Tesla's new core business? CES Grid Storage Device For SuperCharger/SuperSwapper

[rolosrevenge]

Some data might be helpful in your revenue models. Any grid storage device will be on the wholesale market, so forget about your retail rates. The gap between on- and off-peak wholesale prices is far smaller than you are supposing. Here are numbers for purchase today of annual strip power products for 2014:

	PJM	PJM	CAISO	CAISO	CAISO
	East Hub	West Hub	CA/OR Border	NP15	SP15
On-Peak	52.71	47.93	40.98	45.84	51.90
Off-Peak	35.75	33.43	29.33	35.72	38.60
Delta	16.96	14.50	11.65	10.12	13.30

These prices are $/MWh, so we're looking at 1.0 to 1.7 cents/kWh of price swing. Admittedly, you can do better than this: there are some hours when the prices go way above the average on-peak price. But you'd be lucky to get $0.03/kWh on average with storage.

OTOH, there is additional value in providing synchronous reserves and regulation; these are typically priced around $4/MW/hour, though, so we're not talking big bucks (except in times of extreme scarcity).

If you are banking on scarcity to make your bucks, remember that the scarcity you're hoping for begins to disappear as you add more resources to the bulk power system that can respond to the potential shortfalls.

My analysis of ERCOT shows that there are enough large price swings plus high enough ancillary services prices to make a couple million a year playing the optimal storage game. It could help increase Tesla's revenues but not be one of the larger sources of revenue. At most $10-$20 million a year if you covered all of the nation.

 

[DonPedro]

Some data might be helpful in your revenue models. Any grid storage device will be on the wholesale market, so forget about your retail rates. The gap between on- and off-peak wholesale prices is far smaller than you are supposing. Here are numbers for purchase today of annual strip power products for 2014:

	PJM	PJM	CAISO	CAISO	CAISO
	East Hub	West Hub	CA/OR Border	NP15	SP15
On-Peak	52.71	47.93	40.98	45.84	51.90
Off-Peak	35.75	33.43	29.33	35.72	38.60
Delta	16.96	14.50	11.65	10.12	13.30

These prices are $/MWh, so we're looking at 1.0 to 1.7 cents/kWh of price swing. Admittedly, you can do better than this: there are some hours when the prices go way above the average on-peak price. But you'd be lucky to get $0.03/kWh on average with storage.

OTOH, there is additional value in providing synchronous reserves and regulation; these are typically priced around $4/MW/hour, though, so we're not talking big bucks (except in times of extreme scarcity).

If you are banking on scarcity to make your bucks, remember that the scarcity you're hoping for begins to disappear as you add more resources to the bulk power system that can respond to the potential shortfalls.

There are gas-powered power plants that stand by to rev up on 15 minutes notice to provide those last MW required during the annual global peak - they are active a few hours per year. The cost of this marginal power is astronomical. The need for such balancing capacity will increase sharply when integrating renewables into the grid. This is why grid storage has such a high economical power. Consumption-side on-peak/off-peak rates (whether retail or wholesale) have no meaningful granularity, and will not reflect the economic benefit of grid storage.

 

[CapitalistOppressor]

Some data might be helpful in your revenue models. Any grid storage device will be on the wholesale market, so forget about your retail rates. The gap between on- and off-peak wholesale prices is far smaller than you are supposing. Here are numbers for purchase today of annual strip power products for 2014:

	PJM	PJM	CAISO	CAISO	CAISO
	East Hub	West Hub	CA/OR Border	NP15	SP15
On-Peak	52.71	47.93	40.98	45.84	51.90
Off-Peak	35.75	33.43	29.33	35.72	38.60
Delta	16.96	14.50	11.65	10.12	13.30

These prices are $/MWh, so we're looking at 1.0 to 1.7 cents/kWh of price swing. Admittedly, you can do better than this: there are some hours when the prices go way above the average on-peak price. But you'd be lucky to get $0.03/kWh on average with storage.

OTOH, there is additional value in providing synchronous reserves and regulation; these are typically priced around $4/MW/hour, though, so we're not talking big bucks (except in times of extreme scarcity).

If you are banking on scarcity to make your bucks, remember that the scarcity you're hoping for begins to disappear as you add more resources to the bulk power system that can respond to the potential shortfalls.

This is true, which is why I think the model will end up generating relatively high returns per MWh stored initially, which will then decrease as capacity increases, and CES moves up the chain (down the chain?) from grid buffering to load leveling.

But the other thing to keep in mind is that the need for CES increases in direct proportion to renewable adoption. So there is a synergistic relationship between wind/solar and CES. CES actually increases the value of the renewable energy, which must be sold at a discount because of the technical problems that a lack of storage creates. (I'm not sure how this discount works in the marketplace, but it must exist because of the technical resources that must be devoted to solve the variability problems)

Without CES, there is a real upper limit on how big of a market share renewables can get, regardless of how much the notional per/MW cost of renewables falls (or conventional increases). With CES, renewables have more market power, which allows them to capture market share, which then allows you to monetize a larger CES capacity.

Current rates, or discussions of the different flavors of peaker power (there are sources that operate for hours each year, and larger sources that operate for portions of each day, etc) are relevant for the initial business. But the end state is the renewable+CES combo going head to head with conventional sources. If/when the combo is able to profitably offer a lower price than conventional sources, it will displace conventional sources and earn vast sums of revenue for CES operators, and vaster sums for the renewable generators.

In the mean time, there are existing mandates and subsidies for renewables that offers a market opportunity for CES. The minimum opportunity lies in eliminating the costs incurred by the various workarounds being implemented to support renewables. If Tesla's CES is more economical than the current kludge, then they should be able to do quite well.

So, aside from simple rate arbitrage, I see two opportunities at the moment. Fixing the renewable kluge, and replacing peaker sources. How to properly characterize the size of those opportunities, now and in the future, seems to be the path we need to take. Do we see other additional opportunities?

(Edit: it occurs to me that renewables are mandated, so we wont necessarily see a discount in the "market" price, but the costs are still there, and thus the opportunity)

 

[rolosrevenge]

Ancillary markets: Frequency regulation up and down, responsive (spinning) reserves, non-spinning reserves are rather lucrative and can easily be performed with batteries. You could also supply black start services and voltage regulation as well as backup power to the shopping complex.

 

[CapitalistOppressor]

Here's a simple to read explainer -
Explainer: Base Load and Peaking Power | Explainers | KCET

Here's nugget of a datapoint -

'Peakers' plants provide electricity when it's hot, but at the highest price | NJ.com

It costs ratepayers about $13 million annually for PSEG to keep the plant operational. That cost could vary depending upon how many days of the year PJM decides the plant is available to operate. The more days it’s available, the higher the charge to ratepayers.

And a research opportunity (PSEG Fossil) for a company in the peaker business and is operating the plant in question.

Here is a Greentech article from 2010 on the subject. I strongly recommend reading it. The link for the Strategen and CEC papers are broken, so we should find them -

How Storage Can Help Get Rid of Peaker Plants : Greentech Media

Key nugget -

Using assumptions taken from a recent California Energy Commission (CEC) study, our analysis found a levelized cost of generation for the simple cycle gas-fired peaker plant of $492 per megawatt-hour (MWh), or $203 per kilowatt-year (kW-yr). In comparison, the energy storage plant demonstrated significant savings, with a levelized cost of generation of $377 per MWh ($155/kw-yr).

This study was apparently done between 1988 and 1996, using lead acid batteries. Electricity rates have changed (at a guess they are much higher, but I am biased because I live in California, and I remember Enron). 30 Model S batteries is probably ~2,670kWh of storage (based on my simulation of the Model S battery anyways). Elon mentioned that ~500kWh of storage could generate 1MW of power at need. So lets give the 30 battery plant a peak output of 5.3MW, and the ability to average 1MW over the year.

That gets us to the $203/kW-yr vs $155/kW-yr estimate which seems to imply that you can make ~$48/kW-yr, or ~$48,000 for the 1MW plant in this example. If each pack is $17,000 (my model, my rules, feel free to substitute your assumptions) then the cost for the batteries for the plant is $510,000. As a stand-alone product, this plant needs to be operational for a very long time just to recoup the initial investment (~11 years). But for Tesla, this is all free money because they need to build these things anyways for their automotive business.

BTW, this model is literally being transferred from the back of a napkin into my iPad in real time, and I haven't even fully read the Greentech article and examined its graphs (I was actually just trying to find some quick links, lol), so be gentle

- - - Updated - - -

I wonder how this patent might effect Tesla's plan in grid storage leveling, it seems to have an earlier priority date. METHODS AND SYSTEMS FOR BIDIRECTIONAL CHARGING OF ELECTRICAL DEVICES VIA AN ELECTRICAL SYSTEM

Interesting bit of IP. Seems focused primarily on maximizing the value of the trade between profits and battery degradation. Probably some overlap with Tesla, but I don't see either Tesla's IP or this IP being particularly critical. In the case of Tesla, its just a roadmap to their thinking (and proof that they have been at least thinking about the concept).

 

[Kraken]

I had a conversation today with some solarcity employees today. I get the feeling like they basically feel like solarcity/tesla is already one company in a lot of ways. Anyway, they were discussing how they figure tht the battery packs could basically be sold seperately for helping people with solar systems go off the grid. I told them to count me in... Anyway, it would seem that there could be some serious revenue waiting for Tesla just by selling the batteries. After my conversation with them, it almost felt like Tesla's cars could all just be a front while Elon masterminds his way to take over the energy industry. i know that's not true, but there is more potential then I ever realized...

 

[Johan]

I had a conversation today with some solarcity employees today. I get the feeling like they basically feel like solarcity/tesla is already one company in a lot of ways. Anyway, they were discussing how they figure tht the battery packs could basically be sold seperately for helping people with solar systems go off the grid. I told them to count me in... Anyway, it would seem that there could be some serious revenue waiting for Tesla just by selling the batteries. After my conversation with them, it almost felt like Tesla's cars could all just be a front while Elon masterminds his way to take over the energy industry. i know that's not true, but there is more potential then I ever realized...

I believe you might not be far off. In fact I just posted in another thread (Elon's multi-business strategy) on this: What if in 5 years the Nummi factory is in full use with 30% of it manufacturing cars and the other 70% of it spitting out battery packs for a wide range of uses: automotive both for Tesla and other car manufacturers, energy storage for individual's with solar power, large grid storage/grid balancing applications for utilities, batteries for electric airplanes etc. etc. etc??? With economics of scale the cost would only come down further...

 

[Robert.Boston]

A few more points about reality vs. theory on the power grid:

1. The cost of peaking plants is not covered by energy prices during demand spikes in US markets (except, maybe, in Texas). The system operators / utilities sign enough "capacity contracts" to meet expected peak load + a planning reserve margin; these capacity payments cover peaking plants' fixed costs. During the operating day, plants are limited to offering energy at their incremental operating cost (fuel + O&M) plus a small margin (~10%). Some areas have put in a scarcity adder that forces energy prices up during scarcity conditions. These kick in rarely--here in New England, about 3 hours a year.
-- Takeaway: if a resource doesn't qualify as a capacity resource, it won't get paid the big bucks that these peaking plants cost (cited by @CapitalistOppressor). Storage can qualify as a capacity resource, but it's a more complex process than just showing up to provide peaking power when it's convenient for you.

2. While intermittent resources do cause more system strain, and therefore the value of grid storage should be going up, current market designs are masking this value. The California ISO is "buying" a FlexiRamp product from generators at confiscatory prices, while PG&E is refusing to allow CAISO to schedule its hydro system to provide any flexibility. When Bonneville had the temerity to attempt to value its integration costs for wind, the value it came up with (~$15/MWh) that political forces kicked in and forced BPA to back-pedal.
-- Takeaway: although reserves should be increasingly valuable, this economic reality may not be reflected in prices.

3. Demand response providers will hop into the regulation market if it ever gets lucrative. Currently, DR isn't allowed to provide reserves in most markets. This will be changing over the next few years; New England, e.g., is scheduled to bring on DR reserves in 2017. This change would likely strip out the high-value hours for grid-tied storage.

 

[DonPedro]

A few more points about reality vs. theory on the power grid:

1. The cost of peaking plants is not covered by energy prices during demand spikes in US markets (except, maybe, in Texas). The system operators / utilities sign enough "capacity contracts" to meet expected peak load + a planning reserve margin; these capacity payments cover peaking plants' fixed costs. During the operating day, plants are limited to offering energy at their incremental operating cost (fuel + O&M) plus a small margin (~10%). Some areas have put in a scarcity adder that forces energy prices up during scarcity conditions. These kick in rarely--here in New England, about 3 hours a year.
-- Takeaway: if a resource doesn't qualify as a capacity resource, it won't get paid the big bucks that these peaking plants cost (cited by @CapitalistOppressor). Storage can qualify as a capacity resource, but it's a more complex process than just showing up to provide peaking power when it's convenient for you.

2. While intermittent resources do cause more system strain, and therefore the value of grid storage should be going up, current market designs are masking this value. The California ISO is "buying" a FlexiRamp product from generators at confiscatory prices, while PG&E is refusing to allow CAISO to schedule its hydro system to provide any flexibility. When Bonneville had the temerity to attempt to value its integration costs for wind, the value it came up with (~$15/MWh) that political forces kicked in and forced BPA to back-pedal.
-- Takeaway: although reserves should be increasingly valuable, this economic reality may not be reflected in prices.

3. Demand response providers will hop into the regulation market if it ever gets lucrative. Currently, DR isn't allowed to provide reserves in most markets. This will be changing over the next few years; New England, e.g., is scheduled to bring on DR reserves in 2017. This change would likely strip out the high-value hours for grid-tied storage.

OT: The more I read about the US economy, the more surprised I am how less it is like a free market economy than Norway. We're supposed to be communists over here (as classified by the US right wing), but we out-free market you guys any day of the week. In this case, electric power is traded on a common Nordic power exchange, allowing prices to accurately reflect scarcity at any given point in time. As consumers, we can buy at spot prices (+small admin fee), or we can buy fixed-price (or other) packages from any utility. The dealership issue (TX, NC and so forth) is another case in point.

 

[Robert.Boston]

Yep, the US power market is pretty screwed up in many ways. Far better than it used to be. There are two layers: First, the federal layer the regulates the wholesale power markets and requires that all prices be "just and reasonable" which, if there's any risk of there being market power, means something linked to marginal costs. Second, the state layer, which determines what prices consumers actually see. State regulators are generally very paternalistic and use retail tariffs to accomplish many goals, including wealth redistribution and subsidies. It's a mess of cross-wired incentives.

As a great example, recently someone posted how his local utility rates, which are tiered depending upon your monthly usage, made it more profitable for him to put PV on the north side of his house (which has a high load) than for a neighbor to put PV on the south side of his (because he has a lower load). This violates at least two fundamental economic principles.

 

[Citizen-T]

You'll find that most 100ish year-old industries in the U.S. have very anti-capitalist characteristics. Another two I can think of in addition to autos and energy: railroads and landline telephone operators.

 

[Rla]

Wow, awesome thread. Love it that TM is also going into the grid storage space. I have been looking at developments for CES for a few years now, and nothing big is happening. Most solutions do target the pov of the classic energy producers, few are catering for the CES scenario.

I must say that this does re-enforce my guess that the 'under your nose' announcement could well be that all Model S cars have a 85kWh battery. It provides TM with a simplified production line and supports the SuperSwapper idea. All batteries being equal, the stock required at a swapping station also becomes much less of an operational hassle. Furthermore it could (can it really?) allow a current 60kWh Model S owner to upgrade to a 85kWh, just by buying the upgrade. (Swipe credit card, wait for the OTA update sent to the car, and tadaa, extra range)

Given that an 80% 85kWh battery pack still has enough power for a 60kWh car, the lifetime for the current battery packs move into the timeframe of the GenIII car, allowing Model s drivers to upgrade to newer more powerful packs while TM can repurpose the 'used' ones for the GenIIII.

 

[rolosrevenge]

As a great example, recently someone posted how his local utility rates, which are tiered depending upon your monthly usage, made it more profitable for him to put PV on the north side of his house (which has a high load) than for a neighbor to put PV on the south side of his (because he has a lower load). This violates at least two fundamental economic principles.

The primary being that things facing south are always more profitable ... :tongue:

 

[EarlyAdopter]

After reading this thread I think "the way for the Tesla Model S to be recharged throughout the country faster than you could fill a gas tank" is through grid tied batteries at the supercharger stations.

What if Tesla has found that their packs are capable of safely charging and discharging at 6C (10 mins). This normally would require a massive substation feed from the utility. Not going to happen at the hole-in-the-wall locations of the Supercharger stations. But if they've got battery packs at the stations this won't be a problem as the battery will be the high current buffer. "Trickle" charge the station batteries from the grid at night and solar during the day. Then dump a massive charge into a car when it pulls up. If too many cars happen to stop by and deplete the batteries, no worries - just fall back to 120kW fed from the grid.

Elon just hinted at this at the shareholders meeting.

 

[deonb]

Elon just hinted at this at the shareholders meeting.

He did say that they have the ability to upgrade the SuperChargers. And since they have the grid storage batteries there it gives them the ability to put out more than 120kWh. And as I said before, as far all the futures in the multiverse, this is my favorite one.

HOWEVER, when you look at Elon during the stockholder meeting, he spoke 3 times about fast charging (only #2 is an exact quote, the rest I took down live, but should be close):

1) He was talking in regard to getting a 99% score from Consumer Reports - the only American car that has ever achieved that. And mentioned (which I jotted down as):
Elon: The reason we didn't get 100 was because of charge time and range. I'm wondering if we address those issues (smile) - whether a score of 100 is possible. I don't know if they'll give us a 100. But we will try and address those issues. (Smirk). (Smile). (Grin)

2) During Q/A:
Q: You recently tweeted about that it may take less time charging a Model S would take actually less time than filling a tank of gas. Is the recent announcement that it is 20 minutes. Is that it now?
Elon: No. That is not it. That is not it. I would say, you should... I mentioned there will be an announcement on June 20th, in fact, not an announcement, but a demonstration. Live demonstration.

3) Q/A question about the SuperCharger speed:
Q: Is there any concern that with the rapid rollout, SuperChargers will be obsolete by the time you deploy them?
Elon: We're going to upgrade the SuperCharger technology over time. [He mentioned the stationary storage again]. Combined with Solar - we charge battery packs with solar power during the week. That unlocks even more power potential by drawing from the batteries. In the future we can potentially go above 120 kW, without having to necessarily lose our existing hardware investment. Long term goal is even with Zombie Apocalypse, the SuperChargers will still be functional.

On #1 and #2 he was positively giddy. He was like a child that bought a Christmas present for her dad and can't wait for him to open it and see the look on his face. [ Aside: Don't you just absolutely love the excitement that this man has for his field? ]

On #3 he was much more subdued, like he had to think about the question for a while. And he answered this in the tone of the engine noise question question he received with his directional pleasant sound solution solution. Yes, it's something that they're thinking about doing, but I don't get the feeling it's something he's just bursting to tell us about.

So sadly, I think one of the two battery-theories are going to win out on this one. But I'll keep rooting for charging!

 

[kenliles]

So sadly, I think one of the two battery-theories are going to win out on this one. But I'll keep rooting for charging!

What if it's a non-cable special connector for flash charging from a battery source? (Perhaps even also with flush coolant connectors)

 

[CapitalistOppressor]

What if it's a non-cable special connector for flash charging from a battery source? (Perhaps even also with flush coolant connectors)

To me, that is the only "SuperCharging" option that ever made sense. Leverage the fast removal and replacement of the battery that is possible with SuperSwapper, and instead of swapping, just hook up a mondo charge connecter directly to the battery, and a SuperCooling system directly to the cooling ports, and flood the battery with electricity/coolant until it is charged.

Then replace the battery in the car, and drive off.

I don't know if its possible from a technical, charging point of view. But as an engineering fix, the hardware and processes are even simpler than SuperSwapping (no need to move batteries around). You still require massive CES infrastructure as well, because otherwise you need a miniature power plant on site to provide the massive power quantities available.

But I've never seen comments from users that show this is possible without destroying the battery. If you can't resolve charging issues with massive cooling systems, than its pointless to consider, which leaves SuperSwapping or the Aluminum Air concepts.

I gave up on arguing with the AA guys because I realized you need CES regardless of the method you choose. The electrical grid just isn't equipped to allow for millions of automobiles to do MW levels of charging on a Memorial Day weekend. You need to store the energy on site, and then give it to the cars mechanically (by swapping, either conventional packs, or AA) or by SuperCharging (at whatever really high speed you can do).

 

[kenliles]

To me, that is the only "SuperCharging" option that ever made sense. Leverage the fast removal and replacement of the battery that is possible with SuperSwapper, and instead of swapping, just hook up a mondo charge connecter directly to the battery, and a SuperCooling system directly to the cooling ports, and flood the battery with electricity/coolant until it is charged.

Then replace the battery in the car, and drive off.

I don't know if its possible from a technical, charging point of view. But as an engineering fix, the hardware and processes are even simpler than SuperSwapping (no need to move batteries around). You still require massive CES infrastructure as well, because otherwise you need a miniature power plant on site to provide the massive power quantities available.

But I've never seen comments from users that show this is possible without destroying the battery. If you can't resolve charging issues with massive cooling systems, than its pointless to consider, which leaves SuperSwapping or the Aluminum Air concepts.

I gave up on arguing with the AA guys because I realized you need CES regardless of the method you choose. The electrical grid just isn't equipped to allow for millions of automobiles to do MW levels of charging on a Memorial Day weekend. You need to store the energy on site, and then give it to the cars mechanically (by swapping, either conventional packs, or AA) or by SuperCharging (at whatever really high speed you can do).

agree with your analysis- and agree it's the best option if they cold pull it off-- thanks cappy;
and thanks for your great analysis of the battery costs (other thread) while I'm at it...

 

[CapitalistOppressor]

A few more points about reality vs. theory on the power grid:

1. The cost of peaking plants is not covered by energy prices during demand spikes in US markets (except, maybe, in Texas). The system operators / utilities sign enough "capacity contracts" to meet expected peak load + a planning reserve margin; these capacity payments cover peaking plants' fixed costs. During the operating day, plants are limited to offering energy at their incremental operating cost (fuel + O&M) plus a small margin (~10%). Some areas have put in a scarcity adder that forces energy prices up during scarcity conditions. These kick in rarely--here in New England, about 3 hours a year.
-- Takeaway: if a resource doesn't qualify as a capacity resource, it won't get paid the big bucks that these peaking plants cost (cited by @CapitalistOppressor). Storage can qualify as a capacity resource, but it's a more complex process than just showing up to provide peaking power when it's convenient for you.

If Tesla were to promise a certain quantity of power be available on demand by the utility, would that allow them to be eligible for a capacity contract? Let's set aside the small potato plants we will see during the initial rollout, which only need a small capacity to support a national vehicle fleet of under 100k, and look ahead to 2025.

Assume 32 GWh of storage are built to support a vehicle fleet of 5 million Tesla's (purely hypothetical), and that capacity is broken up into five hundred 64MWh plants, Each of which is capable of 128MW of output. If only half of the plant capacity is needed to support the automotive fleet, could Tesla contract with the utility to offer 34MWh of capacity, on demand, year round? With the understanding that it's a finite resource, and needs to be replenished when exhausted?

Or do you sell this capacity at the network level, by looking at the regional capacity of the CES stations?

2. While intermittent resources do cause more system strain, and therefore the value of grid storage should be going up, current market designs are masking this value. The California ISO is "buying" a FlexiRamp product from generators at confiscatory prices, while PG&E is refusing to allow CAISO to schedule its hydro system to provide any flexibility. When Bonneville had the temerity to attempt to value its integration costs for wind, the value it came up with (~$15/MWh) that political forces kicked in and forced BPA to back-pedal.
-- Takeaway: although reserves should be increasingly valuable, this economic reality may not be reflected in prices.

Yes, I was concerned about the possibility that these costs are hidden away from the "market". Nevertheless, everyone is is aware of them, and regulators can adopt new rules to facilitate the market entry of a major new CES capacity that is going to come into existence regardless of what the energy market thinks (thanks to the core charging requirement).

3. Demand response providers will hop into the regulation market if it ever gets lucrative. Currently, DR isn't allowed to provide reserves in most markets. This will be changing over the next few years; New England, e.g., is scheduled to bring on DR reserves in 2017. This change would likely strip out the high-value hours for grid-tied storage.

I'm not familiar with Demand Response. Reading up on the Wikipedia entry here -

Demand response - Wikipedia, the free encyclopedia

 

[Bgarret]

Can anyone enlighten me about the purpose of the "clusters" of superchargers in the projected 2015 supercharger map that are not located near large metropolitan areas? In Michigan, there is a cluster of 3 proposed superchargers in South Central Michigan, - on the rough map, near Jackson, Mi. - headquarters for Consumers Energy. There are also 2 clustered near Luddington, Mi - population ~20k on the NW side - near Consumers stored power facility. There are similar clusters near Burlington, VT, in Massachussets, and in Georgia, near Augusta. There look to be 2 just North of Indianapolis, roughly the site of a major wind farm, and a bunch near Jacksonville, Fla and strung across Az (solar?). I get locating along Interstates and near population centers, but these seem to be built in the middle of nowhere. Not sure if I'm looking for patterns in randomness. Thoughts appreciated.