Categories: Battery/StorageSolar

My Tesla Powerwall: Tales of an early adopter

Published by
Unleash the Powerwall
As per the tweet I sent the other day, my solar system has now generated over a MWh (Megawatt Hour) since it was installed in late January with the Powerwall. Essentially that means my panels have generated…
wait for it…
zoom in to close up …

One Million Watts!

Yep, that’s lame, but I’ve been waiting to kick that off for weeks now. And some would say that picture has a striking resemblance to me.
On average that makes it a bit over 25kWh per day solar power generated, which is quite good considering we had a few cloudy days. Record day so far is about 33kWh generated, which might get broken around the summer solstice next year when the days are a bit longer.
March in Australia generally resembles February, but at this stage it is even warmer than usual in Sydney. You can read a really good summary at The Conversation about why its happening this year in particular.
I’m also following 9th Dan Chart-Master, Ketan Joshi on twitter, to see what statistical wizardry he’ll come up with next in terms of climate science and weather patterns.
This record period above 26oC (79oF) has certainly provided some amazing solar power statistics for the first week in March, as follows:

Fairly healthy rates for import and export, and its just a shame I’m yet to take advantage of feed-in tariff. I’m still awaiting changeover to Diamond Energy so I can take full advantage of Reposit Power with the Powerwall later this month. SO excited. GridCredits!!!
If I was on my current provider’s plan (which I’m not because I’d have to sign a new agreement), I’d have imported about $1.70 worth of power after all discounts, and made about $4.40 in feed-in tariff. Throw in 7 days of connection charges of $4.94 after discounts, and I’m $2.24 in the red, theoretically.
However, I’ve consumed 110.17kWh of my own making, or nearly 16kWh per day. That is actually pretty low against the household average over the year (~20) and if I didn’t have solar panels, that 110.17kWh + connection fees would be somewhere in the order of $31.70.
So really, when you think about it, I’m almost $30 to the good!
Importing just over 1kWh per day looks pretty spesh, but I’m here to tell you it could have been even better, and will be once this warm spell expires. Let’s go to the chart…

Right at the start of you can see a little red block. This was an anomaly where the battery management firmware was confused by 29th February (Leap Day). Yeah, as a developer I can tell you it happens, occasionally. I’m assured its fixed, and because we’re on daylight savings time right now it started at 1AM on the 29th Feb and went through until 1AM on 1st March. Cool.
There is a big spike on 3rd March which was air conditioning going on for a very warm afternoon. On the morning of 7th March the battery finally went to sleep after the previous evening’s efforts cooking dinner and dessert.
Why burn stored energy doing all that baking in the evening? Firstly, it was a work day, but mostly, let’s just say that you can’t put obsessive battery monitoring ahead of my wife’s roast chicken, followed up by chocolate cake. You just don’t.
Our electric oven is running a bit out of kilter at the moment, judging by the noise coming out of the fan assembly. I’m getting that rectified next week, which should decrease duration of usage, by providing a more efficient cooking cycle. Moar cakez!
I recently added a new network attached storage (NAS) for backup duties, after the old one finally threw a shoe – i.e. one of its RAID partitions. The new one has power recovery so can handle being on a timer, along with the TV and games console running from that outlet. Only a few watts saved, but every bit counts when you’re trying to get the battery through the night as often as possible, under varying circumstances.

Powerwall Zen

As I’ve learned a bit more about the Powerwall, I’ve also tweaked some of the larger hardware pieces to exist in harmony with each other and the solar system in general.
For example, the pool pump needs to be running for a certain number of hours per day to ensure cleanliness, and takes about 0.8kWh at standard operation. I’ve worked out the pump capacity (Qmax) and rate of flow correctly, to ensure I use only the minimum hours per day. I also figured out that running it later in the day allows the battery to top up earlier as the panels start cranking up the flow mid-morning.
In partnership with this, the timer on both our washing machine and dishwasher are set for the early afternoon, when the sun is at its highest. I leave a gap in the pool timer of 2 hours in the middle of the day so both these devices can run without issue, even when the sun isn’t so bright.
As always, a solidly overcast day can throw some of these plans out of whack, and that’s why a grid connection is a must. Beyond keeping the warranty of the Powerwall at 10 years (off-grid is only 4), in suburbia there is simply no reason to disconnect, with the grid as your fallback.
And I’m only a couple of weeks away from generating GridCredits, getting the battery topped up before the 2PM peak boundary makes sense both financially, and from the data it will provide to advance our knowledge of micro-grids in a modern urban environment.
Now, if you’ll excuse me, I smell cinnamon cupcakes baking…

The Pfitzner household is equipped with 20 x 250W Phono Solar panels in two arrays of 10 panels, for a generation capacity of 5kW. Both arrays address a north-west aspect, with the western set slightly shading the other in the late evening. There is no shading during prime generation hours, and summer hours are expected to keep the array at or above 1kW between the hours of 0900-1800, based on readings to date.
The panels connect to a SolarEdge SE5000 inverter, with P300 Power Optimisers on all panels, providing maximum throughput of 5kW to match the PV array. The inverter, in turn, connects to the StorEdge battery controller that controls flow between the Powerwall and the house. The Powerwall is a “Daily” model, Lithium-Ion store of 6.4kW capacity, which from full will provide the house enough power to get through most nights, depending on use.

This post was published on March 16, 2016 9:51 am

View Comments

  • You mean, of course "One Million Watt-Hours," or, more elegantly, "One Megawatt-Hour."
    Good work all the same.

  • With out knowing exactly how many days in this period, but I've estimated 18kwh/day average PV production not 25kwh.
    The Power Wall must be grid connected and can't operate off grid at all. What BS are you spinning here.

    • What makes you think the 7KWh powerwall cannot operate off grid? I wonder why Elon Musk said " or you can go off grid, if you want". Perhaps you know more about the Tesla 7 kwh battery than Musk himself.

      • People don't realize the Power Wall is just a battery. To go off grid you would need an Inverter/charger + a battery. That's what Musk left out. So for just a battery it's fairly pricey.
        Other offerings on the market come with battery and inverter/charger and can operate when the grid goes down for a better price. The Power Wall can't do that.

        • Might be useful if you actually have some other options listed or linked. Otherwise you just look like you're here to have a crack without contributing.

          • Had a look - their website was skinny on details. There was an article in The Guardian and I think Redback's spokesman needs to do his research.

          • http://www.theguardian.com/sustainable-business/2016/feb/23/half-the-price-in-half-the-time-solar-storage-innovation-harnesses-new-energy-frontier
            "Redback’s founder and managing director, Philip Livingston, said it would take an average 5.6 years to earn back the Ouija’s $9,000 price tag through electricity bill savings, versus 11 years on the Powerwall’s $15,000 to $17,000 installation cost."
            ...
            “So effectively, a few guys and a small truck can get it installed rather than four guys and a crane, which is what you need for a Powerwall. ..."
            Both claims either ridiculous showmanship, or a complete lack of research. Cost is out by several thousand dollars in the first instance, and a crane for 95kg???
            Don't get me wrong, the smart tech aspect of Redback is a massive plus, and I'd like to see them get going. But they're not even in the market yet and they're slagging the opposition.

          • Redback 40kg, Power wall 100kg need I say more. Redback is well and truly in the market. And as I said watch the video!

          • Even if it was 100kg (and it isn't), it isn't anywhere near "four guys, and a crane". My Powerwall was mounted by two guys, using the box that came with the unit (it has a mounting arrangement).
            Results may vary depending on strength of installer :)
            Back to Redback: according to the Guardian article, its 40kg for the inverter (my SolarEdge is 33kg), and 25kg for the battery enclosure. Easier to mount in bite-sized pieces, without any doubt whatsoever. That's an advantage.
            The article goes on to say: "and each battery you place inside it is 20kg". So that 40kg doesn't include any storage at all, it would seem.
            The Redback spec sheet doesn't mention batteries at all, just a storage range: http://redbacktech.com/wp-content/uploads/2015/08/RBT004_data_sheet_Inverter-System-v11.pdf
            Given you're the one who introduced me to Redback, you might know the following:
            - Does the $9000 Ouija cost include any storage capacity?
            - If not, are they offering to supply batteries, and at what cost per kWh?
            - Does the $9000 Ouija cost include solar panels? I only ask because for $15990 I got my Powerwall, and 5kW of solar panels (power optimisers) installed. So maybe Livingston got his wires crossed on the cost thing. I don't quite know.
            BTW I'm watching the Powerwall Launch video and he says you can go completely off-grid with solar panels with a "complete and affordable solution". You can construe that as misleading, unless you consider that everyone who knows anything about solar panels knows you require an inverter.
            And in fact, the attached image from the display shows an inverter on the graphic.
            It still isn't $15000+ for the Powerwall by itself as Livingston states.
            http://youtu.be/NvCIhn7_FXI?t=8m55s

          • Yes it does weigh very close to 100kg. You say you have a Power Wall, then you should know that. Why lie. Redback is battery agnostic and can use Li ion or lead acid, perhaps even Redflows battery.

          • Storage agnosticism for Redback is great. Not sure the battery enclosure would hold a ZCell, so you're saving 25kg there.
            Assuming you got the ZCell ($17.5k - $19.5k), that would be a 10kWh system for somewhere around $28k, but still has no panels. Add another ~$5K to get 5kW of decent units in a package scenario. Total cost somewhere north of $32K.
            Total weight somewhere north of 280kg - ZCell is expected to be approximately 240kg according to https://faq.zcell.com/content/6/35/en-us/what-are-the-physical-measurements-and-what-is-inside-the-box.html
            For that money I can get two Powerwalls (12kWh) installed with 10kW of power optimised panels and appropriate inverters.

          • You stated in a previous post "no where near 100kg" I'm sure nobody else would consider for a second, that 97kg is no where near 100kg, so you lied!
            So you would make a killing in grid credits would you. EA is only paying 4.8cents/kwh, some others, a little more and some zero. So @ 4.8c/kwh and you exported 300kwh for the month you would make the pitiful amount of $14.40 WOW.
            You have come to show no credibility. Now that scares me off!

          • LOL read it again - my words:
            Even if it was 100kg (and it isn't), it isn't anywhere near "four guys, and a crane".
            So I didn't say " nowhere near 100kg" at any point.
            If you don't like Tesla, then OK. Just don't make things up, then sling around words like "credibility".
            I get 8c from Diamond Energy (having left Energy Australia) which at the moment makes my usage completely neutral, and even reduces my connection fee slightly.
            Once the Reposit Power GridCredits start to take effect, I'm looking at zero connection fee too

          • On credibility, you said Redback's website was "skinny on detail" and also stated, "their not even in the market yet" Both BS statements!
            And please elaborate on " makes my usage completely neutral" re: 8c

          • Still deflecting? You haven't answered a single one of my questions, or even apologised for your gross inaccuracy and outright fabrications.
            However, I will answer the cost neutral one, because I have the facts to do so, and because I'm really looking forward to what you come up with next.
            I will pay about 20c per kWh (after GST and discounts) for any power I import.
            I will get 8c per kWh for power I export.
            Therefore I need to export about 2.5 kWh for every one I import, to keep the cost at zero.
            Still with me?
            Meter this morning reads 140 Import, 419 Export, since it was installed on the day I moved to Diamond Energy. That's almost exactly 3:1 in my favour.
            I'm ahead at this point, and will even see a small reduction on my daily connect fee.

          • Haven't answered any of your questions? Rather you haven't answered mine!
            Your rather insolent, saying I have been grossly inaccurate and that I have been fabricating. Your the one doing that pal. Now go away little man.

        • I'm not sure quite where you're going with this point - Tesla never said it was a boxed solution. It is a battery, and has never been marketed as anything else. Of course you need some form of interface for it. The fact that other manufacturers are supplying it will reflect in the total cost of ownership

          • In fact Elon Musk, made a point about going off grid with it. That's the main point, the other is there are all in one systems out, that will do what I said above, but cheaper and better performance. Redback's web site isn't light on detail either.

      • The battery is too small for off-grid use, but for 'back-up' during outages,
        the user must arrange 'emergency loads' to be connected to a separate circuit.
        When the grid fails, a switchbox connects that circuit to the inverter.
        Requires addtional switchbox and Solaredge's autotransformer, according to both Tesla's website, and SolarEdge's user manual. Those who may find a way to use the battery off-grid, may be interested to know that only one Powerwall may be connected to the SolarEdge inverter. To add more, either switch one battery in circuit at a time, or add another inverter. Additonal 'StorEdge Interfaces' are required for each additional battery.
        Those are things Elon Musk knew, but you didn't.

    • Original article was written at 39 days post install. OneStep have generously reproduced it, which obviously changes the calculation.
      Powerwall can be off-grid with the right fit out, but it isn't designed for that. Also drops the warranty to four years, and as Peter has pointed out below, is really not big enough to do so. Unless you hooked up more than 1.
      VRLA is a better fit for off grid.

      • On the off grid bit your correct needs an inverter/ charger that is compatible. On the VRLA battery it all depends which ones your talking about, but in essence a good true deep cycle gel battery is good for both if weight and size ain't a problem and cheaper too.

  • LG Chem / Eguana Technologies AC Battery is supposedly more efficient not to mention a better value, wonder if you have it available in Australia?

  • Nice informative charts. Do they come from the powerwall or your electricity provider?

  • The calculations are for solar+battery, but do not separate battery costs from the total.
    I infer from the text that the imported energy cost of the unsigned 'current provider's plan' would be $0.227/kWh, and FiT $0.05/kWh.
    The battery's capacity is 6.4kWh. If fully-cycled, could earn $1.45/day in avoided import costs, but charging would consumme at least 7kWh in solar energy that could have been sold as FiT = $0.35, meaning the battery could provide a net gain of $1.10/day.
    Battery cost is ?? Perhaps Choice's estimate of $9,500. Seems reasonable when the battery system costs of around $16,500, and 5kWp installed solar, are compared.
    At $1.10/day, simple payback on the battery would be 23.66 years. Better to pay the avoided grid import costs, and reduce capital outlay by the cost of the battery, and then calculate the gains made from solar alone.

  • Nick, can you answer a few questions?
    1. Do you have a measurement of the true capacity, is it 6.4 kwh?
    2. Have you measured the efficiency...is it 92% of 6.4 kwh. In other words do you input 7 kwh to get 6.4 kwh out? Or do you input 6.4 kwh and get 5.9 kwh out?
    3. I notice in the warranty it is slightly strange in giving minimum guaranteed kwh.
    a) First 740kwh @ 5.4 kwh
    b) Second 1087 kwh @ 4.6 kwh
    c) Third 2368 kwh @ 3.8 kwh.
    This totals about 18,000 kwh over 4194 cycles or 11.5 years. I am a little perplexed by this, any thoughts?
    4. My guess is that the battery will far exceed these numbers but Tesla is cautious on the warranty. Secondly the numbers are abrupt step downs rather than smooth curves. I decided to recalculate with a smooth line from 6.4 kw down to end of life at 3.8 kwh and this gives warranty lifetime of 19,746 kwh. Sounds better and more likely.
    5. My realistic guess of total lifetime kwh is 5000x 6.4 x (average of 80% decline over life) =
    25,640 kwh. Do you agree?
    Thanks for your time...the rest of the battery geek world is mighty curious. You will help resolve some passionate online arguments.

    • Capacity loss is summarised by the numbers in the warranty, which represent points on the smooth curve you mentioned. Total stored energy may vary a bit depending how well that curve is fitted the known values, but 18MWh in 11 to 12 years is close enough, but less within 10 years.
      Another approximation can be obtained by assuming some fixed-percentage loss per year. I have plotted a chart where that rate is 1%, 2%, 5% and 6%. The (obvious) curve produced from the warranty, rather than the almost straight line, says the Powerwall is in the higher loss bracket, so little will change the stored total found by fitting the actual values.
      Efficiencies are expanded upon in another document;
      2kW (92.5%), 3.3kW (92.0%) beginning of life,
      2kW (90.0%) 3.3kW (87.0%) end of life.
      Energy is given as 6.4kWh @ 2kW output, with the additional
      "For all initial ambient conditions > -20°C and 5.0 kWh +/-1% measured at 400V to 450V DC link when discharged at a rate of 2kW from 100% SOE at BOL."

Share
Published by

Recent Posts

Solar Insiders Podcast: A roller coaster year in review – and the keys to a smoother 2025

In our final episode for the year, SunWiz's Warwick Johnston on the highs and the…

December 20, 2024

“Nightmare:” Energy tariffs that are blowing out bills, blindsiding consumers

Regulator report finds that little-understood but increasingly common demand tariffs can add up to $800…

December 20, 2024

Hidden cost of rooftop solar? Actually, networks spend next to nothing on managing exports

Have you heard the one about non-solar homes paying the cost to networks of accommodating…

December 19, 2024

With just $500 of rooftop solar modules, you could charge your EVs for 20 years

Four good quality solar panels - costing around $500 - would produce enough power for…

December 19, 2024

“It makes no sense:” How fossil gas industry is blocking electrification and consumer savings

The gas war still burns: “We need to think about how to stop misinformation going…

December 17, 2024

SwitchedOn Podcast: Pathway to electrification being blocked by fossil gas

IEEFA's Amandine Denis-Ryan explains why new gas supplies are not needed in the long term,…

December 16, 2024