Some examples of long-duration energy storage include:
• Pumped hydroelectric storage: Electricity is generated when water is released from a reservoir and travels downhill through a turbine. The water is pumped uphill during lower energy demand. According to the British Hydropower Association, the UK’s pumped hydropower capacity
was 2.8 GW or 32 GWh (see box 2 for an explanation of the difference)."
Ummm. 32GWh of pumped hydro??? That didn't pass the sniff test.
The late Prof Sir David Mackay's pumped hydro table in 'Without the hot air' informs we've 26.7GWh.
Since the HoC referred to BHA, I looked at its site
We are informed by "... the leading trade membership association solely representing the interests of the UK hydropower industry." that:
"There are four operational PSH plants in the UK:
Dinorwig (1983) 1.7 GW, 10.4 GWh
Foyers (1974) 300 MW, 6.4 GWh
Ffestiniog (1963) 360 MW, 7.6 GWh
Cruachan (1966) 440 MW, 7.6 GWh"
It claims Ffestiniog stores 7.6 GWh whereas the ex-Chief Scientific Advisor to the UK Department of Energy and Climate Change stated Ffestiniog has an energy storage capacity of 1.3GWh.
The only way to settle that was to contact Ffestiniog's owners.
A change/error of 0.1GWh in a 16-year old pamphlet is tolerable; a 6.2GWh / 590% exaggeration in 'THE' trade association's webpage really isn't.
Worse, when a member of the HoC research team was personally informed of its error, all that resulted was an auto-reply that the email had been received. They clearly did sweet FA, their error is still visible to the world today.
The Climate Change Committee (CCC) is a group of "true believers" whose utterings are more religious than scientific / expert, so I would need a truck-load of salt with any numbers they've produced.
I tried doing a 'what if' type of analysis using the last four years' of GB data, which you can find in link2.
Any mix of technologies is going to need to be affordable and effective. I don't think we know what that mix of technologies will be, so you can see the effect of a range of MWh:
100,000 MWh LDES makes hardly any difference to fossil fuel requirement etc.
500,000 MWh LDES is obv better, but look at the scale comparators;
7,300,000 MWh LDES (Fig 3) reduced curtailment to zero for 2023 weather, but still left Cumulative shortfall over the year of nearly 65 million MWh (65 TWh).
That's all using HMGov's 'ambitious' levels of installed W & Solar *capacity* by 2030 which DNZ is legally bound to deliver. HAH!
10 gigawatt hours (GWh) by 2030 doesn't sound much because it isn't! ;-)
Considering that a Tesla Megapack 1.9 MW Power; 3.9 MWh Energy would cost a Californian utility $1,030,680 (i.e. roughly £0.8bn for 2GW / 4GW) or a 1 MW Power; 3.9 MWh Energy $963,540 (i.e. roughly £0.76bn for 2GW / 4GW) PLUS land value/cost + installation costs.
Chris, historically the pump storage uses night time “cheap” electricity to pump the water from the lower to the upper reservoir. The operators tend to be wait for peak electricity demands to release the energy over a 30-60 minute window. This maximises revenue and is not designed to support a government’s macro “green” ethos.
Ian, thank you, and indeed that has historically been the case.
Whether NGESO will get increased powers to mandate different modes of operation of energy storage if it's crucial to keeping the lights on? Presumably under different modes of compensation / subsidy if a juicy chance of making a killing on market prices is overridden by the grid operator...?
Chris, thanks for this post. An inability to differentiate between power and energy is a common problem. I often read the capacity of energy storage systems being given in W. This seems to occur even in places where you would have thought they would know better. I have always assumed they mean Watt hours but this may not be the case, though I don't know practically what an energy storage rating in Watts would mean.
I think if you asked every adult in the UK, probably less than one in five would be able to tell you the difference between power and energy yet it is written on everyone's energy bills.
I think a good question in parliament would be to ask Ed Milliband if he could say what the difference is between power and energy. If he answered yes, the follow up question would be, how much "zero carbon" generating power and how much energy storage in nominal GW and GWh do we need to fulfill the government's promise of zero carbon electricity generation by 2030. These are numbers which someone must know but I haven't seen written down anywhere. Then you could ask him how much we have now.
I looked up Barbour ABI and they are market analysts. There is no guarantee that anyone there knows what the difference between a MW and MJ is. They might know that Watt was Scottish and Joule was English. They would probably think you were being picky if you told them storage capacity in MW is meaningless.
By the way. Have you in your various calculations come up with an estimate of these numbers?
I entirely agree with you, ask 100 random people in the street what is the relevance of MW and MWh in the context of electrical energy storage, the great majority is unlikely to have a clue.
However, DESNZ (really DNZ) are *responsible* for planning the UK's energy 'transition'.
They *really* need to know and understand this basic stuff.
Please also see Ron Hughes' comments beginning
"Hi Michael
The House of Lords Sci & Tech Committee report ..." under this post.
On Gridwatch today wind and solar were less than 6% of total. Without FF there would be a hell of a lot of what is euphemistically called load shedding.
The House of Lords Sci & Tech Committee report "Long-duration energy storage: get on with it." published 13th March 2024 states in Box 1:
"Units of energy and power, and scale of existing energy storage in the UK"
"The UK currently has a relatively small amount of low-carbon energy storage deployed, mostly in the form of pumped-storage hydroelectricity (see box 2 for explanations of different forms of energy storage) at 2.8 GW of pumped-hydro, with a storage capacity of 26.7 GWh. In 2022, 2.6 TWh of electricity was used to pump water uphill in these storage facilities, which then generated 2 TWh of electricity in 2022.12 There is also a growing capacity of battery electricity storage at grid-scale, which is discharged more frequently, usually across minutes to hours, and has been estimated at 2.4 GW (2.6 GWh) of capacity across 161 sites.
Comparing power capacity only, and not the length of time the power can be supplied for (which depends on the total energy stored), can be misleading. From these figures, if all the UK’s grid-scale batteries were fully charged and discharged at once, they could provide 2.4 GW for just over an hour; if the UK’s pumped-hydro was fully charged and discharged, it could provide 2.8 GW over nearly ten hours. Balancing the energy system requires matching the instantaneous power demand (GW) and also meeting the energy demand over time (GWh)."
Actually, I've just realised that even "... if the UK’s pumped-hydro was fully charged and discharged, it could provide 2.8 GW over nearly ten hours." is erroneous!
Foyers can discharge its 6.3GWh at only 0.3GW
Cruachan can discharge its 6.3GWh at only 0.4GW
Ffestiniog stores 1.3GWh & discharges at 0.36GW for 3.6hrs
Dinorwig stores 9.1GWh & discharges at 1.8GW for 5.1hrs
[NB I've used PH data from the late Prof David Mackay's "Without the hot air" for simplicity. Chris yesterday pointed out a discrepancy with Crauchan's power to energy ratio.]
So, for first 3.6 hours, discharge would be at 2.86GW; hours 3.6 to 5.1 would be at 2.5GW, and hours 5.1 to 10 would be at 0.7GW!
I've just read thru that Lords Sci & Tech Committee Chapter 2...
Re: The Royal Society, I read their Large-scale-electricity-storage-UK report and was not impressed.
"The models were based on real weather data in the 37 years 1980 to 2016" [page 6] sounds good, but
"based on actual weather data in a typical year (1992) scaled to 570 TWh/ year averaged over 37 years"[page 10] is rather different.
"and the renewables.ninja model of wind and solar supply that is based on 37 years of real weather data (1980 to 2016). " [page 12] is different again.
What killed that Royal Soc report for me was Table 1 [page 18]:
I can speculate that the older BESSs are 1-h, some maybe 2-h. And I've seen at least one recent project reported in the trade press stating its MW/MWh numbers which is a 2-h design.
It's essential data for any planning of storage resources and it's missing from the database.
In fact, it is possible to make a few back of the envelope calculations.
Supposing we translate November 2024 magically into the future and say that 100% of the CCGT generation was supplied by wind.
In November average CCGT generation was around 12.5GW and average wind generation was 7.0GW.
Supposing that was all wind that would be 19.5GW. That would make around 66% of electricity generated by wind. This assumes we have around three times as much wind generation as today (based on the amount of energy being generated by wind in November). Doesn't sound impossible.
Supposing that wind now dropped by 90% for 1 day. A perfectly common scenario. Suppose that all other sources stay constant. I don't think this is an unreasonable supposition; nuclear is maxed out, bio is maxed out, solar and hydro are negligeable (in November), I don't think it is reasonable to expect the interconnectors to supply much more.
In this scenario, stored energy systems would have to supply 421GWh at an average 17.6GW.
Another scenario might be a drop in wind of 66% for a week. Again, if we suppose nothing else is capable of filling the hole, storage systems would have to supply 2195GWh at 13GW.
While these look like monster numbers compared with the 10GWh that the government expects to be required by 2030 according to the BESS research briefing linked by one of the other posters it can be put in perspective. If every household in the UK had a battery with a capacity of 15kWh that would meet the requirements for the first scenario. It would need 77kWh for the second scenario but that is only the size of a highish end EV battery.
The cost would be a cool trillion pounds just for the batteries, never mind the network, but didn't the government spend that to bail out the banks in 2008?
Re: "Another scenario might be a drop in wind of 66% for a week. Again, if we suppose nothing else is capable of filling the hole, storage systems would have to supply 2195 GWh at 13GW. [24x7x13=2,184 GWh]
While these look like monster numbers compared with the 10GWh that the government expects to be required by 2030 ..."
I honestly don't think HMGovt 'expects' 10 GWh by 2030.
I suspect it has been asked to whisk up a number based on 'anticipated advances in technology', 'anticipated falls in costs', etc.
Lo, 10, nice round number... in giga-watt thingumies, that'll be good enough for this report, we'll be well away on the path to a knighthood before anyone finds out...
Because a very similar scenario happened in early November:
UK*: Wind below about 5 GW from mid-01 Nov, low to very low until mid-10 Nov."
10 GWh is 10,000 MWh = 4 times what we think is the likely total storage capacity of the listed battery schemes... which took roughly 10 years to install.
I round the figures from my spreadsheet. That's why you don't get the same result when you multiply them. It bores me to write decimal places. It would also bore me to redo the calculations in terms of the rounded figures. I'll bear it in mind for future posts. What look like errors in arithmetic reduce confidence.
In the Royal Society paper you linked, they comment that their storage numbers would have been much smaller if they hadn't considered 2009 to 2011, years in which, apparently, the wind didn't blow much. So calms are not just measured in days or weeks.
I found their numbers believable. On the other hand I don't expect any government would want to consider them. 100TWh of storage and an extra 50% of installed generating capacity to keep it full? You have got to be joking Mr. Newton!
Other outfits that don't have a clue are:
House of COMMONS Library researchers, and (remarkably), the British Hydropower Association (BHA).
Earlier this year (19/4/2024), the former published their Research Briefing on Battery Energy Storage Systems
https://researchbriefings.files.parliament.uk/documents/CBP-7621/CBP-7621.pdf
Page 10:
"Other types of energy storage systems
Some examples of long-duration energy storage include:
• Pumped hydroelectric storage: Electricity is generated when water is released from a reservoir and travels downhill through a turbine. The water is pumped uphill during lower energy demand. According to the British Hydropower Association, the UK’s pumped hydropower capacity
was 2.8 GW or 32 GWh (see box 2 for an explanation of the difference)."
Ummm. 32GWh of pumped hydro??? That didn't pass the sniff test.
The late Prof Sir David Mackay's pumped hydro table in 'Without the hot air' informs we've 26.7GWh.
Since the HoC referred to BHA, I looked at its site
https://british-hydro.org/pumped-storage-hydropower/
Scroll down to, and expand 'Key statistics'.
We are informed by "... the leading trade membership association solely representing the interests of the UK hydropower industry." that:
"There are four operational PSH plants in the UK:
Dinorwig (1983) 1.7 GW, 10.4 GWh
Foyers (1974) 300 MW, 6.4 GWh
Ffestiniog (1963) 360 MW, 7.6 GWh
Cruachan (1966) 440 MW, 7.6 GWh"
It claims Ffestiniog stores 7.6 GWh whereas the ex-Chief Scientific Advisor to the UK Department of Energy and Climate Change stated Ffestiniog has an energy storage capacity of 1.3GWh.
The only way to settle that was to contact Ffestiniog's owners.
https://i.postimg.cc/FKmHPmPX/temp-Image-Ta-TNl-P.avif
A change/error of 0.1GWh in a 16-year old pamphlet is tolerable; a 6.2GWh / 590% exaggeration in 'THE' trade association's webpage really isn't.
Worse, when a member of the HoC research team was personally informed of its error, all that resulted was an auto-reply that the email had been received. They clearly did sweet FA, their error is still visible to the world today.
Thanks for the link to the Research briefing on BESSs.
"government expects the demand for grid energy storage to rise to 10 gigawatt hours (GWh) by 2030"
That doesn't sound like much. With total demand at around 40GW it sounds like BESS is only expected to supply a small part of energy storage.
Have you seen anywhere an estimate of the total energy storage for electricity generation needed by 2030 to meet Ed Milliband's rash promise?
Hello Andrew,
Re: "an estimate of the total energy storage for electricity generation needed"
If you read The Royal Society's report from last year, 60-100 TWh.
But I cast doubts on their ability to correctly define the problem in link1.
link1] https://open.substack.com/pub/chrisbond/p/dunkelflaute-dunkelcobblers?r=om40y&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
The Climate Change Committee (CCC) is a group of "true believers" whose utterings are more religious than scientific / expert, so I would need a truck-load of salt with any numbers they've produced.
I tried doing a 'what if' type of analysis using the last four years' of GB data, which you can find in link2.
link2] https://open.substack.com/pub/chrisbond/p/intermittency-aka-diminishing-returns?r=om40y&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
Any mix of technologies is going to need to be affordable and effective. I don't think we know what that mix of technologies will be, so you can see the effect of a range of MWh:
100,000 MWh LDES makes hardly any difference to fossil fuel requirement etc.
500,000 MWh LDES is obv better, but look at the scale comparators;
7,300,000 MWh LDES (Fig 3) reduced curtailment to zero for 2023 weather, but still left Cumulative shortfall over the year of nearly 65 million MWh (65 TWh).
That's all using HMGov's 'ambitious' levels of installed W & Solar *capacity* by 2030 which DNZ is legally bound to deliver. HAH!
Hi Andrew
10 gigawatt hours (GWh) by 2030 doesn't sound much because it isn't! ;-)
Considering that a Tesla Megapack 1.9 MW Power; 3.9 MWh Energy would cost a Californian utility $1,030,680 (i.e. roughly £0.8bn for 2GW / 4GW) or a 1 MW Power; 3.9 MWh Energy $963,540 (i.e. roughly £0.76bn for 2GW / 4GW) PLUS land value/cost + installation costs.
https://www.tesla.com/megapack/design
Consequently, the question to nearly all punters will not be "How many would you like, sir?" but rather "What's your budget, pal?"
Chris, historically the pump storage uses night time “cheap” electricity to pump the water from the lower to the upper reservoir. The operators tend to be wait for peak electricity demands to release the energy over a 30-60 minute window. This maximises revenue and is not designed to support a government’s macro “green” ethos.
Ian, thank you, and indeed that has historically been the case.
Whether NGESO will get increased powers to mandate different modes of operation of energy storage if it's crucial to keeping the lights on? Presumably under different modes of compensation / subsidy if a juicy chance of making a killing on market prices is overridden by the grid operator...?
For example in Australia:
https://reneweconomy.com.au/aemo-warns-big-batteries-to-expect-breakfast-call-to-protect-the-grid-from-rooftop-solar-juggernaut/
Chris, thanks for this post. An inability to differentiate between power and energy is a common problem. I often read the capacity of energy storage systems being given in W. This seems to occur even in places where you would have thought they would know better. I have always assumed they mean Watt hours but this may not be the case, though I don't know practically what an energy storage rating in Watts would mean.
I think if you asked every adult in the UK, probably less than one in five would be able to tell you the difference between power and energy yet it is written on everyone's energy bills.
I think a good question in parliament would be to ask Ed Milliband if he could say what the difference is between power and energy. If he answered yes, the follow up question would be, how much "zero carbon" generating power and how much energy storage in nominal GW and GWh do we need to fulfill the government's promise of zero carbon electricity generation by 2030. These are numbers which someone must know but I haven't seen written down anywhere. Then you could ask him how much we have now.
I looked up Barbour ABI and they are market analysts. There is no guarantee that anyone there knows what the difference between a MW and MJ is. They might know that Watt was Scottish and Joule was English. They would probably think you were being picky if you told them storage capacity in MW is meaningless.
By the way. Have you in your various calculations come up with an estimate of these numbers?
Hello Andrew, thank you for your comments.
I entirely agree with you, ask 100 random people in the street what is the relevance of MW and MWh in the context of electrical energy storage, the great majority is unlikely to have a clue.
However, DESNZ (really DNZ) are *responsible* for planning the UK's energy 'transition'.
They *really* need to know and understand this basic stuff.
Please also see Ron Hughes' comments beginning
"Hi Michael
The House of Lords Sci & Tech Committee report ..." under this post.
On Gridwatch today wind and solar were less than 6% of total. Without FF there would be a hell of a lot of what is euphemistically called load shedding.
Thank you, David.
Currently (2:15pm) GridWatch is showing 4.24% Wind & 0.78% Solar.
Nearby Denmark, https://energy-charts.info/charts/power/chart.htm?l=en&c=DK&interval=month&year=2024&month=12&source=public
at 11:00 shows Wind around 500MW (around 10% of Load), Solar about half that.
Germany is showing the beginning of the lull:
https://energy-charts.info/charts/power/chart.htm?l=en&c=DE&interval=month&year=2024&month=12&source=public
But "interconnectors are the answer", lol.
Do they give any estimate of the total MWh electricity storage in UK?
Hi Michael
The House of Lords Sci & Tech Committee report "Long-duration energy storage: get on with it." published 13th March 2024 states in Box 1:
"Units of energy and power, and scale of existing energy storage in the UK"
"The UK currently has a relatively small amount of low-carbon energy storage deployed, mostly in the form of pumped-storage hydroelectricity (see box 2 for explanations of different forms of energy storage) at 2.8 GW of pumped-hydro, with a storage capacity of 26.7 GWh. In 2022, 2.6 TWh of electricity was used to pump water uphill in these storage facilities, which then generated 2 TWh of electricity in 2022.12 There is also a growing capacity of battery electricity storage at grid-scale, which is discharged more frequently, usually across minutes to hours, and has been estimated at 2.4 GW (2.6 GWh) of capacity across 161 sites.
Comparing power capacity only, and not the length of time the power can be supplied for (which depends on the total energy stored), can be misleading. From these figures, if all the UK’s grid-scale batteries were fully charged and discharged at once, they could provide 2.4 GW for just over an hour; if the UK’s pumped-hydro was fully charged and discharged, it could provide 2.8 GW over nearly ten hours. Balancing the energy system requires matching the instantaneous power demand (GW) and also meeting the energy demand over time (GWh)."
https://publications.parliament.uk/pa/ld5804/ldselect/ldsctech/68/6805.htm#_idTextAnchor007
Actually, I've just realised that even "... if the UK’s pumped-hydro was fully charged and discharged, it could provide 2.8 GW over nearly ten hours." is erroneous!
Foyers can discharge its 6.3GWh at only 0.3GW
Cruachan can discharge its 6.3GWh at only 0.4GW
Ffestiniog stores 1.3GWh & discharges at 0.36GW for 3.6hrs
Dinorwig stores 9.1GWh & discharges at 1.8GW for 5.1hrs
[NB I've used PH data from the late Prof David Mackay's "Without the hot air" for simplicity. Chris yesterday pointed out a discrepancy with Crauchan's power to energy ratio.]
So, for first 3.6 hours, discharge would be at 2.86GW; hours 3.6 to 5.1 would be at 2.5GW, and hours 5.1 to 10 would be at 0.7GW!
So it's not only DESNZ that doesn't have a clue!
Good info and really useful link, ty Ron.
I've just read thru that Lords Sci & Tech Committee Chapter 2...
Re: The Royal Society, I read their Large-scale-electricity-storage-UK report and was not impressed.
"The models were based on real weather data in the 37 years 1980 to 2016" [page 6] sounds good, but
"based on actual weather data in a typical year (1992) scaled to 570 TWh/ year averaged over 37 years"[page 10] is rather different.
"and the renewables.ninja model of wind and solar supply that is based on 37 years of real weather data (1980 to 2016). " [page 12] is different again.
What killed that Royal Soc report for me was Table 1 [page 18]:
"Weather stress events.
Summer wind drought – frequent
One full day of very low wind speed in summer
One or two per year
Summer wind drought – infrequent
Up to four weeks of very low wind speed in summer
Once every 10 years
Winter wind drought
Up to a week of very low wind speed in winter
Every few years"
I included my comments on that here: https://open.substack.com/pub/chrisbond/p/dunkelflaute-dunkelcobblers?r=om40y&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
The CCC? I believe them implicitly [sark].
Correction "Cruachan can discharge its *10GWh* at only 0.4GW"
Hi Michael, not in that database they don't.
I can speculate that the older BESSs are 1-h, some maybe 2-h. And I've seen at least one recent project reported in the trade press stating its MW/MWh numbers which is a 2-h design.
It's essential data for any planning of storage resources and it's missing from the database.
It is likely deliberate.
I have noticed that Greens often use the unit “MW” for electricity storage instead of “MWh” either because:
1) they do not understand how energy works (which is inexcusable for people who make proposals to transform our energy system), or because
2) they do not want people to know how long the electricity storage actually lasts because they know it is far too short (which is very unethical).
In fact, it is possible to make a few back of the envelope calculations.
Supposing we translate November 2024 magically into the future and say that 100% of the CCGT generation was supplied by wind.
In November average CCGT generation was around 12.5GW and average wind generation was 7.0GW.
Supposing that was all wind that would be 19.5GW. That would make around 66% of electricity generated by wind. This assumes we have around three times as much wind generation as today (based on the amount of energy being generated by wind in November). Doesn't sound impossible.
Supposing that wind now dropped by 90% for 1 day. A perfectly common scenario. Suppose that all other sources stay constant. I don't think this is an unreasonable supposition; nuclear is maxed out, bio is maxed out, solar and hydro are negligeable (in November), I don't think it is reasonable to expect the interconnectors to supply much more.
In this scenario, stored energy systems would have to supply 421GWh at an average 17.6GW.
Another scenario might be a drop in wind of 66% for a week. Again, if we suppose nothing else is capable of filling the hole, storage systems would have to supply 2195GWh at 13GW.
While these look like monster numbers compared with the 10GWh that the government expects to be required by 2030 according to the BESS research briefing linked by one of the other posters it can be put in perspective. If every household in the UK had a battery with a capacity of 15kWh that would meet the requirements for the first scenario. It would need 77kWh for the second scenario but that is only the size of a highish end EV battery.
The cost would be a cool trillion pounds just for the batteries, never mind the network, but didn't the government spend that to bail out the banks in 2008?
Hi Andrew,
Re: "Another scenario might be a drop in wind of 66% for a week. Again, if we suppose nothing else is capable of filling the hole, storage systems would have to supply 2195 GWh at 13GW. [24x7x13=2,184 GWh]
While these look like monster numbers compared with the 10GWh that the government expects to be required by 2030 ..."
I honestly don't think HMGovt 'expects' 10 GWh by 2030.
I suspect it has been asked to whisk up a number based on 'anticipated advances in technology', 'anticipated falls in costs', etc.
Lo, 10, nice round number... in giga-watt thingumies, that'll be good enough for this report, we'll be well away on the path to a knighthood before anyone finds out...
Because a very similar scenario happened in early November:
https://chrisbond.substack.com/p/northern-europe-reality
"from Figure 1:
UK*: Wind below about 5 GW from mid-01 Nov, low to very low until mid-10 Nov."
10 GWh is 10,000 MWh = 4 times what we think is the likely total storage capacity of the listed battery schemes... which took roughly 10 years to install.
I round the figures from my spreadsheet. That's why you don't get the same result when you multiply them. It bores me to write decimal places. It would also bore me to redo the calculations in terms of the rounded figures. I'll bear it in mind for future posts. What look like errors in arithmetic reduce confidence.
In the Royal Society paper you linked, they comment that their storage numbers would have been much smaller if they hadn't considered 2009 to 2011, years in which, apparently, the wind didn't blow much. So calms are not just measured in days or weeks.
I found their numbers believable. On the other hand I don't expect any government would want to consider them. 100TWh of storage and an extra 50% of installed generating capacity to keep it full? You have got to be joking Mr. Newton!