The energy world is fixated on the “huge” amounts of battery storage presently being installed to back up slowly-increasing levels of intermittent renewables generation. The feeling seems to be that as soon as enough batteries are installed to take care of daily supply/demand imbalances we will no longer need conventional dispatchable energy – solar + wind + storage will be able to do it all. Here I take another look at the realities of the situation using what I hope are some telling visual examples of what battery storage will actually do for us. As discussed in previous posts it will get us no closer to the vision of a 100% renewables-powered world than we are now.
*Note: “Battery storage” covers all storage technologies currently being considered, including thermal, compressed air, pumped hydro etc. Batteries are, however, the flavor of the moment and are expected to capture the largest share of the future energy storage market.
This post is all about the difference between pipe dreams and reality. Prof. Mark Jacobson of Stanford University et al. have just published a new study that responds to the critics of their earlier 2017 study. The new study is paywalled, but Stanford’s press release describes the basic procedures used:
For the study, the researchers relied on two computational modeling programs. The first program predicted global weather patterns from 2050 to 2054. From this, they further predicted the amount of energy that could be produced from weather-related energy sources like onshore and offshore wind turbines, solar photovoltaics on rooftops and in power plants, concentrated solar power plants and solar thermal plants over time. These types of energy sources are variable and don’t necessarily produce energy when demand is highest.
The group then combined data from the first model with a second model that incorporated energy produced by more stable sources of electricity, like geothermal power plants, tidal and wave devices, and hydroelectric power plants, and of heat, like geothermal reservoirs. The second model also included ways of storing energy when there was excess, such as in electricity, heat, cold and hydrogen storage. Further, the model included predictions of energy demand over time.
Scenarios based on the modeling data avoided blackouts at low cost in all 20 world regions for all five years examined and under three different storage scenarios.
What’s the energy mix that leads to this happy ending in no fewer than 139 of the world’s countries? The lead-in figure of Jacobson et al’s 2017 report, reproduced below as Figure 1, tells us. Rounded off to the nearest percent it’s 5% hydro + geothermal, 37% wind, 58% solar and not a kilowatt of nuclear.
Figure 1: Jacobson et al’s global energy mix, plus propaganda
In contrast to Jacobson et al, who compare this energy mix with computer-generated demand scenarios that foresee the replacement of fossil fuels with wind and solar somehow lowering demand by 42.5%, I have taken my usual approach of comparing an energy mix with real-life grid data, which raises the question of which real-life data to use. Well, Stanford University is in California, and I happen to have quite a lot of grid data from the California Independent System Operator (CAISO), so I used that. And California is also a good example to use because it’s heavy into solar and battery storage, or at least would like to be.
So what’s the problem with energy storage in California? It’s widely perceived to be the now-famous California duck curve, which shows how rapidly increasing solar generation could within a few years increase afternoon ramp rates to the point where existing gas-fired and hydro balancing facilities are no longer able to handle them:
Figure 2: The California duck curve
But while this could indeed be a problem in the future it isn’t at the moment. We begin our analysis of the real-life CAISO grid data with Figure 3, which plots hourly generation against demand for three days in early March 2015. With the help of imports from surrounding states California had no difficulty matching generation to demand over this period, with most of the load-balancing handled by gas-fired generation:
Figure 3: Actual CAISO generation by source and demand (black), hourly data, March 3, 4 and 5, 2015
Figure 4 now shows what Figure 3 would have looked like with the Jacobson et al renewables generation mix (5% hydro+geothermal, 37% wind, 58% solar) in place. It looks more like the Shanghai skyline than a duck:
Figure 4: Generation and demand, hourly data, March 3, 4 and 5, 2015, Jacobson et al generation mix. Generation is scaled to match demand over the period.
In this case CAISO would have considerable difficulty balancing generation against daily demand, and since a) the imbalances are caused almost entirely by solar and b) when it’s dark in California it will be dark in the surrounding Western US states too there will be little or no surplus energy available. So balancing will have to be done by storing the daytime solar surpluses for re-use at night. How much storage would be needed over the three-day period considered? According to Figure 5, about 300 GWh, the equivalent of over 2,000 Big South Australian Batteries (BSABs):
Figure 5: Storage balance for Figure 4, hourly data, Jacobson et al generation mix
This, of course, is not a real-life case. No sane grid operator, nor even the California state legislature, would allow imbalances and ramp rates of this magnitude to develop in the first place. But as we shall see they are good for illustrative purposes, so we will go with them.
Figure 6 now presents another generation vs. demand comparison using the Jacobson et al generation mix, but instead of covering three days it covers three years – 2014, 2015 and 2016. The comparison replicates all the basic features of Figure 4 except that the generation peaks are less extreme. Because wind and solar generation both peak in the summer in California – there are no seasonal offsets – imbalances occur seasonally as well as on a daily basis:
Figure 6: Generation and demand, monthly means, 2014, 2015 and 2016, Jacobson et al generation mix. Generation is scaled to match demand
The key question now arises: why should the summer surpluses not have to be stored for winter re-use, just as the daytime surpluses are stored for re-use at night?
No reason at all. They clearly will have to be.
And how much storage would be needed? According to Figure 7 about 25 terawatt-hours’ worth during 2015, based on hourly grid data. (The graphs used to develop the storage balance are shown at the end of the post):
Figure 7: Storage balance, hourly data, 2015, Jacobson et al generation mix
But wait a minute. Where are the huge daily swings that we saw in the Figure 5 storage balance plot?
They’re the little wiggles under the black arrow.
As noted earlier this is not a real-life case, but should it wish to go 100% renewable California will clearly have a seasonal energy storage requirement which vastly exceeds its daily “duck curve” requirement. And what does California, which claims to be a world leader in energy storage, propose to do about it?
Well, in 2010 it passed an energy storage mandate, the wording in which (offpeak, peaking powerplants, peak load requirements) left little doubt that its basic intention was to flatten out the daily duck curve when more solar comes on line:
SECTION 1.
(b) Additional energy storage systems can optimize the use of significant additional amounts of variable, intermittent, and offpeak electrical generation from wind and solar energy
(c) Expanded use of energy storage systems can (avoid or defer) the need for new fossil fuel-powered peaking powerplants
(d) Expanded use of energy storage systems will reduce the use of electricity generated from fossil fuels to meet peak load requirements on days with high electricity demand
The mandate went on to confirm that this was indeed its intention by calling for 1.325 gigawatts of energy storage without specifying how many hours the gigawatts were to last for. Apparently this was unimportant. According to recent reports California is about to call for two gigawatts more “storage”, with gigawatt-hours again unspecified. It‘s questionable whether California even understands what energy storage is.
Now there’s no question that high levels of intermittent renewables generation will require fast-frequency-response capabilities to ensure grid stability during the day, but what is California doing about seasonal storage, which makes up 99% of its total storage problem?
Absolutely nothing. It has yet to recognize its existence.
And the same goes for everyone else, including the UK, where proposed revisions to the energy storage market concentrate almost entirely on “fast frequency response” (I remember reading somewhere that according to National Grid any storage exceeding 15 minutes in duration will be superfluous but can’t find the reference).
People may be wondering why I’ve been spending so much time recently writing about energy storage problems. Well, this is why. Go back to Figure 7 and imagine what it would look like with the little wiggles gone. To all intents and purposes it would look exactly the same. And these little wiggles are all the growing rush for battery storage is going to remove.
Backup data for Figure 7 storage calculation:
Figure 8: Generation by source, hourly data, 2015, Jacobson et al generation mix
Figure 9: Total generation (red) vs. demand (blue), hourly data, 2015, Jacobson et al generation mix
Figure 10: Generation surpluses and deficits from Figure 9, hourly data, 2015, Jacobson et al generation mix
“According to recent reports California is about to call for two gigawatts more “storage”, with gigawatt-hours again unspecified.”
I could supply that quite cheaply. A couple of capacitors! JET in Oxford has two flywheels which can deliver 400MW each. (For about 10 seconds, but who cares!).
Roger – To be fair though, you said this is about all storage, including pumped hydro. Perhaps a reference to the Baja California “Andrews Pumped Storage Scheme”?
The other point to note, is that if batteries can be used to iron out daily fluctuations in renewables, then it means that CCGT can work at optimum efficiency to even out the seasonal variations.
Who in their right mind is going to have CCGT sitting idle for 6-9 months of the year to be switched on when needed for the “seasonal variations”.
I wonder how much 2,000 Big South Australian Batteries (BSABs) will cost?
This drive for CO2 reduction is totally insane.
Musk said the BSAB cost $50 million. So 2000 BSABs cost $100 billion.
And how long do they last? We all know batteries degrade.
You get calendric aging with lithium batteries using organic solvents. There are some companies such as Alevo that use sulphur.
Another option is hydrolise sea water to get hydrogen or ammonia. That can easily be stored and then burnt in a turbine.
https://pubs.acs.org/doi/abs/10.1021/acssuschemeng.7b02070?src=recsys&journalCode=ascecg
From Memory the original PR stuff when Muck proposed the battery mentioned a figure of around $100M.
I was never sure whether that was USD or AUD.
Whichever I assume that the number was “then” and “retail” whereas the risk Tesla took on would be mostly “now” and trade price plus a “loss of opportunity” charge if they were so busy that the South Aus project lost them other sales or delayed cash flow.
Presumably his “risk” number did not site preparation, connections and so on. – so it’s possible that both numbers, in whatever currency, were more or less reasonable in the context of the overall project so long as neither number was totally fabricated by any party to the plan.
From Memory the original PR stuff when Musk proposed the battery mentioned a figure of around $100M.
I was never sure whether that was USD or AUD.
Whichever I assume that the number was “then” and “retail” whereas the risk Tesla took on would be mostly “now” and trade price plus a “loss of opportunity” charge if they were so busy that the South Aus project lost them other sales or delayed cash flow.
Presumably his “risk” number did not site preparation, connections and so on. – so it’s possible that both numbers, in whatever currency, were more or less reasonable in the context of the overall project so long as neither number was totally fabricated by any party to the plan.
Bob,
Turnkey is about two times $50 million.
Bob,
This URL shows the turnkey capital cost likely would be well over $100 million.
Tesla made a low bid and a 100-day schedule, because it wanted to make sure to get the job.
Tesla BOUGHT the job at its cost.
That cost likely would not be for projects going forward.
wanted the contracthttp://www.adelaidenow.com.au/news/south-australia/south-australian-government-refuses-to-say-how-much-worlds-biggest-lithium-ion-battery-will-cost/news-story/77991b5552efa947928e3b105b249286
The $US50 million was the ex factory cost and transport to remote South Australia , installation and commissioning would double the cost
Whoever wins the capacity auction.
At the moment in the UK, contracts are going to diesel and OCGT, as they can help out on an hourly bais.
Suggest we all re-read the late Sir David Mackay’s comments at:https://www.withouthotair.com/c6/page_41.shtml re the UK and solar energy:
“Fantasy time: solar farming
If a breakthrough of solar technology occurs and the cost of photovoltaics
came down enough that we could deploy panels all over the countryside,
what is the maximum conceivable production? Well, if we covered 5% of
the UK with 10%-efficient panels, we’d have
10% × 100 W/m2 × 200 m2 per person
≈ 50 kWh/day/person.
I assumed only 10%-efficient panels, by the way, because I imagine that
solar panels would be mass-produced on such a scale only if they were
very cheap, and it’s the lower-efficiency panels that will get cheap first.
The power density (the power per unit area) of such a solar farm would be
10% × 100 W/m2 = 10 W/m2.
This power density is twice that of the Bavaria Solarpark (figure 6.7).
Could this flood of solar panels co-exist with the army of windmills we
imagined in Chapter 4? Yes, no problem: windmills cast little shadow, and
ground-level solar panels have negligible effect on the wind. How auda-
cious is this plan? The solar power capacity required to deliver this 50 kWh
per day per person in the UK is more than 100 times all the photovoltaics
in the whole world. So should I include the PV farm in my sustainable
production stack? I’m in two minds. At the start of this book I said I
wanted to explore what the laws of physics say about the limits of sus-
tainable energy, assuming money is no object. On those grounds, I should
certainly go ahead, industrialize the countryside, and push the PV farm
onto the stack. At the same time, I want to help people figure out what
we should be doing between now and 2050. And today, electricity from
solar farms would be four times as expensive as the market rate. So I feel
a bit irresponsible as I include this estimate in the sustainable production
stack in figure 6.9 – paving 5% of the UK with solar panels seems beyond
the bounds of plausibility in so many ways. If we seriously contemplated
doing such a thing, it would quite probably be better to put the panels in
a two-fold sunnier country and send some of the energy home by power
lines. We’ll return to this idea in Chapter 25.”
In his view, the best use for solar in the UK was a direct solar water heating via either panels or indirectly via ground-source heat pumps GSHP.
Even before his untimely and desperately unfortunate final illness and eventual demise, Prof MacKay had been sidelined at DECC albeit his role of Chief Scientific Advisor, which he gave up about 2 years prior to passing away . The powers that be just do not see any of what this forum regularly covers with such technical distinction. They are in thrall to a fantasy world and everyone inhabiting it are of the same mindset. It seems impossible to exert any traction on this closed world.
The cheapest PV is thin-film, and the largest manufacturer of this is First Solar. Their production panels’ efficiency is already up to 17+%:
http://www.firstsolar.com/en-EMEA/-/media/First-Solar/Technical-Documents/Series-6-Datasheets/Series-6-Datasheet.ashx
And they have the aim of going significantly higher still:
https://www.technologyreview.com/s/600922/first-solars-cells-break-efficiency-record
Most of the PV farms in the UK use silicon, which continues to be more efficient than thin-film.
So Mackay’s land requirement can be halved, in addition to slashing his costs by factors he never imagined. I think his book has now had its time and should be filed under ‘History’.
Covering 5% of the UK with solar panels is utterly insane when you consider the fact that currently less than 2% of the land in the UK is developed.
Roger while I agree that the duck curve is not a problem at the moment. pet hobby horse digression.
But consider that the 13 GW per 3 hour period was exceeded in January. 13290MW/3hr from:
http://www.caiso.com/Documents/MonthlyRenewablesPerformanceReport-Jan2018.html
For some reason I am not currently able to follow the links provided or for that matter properly load the page. Shrug, pretty sure I broke something. And it is a shame because there is a lot of good info on those pages.
Then again solar and wind install rates have slowed considerably last year. Wind added a grand total of 48 MW and solar added 1,655 MW of net summer capacity from Nov 2016 to Nov 2017. Wind comprising 38.6% of the total. I guess that off shore wind will need to get busy. 😉
Jacobson needs to get out of the simulation and live in the real world. Really 9.72% from concentrated solar anyplace than a desert is … words fail me.
I start to wonder if he just took the current California grid and made up a model to support it. Fudging things to fit.
Thank you for the analysis,
T2M
T2M: You’re welcome.
The 13 GW/3 hour ramp rate was also exceeded on June 19, June 26, July 7 and August 27, 2017 (I have no data for 2018). These all occurred during the ramping-down period between 9pm and midnight. There were no ramping-up exceedances.
The down ramps are interesting.
Did they show in the 5 min or hourly data. Just curious.
T2M
The data are hourly,
Roger
The number for the up ramp was quoted from the website that uses 5 min data. Hourly data shows highest ramp of 12,411 MWH and lowest of -8,133 MWH. IF, I did my sums right.
Also I copied the January Renewable Watch hourly data to the file sharing site:
http://www.mediafire.com/file/ilmf07leskr5hl5/CASIO_Renewable_Watch_Downloader_2018_January_-_Copy.txt
It is a text file but will open in a spreadsheet program. I should rename that I update it monthly, “around” the first. What would be a good title? Only reason I mention it is you said that you didn’t have 2018 data.
I also uploaded the 5 min data from June 1st to Nov 30th of 2017 to the site:
http://www.mediafire.com/file/2al7ikyo4mv4okb/CAISO%205%20Min%20June%202017%20to%20Nov%202017.xlsx
Somehow I lost the longer file for data back to 2014, if I recall correctly. I can’t find it on the website right now. :-(.
T2M
Thanks T2M. I’ll take a look at your data when I get time, which unfortunately is in short supply at the moment.
Are up-ramp rates a bigger problem than down-ramp rates?
Roger,
For up and down ramps severity based on the CA climate I would say they are probably seasonal up in winter and down in summer. I haven’t crunched any numbers, still debating exactly how to go about it.
As always I just put the data I find out there for people smarter and more able to communicate than me to use. If they choose to!
Definitely understand time issues!!!
T2M
Roger a 3+ year graph of daily Min and Max Net Load Ramps for CAISO based on the 5 min data.
http://prntscr.com/ii8o7l
Found the file hiding under a production an curtailment name instead of 5 minute data.
Too may things going on for my brain to remember.
T2M
Why does demand get higher in the summer months, the opposite of what happens in Europe ?
Air con, same in Oz – they fight the heat, we fight the cold…
Same for Japan, by the way.
Same reason that solar works so much better. Farther south with higher ambient temperatures and more sunshine. Half South California (where most people live) is desert. When it is 110 degrees F in Palm Desert air cooling is pretty much a requirement for average human survival. A wet season goes from Oct to March adds to the problem. The list is really long these are just highlights.
When I lived there I had natural gas hot water heater and space “heating” reduces winter electric usage even more. Said space heater was a joke in comparison to what I grew up with in Western New York state.
All that said it is hard to quantify how much is used for cooling, ice making and such.
Of course it gets even worse when a state only is 200 or 300 miles north to south instead of 800 miles. Like say Virginia and points south where the humidity is much higher especially in the summer wet season.
Hope this explains some of it,
T2M
Evaporative cooling works well in dry desert climates. It requires much less energy to run than air conditioning. Of course it will not provide a winter indoor climate in the middle of summer that air conditioning can if that is your thing.
I used evaporative cooling for many years to keep cool in the Arizona summers. Because of the low humidity it generally works well. But evaporative coolers get clogged with scale, and anyone who has ever cleaned one out will recognize the benefits of air conditioning.
Air conditioning generates that extra demand.
I have been harping on this same issue for years now – daily demand changes are very difficult to iron out, but seasonal variations are almost impossible. Seasonal variation can ONLY be ironed out with immense physical energy storage devices. Over an over I have run the numbers, and solar and wind and batteries must be very nearly free for this to pencil out.
It won’t work. None of what they are proposing can possible work the way they describe. I wish it could, but it won’t.
Sufficient short-term storage to match supply to demand over mere minutes is already an extremely valuable service, especially if the response is near-instantaneous. It can eliminate operational requirements for spinning reserve, and for running thermal generation equipment at part load for extended periods. It obviates any concerns about sudden variation in output from solar or wind farms due to clouds or gusts (which spinning reserve already does, of course, but at the cost of some thermal or hydroelectric capacity).
Actually smoothing out the diurnal duck curve is harder, requiring storage sufficient for a day. This is exactly what pumped hydro and CAES are expected to be used for. Batteries don’t yet compete in that market; though no doubt they eventually will.
There are two ways to deal with the longer-term question of seasonal variation. One — only possible because solar and wind generation are becoming so cheap — is simply overbuilding. Even if there are hundreds of days per year when additional capacity contributes nothing whatsoever to requirements, if there are still hundreds more during which (with the help of the diurnal storage), more hardware can help meet demand in winter time, then it’s likely that more will be built. Your figure number 6, with total renewable generation equal to total demand over the course of the year, has solar output in January one third of the way towards meeting January demand. By the time sufficient solar generation has been built to meet that point, would additional increments really be so expensive that none more would be built? I doubt it. Solar equipment and its installation will be trivially inexpensive by that stage. In the UK there would certainly be concerns about finding sufficient space to put it, but I don’t think that will be the case in the south-western USA. California already imports power from as far away as New Mexico and British Columbia: within that radius is more than enough open space to site solar generation to power many times the entire world’s [traded — ie. excluding photosynthesis and passive solar warmth] energy consumption.
That large and probably growing excess in the spring and summer months would attract dispatchable or seasonal power loads. One of those, electrolysis and fuel synthesis, is ideally suited for seasonal storage. Whether in the form of hydrogen, methane, methanol, ammonia or even alkanes, synthetic fuel made using renewable energy can be stored and transported just as readily as the same chemicals made using fossil energy, and can serve identical purposes including being burned to meet seasonal generation shortfalls.
The fact that fossil fuels have remained stable in vast quantities in natural geological storage for millions of years is sufficient proof that identical products, sourced from renewable energy, can be stored for six months or so.
There are, to be sure, greater energy losses from synfuel manufacturing and burning than there are from batteries or pumped hydro energy storage. But when intermittent capacity is very cheap, and overbuild is happening anyway, and when these chemicals are only gradually substituting for the identical fossil-sourced product, this is unlikely to be a show-stopper.
What a lot of rubbish! Your only sentence that makes sense is sentence 1 of paragraph 2.
Want proof that it’s all rubbish? Read every blog post on this site.
Geoff:
I appreciate Jonathan’s contributions for the same reason that I appreciate Euan’s site – both the posts and comments are generally well written and thought out. There is very little emotion and rudeness; what is written is usually based on evidence or at least a sound thinking process.
The question of energy in the world today is so important because it affects everyone and the decisions that we make as a race today may affect us for good or bad for a long time. Sadly, many websites that discuss this important issue are dominated by those who believe that their preferred solution is so great that nothing else need be considered and that anyone who thinks otherwise is a fool who should be shouted down or intimidated if they don’t agree. Such places become emotive ‘echo chambers’ where no-one is challenged by the complexities, intricacies and inconveniences of the subject.
I find myself in the middle in this nuanced debate: wanting cleaner technologies but recognising that they have many short-comings. When I look for information, I often find myths repeated again and again without challenge and wonder how many ‘unknown unknowns’ there are. At energy matters, I find lots of interesting news and some very ‘inconvenient truths’. However, I know that things can be missed, even by researchers as knowledgeable and careful as Euan and Roger, so I want to see challenges to their work too.
So please don’t respond to a long, well written comment with a short derogatory one. Please don’t tell the writer to look at everything else on this site. I’d love to know what you find wrong with what Jonathan has written – so please go through the points carefully, challenging each one so that I can learn some more. References are always helpful where available.
Let’s keep the tone up, eh?
Except Pumped hydro is part of the proposed solutions.
How much did solar and wind cost yesterday? They only had 9 GWHs (~5% of what was produced) curtailed and still had negative prices for 7 straight hours on all three CAISO trading hubs.
http://www.caiso.com/Documents/Wind_SolarReal-TimeDispatchCurtailmentReportFeb18_2018.pdf
Guess they should have that hydrogen producing system on line then. Then it might make sense to install more in the meantime …
Oh they already have 2.8 GW of hydro in the CAISO system.
Color me unimpressed with people who get (over)paid even if the product isn’t used and infests the rest like rats in a food store house.
T2M
Dang two edits.
Previous post was in response to Jonathan Maddox.
And the first sentence was suppose to read, “Except Pumped hydro is NOT part of the proposed solutions.”
oops,
T2M
Because photosynthesis efficiency in living matter is <=2%, it must be the solar energy captured in fossil fuels over past billions of years was so immense and astronomical.
Now, whatever an energy-generating device we construct today using fossil fuels, the total energy expended in its construction will never be matched by the useful energy the device will ever produce, a recently circulated thesis inspires;
https://the-fifth-law.com/pages/press-release
You will have to do a lot better than that press release to convince anyone on here that what you say only applies to Wind Turbines & Solar Panels and not other methods of Energy Generation.
The new proposed relationship in thermodynamics is a universal, applicable to all energy-generating devices, from ICEs to hydro, nuclear, renewable, geothermal, the Sun and beyond.
The press release wast just about to register the event of that humans have been aware of this constraint at least since 2017!
2018 Jacobson study:
http://web.stanford.edu/group/efmh/jacobson/Articles/I/CombiningRenew/WorldGridIntegration.pdf
Thanks Zigak: Lots of graphs there for me to get my teeth into. Back later.
In a decade or so, adding a day of (average) storage to substations and transformers at their time of replacement will likely pay for itself (due to higher use rate of lines, less stress on transformers, and lower reserve needs for maintenance crews and equipment).
Adding a mostly electrified transportation fleet to the equation (with a few days worth of storage on average) and large incentives to charge off peak will greatly improve the situation (which many consumers will purchase storage to take advantage of). These two factors combined can solve the intermittent problem excepting the seasonal problem. Although seasonal pricing will definitely ameliorate it somewhat (especially for businesses).
If Hoover Dam basically only ran during summer months and refilled to capacity the rest of the year it would make a pretty good start on the seasonal problem in the SW. Running HVDC infrastructure to SoCal and LV would also allow significant optimizations in efficiency, especially adding in massive solar installations en route (only pulling from the dam when the lines aren’t maxed out by the active producers) and tying into a wind HVDC from the south or east. Eventually HVDC could also connect the massive hydro resources in the NW to NoCal in the same manner. If all interconnected, the NW could run primarily on cheap solar and wind year round and its hydro resources mostly saved for the summer in the SW.
If trends for solar and wind and batteries and DC distribution costs and efficiencies continue (and financing costs remain super low) it seems to me the west half of the US could be 100% renewable at similar retail prices to SoCal (albeit with an entirely different rate scheme which does a much better job of shifting demand). A much less optimal system could be designed for the eastern half, similarly using all hydro resources (mostly in Canada) for winter and relying on western grid surplus year round (in addition to significant offshore and super high hub heights for onshore plus significant solar in the southeast). In that system, having a couple days worth of storage at substation and transformer (and producer and consumer) probably would be needed to cover overlaps in weather + western surplus. We’re probably talking 50% over CA rates averaged over the year.
Seems like the Jacobson litigation is still ongoing:
https://eos.org/articles/scientific-row-over-renewables-leads-to-free-speech-legal-fight
http://www.sandiegouniontribune.com/business/energy-green/sd-fi-jacobson-lawsuit-20171204-story.html
The Jacobson’s study uses renewable oversupply to feed different flexible loads, representing diverse forms of storage (hot water and cold water thermal energy storage, underground thermal storage, ice storage, hydrogen electrolysis and storage, batteries, etc.). Considers the Demand Response too. There is a huge “artificial” peak load increase to absorb excess of renewable generation and the power system is used more to feed storage loads than the traditional consumption. Anyway, batteries continue to be defined of low storage capacity (2 hours) and Pumped Hydroelectric Storage (14 hours). It seems that the flexible load is the main tool for matching demand with supply. Nothing is said about system capacity to cope with sudden changes of generation (ramps) and renewables forecast errors.
The CPUC Reference System Plan (California) for 2030 does not follow such radical and doubtful approach. This Plan also considers batteries as a short – duration storage (~1 hour) and increases a little the natural gas generation and peaking plants. Therefore, the new battery storage is just an ancillary service provider. Under a structural point of view California remains based in gas generators to backup intermittent renewables.
http://cpuc.ca.gov/uploadedFiles/CPUCWebsite/Content/UtilitiesIndustries/Energy/EnergyPrograms/ElectPowerProcurementGeneration/irp/AttachmentA.CPUC_IRP_Proposed_Ref_System_Plan_2017_09_18.pdf
California also cheerfully consumes the the fossil fuel generation of other states.
Jacobsen’s graphic shows 52 million jobs created. The US could do that tomorrow by outlawing all mechanised farm equipment and forcing farmers to use only manual labour. On top of that, make sure that the farmhands are paid $15 an hour and food will become so expensive they’ll sell it only to the nomenklatura at Tiffany’s.
@ Robert Hawcroft , Alevo went bankrupt last autumn (2017).
I profoundly disagree with your assertion that “Another option is hydrolise sea water to get hydrogen or ammonia. That can easily be stored and then burnt in a turbine. ”
Presumably you mean “hydrolysis of sea water”?
The round-trip efficiency of the process you describe in the very best turbines ever built may be 25% and the capex is off the chart.
PLEASE LET’s stay serious in this forum
Unfortunately, when “batteries” are mentioned/lauded in relation to running the world on “green” energy, the writer most often means “lithium ion” batteries which, because of their ubiquity (and absolute necessity) in hand-held electronic devices are possessed by almost everyone on on Earth.
These wonderful batteries have been around for some 50 years because of their thus far incomparable energy/weight ratio. No other battery chemistry comes close.
Unfortunately, for this reason, they are intrinsically unsafe as demonstrate still by the catastrophic fires they cause when something goes wrong. Airlines are not allowed to transport lithium batteries in bulk.
They come in a wide variety of chemistries, involving, variously, lithium, cobalt, manganese, titanium, iron, phospherous and aluminium and always, when at a larger scale than a few Ah, in packs, most often of small, cylidrical cells. The larger the battery pack, the more cells, of course.
Particularly high energy/weight, energy/volume ratio chemistries are required in hand-held devices, cars, drones, etc,. Cathodes for these batteries invariably requirire high fractions of cobalt which is relatively scarce. Both the price of battery-ready lithium and cobalt have increased hugely in recent years and the ever rising cost and availability of these metals is well dealt with in articles like this https://seekingalpha.com/article/4145849-tesla-artfully-dodging-discussions-technology-metal-costs-supply-chain-risks?app=1&auth_param=ggqr:1d85t19:d8580717f8462b80670450abfeb6e1d9
All lithium ion batteries “wear out” as any smart phone (or EV) user knows only too well.
Simply because of the many lithium ion battery chemistries there are, recycling these is a huge technical challenge. As I write, I believe that there is still no commercial (ie profitable) process for recycling lithium batteries, so that their end-of-life destiny is in a toxic (ie chemically reactive) dump.
Because the Belgian metals giant Umicore has owned and operated a 7000 t/y process (http://pmr.umicore.com/en/batteries/) for several years but has chosen not to expand production beyond this, despte the now, many-hundreds-of-thousands-of tons-per-year opportunity, rather makes my point.
So here we are, exposed to nonsensical claims by “green” lobbyists, claiming that the world can be largely run on wind and PV, while the materials that are necessary for making all these good things happen are ever scarcer, more expensive and unobtainable through recycling, the very acme of sustainability.
If any reader can correct me, I would truly be grateful. I do declare that I am and always have been a “tree hugger” and an incorrigible “peak oil” believer and grumpy Malthusian who sincerely wishes the facts were otherwise.
Minor nit.
LiFePO4 batteries are “supposedly” safe. Safe enough that a .50 cal round through a pack doesn’t cause any safety issues. No longer much use as a battery though.
Of course the energy density is much lower but they are claimed good for 2000 cycles. Switching my campground solar to that this year. I’ll see how they work.
I have no clue on the the recycle ability.
Just saying that generalizations are generalizations.
T2M
Thinkstoomuch!
Much appreciated!
You may be right about lithium, iron phosphate chemistries, where BYD is the largest player, mostly in China.
I focus mostly on the most power dense chemistries which form the majority of the market, use a wide range of cathode metals like cobalt and present the biggest challenge for recycling.
Could you perhaps dig into recycling of BYD batteries and report what you find, please? You are such a diligent seeker after truth, you won’t be able to resist that challenge! 🙂
The use of shorter life chemistries for a grid battery still astounds me. If a battery costs twice as much for twice the life labor costs are half better bargain. But that is my simple mind not understanding the financial complexities.
What batteries aren’t made in China anymore. Just like I pointed out that a aluminum smelter closing in New Zealand “due to oversupply” was not due to over supply. The world is using more every year “oversupply” was an inaccurate fact. It was in “oversupply” because someplace else made it cheaper. They just didn’t say that electricity is cheaper when generated with coal(other things too) than the cost in New Zealand.
Just like people saying that you can recover 60% of the battery using current recycling techniques. Ignoring that the other 40% includes all the active elements in the battery. The lithium and whatever it reacts with. Which is the issue that is pointed out in the inability to recycle.
As far as your challenge ain’t doing it. :-p
The same problems with recovering stuff is still there. I typed before I thought(common problem on my part). Lithium’s location on the periodic chart doesn’t change. Thank you I can find my own windmills and my motorcycle makes a poor weapons platform. : lol :
T2M
Hugh:
A brief historical perspective on lithium batteries.
When I first got into mining, the company I worked for (Kennecott) was trying to sell its Missouri lead mine. Why? Partly because lead was nasty stuff but largely because lithium batteries were going to replace lead-acid batteries any day now.
The date? 1969.
Later I got the company involved in Salton Sea geothermal, but not because of the geothermal. It was because Salton Sea brines contained lots of lithium that was potentially extractable, and lithium batteries were going to replace lead-acid batteries any day now.
The date? 1980.
Plus ça change
I’ll also take this opportunity of thanking you for the link you sent me showing that National Grid doesn’t care about battery storage that lasts for more than four hours.
Thanks Roger!
As regards hand held devices, modern electronics could not be so ubiquitous and affordable without lithium ion batteries. On that we can agree. Your mining chums have done very well out of them.
My only quarrel with the “green” optimists, like Jacobson, is that they make such unreasonably optimistic extrapolations from so little data. Even 5 minutes of thought together with known facts should stop them rushing to print, twitter, Facebook and the like.
Lithium ion batteries are economical to recycle currently by smelting the battery and retrieving the metal alloy and dumping the slag (lithium, plastics, etc). This works with any lithium ion chemistry as far as I know. If you have a large volume of a known chemistry (and ideally, casing and TMS) you can do far better, for instance targeting lithium cobalt compounds that can be directly reused. The problem is as you said, there are many chemistries and at the normal points of recycling they do not have the volume necessary to accommodate them all nor enough OEM knowledge to properly identify chemistries from many sources that are (also) constantly being tweaked.
Ideally IMO, each business entity involved in the sale would bear responsibility for recycling and pay a fine based on the amount and composition of disposed material. A person would be able to return a battery to the dealership they bought the EV from or direct to the manufacturer. The manufacturer would either recycle the battery directly or return it to the battery manufacturer (who should have a domestic recycling plant to avoid the maximum disposal fee).
I think it’s also useful to note that significant value is to be realized before recycling. An EV pack may no longer have sufficient range for its driver after say five years but will have sufficient range for others for say 15. Furthermore, even after removed from the vehicle it has significant value as a backup battery (these are replacing lead acid installations presently). Value can be enhanced further by replacing bad cells (and even more via rebalancing). Given all that, it is likely manufacturers would merely store bad cells for the next decade or so until enough volume built up for a particular chemistry to recycle it. They have all the info needed to optimize recycling, just not the access to enough of the batteries nor environmental incentives.
Thank you meliorismnow!
Please give me chapter and verse for this claim which I reject
Even the incorrigibly optimistic Musk (I suppose?) claims at https://hornsdalepowerreserve.com.au/faqs/ that “Tesla will safely remove all batteries from the site when the facility is decommissioned. Tesla will recycle all returned battery packs and modules at its Gigafactory in Nevada, United States, where over 60% of the materials, especially critical minerals, will be recovered for reuse.”
TESLA information is cluttered with such claims which are frequently altered.
I don’t understand your question/request. Do you want me to respond to that FAQ answer? It seems smart for Tesla to offer recycling for its batteries and it certainly has the scale to do so for the few chemistries it currently offers (this one is Samsung made). I’m not sure if Tesla can make money on 60% recovery (assuming Australia won’t pay recycling fees) but assuming they have fifteen years to develop and test techniques it certainly seems plausible. On the plus side, nickel, cobalt, and lithium prices should continue to increase. On the negative side alternative chemistries may obviate these over a fifteen year period (grid storage may migrate to vanadium or molten salt etc while EVs and gadgets may migrate to solid state). In that case, smelting to retrieve just the metals might be more economic.
Jacobson makes his own weather, as well as his own assumption on the degree of demand management that is supposed to be possible. Both are highly suspect. It would take some detailed examination of his data to show the extent of these hidden assumptions.
In S. California, 35 deg north, mild winters and peak demand in the summer, it is at least worth thinking about solar PV. In the UK, 55 north, average solar LF in the winter is 2-3% and can be <1% for 2 or three successive dull days. As McKay says we'd be pushed to find space for enough to panels to meet current electricity demand, let alone what would be required in the fantasy world of electricied heating and transport.