Tesla Powerwall: Engineering Analysis

1. May 5, 2015

rollingstein

Just thought I'd post this on here to see what comments people have:

http://www.teslamotors.com/powerwall

Tesla has announced a Li-ion battery "Powerwall" priced at $3500 for 10 kWhr storage. What payback to people perceive? Gimmick, niche market or disruptive technology? 2. May 5, 2015 russ_watters Staff: Mentor So, I'm not sure if you are aware of the purpose, but the idea is to store energy at night when it is plentiful and cheap and use it during the day when it scarce and expensive. The idea is already used on a utility scale to reduce the number of power plants needed and on a commercial scale to reduce costs (and to help the utilities). The problem for now is that there is no residential market that I'm aware of for this concept. In order to make it useful, your electric rate would have to vary from day to night (as it already does commercially). So this is a product without a market (except for the secondary purpose of being a whole house backup). 3. May 5, 2015 anorlunda Staff: Mentor It should be very welcome to people who really live off the grid. But that is probably not a large market. 4. May 5, 2015 rollingstein I think I'm aware of the purpose just unsure evaluating the commercial payback. The competition there should be from diesel / propane generators or conventional lead acid. http://www.pge.com/en/mybusiness/rates/tvp/toupricing.page?WT.mc_id=Vanity_tou I think many areas already have time of use pricing for residential customers. Last edited by a moderator: Jul 10, 2015 5. May 5, 2015 MarkJW Not sure 7 kW is enough to run air conditioning and the rest of the house, and AC needs do not stop the moment the sun sets. Still an interesting offer. 6. May 5, 2015 russ_watters Staff: Mentor 7. May 5, 2015 russ_watters Staff: Mentor 7 kWh: 2 kW. To be fully off-grid in a decent sized house with air conditioning and electric appliances, you'd need about 5 of them. If your cooking and hot water are natrual gas, you might get away with 3. 8. May 5, 2015 Bandersnatch Can't it be used for the opposite? Store energy during the day from solar cells, and use it during the night? I thought that was the main selling point, but admittedly I'm not following it very closely. 9. May 5, 2015 anorlunda Staff: Mentor The US Energy Information Administration says that the average US household consumption is 10.9 kWh/day. If so, then 7 kWh for the hours when the solar panels are not producing seems about right. I think the difference might be the definition of "decent sized house" which may vary an order of magnitude in many places. Also, climate makes a difference; many areas live with little or no air conditioning and those contribute to the national average. I do think that the Musk's intention is that only one such battery bank would be needed for the household. 10. May 5, 2015 rollingstein You are right. Wrong link. Here's another page on the PGE site: http://www.pge.com/en/about/rates/rateinfo/rateoptions/index.page "There are two basic rate options for single-family, separately meteredresidential customers. E-6 is an optional time-of-use rate schedule for individually metered customers who can minimize their loads during defined time periods." Also here are other utilities that seem to offer time of use plans: SCE Introduces a New Residential Time-Of-Use Rate Plan https://www.sce.com/wps/portal/home...0mKkYfEEvibj1Q!!/dl4/d5/L2dBISEvZ0FBIS9nQSEh/ http://www.srpnet.com/prices/home/tou.aspx#save (Looks like Residential Tarriff but not 100% sure) https://www.portlandgeneral.com/residential/your_account/billing_payment/time_of_use/pricing.aspx [Broken] https://www.nvenergy.com/home/paymentbilling/timeofuse.cfm Last edited by a moderator: May 7, 2017 11. May 5, 2015 russ_watters Staff: Mentor Hmm...the website lists both uses together, and now that I google, a lot of articles focus on combining it with a solar installation. But as it happened, this is the article I first saw, which emphasizes the time-of-use pricing: http://money.cnn.com/2015/04/29/investing/tesla-musk-battery/?iid=EL Bad math aside (does it save 25% or 75%?), it glosses over the availability issue for time of use rates. 12. May 5, 2015 russ_watters Staff: Mentor I stand corrected -- I hadn't heard of it being offered residentially before. 13. May 5, 2015 rollingstein That's what I had thought too. If this is a cost effective plan to store utility generated power why isn't the utility itself doing it in a scalable manner & earning economies of scale? I thought that other than pumped storage none of the other storage techs were cost effective yet. If it ain't cost effective at scale how is it turning out to be cost effective at consumer level to store grid power? Right. I think the design allows eight to ten units to be chained together & to work as a monolithic electrical unit. My concern though, whether buying a single unit or ten, is what sort of customer this is going to pay off for. Note that neither solar panel nor inverter nor installation is included in the$3500

Last edited by a moderator: Jul 10, 2015
14. May 5, 2015

Staff: Mentor

Because for the most part, the utilities are selling energy, not buying it. Applying this idea at the utility wouldn't change anything for the utility because for the most part the electricity made at night costs the same amount to make but just isn't worth as much to the consumer. The primary benefit of something like this for utilities is saving in the capital cost of building new power plants that just run 100 hours a year (a few hours a day in the middle of summer) or if they have a high nuclear fraction and have trouble throttling at night. But barring that, the energy cost savings all goes to the consumer.
For customers, what makes storage cost effective is a wide variation in rates between day and night. That hasn't been common until recently. There are other energy storage technologies, such as ice and cold water storage, that are starting to make a comeback now that there are more such rates. It also used to be that you could get a special electric heating rate that was very cheap, because again most heating happens at night when the electricity is plentiful and the utilities have to dump it (occasionally, the spot rate will go negative: the utility will pay you to take the electricity).

15. May 6, 2015

rollingstein

Hmm...Not sure that I understand. e.g. Say you are an utility then during the daytime peak you are bringing on load several peaking power plants. Their cost of production is much higher than base load generation. Plus the Capex & maintainance needed for peaking power plants. Isn't that the fundamental reason for the variation between day / night rates?

This is the niche that pumped storage satisfies for the utilities I thought.

If indeed battery storage were cost effective, why might I not, as a utility, invest in massive battery storage farms & thus use my cheaper cost of generation base load plants?

My point is, if the cost of batteries does not make sense yet on scale why is it making sense in retail?

16. May 6, 2015

It appeals to customers who already spend $25,000 or more (pre-subsidy) for solar panel/inverter installations. When you are investing in such an installation, the incremental cost to add capacity beyond your daytime demand is small. In some places (not all) you can use net metering to sell energy back to the utility, but those net metering deals are full of problems, and are probably not sustainable. It can be more appealing to increase investment 20% for batteries, and to store excess energy during the day for you to consume later. For those dreaming of going entirely off grid, it makes it one step easier to have a battery option . rollingstein is correct regarding this conversation. I have no idea how the idea of utility customers shifting load time-of-day got into this discussion. There are simpler, nearly free, easier ways to do that. Setting your dishwasher and laundry to run at night on timers, and to program your climate controls to do more off-peak. Since very few consumers do even that, it is silly to imagine them investing thousands of dollars as their first step in off-peak shifting. To show the whole world that you are "green" hanging your laundry on a clothesline is much more visible than a battery in a closet. Bur rollingstein missed this: General Electric, has a big factory since 2012 to make batteries for utility and industrial applications. They don't mess around with consumer markets. But there are hundreds of reasons why utilities and industries may want such batteries, time-of-day shifting is only one of them. I think one of their larger customers have been wind farms in western states, where the wind blows mostly at night. In the eastern parts of USA, wind tends to blow more in the daytime. I also want to point out that in many parts of the country, ownership of electric generating plants has been divorced from utilities that sell power retail to consumers. I used to work at one of the ISOs that operate the grid and the energy markets. Generators sell their power to the ISO, and utilities buy their power from the ISO, on a minute-by-minute open auction system. The wolesale price of electricty thus varies minute-by-minute and by location on the grid. Retail customers do not participate in the volatile wholesale market, they get fixed rates set by the state public service commissions. Wholesale powerplant owners (renewable or other) no longer have a stake in the economics of the utility, each of them has to eke out its own profit or die. In addition to raw energy, the ISO buys capacity to do regulation of frequency and voltage, to provide standby reserve power capacity, and even to provide capacity to black-start in case of a blackout. During extreme peak times, the ISO even buys voluntary curtailment of load demand as an alternative to increasing generation. All of these secondary markets have value, in aggregate several billion dollars worth per year nationwide. Battery technology can be profitable in various secondary niches. I do not believe that the 7 kWh Musk battery is targeted to play any role in these wholesale electric markets or to compete with GE's utility/industrial batteries. Please let's not muddle this conversation by mixing utility/industrial/residental things that operate mostly independent of each other. The 7 kWh Musk battery is targeted at residences. 17. May 6, 2015 rollingstein @anorlunda I totally agree that batteries have hugely useful industrial functions. They are ubiquitous. This big GE battery factory is no surprise. What would surprise me though is if a large fraction of that factory's production went to utility peak shaving applications. Does it? What I wasn't aware of were any non-test-of-concept, large scale deployments of batteries as a utility peaking device. Based on what you wrote, wind farms are already doing that? If so, that's very interesting. I'll read up more on that. Do you know if the wind farms are using Lithium or Lead Acid? 18. May 6, 2015 anorlunda Staff: Mentor the side The technology keeps shifting. GE started with lithium, then went to sodium, now they're talking about fuel cells. The point is that these devices are the size of a small RV, cost up to millions of dollars, and are not relevant to the Musk battery discusssion. By the way, I read somewhere that 10,000 - 20,000 households in Japan are already using fuel cells. For some reason, they are not marketed in America. But be it solar, or batteries, or fuel cells, the future sounds exciting. 19. May 6, 2015 rollingstein I think they are relevant in this sense: If at least on scale a particular "device" or technology makes economic sense then I'm willing to consider that it might also perhaps make sense in small scale. But if the technology won't be cost effective at scale then it is hard to think why it can be specifically effective at household level. So although the devices are far different one sort of sets one extreme of viable operation for the other. 20. May 6, 2015 cpscdave I'm interested to see how the energy utilities are going to react to this. Assuming a wide scale adoption of the batteries, it would help the utilities by smoothing out the power demand over a 24hour period. Which will them to rely on their more efficient, cleaner and cheaper plants. I could also see not having to scale up and scale down generation as much will help reduce wear on their equipment. On the other hand a half decent solar panel & this would drastically reduce the consumers reliance on the grid. 21. May 6, 2015 rollingstein Is reducing reliance on the grid axiomatically good? Most grid connected consumers in the US enjoy amazingly high uptime. I'm pretty sure you'd have to spend a lot more to get that sort of reliability when you go off grid. 22. May 6, 2015 insightful Regarding payback, if the 7kWh unit costs$5000 installed and saves 7kWh/day peak-cost power, for a 10 year payback you'd need a differential of 5000/7/365/10 = \$0.20/kWh. This is just to break even (no savings). Color me unimpressed.

Last edited: May 6, 2015
23. May 6, 2015

rollingstein

Besides, doesn't that assume approx. 3500 charge / discharge cycles. Are these batteries capable of that life?

24. May 6, 2015

insightful

Tesla warrants them for 10 years.

25. May 6, 2015

Staff: Mentor

1. The cost to produce a kWh will be the average over the entire set of operating power plants. So a coal plant operating 100% all the time uses the same amount of coal and cost the same amount to make a kWh regardless of time of day.
2. I don't think peaking power is much more expensive than from other sources, but obviously it will depend on the particular sources: it certainly isn't four times more expensive (after averaging it in with baseload) per Tesla's number. And as I said before, the price of electricity at night sometimes actually goes negative!
Right: building and maintenance is the big cost for peaking plants. Not energy production.
The price of any product is driven by a balance between supply and demand. Energy supply availability is fixed (not counting solar), meaning you can run a peaking plant at the same cost any time of day. Demand is what varies and forces shutting off power plants or dumping electricity at night or turning-on a peaking plant during the day. So you are charged during the day for the existence of the peaking plant and energy is dumped onto the market at night because base load plants don't throttle well.
Because the difference in the value of the energy to the consumer is greater than the difference in cost for the utility to produce it.