More Guest Blogs
By W. ROCKY NEWMAN
One year ago Tesla Motors announced plans to build its Gigafactory to produce huge numbers of batteries, giving life to the old saying, “if you want something done right, do it yourself.”
By making electric car batteries that Tesla used to buy from others, CEO Elon Musk adopted a strategy made famous by Henry Ford – build a vertically integrated company that controls the many stages of production. By integrating “backward” into its supply chain, Musk is betting Tesla can improve the performance and lower the costs of batteries for its vehicles.
Now, Musk wants Tesla to acquire SolarCity for similar reasons, but with a slightly different twist.
SolarCity is one of the largest installers of photovoltaic panels, with some 300,000 residential, commercial and industrial customers in 27 states. The proposed merger with SolarCity would vertically integrate Tesla forward, as opposed to backward, into the supply chain. That is, when people come to Tesla stores to buy a vehicle, they will be able to arrange installation of solar panels – and potentially home batteries – at the same time.
This latest move would bring Tesla one step closer to being the fully integrated provider of sustainable energy solutions for the masses that Elon Musk envisions. But does it make business sense?
The real issue in my mind comes down to batteries and innovation.
Creating demand and scale
Although installing batteries is not a big part of SolarCity’s current business, the company is a potentially large consumer of Tesla’s batteries from the Gigafactory. Tesla makes Powerwall batteries for homes and larger Powerpack systems for commercial and industrial customers.
Any increase in the flow of batteries through the factory gives Tesla better economies of scale and potential for innovation. Innovation comes with the accumulated experience gained from building a key component of its electric vehicles as well as Tesla’s energy storage systems. As the company manufactures more batteries, it will find ways to innovate around battery design and production.
Because batteries are the single most important component to driving range between recharges and the cost of cars, the synergy between battery and electric vehicle production is obvious.
Similar synergy could be argued between SolarCity and the Powerwall stationary battery business. SolarCity’s integration into Tesla would likely give a combined company more opportunities to pair Powerwall systems with solar panel installations in the near term. And more demand for batteries pulled by SolarCity’s marketing to solar customers should make the Gigafactory more profitable once it is up and running.
Some analysts claim the proposed merger is a bailout of SolarCity, where Musk sits on the board along with several friends and relatives.
But all of these issues of scale and cost are relevant reasons to argue for the merger in the near term. The most important question, though, could be further down the road at the Gigafactory in Nevada.
Manufacturing feeds innovation
Even as Tesla makes another move to increase accumulated demand for batteries, it will also be ramping up the pace by which it accumulates opportunities for innovation in the design of those batteries. Greater demand means greater production scale and scale leads to innovation and cost reductions.
Better battery design and lower cost make products like electric vehicles and, potentially, photovoltaic installations more economically feasible to a broader market. This approach worked for Ford more than 100 years ago. There is no reason to think it won’t work for Musk today.
Historically we have seen this in many industries. Examples include Ford and the Model T 100 years ago, Texas Instruments and others making hand-held calculators in the 1970s, as well as personal computers since the 1980s.
Maybe the best example would be home video recorders. When home recorders were first introduced in the 1970s, Sony sold them for $1,300, which would be about $4,600 in today’s dollars. As the demand for home recording increased, and an industry standard emerged to consolidate demand on one format, scale accumulated and both design and cost innovations resulted in vastly improved functionality at prices well below $100 by the early 2000s.
This was all made moot by the introduction of the DVD, which made the VCR obsolete. DVD players later took the same cost reduction trajectory. While there is no guarantee the results will be the same with battery design and cost, increased scale is the key to exercising that potential.
Will consumers buy it?
Another catalyst in this situation will be the increasing availability of “time of use” rates for grid-sourced electricity. For years, utilities have instituted rate plans that charge more for usage during peak hours (typically midday) versus steeply discounted non-peak (nighttime) rates. This allows consumers the option to shift demand to non-peak hours or simply reduce peak consumption to lower their electricity bills.
With time-of-use plans, Powerwall batteries can be used as a means to store grid power drawn during off-peak periods and use it to supplement peak-time consumption. For example, a Powerwall system capable of storing 10 kilowatt-hours (kWh) of electricity costs about $3,500.
By my calculations, using $0.25 per kWh off-peak rates to charge a battery and then using it to supplement $0.34 kWh peak rates would save 3,650 kWh of peak demand rates. That translates into saving $324 per year or a 9% return on investment on the system. Powerwall batteries can also be used as back-up in case of disruptions to the power grid.
But pairing a Powerwall with a solar panel installation could be where the biggest advantages can be found. The typical U.S. household uses about 1,000 kWh per month. A homeowner can have a 5 kW solar system installed for about $10,000 (after tax credits) and depending on the climate, generate most of their monthly grid needs.
With many climates in the U.S. getting between four to seven hours of sun per day, a 5 kW solar system could generate as much as 1,000 kWh per month. By adding a 10 kWh Powerwall, the user can store excess solar generated power or non-peak grid-sourced power and use it when most advantageous.
Starting the peak period every day with a fully topped-off Powerwall could save, by my estimate, as much as 3,650 kWh of peak usage per year. Added to the solar energy generated and used immediately, many users could avoid peak demand use from the grid altogether — at least on sunny days. This, depending on local rates and climate, could easily generate a return of 20% or higher.
Focus on innovation
So will the Tesla name and potential for seamless integration between the home energy storage and car markets (made possible with the merger) increase demand for SolarCity’s solar panels? It seems logical. Will the same integration as well as increased market share for SolarCity increase battery demand? Also logical. Will this advantage also hold for industrial users who tend to value sourcing both from the same vendor? Again, logical.
But the real game-changing question I see is: Will the increase in accumulated demand for batteries generate the opportunities for innovation?
By ramping up production at the Gigafactory, will Tesla find ways to lower battery cost and improve battery performance with better energy density (energy storage capacity relative to weight)? And will that lower cost and better performance attract more first-time users, broadening the market and increasing the likelihood of even better and cheaper batteries?
If you can extrapolate from Henry Ford, calculators, personal computers, and even VCRs, you would also say this is very likely. As battery design improves and cost falls, the Powerwall battery should become more attractive economically and the the forthcoming Tesla Model 3 electric car should be affordable for an ever-wider demographic. I signed up for a Model 3, along with about 400,000 other people.
Is this the synergy that Musk is really banking on? I think so.
W. Rocky Newman is a professor of management at the Farmer School of Business at Miami University. This post originally appeared at The Conversation.
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
14 Comments
OK, where in the US...
...does this statement square with local utility market rate reality:
"...using $0.25 per kWh off-peak rates to charge a battery and then using it to supplement $0.34 kWh peak rates..."
????
The national average FLAT residential rate is barely half the 25 cents used as the off-peak example. Only Hawaii, some island grids and remote Alaskan villages powered by diesel gensets would have those sort of rates.
https://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a
Using that example as some sort of paradigm without discussion of how & where those numbers are even possible feels like a straw-man type of argument.
In Australia where the rates are 2x higher than the US and PV costs about half what it does in the US, in locations where the remuneration of exports of rooftop PV is paltry or abusive, residential storage for self-consumption of site solar production is just now starting to make financial sense. But most of the US is still a LONG way from financially rational local storage. With the learning curves of both PV and batteries that may change in a decade or so, but not by 2020. For now it only makes sense under commercial rate structures where the ratepayers are assessed demand charges, as a means of lowering the peak draws to limit the demand charges.
If electric vehicles proliferate in the next decade there will be surplus of used batteries suitable for grid storage, which may be at an attractive price point for re-purposing as behind-the-meter grid storage, and far cheaper than new purpose-made grid batteries coming out of gigafactories.
Dana,
Not where I live. My
Dana,
Not where I live. My weighted average with TOU is $.13 which is $.07ish off-peak and $.22ish peak. I am drawing about 8,000 kwh/year off-peak to charge my car.
Now I am one of those crazy folks who is planing to go with battery back-up for my non-certified Passive House. Can I justify it financially? Maybe. After Sandy and Irene, everyone installed a whole-house generator, including myself. Average installed price was probably about $10k. Nice to have, but it creates more work for me. I have to change the oil, sparks plugs, air and oil filter once a year. Most people contract the service out at around $350/year. It burns NG for 10 minutes a week to keep the battery charged, it looks ugly, it is noisy, blah, blah, blah. I would rather spend the money on a battery that I don't have to do anything with and with the low energy demands of a passive house, I can probably live through another Sandy fine, and my triple pane windows will keep the noise from my neighbors' generators out of my house. For me, on a dollar for dollar basis, I would rather have battery back-up then generator back-up.
Funny, I was at town hall the other day filing for my demo permit and a lady walked-in and asked for a permit application for a back-up generator. I thought to myself, if this was 2 years down the road, I wonder if she would have walked-in and asked for a permit for battery back-up instead of a generator.
Predictions on the adoption rate for residential batteries
In case any GBA readers missed it, this article includes my report and predictions concerning the current and future cost-effectiveness of residential batteries: Batteries for Off-Grid Homes.
7kW powerwall
The 10 kW powerwall has been discontinued. Also, it was only meant for infrequent use. Not daily load shedding.
The 7 kW product is still available at the listed $3,000. This one was designed for daily use.
I agree these stated rates appear high. very high. But the difference of 9 cents might still be close enough for the payback calculation.
"...using $0.25 per kWh off-peak rates to charge a battery and then using it to supplement $0.34 kWh peak rates..."
The problem I'm seeing is that Utilities are very opposed to the idea of buying back their own power at inflated rates. They are getting laws written to reduce this payback. They're are even fighting to stop net metering for solar. I see the Utilities getting the rates fixed in a way the will encourage most people to have a combination of solar and batteries, and the solar panels will almost never feed power back into the grid, and then they'd charge a huge connection fee just because.
You seem to assume an infinite battery life...
Your calculations do not seem to take into account the fact that you are wearing out your battery pack, that it will probably only last ten or fifteen years (ok, I'm pulling those numbers from a dark place, but probably roughly accurate) at that level of discharge. If it lasts for 15yrs, your cost is about $200/year, cutting your payback considerably. If lifespan is much under ten years, your return in the example given goes to zero.
There is also the cost of the power required to keep your batteries charged. All batteries have a discharge rate and you have to account for that.... and many battery chargers have a horrendously wasteful power factor.
There certainly are some good reasons to have a battery system:
>low maintenance compared to a generator, so that is probably a good payoff.
>If utilities start to charge us a demand charge, a battery pack that can shave our peaks might pay off handsomely as well.
The powerwall is still glittering
The powerwall is still glittering in the distance. Very few have been installed though you can see all the details of a fully installed PV powerwall system in Australia that has been functioning for over 6 months at "unleash the powerwall".
Something strange happened to the whole powerwall thing. Debacle might be the word I am looking for. However I am not so jaded that all hope has passed. The new powerwall2 is supposed to be released in November. It includes an integrated inverter designed and built by Tesla. This should reduce price, installation costs, complexity and shipping. And if it all works out it may have a future in North America. We will still need the ITC 30% reduction for it, as we do with PV, but the numbers improve drastically with all these new factors....if, of course, Tesla delivers.
Infrastructure Investment
Isn't the longer goal zero net energy neighborhoods and districts? Utilities have demand reduction programs to reduce the need for new generation capacity. Especially as the grid gets cleaner - California has a goal of 30% renewables by 2020 and 50% by 2030 - should we begin discussion of how much battery capacity it makes sense to install at the individual building level? Peak reduction makes sense. I am not sure at a societal level it makes sense to install enough batteries to fully go off grid. I'm not advocating for anything! I don't have enough expertise. Just asking whether this should be part of the discusion.
The amount of resources to go off grid with PV & batteries...
...as opposed to grid-tied/micro-grids st huge, and generally uneconomic. No serious analyst believes that autonomous single-house energy units makes any sense at all.
But the notion that power should be primarily served from large centralized power generators the way it has been for the past century is also nonsensical. The more distributed energy resources there are on a grid, the more reliable it can be become, simply because there are few single points of failure that would have consequences to other grid users. If a hurricane takes down a gigawatt of transmission line connecting a large central power plant to the grid it is a major perturbation all at once, a threat to grid stability. But if a gigawatt or even ten gigawatts of small scale rooftop solar in a region get blown away in a storm, it doesn't happen all at once, and it's a reliability problem. With some amount of distributed storage and some smarts to support "islanding" portions of the grid during major events, even large disruptions can be smoothed out.
The amount of battery it takes to have a stable distributed grid really depends on the local details, and even in storage mandate states such as Oregon, Massachusetts, and California there are many details to be worked out as to what type and how much battery gets installed, what services the batteries will be providing, and which batteries will be under the grid operators or local utility's direct control. Batteries can have quite a bit of value to the grid other than raw energy storage, and are becoming the go-to technology for frequency and voltage control in lieu of fast ramping peaking generators, which burn fuel to be available "just in case". Market structures have to be created to get the full value out of distributed batteries on the customer's side of the meter, but the regulatory environment is changing pretty fast in many states, though none are moving at quite the pace of New York. New York does not have grid storage mandate, but market structures are being worked out for properly remunerating distributed grid resources of all types.
Urban NYC has problems to work out before large amounts of energy can be safely stored in distributed lithium battery packs (just as you can't store a million gallons of gasoline in high density neighborhoods), but as safety regulations get hammered out it's likely that more single building scale storage. Single building high-rise microgrids already exist in NYC, but they are not including 50 tons of lithium battery pack on the 12th floor just yet...
With investments in more grid infrastructure simply shipping the output of variable renewables around and utilizing demand response markets would eliminate the "need" for grid storage in an all-renewables environment, but the bang/buck for those investments needs to get evaluated relative to storage solutions as well. It's a moving target, with shifting regulations and rapidly evolving cost structures. There aren't likely to be any "one size fits all" solutions to the energy distribution & storage issues as ever more variable renewables come on line, and it won't look the same everywhere.
Just one current example of how a change in the regulatory environment and who can participate in which market changes the picture, and the likely shape of future development:
http://www.utilitydive.com/news/after-tariff-changes-storage-to-play-larger-role-in-caiso-demand-managemen/425698/
Bill - I think Tesla thought
Bill - I think Tesla thought there was a bigger market for battery back-up to replace generators (which is one of the reasons I want batteries), hence the 10kw non-cycling option, but I believe they have found there is a bigger market for load-balance, TOU arbitrage, etc.
Skip - Batteries wear like everything else, but they have much less moving parts which limits maintenance requirements and they are scalable. My Generac starting giving me "governor default" alarms right after the warranty expired. Now I am getting a low oil pressure warning and it shuts down after 30-seconds of operation. Checking the Generac boards, it appears to be a common issue caused by a faulty panel. I am sure that is going to cost me $300 to fix.
Ven - I had a chance to speak with one of the higher up Tesla battery guys at an energy conference a few months ago. He was pretty clear that residential PowerWall is not their target. Their primary target is the Commercial/Utility market where their battery (called PowerPack) is used for Demand/Recovery/Load Balancing/Arbitrage, etc. The 7kw pack is dwarfed by the utility grade packs Tesla is selling. One of the presenters at the conference talked about electric service providers currently using batteries to buy off-peak power, sometimes at negative rates, and then resell at peak rates during the day. Another guy talked about a major utility building solar farms with battery storage and that it was cheaper for them than trying to build a new fossil plant that could burn when the sun doesn't shine. The second market Tesla is going after is the residential arbitrage market, such as Australia and California. Carbon cutters like me are at the bottom of their list, but he said product should be available by the end of this year and I won't be installing batteries until next summer anyways. Tesla is just one provider. Sonnen in probably the biggest in the residential market and they provide a fully integrated unit with inverter and software. They also claim to have an active market place that allows the owner to play arbitrage and sell excess power stored in the battery, not just to utilities but to other end-users. LG and Siemens also have battery options. There is a lot of capacity coming online, Tesla is only about 25% of that and pricing is going to drop.
Bill/Dana - I think the end goal is carbon zero, because net zero still requires utility power and without storage (be it battery, trains going uphill, reservoirs, etc.) that power cannot be carbon zero because there is not enough renewables energy that can be generated from dusk to dawn and nuclear capacity has probably peaked. I think we can take a big chunk of the residential carbon out of the equation by building new communities that are 1) very energy efficient- near climate specific passive house standards, 2) locally solar powered, whether that be rooftop or centralized in a section of the development set aside for panels, and 3) tied to a means of storage, probably a centralized PowerPack-like plant. Any excess power can be sent back to the utility. You can allocate a certain amount of power to each home based on the size of the array and if they use less than their allotted amount, they can get cash back. So you create an incentive to change behavior. EV's would probably have to be set up with a separate meter and a fixed rate because you don't want to discourage EV use.
Perhaps SolarCity and Tesla can making a profit first
These companies are becoming TBTF out of hubris if nothing else.
It's easy to overestimate the storage requirements (@ #9)
"Bill/Dana - I think the end goal is carbon zero, because net zero still requires utility power and without storage (be it battery, trains going uphill, reservoirs, etc.) that power cannot be carbon zero because there is not enough renewables energy that can be generated from dusk to dawn and nuclear capacity has probably peaked."
The notion that storage is needed because "Sometimes the wind doesn't blow and the sun doesn't shine" is patently false. The sun is always shining (somewhere) and the wind is always blowing (somewhere). It's a matter of scaling geographically, and having the grid capacity for moving it around. And the problem isn't not having enough renewables from dawn to dusk- its all about finding enough load when there is excess. Right now Iowa is getting more than 30% of it's total grid power from wind, but without curtailment of the wind power output on windy nights is due the ability to export power to neighboring areas over the grid. The key is finding the optimal renewables mix for the local grid-region to minimize curtailment of output, and storage only comes into play when the curtailment levels would render the projects uneconomic. There's nothing magic about storage, and as renewables keep getting cheaper, the amount of curtailment that is tolerable increases.
Even with just the existing transmission grid it's possible to get to zero-carbon using intermittent renewables with only modest amounts of grid storage- orders of magnitude less storage than if treating it as a single location nano-grid the size of a single house (aka, an off-grid house). See:
http://ac.els-cdn.com/S0378775312014759/1-s2.0-S0378775312014759-main.pdf?_tid=30cdb440-7532-11e6-bdbb-00000aacb361&acdnat=1473277091_3266798c9cdc463353c8ffb34d6e0b94
There will still be transmission capacity and interconnect upgrades that are more economic than curtailment or storage, but you don't need baseload power to make a 100% grid that's still cost effective.
Inflexible large scale generators such as thermal coal or nuclear increase rather than decrease the storage requirements once intermittent renewables are supplying over 50% of the total grid power, due to the painfully slow ramp times. In CA that inflexibility has been a problem as far back as the 1960s, when pumped storage projects were built to be able to manage the overnight excess power as more nukes came on line. Part of the deal for building the Diablo Canyon power plant while it was still in the planning stages stipulated simultaneously building a pumped storage facility to be able to take that much overnight excess, which became the Helms Pumped Storage Project. Diablo Canyon is going away, but Helms isn't.
The new kids on the block are large scale demand response programs being bid into both grid balancing and capacity markets, and it's likely that smart chargers for electric vehicles will become a significant factor as more renewables get built out. (FERC Order 745 requires among other things, that demand response be allowed to bid into capacity markets, which was fought by the power generators all the way to the US Supreme Court, but is now settled.) With a ready made load like a fleet of EVs a lot less curtailment or the amount of dedicated grid storage required to manage the excess intermittent renewables output, but demand response alone is soon going to displace fossil fired peakers in CA:
http://www.greentechmedia.com/articles/read/big-changes-proposed-for-demand-response-in-california
The current lifecycle cost of energy of midwestern wind power is already cheaper than combined cycle natural gas- you can over-build wind and still absorb curtailment losses before it hits parity with fossil burners. The LCOE of distributed small scale solar is still more expensive, but in some US locations large scale solar is already at parity with combined cycle gas, and expected to be cheaper than wind by 2025 (even though both will be cheaper by then.) The cheaper renewables become, the cheaper it is to build out enough to guarantee that the grid stays up without resorting Godzilla-scale battery storage to manage a zero-carb grid.
The learning curves of wind, solar, and storage are all pretty steep, and the financial analysis of what's optimal is guaranteed to change, but the story on renewables in the US is really the substantial size of the resource relative to the size of the load. In a recent NREL technical analysis published earlier this year, it's possible to deliver about half the total power used in the US with rooftop solar alone utilizing the existing roof pitches suitable for solar. This would of course be producing over 100% of the mid-day load on sunny days, but that doesn't mean that batteries are the solution, or necessary for covering the overnight load (unless you were for some reason planning on an all-PV grid, which is just silly.)
http://www.nrel.gov/docs/fy16osti/65298.pdf
Of course building that all out isn't going to be economic, but a lot of it is, and the more that goes up, the greater the resilience of the grid as a whole. But getting to zero carbon really a matter of managing the surplus, of which grid storage is only a bit player. For now, simply targeting applications served by low capacity-factor low efficiency fast ramping peakers and deferring substation upgrades the best financial case for grid storage on the utility's side of the meter.
TBTF? Hardly! (@ Chris M)
Don't be fooled by the ridiculous market cap that puts Tesla at half the size of GM!
Tesla has fewer than 15,000 people on the payroll (which is only about 200,000 fewer employees than GM). Total employment to date at the battery factory is maybe 2000, and is targeting 6000 (in your dreams) when it's in full production. Tesla is definitely not in the same class, not even close. Solar City may have about 1/3 of the rooftop solar market for now, but the rooftop market is smaller than the utility scale solar market. Their workforce is comparable in size to Tesla. So, combined the company is still on the order of 30,000 employees, or about 1/7 the size of GM, by total employment.
If they failed tomorrow (together, or separately) Panasonic would likely still employ the folks at the battery factory (worldwide the market for EV batteries is still huge, and Tesla is a high profile but still small player in the EV market). Most of the engineers at Tesla wouldn't have to even move to find comparable jobs in CA, and most of SolarCity's employees would be picked up in a heartbeat by other solar companies- it's still a high growth industry.
The are nowhere near being TBTF right now, but maybe in a decade, if their wildest dreams come true.
Making a profit when growth rates are that high isn't always realistic for booming yet highly competitive industries. The capital required for sustaining growth while retaining market share definitely takes a toll. It's not clear that the merger makes sense, but Elon Musk has a demonstrated knack for inventing creative business models and making them work. I'm not counting them out just yet (though I have done so in the past only to be proven dead wrong.)
Dana,
Interesting analysis of
Dana,
Interesting analysis of the cost minimization model. Basically their conclusions are that inland and off-shore wind are our most-cost effective way to get to carbon zero, if we desire to go that far. At that level, there will be a lot of overcapacity, but that is not surprising. Now that is a theoretical analysis that may or may not happen. What I know is that I live in a state with lots of shoreline, but less than 2% of my local electricity is generated from non-hydro renewables. The majority of my generation comes from nuclear - 40%, nat gas - 30% and then goal - 25%. The nuclear plants have limited life left in the them and I don't consider fracked gas to be so environmentally friendly. My state is going to need additional sources of electricity as that nuclear winds down, wind would seem to make a lot of sense. Ironically, despite having excess wind capacity in Iowa, Des Moines generated more than 62% of its electricity from coal, according to Power Profiler. Now that is today and maybe it is just a function of wind still ramping up.
In about a year's time, I will have a house that is close to carbon zero and I don't need a lot of battery capacity to get there because the majority of my load will covered by solar during the day including charging the EV, making hot water, washing clothes, dishes, etc. This may not be the optimal strategy in the long-run, but it is something in the meantime and it saves me from having to get another stinky generator.
As for Tesla, what happens to them is anyone's guess but Musk is tough to bet against. I can say that they make a fantastic car, by the far the best car I have ever owned. Every time I get in a rental car I am reminded of how superior electric cars are. Because of that, I think the world is going EV. It is a game changer in my mind and the cat is out of the bag. That does not mean Tesla will succeed, but it does mean that consumers will demand real EV's and I think car manufacturers will have to oblige. Now If Tesla finds itself in bankruptcy, which would be because it was too ambitious and not because it makes a bad product, there will be buyers lined up for the assets. I can't say I see what is so special about Solar City. They are just a service provider in my mind. Can Tesla do more with those assets than someone else? Probably. Does that mean it is a good buy? I don't know.
@Dana. I was thinking from a political perspective.
They are the darlings of DC.
Log in or create an account to post a comment.
Sign up Log in