ITF report finds self-driving shared vehicles could take up to 90% of cars off city streets; total kilometers travelled increases
Allison Transmission introduces more fuel-efficient xFE transmission models for buses; up to 7% improvement

Tesla launches line of Li-ion stationary storage systems for homes and businesses: POWERWALL and POWERPACK

Multiple POWERWALL units may be combined. Click to enlarge.

As widely expected, Tesla CEO Elon Musk unveiled the company’s new product line: scalable stationary battery systems for homes, businesses and beyond. Available for immediate order on the Tesla Web site ( or is the modular wall-mounted POWERWALL system, which is targeted at homes and perhaps some small commercial applications, Musk said. Delivery is projected for sometime later this summer (3-4 months from now.)

The sleek, sculpture-like Li-ion packs come in 7 kWh ($3,000) and 10 kWh ($3,500) configurations. Both are guaranteed for 10 years (with an optional 10-year extension), and can be combined to up to 9 units—i.e., up to 90 kWh of storage. The packs contain all the integrated safety systems, the liquid thermal control and the DC/DC converter, and work with solar systems straight out of the box, Musk said.

Tesla POWERWALL unit. Click to enlarge.

Round-trip DC efficiency is 92%, power output is 2 kW continuous, 3.3 kW peak. Voltage is 250-450 volts, with 5 amp nominal current, 8.5 amp peak output. The POWERWALL units are rated for indoor or outdoor installation, and operate in the range of -4°F to 110°F / -20°C to 43°C. The POWERWALL requires installation by a trained electrician; the AC-DC inverter is not included.

The 220 lb (100 kg) unit measures 52.1" x 33.9" x 7.1" (130 cm x 86 cm x 18 cm), and comes, noted Musk, in different colors.

The fact that it is wall-mounted is vital. You don’t need a battery room filled with nasty batteries. A normal household can mount these in the garage or outside the house.

—Elon Musk

The POWERWALL units will initially be assembled at the Fremont plant, Musk said, and the ramp will be slow. a plans to transition production to the GigaFactory in Nevada next year, resulting in a higher ramp.

For larger scale systems, Tesla is proposing the industrial POWERPACK, which, Musk said, is “designed to scale infinitely. This could be a GWh solution.” Musk said that a Tesla is already working with a utility that wants to do a 250 MWh pilot. (The entire evening livestream event was powered by POWERPACKs.)

At the conclusion of the streamed announcement, Musk said that solar panels and batteries represented “the only path I know” that could transition the world to sustainable energy. He suggested that 2 billion POWERPACKs could fully transition the world to a solar/battery future, and observed that it is “within the power of humanity to do so.



Reading up a bit more apparently the SGIP covers 50% of the battery costs, not the total installation, and:

'You're right about the subsidies in California, although I'm not an expert at the specifics either.

From my understanding most of these cover 50% of the battery costs but not the installation. Including the inverter capacity, I'm putting $1500 as extremely cheap. I'd expect actual installation to run closer to $3k+.

With a 50% battery rebate and mid-point of installation cost, we're still looking at $4k. Keep in mind 10kwH is not enough for a long blackout... Perhaps good for 3-4 hours max (2500w/hr for the whole house-- for example most computers use 250-500w/hr). A central air condition uses around 3-4k w/hr. So yes, summer blackout- good for maybe 3 hours of full house power.'

(Mintzmyer, comment)

However, as I note in the comment immediately below:

'The 10kwh unit CANNOT cycle once a day.
It is rated at only once a week:

'Powerwall comes in 10 kWh weekly cycle and 7 kWh daily cycle models.''

Pretty useless, and uneconomic even with 50% of the battery costs paid, IMO.

Account Deleted

@Mahonj in Musk's presentation he actually say that he imagines that people living in the equatorial belt will forgo ever building a grid and central power plants because it is cheaper and fully sufficient using solar cells and battery backup power just like those countries have skipped laying telephone lines and gone directly to wireless smart phones. As you know very well the sun shines equally much every day during the year in the equatorial belt so this is possible.

In the Northern and Southern hemisphere the sun have seasons so we either need a grid with central power plants or we could start building small scale underground heat sinks storing 100 of kWh of energy per cubic meter of heated sand in insulated containers for storing seasonal energy. The heat sinks are heated up during summer with excess solar power and the energy is subsequently used during winter for electricity and heat production. Really for security reasons we should not have central power plants and a grid as it can easily be sabotaged and we cannot afford just a little downtime on our power systems. They need to be running all of the time especially during natural disasters and terror attacks.

Account Deleted

After reading and thinking a little more I think that both the 7kwh daily cycle pack and the 10kwh weekly cycle pack may actually use the same cells as the Model S. Both systems weight 100 kg and are identically dimensioned. The only difference may be in the control software. Each battery pack could contain 10kWh of standard Model S cells. The reason that the nominally rated 7kwh unit can do 365, 7kwh cycles per year for at least 10 years is that it is limited to 70% SOC at the cell level so that the cells will last much longer. Therefore, most house owners with solar power will probably just need to install one 10 kwh and one 7 kwh unit in order to go completely off-grid most of the year. The grid is dirty and always will be as long as it contains nuclear and fossil fuel burning power plants. Therefore, the more people that can go off-grid for a longer time during the year the better for everybody on our planet.

Nick Lyons

Enphase is developing a somewhat similar system (wall mounted, modular, monitoring, time-load-shifting, etc), but they include inverters in every module--AC in AC out. It will be interesting to compare when they are both on the market.


My order of merit for current cells for home storage would be:

1.Toshiba SCiB.
Nearly ideal if the price is right.
The only cell which can get great cycle life at a wide range of temperatures and produce 2kw per 1kwh
Cost is the only issue

2.BYD LiFePo.
Environmentally friendly, cheap, robust.

3.Tesla/Panasonic NCA.
Designed with high energy density as the priority for cars.
Needs sophisticated management and cooling for its volatile chemistry.

If you were going to pick a battery chemistry from scratch for home storage, it would not be NCA.
Tesla reckon it can reduce costs by volume in coordination with its car business is the reason for the choice.

Well, BYD will have 16Gwh of capacity by the time the first stage of the GF is online with 7Gwh.

That seems the best balance to me until supercapacitors with lead acid battery energy density get here:

With of the order of a million cycles, they would simply be built into the house.

We will have to wait a bit until we can evaluate Emphase's offering, as Nick says.


The Achilles heel of Tesla appears to be their ability to turn a profit. I wonder if they will be able to do so with the Power-wall? There are some good and suitable products made by other companies and as Musk points out in his presentation, the market is enormous if the batteries can be produced at a suitable price, but so far they have not had much success. I think the key point that Tesla understands is that well off individuals can easily afford to pay $350/kwh while utilities are more cost constrained. It will be interesting to watch how this plays out but it seems to me the most significant aspect of the presentation is how Musk is able to get the attention of the mainstream media. He didn't say anything in his presentation that all of the posters at this site don't already understand.

More important than building batteries and cars, Tesla appears to be building a very strong brand and I expect that when they reach the point where they need a financial savior, the brand will be what the savior is buying. If you believe the future is renewables and batteries then the brand could be very valuable. Just like Apple.

I think the residential fuel cell in Davemarts post would combine well with solar panels and a battery, espescially in sunny northern regions like Calgary so perhaps Musk will need to retract his "fool cell" comments when they add that to their product line.


Hi Calgarygary.

My position has always been that one way or another we will deploy the full range of technologies, with the mixture depending on how they progress.

Batteries are already useful at the grid level, with BYD already having deployed units with the biggest 36MWH.

Tesla are also going after that, and are bidding for a large 250MWh contract.

Batteries are needed in the grid because of the increasing amount of renewables, with both solar and wind varying in a second.
Frequencies would fluctuate wildly damaging equipment without batteries etc to cushion them.

They are also getting to the stage where they can displace spinning reserve, which is great, as that involved running gas turbines on tickover wasting fuel so that they were ready to speed up and provide power very quickly.

Its an exciting time in the energy industry with lots of options becoming available.

I would love to see more nuclear for baseload, which would cut a lot of the expense and messing around, but we work with what we have.


Congratulations Tesla.
-Attractive design and price points for the launch products in a product line.
-Modular and expandable.
-Until the gigafactory is producing them they won't be making many of them anyway.
-Tesla star power is bringing tons of free, positive publicity to a behind the scenes technology.
-It will only get cheaper from here as competition, scale and technology improvements lower the cost, add improvements.
-Works now for high cost areas (Hawaii, California, etc...) with high cost power and subsidies for grid batteries.
-American designed and assembled, made with American materials (at least in the first gigafactory).

So, it doesn't work for everybody right does for some and that is great!

+1 Chops

Davemart, my comment about LG Chem and Equana was:

(7kW x 0.6 for best battery life)

4 kWh x $0.34 x 365 = $496.4 per year. They don't mention the price, but if it's in the neighborhood of $3k, six years later you start earning ~$500 per year every year.


They still have dull (rainy) days near the equator. With storage like this, you can increase the number of days per year that you can go off grid, but getting it to 365 is very hard.

The grid is a great leveller, it can shift power to wherever it is needed, and it is not that expensive (when you divide by the number of people is serves).

What you might find are mini or micro grids where most of the energy comes from solar+batteries but you have a generator for runs of dull days.

I think Must is trying to build demand for batteries that his Gigafactory will start producing in a couple of years. Once it starts generating the number of batteries predicted, you had better have a use for them and the two choser are EVs and domestic storage.

Also comparing mobile phones (which are about information) and electric power (which is just about power and energy) is a bit of a cod as they are different things. Phones and cell tower controllers are (were) subject to Moore's law. PV cells and batteries aren't. (PV cells have certainly got much cheaper very rapidly in the last 10 years) but batteries are improving rather slowly.

But at least Musk is putting it out there and making people think (and argue).


I am not going to comment on the costings for a system which we have absolutely zero information about what they are charging, nor do I see how anyone sensibly can, save perhaps to mention that your notion that the unit including not only the battery pack but installation and the inverter can be had for $3000 seems utterly incredible.

Looking at the costs for the Powerwall, which we do have some although inadequate figures for, then it is difficult to see what it is actually for.

If you live in an area where most will be sold, such as California, then there are two cases:

1.You have or are buying a solar array.
In that case any excess during the day can be sold to the grid, so you don't need storage to arbitrate rates.

2.You do not have or want a solar array.
In that case you would have to pay for the inverter and installation.
For the daily cycling unit that is 7kwh.

Note that we don't know that this is the figure which can be 100% cycled, or if you need a reserve.
So lets make the most favourable assumption that you can cycle 100% low to high every day for the 10 year warranty, which we don't know even guarantees capacity or just against defective workmanship.

But what the hell lets assume that the battery not only has 100% DOD but zero capacity decrease over 10 years.

And lets put the inverter and installation at a very, very low $2000, so total $5000.

Lets not bother costing any loan interest either, so you have 7*3650 = 25,550kwh over ten year.

That comes to near as darn it 20 cents/kwh to arbitrage expensive to cheap electricity.

That is with every single assumption made favourably.

Real world is going to be much worse.

So lets look at the 10kwh weekly cycling unit instead.

Rated at 520 cycles over 10 years lets not be silly and talk about arbitrating power rates.

Lets not talk about energy efficiency either, as you are continually topping up the battery.

For the money there, which is $3500 plus installation, a two man job with this ludicrous on the wall installation of a 100kg unit, plus an inverter if you have not got a suitable one with a solar array, you then have a unit which provides power at an average draw of 2kw, max 3.3kw.

Regardless of the bull in the presentation, this is not remotely enough to power a American home under normal usage.

So for $4500 or so plus possibly an inverter you get a heady 2kw available for a whole 5 hours.

Don't try to make toast and boil a kettle at the same time out there folks!

Not a lot of use in hurricane or earthquake zones either, where nature is often inconsiderate enough to cause power loss for more than 5 hours.

Put the money for this bling towards buying a Mitsubishi Outlander PHEV when they are available in the states instead, where they include power out and you can have some minimal level of power from the battery not only from 12kwh but from all the petrol you have got.


I forgot the Californian SGIP in the above figures.
That I believe is 50% of the battery but not the inverter and installation costs.

Stick that in and the figures still don't work, especially if you make more realistic assumptions than those wholly favourable to the Powerwall ones I made.

It should also be noted how expensive the notion of adding small modules as needed is in installation costs and possibly in upgrading the inverter, which are the things not covered AFAIK by SGIP.


Thinking further, if you have a place that is currently off the grid, this would be a great boon. In that case, you are using 0 KwH / day, every day and something which gives you 350-360 days of electricity per year would be a major benefit.

Then, you have the problem of scaling and cost.
You need to be able to scale right down to a single or half sized solar panel (1.6 x 0.8 metres in size, approx 220 W), would probably benefit from 0.5 or 1.0 KwH storage.

Alternately, you have micro communal power where you have a few kw of Solar and a 7 KwH battery powering several homes.

Starting an a US house requiring 30 KwH / day (or a European house at 11 KwH) may not be the way to go, it may be better to start at the bottom with people with no power at all.

Account Deleted

In the equatorial belt there are probably 2 billion people and probably 500 million have no electricity at all and are extremely poor. There may be 500 million of those 2 billion who are connected to a well functioning and stable grid. The last 1 billion that have electricity is on an extremely unstable grid that is on and off multiple times per day and the grid downtime is perhaps 2 to 4 hours on average per day. If you can afford it you supplement your grid power with a diesel generator. I expect there are 10s of millions among the 1 billion that depend on an unstable grid that can afford a solar cell roof and a Tesla backup solution even though we are talking 30k USD for the solar cells (10kW installation) and 10k USD for a Tesla backup system. These people could say goodbye to a grid that is hopeless anyway and save them the cost of buying diesel for that generator and a new generator for every 3000 hours of operation which may be as much as every 4th year.

In the northern hemisphere there are many more wealthy customers for Tesla's energy systems. They already have a stable grid so the main motivation for going off grid during spring and summer will be a genuine concern for the environment and a willingness to do the right thing for our planet. I still do not see any important savings. Going off grid for most of the year will not save you any money. We need lower prices for solar power and battery backup before that happens. Today it will cost you about 40k USD to get a proper solar power roof (10kW peak power) with a Tesla battery backup system. This system may last 20 years and produce about 200,000 kWh over those 20 years so the cost per kwh is 20 cents per kwh. For everybody to order solar power with battery backup we need a 50% price reduction which is achievable within the next 15 years. The global solar power and battery backup industry could become a bigger industry than the global oil and gas industry by 2030. So yes Tesla IMO is definitely in a high growth industry and they are leading it.


I was too optimistic on Powerwall costs.
ECI came in at half the installed price.

The installed cost by Solar City:

'For a 10 kilowatt-hour system, customers can prepay $5,000 for a nine-year lease, which includes installation, a maintenance agreement, the electrical inverter and control systems. Customers can also buy the same system outright for $7,140, Bass said.'

Leaving aside the licence for larceny Solar City leases, which I would expect find a way to snaffle up the SGIP credits, and knocking off the $500 differential for the 7kwh unit, we end up with an installed cost of $6,640, less SGIP.

I will have to see the figures on how SGIP works, but it sounds as though after allowing for that my 20cents kwh is about right, although it does not cover interest on any loan for the system.

As for Tesla's claim:
'Both are guaranteed for ten years and are sufficient to power most homes during peak evening hours'

the actual power provided:

'SolarCity will provide backup power for up to four (4) electrical circuits rated at 120 VAC and up to 20 amps each. The battery backup system can power up to 1,900 watts of backup loads at any given time. Typical backup loads include refrigerators and other kitchen equipment, lights, outlets that power computers, phones and Internet routers, and home security systems. We will never offer backup to circuits greater than 20 amps including most pool pumps, well pumps, electric heaters, electric water heaters, air conditioners, and hot tubs'

So we learn that most American homes don't have air con.


The lattest contracts for NPPs called for $0.16/kWh. When adding high voltage transmission lines, local distribution lines, energy lost and fair profits the end users will probably have to pay between $0.25/kWh to $0.30/kWh.

Considering the above, $0.20/kWh for clean solar power with storage is a rather low fair price.

Account Deleted

Harey the average residential cost per kwh in the USD is 12.29 cents so solar power is still too costly for many who cannot afford the luxury to pay extra for clean power. However, the 12.3 cents is an average. In Hawaii they pay 31 cents per kwh so solar power will save you money. With Tesla's backup power system solar power for Hawaiians is simply a no brainer. Expect Tesla to do extreamly well in Hawaii when they are able to mass produce their Powerwalls next year.


Residents of many major USA cities (NY, San Francisco etc) already pay $0.20+/kWh. When you add the real cost-price for pollution created with CPPs and NGPPs, electricity would already cost over $0.30/kWh in many USA cities.

Various size storage units will soon be available.

How can they b marketed in sunny equitorial places where electricity grids barely exist.

It may be a choice between smart cell phones and solar electricity?

I have noted that many small (basic) solar systems already exist in Asia and Africa, to pump water and othe essential services. Those systems will grow in number and size as they get cheaper?

Hawaii's 31 cents is average, add 5.5 cents for peak; this is what your Powerwall is intended to level out. Not all islands have the same rate. Average on Maui it is $0.38 per kWh and $0.46 on Molokai and Lanai. A 7kWh Powerwall cycled once per day could offset 25% of the average residential load.

Hawaii utilities are already deploying large scale batteries to stabilize solar generation. This will allow individual customers to be part of the solution, and may speed up notoriously slow solar permits because it presents a solution to that problem.

DM ref SolarCity> (Powerwall) installed cost of $6,640, less SGIP (Self-Generation Incentive Program).

Administrator for the program, CCSE, says Incentives available at $1.46/watt.

Even without incentives, if you cycle the system once per day, saving $0.34 per kWH (delta from peak to off peak) the system saves you $8,687.

So you get a four circuit UPS for your house that pays you $2,047 over ten years, and continues to pay you $0.34 per kWh for every useful day after its warranty expires.

California writes me a check to encourage me to be clean, energy independent and stabilize the grid.

Sign me up.


Harvey, 20 cents is for the storage, not the power.



To make sense of saving by arbitrating peak and off peak power you need to post what the peak hours are, and the rates at different times.

If you are using solar, then peak rates when the sun is out are not going to be helped much by storing the power.


OK, an SGIP credit of $1.46/watt gives you 2kilowatts times 1.46 = $2,920

For the 7kwh daily cycling unit at $6,640 that leaves you $3720 to find.


'I keep reading all the posts referring to, and using for calculations, the PowerWall as a 7 KW-Hr device. Even if the PW is charged directly from PV DC, you still have DC to AC inverter losses meaning the daily cycle PW is really only good for about 80% of the rated 7 KW-Hr battery capacity or 5.6 KW-Hr of usable AC. Its like my 10 KW PV array that, at most, provides 8 KW of usable AC to my house. The inverters are hungry.'

For a once daily cycle for ten years we have 5.6*3650= 20,440kwh load shifted.

So just the cost of without interest comes out to 18 cents/kwh without the cost of the excess solar arrays to power it, and without any allowance for reduced capacity over time.

The $0.34 is the peak/off-peak delta in Southern California. If you use the system to avoid peak charges (whatever time of day that is) the system pays for itself. Those cost savings apply whether or not you're time shifting sunlight. The utilities don't want to build expensive infrastructure if they can avoid it. So using more of the existing facilities at a time when there's excess capacity is a good thing even if you don't have solar.

If you do have or get solar, then it's a double win. You can already beat $0.016 off peak. A couple or three of these units and you can be totally energy independent and guaranteed low rates for the rest of your life.

Despite your pessimism DaveMart, these are going to be a big seller in sunny places. Hope you're not short TSLA.



How in the name of Ned is it supposed to be any use if the peak hours are when the sun is shining anyway, so that you need if anything more solar panels, not more storage?

I have shown that this costs at least 18 cents per kilowatt to load shift, so your TOTAL system including the arrays has to cost less than the charge from the utility to make it worthwhile.

Never mind pessimism or jolly optimism, where are your figures?

I am not against battery back up, but that does not mean that the numbers should not be properly examined.

It also does not make a system which will not run air con a proper load shifter in hot places, nor does it make daft design sensible.

Not only is installing any wall units in earthquake territory unnecessarily problematic, as the structural integrity of the wall would have to be checked and will in no case be enhanced by hanging weight for it, when a proper floor standing unit would have done just that, but the wall would probably need structural strengthening to put in many more, and the design means you would end up with a mare's nest of wires and expensive installation for additional units.

So we are not discussing how lovely it is to spread load with batteries, but this particular design and cost.

You have produced nothing to indicate that the cost works anywhere, as you don't want to do the work to show that, preferring to generalise, or that the design is sensible,,as it clearly is not.

So if you want to make the argument, do the work.

The comments to this entry are closed.