Energy From Thorium Discussion Forum

It is currently Jan 22, 2018 1:36 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 33 posts ]  Go to page 1, 2, 3  Next
Author Message
PostPosted: Aug 14, 2015 6:59 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
Well - I have been doing a lot of thinking about an all-nuclear power society, however the peak winter heating loads become enormous - even with Air Source Heat Pumps, which have huge capital cost requirements.
This pushes down the capacity factor of the electricity system and requires very expensive shipping of heat around.

So I started thinking about District Heating and Nuclear CHP - but unfortunately I was of the opinion that this would require loads to be very close to power plants, and this is not really feasible on a large scale with nuclear for practical reasons. (You would never get permission to build a BWR on the Battersea Power station site or at Bankside for example.

That was until I stumbled across a plan to move a thousand megawatts of heat in the form of hot water from the proposed (but now cancelled) Loviisa 3 reactor to the Helsinki district heating system, ~80km away.
I managed to find this paper on the project running a basic simulation - that gives me a pumping power of ~50MWe to move 1000MWt the requisite distance, which is roughly the distance between central London and the site of the old Bradwell Nuclear Power station, or between the current Heysham complex and Manchester or Liverpool.

I was then trying to find out precisely how much electricity generation would be lost in the production of 1000MWt - and then I stumbled across this paper which talks about this very concept from the perspective of the EPR, although the paper says the figures for the ESBWR are very similar.
It turns out that drawing 1000MWt would lose ~125MWe in electricity production - which comes out a total loss of ~175MWe for the thermal power delivered (heat losses are minimal and compensated for by heat created by pipe flow resistance).

Whilst most modern direct-connection (with low capital costs) run on a low temperature scheme of 70C/40C hot water and it would therefore be possible to use the lower temperature case detailed in the second paper - this would only save ~30MWe which would be eaten up in massively increased pumping power because the heat value of each cubic metre of transported water would be drastically reduced (cut in half).
So I suggest the more economic case is the higher temperature water as you need smaller pipes and pumps and the heat can be used for more things downstream if it has to be (it can more easily power high performance absorption chillers for example).

Energy storage can be done far more cheaply than for electricity - indeed you can resort to a simple massive tank of water.
If we assume we use our 120C top temperature to run a secondary side accumulator at 95 Celsius (the maximum possible in a simple unpressurised tank - giving us a 55C temp change) then a MWh of usable heat fits in only 16.5 cubic metres. You can regulate the output temperature using a variable flow shunt that mixes the output water with some water from the return line.
Capital costs for tanks in the ~300 cubic metre range are apparently roughly $1560/cubic metre - but I could see us doing far better at the sort of sizes I am proposing. After all we could build a tank using the plate form concrete technologies developed for the AP1000 and go to truly ludicrous sizes.
Even so, that translates to only ~$25.60/kWh in capital costs, with a cycle efficiency of something like 99.9% - and a fifty year+ lifetime. (A tank built in Pimlico after the war recieves a ten-yearly cleaning and inspection and hasn't required major work yet).

Has batteries beaten hollow I think.

So in summary (tl:dr and so on) - I think CHP LWRs for district heating have serious promise thanks to very low heat production costs, cheap storage and the enormous capital cost of heat pumps.
(Only 175MWe lost for 1000MWt delivered to the district heating hub - even with further distribution losses you should beat heat pumps end to end, and those megawatts are still available at peak times thanks to cheap storage).

Meanwhile during the summer 95 Celsius water from the accumulator used to drive an absorption chiller can produce COPs of around 0.7, allowing for some use in district cooling. Centralised equipment should help reduce the issues with maintenance and capital cost associated with adsorption machines. And while 0.7 sounds low, that is an electric equivalent of 4 and since its from the accumulator it can be generated the night before if necessary.


Top
 Profile  
 
PostPosted: Aug 16, 2015 11:21 am 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Using the waste heat for district heating and hot water is very interesting. The amount of waste heat of LWRs is just incredible.

Do you think it can compete with coal and gas ? (since everything new nuclear is expensive in Europe and America for now)


Top
 Profile  
 
PostPosted: Aug 16, 2015 11:50 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
fab wrote:
Do you think it can compete with coal and gas ? (since everything new nuclear is expensive in Europe and America for now)


For domestic heating? It should be able to manage it if we can get the capital cost of the installations down.
If we assume the district heating network is in place the cost of the heat produced is no more than 17.5% that of the price of electricity, and indeed it will be less than that because storage systems means it will be 17.5% of the off peak price of electricity.

Prices of heat down to $10-15/MWh are probably feasible - but it would require a network already in place.

The critical thing will be reducing the capital cost of the installation of the network - which is I am more interested in direct connections using relatively low temperatures than the traditional high temperature, high pressure approach that puts a heat exchanger in every house for the normal heating. Hot water will obviously still require a heat exchanger but this can be much smaller - in fact it can simply be a coil inside an existing water cylinder if you get clever about the plumbing or tolerate a relatively high return temperature from the hot water supply. (You could route your heating demand through the hot water cylinder on the hot side, with the relatively hot water discharged from the coil doing the loop through the radiators before reaching the return line).

Advances in Smart Meters also mean that the billing costs of the system are nearly zero - and maintenance of a direct cycle system is basically limited to heat meter monitoring and repair.
So the cost is almost all capital really - it can certainly beat heat pumps but I am not so sure about gas fired boilers.


Top
 Profile  
 
PostPosted: Aug 16, 2015 2:03 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
In this website, there are prices of gas in UK
https://www.ukpower.co.uk/home_energy/compare_gas

The lowest price for 13500 kWh of gas is 440 £ i.e 688.38 $

That is 51 $/MWh for natural gas alone for individuals if I am not mistaken.


Top
 Profile  
 
PostPosted: Aug 16, 2015 2:14 pm 
Offline

Joined: Sep 02, 2009 10:24 am
Posts: 507
Are you sure the demand is so high?

A lot depends on how leaky houses are in say - 2040. But by then, a typical house, on a very cold day, might need 50KWh per day. Multiply that by 20 million for the UK, and you get 1000GWh.

Using heat pumps with a COP of 4, that is only 250GWh, or about 20GW, half the time.

That half the time would be offpeak - along with vehicle charging - another 20GW.

So, current offpeak load 40GW + 20 GW heating +20 GW vehicles = 80GW. Add 20GW margin and you have 100GW.

Thorcons, for example, can deliver that cheaply, with only low grade waste heat, perhaps useful for local greenhouse heating.


Top
 Profile  
 
PostPosted: Aug 16, 2015 2:39 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
alexterrell wrote:
Are you sure the demand is so high?

A lot depends on how leaky houses are in say - 2040. But by then, a typical house, on a very cold day, might need 50KWh per day. Multiply that by 20 million for the UK, and you get 1000GWh.

Using heat pumps with a COP of 4, that is only 250GWh, or about 20GW, half the time.

It would probably be greater than 50kWh. The bulk of the housing stock will be of similar standard to now in 2040 I would imagine - retrofits are rather expensive and there is a mis-selling scandal brewing about cavity wall insulation and the like.
The average house expends something like 10000kWh on space heating and then about 1250kWh on hot water - which is obviously spread over the whole year.

We don't need heat for that long in the UK, maybe about four months at high power - which rather drastically increases the peak power.
That is something like 85kWh or so per day.

Even off peak that will easily wash out the night time demand troughs, even with a COP of 4.
There is also Jevon's paradox to take account of - a lot more glass in houses and more, warmer, swimming pools I would imagine.

alexterrell wrote:
That half the time would be offpeak - along with vehicle charging - another 20GW.

So, current offpeak load 40GW + 20 GW heating +20 GW vehicles = 80GW. Add 20GW margin and you have 100GW.


I tried numbers on this using the National Grid's half hourly TGSD figures.
You end up with heating loads and car charging loads that are far larger than the peak/off-peak demand swing. My simplistic simulations end up with peak load being during 'Economy 7' periods.

And there is the fact that a heat pump with a COP of 4 will cost you something like £3500 plus install, and won't have that high a power.

£3500 enough to connect a house to a district heating system if you get a high enough uptake to amortise the main heating pipes and other capital plant.
alexterrell wrote:
Thorcons, for example, can deliver that cheaply, with only low grade waste heat, perhaps useful for local greenhouse heating.


My own simulations often assume a baseline of 100GWe worth of LWRs - because that is enough that we could potentially electrify almost everything.
In a country as small as the UK however the district heat argument does seem persuasive, it significantly improves our reactor's capacity factor and could cut many gigawatts off peak demand if we were to use resistive heating, or reduce capital expenditure compared to heat pumps.


Top
 Profile  
 
PostPosted: Aug 16, 2015 4:48 pm 
Offline

Joined: Sep 22, 2013 2:27 pm
Posts: 262
The main disadvantages of a nuke used for heating are...

A nuclear power plant has high Investment costs and low operation costs.

In Germany the lowest temperature is about -20°C the average winter day is +1 - +3°C. The heating season is from october to may.

That means the occupation of the high investment for the nuke and the hot water grid is low.

It will never pay of!


Top
 Profile  
 
PostPosted: Aug 16, 2015 4:54 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
There is also the hot water market to be considered.
That will be a significant draw throughout the year.
The major capital cost for a modern direct cycle system is also the pipe connections which conservatively have a fifty year lifetime. They could plausibly last the entire length of the reactor and then some.

And this is producing heat at a time where there is spare generating plant (night time) for use the next day during the peak heating demand period.

You could feasibly store days worth of heat if you ahd to.

Essentially the BWR goes to being partially a peaking plant that pumps out vast amounts of hot water at other times. (The reactor does not lose the 175MWe, it is still available at peak generation times).

EDIT:
Capital cost of a heat meter is ~$900 including installation.
In a system with a hot water cylinder that is the only big ticket piece of equipment - if there is no hot water cylinder a heat exchanger will be required but pricing that will require more research so I will assume a hot water cylinder in place as in a large number of homes at current.

Every other cost is pipework, and modern flexible pre-insulated PEX pipes should hold down costs.
A British estimate on capital costs of the main line netwrok at roughy $5/MWh over a fifty year lifetime - it put the cost of the main pipes themselves at roughly $730/m linear length.
That sounds expensive but in towns there are lots of houses in a surprisingly short distance.
At $6000 capital cost per connection (with installation for an ASHP) you can get a lot of pipes in there - and you should also discount the price of the installation because it will last 50 years and I don't thinka nyone thinks an air source heat pump will last that long.


Top
 Profile  
 
PostPosted: Aug 16, 2015 8:57 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
I decided to do some digging on thermal storage as those storage tank prices seemed very high.

I stumbled across a programme in Denmark where they dig a big hole in the ground, line it, fill it water and put a floating insulation liner on top. The price for the last one constructed was €41/cubic metre all up.
41 euros per cubic metre.

Projections indicate that very large 200,000 cubic metre stores would see costs reduced to only €28/cubic metre.
The currently often used polyethylene liner is only really rated to 90C although tests have a 95 celsius lifetime of twenty years, but an elastomeric polymer liner is apparently available that will handle our 95C temperature with absolutely no trouble - it costs about €1/cubic metre more.

I will round up to €30/cubic metre.
That gives us a capital cost of only €495/MWh - roughly $550/MWh.

Losses are a bit problematic though - a sixty thousand cubic metre store loses something like 40% of its total heat capacity in a year (this appears to be a calculated figure assuming it is kept full but I am not sure, I will assume its based on an actual seasonal cycle to be conservative).

If we assume heat loss through the surface scales by the square of linear dimensions and volume scales by the cube (not quite right in this case but close enough) then a 200,000 cubic metre store would lose 18% of its heat capacity.

That isn't amazing but isn't bad - and remember that is a seasonal store, charging and discharging once every year.
Summer heat is going to cost us nothign as plant would likely be idling otherwise - $5/MWh is probably a reasonable esimate considering that yields an electricity price of nearly $30/MWh for otherwise idle wattage.

The liner only costs something like €5/cubic metre and has a twenty year life, and the lifetime of a hole in the ground is almost infinite - the insulating cover is more problematic but I will assume a cost weighted lifetime of 50 years.
That amortises the cost of seasonal energy storage down to $11/MWh if we ignore interest (as ballpark studies normally do).

That gives, at 80% efficiency, a cost of a bulk MWh of winter heat at the distribution station of .... $17.25/MWh.

That is utterly ridiculous.
I wonder if we can get a million cubic metre store.....
That would cut losses to near negligible levels and hold down the amount of land used.

$550/MWh for storage helps this blow heat pumps out of the water I think.
If we were more sane and used it for diurnal cycles then it comes out at something like $300/GWh of heat delivered.


Top
 Profile  
 
PostPosted: Aug 17, 2015 9:59 am 
Offline

Joined: Sep 02, 2009 10:24 am
Posts: 507
E Ireland wrote:
alexterrell wrote:
Are you sure the demand is so high?

A lot depends on how leaky houses are in say - 2040. But by then, a typical house, on a very cold day, might need 50KWh per day. Multiply that by 20 million for the UK, and you get 1000GWh.

Using heat pumps with a COP of 4, that is only 250GWh, or about 20GW, half the time.

It would probably be greater than 50kWh. The bulk of the housing stock will be of similar standard to now in 2040 I would imagine - retrofits are rather expensive and there is a mis-selling scandal brewing about cavity wall insulation and the like.
The average house expends something like 10000kWh on space heating and then about 1250kWh on hot water - which is obviously spread over the whole year.


Very hard to predict. The problem is the UK is only replacing <1% of it's housing stock per year, compared to over 2% in Germany.

However, I have experience with large (by European standards) houses in the UK and Germany. These are both old houses fitted with double glazing and cavity insulation, but nothing special like SIPs). Basically Grade C in the UK, which every house should be able to achieve. Peak requirement is about 100KWh/day on very cold days.

But a new house - with standard 12" insulation, Mechanical Heat Recovery, etc .should be looking at <20KWh

I reckon 50KWh as an average is achievable with the right set of incentives.

Quote:
alexterrell wrote:
That half the time would be offpeak - along with vehicle charging - another 20GW.

So, current offpeak load 40GW + 20 GW heating +20 GW vehicles = 80GW. Add 20GW margin and you have 100GW.


I tried numbers on this using the National Grid's half hourly TGSD figures.
You end up with heating loads and car charging loads that are far larger than the peak/off-peak demand swing. My simplistic simulations end up with peak load being during 'Economy 7' periods.

And there is the fact that a heat pump with a COP of 4 will cost you something like £3500 plus install, and won't have that high a power.

£3500 enough to connect a house to a district heating system if you get a high enough uptake to amortise the main heating pipes and other capital plant.
alexterrell wrote:
Thorcons, for example, can deliver that cheaply, with only low grade waste heat, perhaps useful for local greenhouse heating.


My own simulations often assume a baseline of 100GWe worth of LWRs - because that is enough that we could potentially electrify almost everything.
In a country as small as the UK however the district heat argument does seem persuasive, it significantly improves our reactor's capacity factor and could cut many gigawatts off peak demand if we were to use resistive heating, or reduce capital expenditure compared to heat pumps.


District heating will work better with lower temperatures - so underfloor heating is preferable - as it is for heat pumps. But the expensive bit is putting in place a new infrastructure of insulated pipes.

Tricky retrofit in the UK with a standard ceiling height of 2.4m, easier in Germany with 2.5m standard.


Top
 Profile  
 
PostPosted: Aug 17, 2015 10:02 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
Whilst district heating for heating works better at low temperatures - 55C or lower, this is impractical with normal hot water systems as any simple tanks, large bore hot water pipes and the like will be at risk of contamination with legionella.

70/40C is probably the lowest we can push district heating without the fancy new fit DHW systems they are trialling in Denmark. Although lower return temperatures will likely be achieved in practice.

Additionally the problem with things like Mechanical Heat Recovery is they tend to be expensive in their own right and can cause air quality problems - so I think we should be very careful about using them as an alternative to energy production plant.


Last edited by E Ireland on Aug 17, 2015 10:12 am, edited 1 time in total.

Top
 Profile  
 
PostPosted: Aug 17, 2015 10:06 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Excellent research, gentlemen.

Hot water tanks remain the most attractive overall option for low grade heat storage. Despite all the interesting work on phase change materials and whatnot, it is just too expensive and not as environmentally friendly as simple water tanks.

Hot water tank storage should cost under $200/m3 for 100 megaliter class systems.

Legionella can be eliminated with powerful modern UV treatment. Conventional chlorination is also an option considering the closed loop nature of the system, but you have to be careful with stress corrosion (duplex stainless is fine).


Top
 Profile  
 
PostPosted: Aug 17, 2015 11:11 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
A big problem with legionella is that the domestic hot water supply of the house can become infected if the water is not heated above about 60C for a significant period of time to sterilise it.
This is why the new very low temperature danish systems only allow less than 3l of water in the entire domestic hot water pipe system and no storage at all. (Talking about potable water and not the heating circulation water).

This is going to be very expensive to retrofit if it is even possible and I don't like the idea of chlorinating the water in houses - lots of health issues there with improperly maintained equipment or what not.

A nice standard 70/40(30) C system seems best to me.


Top
 Profile  
 
PostPosted: Aug 17, 2015 1:50 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
E Ireland wrote:
A big problem with legionella is that the domestic hot water supply of the house can become infected if the water is not heated above about 60C for a significant period of time to sterilise it.
This is why the new very low temperature danish systems only allow less than 3l of water in the entire domestic hot water pipe system and no storage at all. (Talking about potable water and not the heating circulation water).

This is going to be very expensive to retrofit if it is even possible and I don't like the idea of chlorinating the water in houses - lots of health issues there with improperly maintained equipment or what not.

A nice standard 70/40(30) C system seems best to me.


There is no need for water storage in the potable side with this system. Your storage is in the closed loop heat transfer water side which will likely be demineralized and UV treated. There is no point in having the thermal store twice! All you need is a bigger HX at each house and the cost of this is offset by the lack of need for a potable hot water tank (from a thermal loss viewpoint, it is very silly to have a million small insulated potable hot water vessels).

I agree chlorinating water potable water is the cheap and filthy way to go. That is not what I meant, I just meant the closed heat transfer water side. That side can have large amounts of cleaning chems if needed, but more likely demineralized water + UV treatment alone (no chems) is the way to go.

I think there may be an interesting application for PCHEs here. They can reduce the delta T to basically nothing. Perhaps more for the PWR big exchanger (the steam-hot water one) than for the individual household exchangers, though. It will help you getting more power out of the turbine.


Top
 Profile  
 
PostPosted: Aug 17, 2015 2:06 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1495
Cyril R wrote:
There is no need for water storage in the potable side with this system. Your storage is in the closed loop heat transfer water side which will likely be demineralized and UV treated. There is no point in having the thermal store twice! All you need is a bigger HX at each house and the cost of this is offset by the lack of need for a potable hot water tank (from a thermal loss viewpoint, it is very silly to have a million small insulated potable hot water vessels).

In a nuclear system we have little reason to go for a lower delivery temperature than 70C as any savings from drawing heat at a lower temperature than 130C are eaten by increased pumping power getting the heat from the plant to the city where it willl be used.
In the interest of allowing a minimum capital cost installation I thought it was prudent to allow those people with hot water cylinders to simply run the new hot water supply through the existing coil on their hot water cylinder, before either returning it to the main or running it through any radiators they need heated.
That way we avoid the capital cost of a new relatively high performance heat exchanger (since it will have to supply high flow rate potable water at a high enough temperature to prevent a legionella infection in the pipework - which has been documented).

New construction would likely have said heat exchanger - although we would have to do a cost judgement as the heat exchanger might have a higher capital cost than a cylinder which is after all a consumer item. Although the heat exchanger is almost certainly more compact.
Cyril R wrote:
I think there may be an interesting application for PCHEs here. They can reduce the delta T to basically nothing. Perhaps more for the PWR big exchanger (the steam-hot water one) than for the individual household exchangers, though. It will help you getting more power out of the turbine.

That depends how cheap we can make them - can we make a ~20kW capable near zero delta-T heat exchanger for less than the price of the hot water cylinder?


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 33 posts ]  Go to page 1, 2, 3  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group