by the temperature difference between the hot and the cold tank, multiplied by the heat capacity of the salt.
> What salt combination is chosen depends
> on the reactor design. High temperature reactors could be used to melt
> NaCl (801C), but most reactors would need a lower temperature salt. I
The single tank system is simpler in design, but is less flexible. Temperature and maximum storage capacity are fixed.
The dual tank system is more flexible for setting upper and lower temperatures.
For wide temperature differences (high efficiency turbines) will require less mass of salt per kWh stored than a single tank system
> suspect at low temperatures (PWRs) the efficiency hit would be too
> great.
Yes. In general, higher temperature is better for efficiency.
But each turbine/generator system is designed for a certain optimal operating temperature.
You would just need to find a storage system that works with the turbine/generator parameters (and primary system temperatures).
>
> The reactor coolant would be pumped through the base of the stack,
> with convection heat melting it.
>
> At the top, helium would be heated to drive combined cycle turbines.
> There isn't a need for a cold cube, as water would do, as in most
> reactors.
>
> The only engineering issue I can think of is salt solidifying around
> the helium pipes. There would also need to be redundant heating tubes
> as repairs would need a full cooling of the stack. (one month down in
> 20 years?).
I'm not sure if you need to design the whole system.
The concept you propose (heat storage at nuclear plants) is good by itself.
But system engineers may want to consider the two-tank system as well, depending on the system requirements.
>
disgoftunwells
Yes - it can go a little above the melting point, but not below. Not
sure what percent could be liquid.
> Then you can indeed use a single tank system.
>
> In the dual tank liquid salt storage systems that I'm familiar with, the salt is always liquid, both hot and cold, and the stored
> heat is determined by the temperature difference between the hot and the cold tank, multiplied by the heat capacity of the salt.
>
> > What salt combination is chosen depends
> > on the reactor design. High temperature reactors could be used to melt
> > NaCl (801C), but most reactors would need a lower temperature salt. I
>
> The single tank system is simpler in design, but is less flexible. Temperature and maximum storage capacity are fixed.
One tank needs to be cold, and there's not much (practical) colder
than the sea.
For solar thermal applications,you could have a "cold" tank full of
water. Where water is scarce (the Sahara), at night time, this could
be pumped around the solar collectors to cool it overnight.
> The dual tank system is more flexible for setting upper and lower temperatures.
> For wide temperature differences (high efficiency turbines) will require less >mass of salt per kWh stored than a single tank system
Only because of the carnot efficiency. The actual stored thermal
energy is mostly in the latent heat of the salt. By keeping the
temperature of this constant, you can maximise efficiency.
In other words, if you have 100% liquid above the melting temperature,
you lose the ability to transfer heat from the reactor coolant to the
salt. (Question, could you dispense the heat exchanger, and pump the
salt directly through the reactor. Issue - the cold spots to pump
through might a solid / liquid mix which doesn't pump). If you rely on
cooling the salt, you lose carnot efficiency.
>
> > suspect at low temperatures (PWRs) the efficiency hit would be too
> > great.
>
> Yes. In general, higher temperature is better for efficiency.
> But each turbine/generator system is designed for a certain optimal operating temperature.
>
> You would just need to find a storage system that works with the turbine/generator parameters (and primary system temperatures).
>
>
>
> > The reactor coolant would be pumped through the base of the stack,
> > with convection heat melting it.
>
> > At the top, helium would be heated to drive combined cycle turbines.
> > There isn't a need for a cold cube, as water would do, as in most
> > reactors.
>
> > The only engineering issue I can think of is salt solidifying around
> > the helium pipes. There would also need to be redundant heating tubes
> > as repairs would need a full cooling of the stack. (one month down in
> > 20 years?).
>
> I'm not sure if you need to design the whole system.
> The concept you propose (heat storage at nuclear plants) is good by itself.
Thanks - but I suspect we need to wait for molten salt or pebble bed
reactors to make it viable. But it seems the idea is sound and a lot
cheaper than pumped storage or batteries. (Interestingly this system
is on a par, weight for weight, with most battery technologies, but
has to be a lot cheaper).