Tesla's Home Battery
NWCCTV
Posts: 3,629
Tesla is expected to unveil a new home battery later today: http://www.usnews.com/news/articles/2015/04/29/new-tesla-battery-could-end-electricity-bills
This could be a major game changer but I want to wait and see if they live up to their expectations first.
This could be a major game changer but I want to wait and see if they live up to their expectations first.
Comments
Let's say off-peak electrical consumption is $50 a month for someone with solar (and that's high). At $13,000, it would take 21 years to pay for the batteries. By then the'll be long dead. These things don't have a great life cycle.
IF your local energy company agrees to buy back peak kWh (your meter spins "backwards"), then you stand to make more money selling your excess energy to the grid, because you're doing it during the highest period of demand. More and more people have so-called smart meters, where the plan is to charge more during peak, and less during non-peak. These batteries are intended primarily for non-peak, when power is the cheapest anyway. At the very least, it doesn't cost you infrastructure to sell back your power.
Presumably the Gigafactory will be pumping out solely Lithium-Cobalts, for the highest energy density, rather than the more rugged Lithium-Phosphates. We'll have to wait and see I guess.
John Abshier
and that is one giant IF
Power companies may initially start doing this for the Green Cred Marketing Fuzzies points, but they quickly change the fine print if any significant number of customers do this.
Then, you find power charges morph into split values, like Lines charges, Daily connection fees, and Power at two rates - the 'smart meters' can bill at any split tariff the companies choose.
All that means you can get best payback on power, only provided you displace full retail kWh with it.
With solar PV with no storage, that means someone at home during the day, who is happy to skew their usage to track the sun-of-the-day... very few fit in that basket.
Add a Battery, and you can shift some of that power to a more typical consumer, who works during the day and cooks at night and at breakfast.
Now you can gain the difference in Tarriffs, minus the round-trip battery losses.
Typical power usage has quite large variations, so the 'sweet spot' will rarely be hit.
Of course, the numbers all change if you cut the cord, but that is quite a large step, and power companies will want their pound of flesh to reconnect you !!
Several problems with this statement.
First, there's no point to solar -- because of the cost -- if all your usage is off-peak. Why spend tens of thousands to save maybe $500-600 a year on evening-hours usage? Solar is the most beneficial if you're at home during the sunny hours to take advantage of it. Batteries can't make up for that because of their cost, both initially and maintenance. No way will these batteries last more than 5-8 years.
Second, stay-at-homes are now averaging about 40% of households. The numbers are higher in some parts of the country, lower in others. But since 2000 the trend has been an increase -- not a decrease -- in at least one family member remaining at home during the traditional workday. So I wouldn't call it "very few."
While it's true many utilities do not currently buy back watts, 99% of them are public utilities, and come under the strict oversight of the government, who likes to legislate such things. The reason we don't see it more is that solar is still a minute part of the grid. As in teeny-tiny. The effort to enforce a buy-back isn't worth the political hassle. At least for now.
(Also, the utilities don't buy at retail. At least none that I know of. They buy at wholesale, or a price reasonably comparable to the amount they pay, or claim to pay, their provider partners.)
Even if you don't sell back the power, at $13K for the batteries there just isn't a sensible ROI. Just taking that cash straight-line you're losing money over putting it into a conservative investment of say, 5%. That's $650 a year, or $54 a month, for doing nothing. Unless you're in the desert, with the AC running constantly, you're about even-steven with your investment return and the the cost of off-peak consumption. IOW, there's just no point in these batteries, at this cost.
Not slamming Musk and Tesla, but this is how they operate. The economics just isn't there today. But it's good PR, and in the meantime, they're building market position if and when the price of the batteries goes down. But it has to go down by at least 50%, or power has to go way up.
I'm putting up some more panels - this frame will have 6kW, and already have 13kW installed. On the best day in summer 13kW will make about 70kWh of electricity, on the worst day in winter, about 15kWh. Average house, ballpark uses 24kWh a day, and if you want to store for a day, 24kWh might cost $24k. Too much IMHO, the cost needs to come down to about 1/10th this to make batteries viable. Tesla is getting closer at ?$250/kWh.
I'm still researching battery technology. The all seem to suffer from the problem that they wear out, then you have to throw them away. The only super long life battery seems to be Edison's original Nickel Iron with cells still working 50 years on and more.
I've gone instead for smart load management. Measure power, transmit this around the place (with propellers and arduinos) and turn loads on and off to match the production. Things like the pool pump, water pumping, heat pump water heating. Export what is left over, but ideally use up energy as it is made. There might even be scope for smart washing machines, dishwashers etc where they can turn on in the middle of the day.
Orient the frames in an arc, or V shape, to catch a broader time span across the day.
Total power would not change much, if at all - could make it worse. If you orient so that only a portion are at a good angle then others are at an even worse angle than they would be otherwise. Each set of panels at a particular orientation would have to be wired separately from other sets, with separate charge controllers, else the disadvantaged panels will be the limiting factor.
Recall a news story about a "genius" student who made a solar panel system with cells randomly oriented like leaves on a tree. There were always cells pointed towards the Sun, so it was reported as a breakthrough.
Any 'gains' are minimal, and you buy into significant wind loading and aesthetics impacts, and even shading issues if you chase large V's. Easier to apply KISS, and add another panel or 2.
A +/-15' angle has moved 3.4% on the cosine curve.The same as adding one panel in 30.
Re strings of panels, generally 200V to 400V DC input on an inverter, panels are 24V under load, so 12-14 panels per string. So certainly can mount strings at different orientations. For maximum energy collection, I think the best answer is to mount pointing due north (or south in the northern hemisphere), and angle = lattitude (35 degrees for me). Might be possible to slightly optimise that, eg I have trees that shade the panels to 9am, so it may be better to point slightly north north west.
Re east and west oriented panels, that works if you want to flatten the curve and match to a particular load. In practice, the curve is hardly ever sine shaped - mostly it is somewhat sine shaped with dips here and there as clouds go over. So at the end of the day, if you have a panel, the best thing to do is point it north/south.
Re batteries, if off grid there is not really much choice. But for on grid load shifting, there are some intriguing optimisations. For instance, instead of charging the batteries with the panels, feed the panels into the grid as AC, then somewhere else, charge batteries from a standard AC to DC battery charger.
Rough figures without subsidies (I'm not claiming any subsidies) - inverter is 10-20c a watt, panels are 60c a watt, frame is 40c a watt, cables are 10c a watt.
So Mr Tesla needs to store power with these sorts of ballpark numbers.
One concept I'd be very interested in is a lease arrangement, where you lease batteries rather than buying them, and the lease includes replacing them at a predetermined interval. Then it truly includes all the costs, including disposal/recycling.
I don't understand you repeating. My argument is based on non-maximal light collection. It inherently expects excess panels installed. What's non-KISS? It's no harder to build one direction over another. The wind argument is a bit of a red-herring. One would hope the panels are well secured already. Shade from other sources is a concern of course, but I was talking generally due to shading being an unknown factor.
Usually they install simply parallel to a roof plane. Anything else is special tooling, and likely special sign-offs too..
Maybe to you ? - but not to an approvals body
Ah yes.. A few seconds with google ...
["In addition, uncertainty about what constitutes a safe and secure installation for a
given wind load can slow or even stop the approval process for PV installations and complicates the training of code officials.
In this report, we provide sample calculations for determining wind loads on PV arrays based on ASCE Standard 7-05. We focus on applying the existing codes and standards to the typical residential application of PV arrays mounted parallel to the roof slope and relatively close (3 to 6 inches) to the roof surface. We do not address other array configurations or building-integrated PV."]
Notice the 'standard' install (of 3-6 inches parallel), and anything else ?? -- kerching !!,
Also, the conversation is about panels mounted in an open paddock or large back yard as per the photo in post #9.
I live in Maine and am currently in the process of installing a 8.5 kW array on my roof. The system will be grid-tied with no storage. Maine has a net metering program that buys any excess electricity at market prices, and you can 'bank' for up to one year. So essentially any excess power I produce will be used to offset any kWh that I buy. Under the current regulations, it would be economically foolish to install a storage system, due to the cost and losses in storage/retrieval/conversion.
I have a geothermal heat/cooling pump that uses a fair amount of electricity in the winter and a swimming pool that used a fair amount of power in the summer (swimming pool = big mistake, will probably fill in as soon as kids are out of the house). I anticipate this array will supply pretty close to 3/4 of my energy use.
Also, each module has a 'power optimizer', a somewhat new-ish technology which optimizes the power point tracking on each individual panel, so orientation/shading of each individual panel is not a big issue (ie if one panel has shading, it does not effect the rest of the string). My setup will not have shading issues, but the best part of the power optimizers is that they allow individual data collection of each panel output as well as the inverter, so any problems can be seen and corrected quickly.
I am installing the system myself because I can and it saves me about 1/3 of the cost. I expect a payback of about 6 years using the simplistic (money spent)/(money saved) formula. This is not a great return, but not too bad and fits ok with other financial risk/reward strategies I have.
Nate
Also known as micro-inverters.
That's excellent by anyones measure of PV installs. I think the general guide is 15 years as the worth while payback time. I'd love to have net metering around here.
Trying to find what battery they are using - interesting article http://batteryuniversity.com/learn/article/bu_808b_what_causes_li_ion_to_die looking at how long different lithium batteries last. Looks like Tesla is using Lithium Nickel Cobalt Aluminium Oxide. Not a battery I've ever heard of, I wonder if you can buy them easily?
Another interesting article here https://www.electricbike.com/lithium-cobalt-manganese/ Lithium titanate looks promising too in terms of long life.
The article doesn't say what the cycle discharge depth is for that 10,000 cycles. Lead-acids, for example, are normally rated for 30% depth only. So, the spec may say it can do 1000 cycles - but that's only for a discharge of 30%, which means the rated full capacity is pretty bogus.
All I'll say is that Lithium Phosphates are hear now and at a similar price per KWh to high quality deep cycle SLA batteries. I guess it depends on how long you want to wait.
1. Do you need to add support to your roof when adding those large solar panels? How does the panel handle up to snow/ice loads?
2. Do you include additional home owner insurance for the panels in case of hail or strong winds?
3. In case of a damaging wind / flood do you need to provide short limiting circuits to prevent rapid (and possibly explosive) battery discharge?
Just never could get the power company person to talk about these.
1: The panels are not very heavy, they are very thin, maybe 3mm thick, so weight on the roof is minor. Wind forces are worse. The panels are made of a strong plastic with some sort of a flexible glass top layer. I was most impressed with the strength the first time I got my hands on one. You can stand on the middle, not that I recommend doing so. The outer aluminium frame has no structural strength, it's purpose is only for clipping to the mounting rails.
2: You'll need to talk with the insurance company. Presumably the installation can be covered as part of contents.
3: Of course circuit breakers are needed at the appropriate places but funnily not on the solar DC supply circuit as this is already current limited. There is many details. If you are going to be doing the mains wiring yourself then just remember that both the insurance company and the power company will likely put you through the ringer to prove you've done it to code. The power company also has to clear the inverter used. Probably best to start small so you can pull the plug if it goes pear shaped. The panels have a fire test rating that you should look for ... oops, I've run out of time ...
However, I did see a solar install on Utube ('Dave the electrical guy' or something like that) in Texas where his residence had a flat roof and they basically built racks for the modules and then weighed the racks down with tons of cement blocks rather than fastening through the roof membrane - but as I said, that was in Texas...
2) My insurance guy says my policy covers the installation just as any part of the house. In my installation, if there is a wind that can remove them from the roof, there is going to be major damage to the rest of the house as well. They are really fastened down with multiple redundancies and are in the plane of the roof with only a small 3 inch between the roof and the panels. My biggest concern is lightning, but I have done what I can with grounding/bonding and the rest is up to the insurance policy.
3) Not using batteries as it seems to be an expensive option that adds little to the value of my grid-tie system (I do have a generator for the infrequent long power outage). Of course as with any electrical install in my house, appropriate current limiting devices are used. This includes standard breakers on the AC side of the system. On the DC side, the power optimizers located at each panel have a 'rapid powerdown' feature that shuts down the DC power from the solar panel if connection with the inverter in the basement is lost/severed. The inverter also has GFI/arc detection on the DC side, which should shutdown the system of there is an imbalance between the (+) and (-).
I talked to the solar sales people, then did the permit application with the power company before actually purchasing anything. That way I was able to run the design and make/models by the power company guy before committing to anything. The power company people as well as the town code enforcement were very pleasant to deal with and there were no real issues.
What is the cost per unit then? Comparison chart here http://www.pv-magazine.com/fileadmin/PDFs/Storage_Database.pdf and looking down the second to last column, some of the figures look very promising. It looks like there is a very competitive and rapidly changing European market.
Wow, that's saying 100% DOD too! That's 40 years at one cycle per day. Impressive. You can expect better than that given the extreme depth rating. Just needs 10x the capacity now.