Home » Commentary » UBS: Tesla Powerwall can deliver 6-year payback in Australia

UBS: Tesla Powerwall can deliver 6-year payback in Australia

Call it the great debate. A new technology that threatens to disrupt a century-old status quo in a trillion dollar industry. Battery storage, teamed with solar, is predicted to do to the energy industry what the internet did to media and mobiles did to the telephone hand set.

tesla storage

There’s really not much debate about that. Consumers, technology developers, retailers, network operators and generators all recognise it. Even the politicians do. The great debate is about when and how quickly this will happen. Some say now and very quick, others are not so sure.

To try and dissect this, and to give as much prominence to differing views, RenewEconomy will run a series of articles looking at various estimates.

We kick off today with investment bank UBS. One of the features of this energy transition has been the close involvement and observation of the financial investment community. This, they believe, is no passing fad. It is fundamental to their assessment risks of hundred of billions, nay trillions, or dollars in energy investment across the globe.

The UBS note on the economics of the Tesla Powerwall comes from their Australian analysts. That’s not surprising, given that Australia will be ground zero for battery storage around the world, thanks to its high electricity costs and massive solar penetration. But it’s also ironic given that the product won’t be available until next year, although some of Tesla’s competitors may try to steal a march beforehand with their own product releases.

The headline conclusion of the UBS team is that the Tesla Powerwall – the 7kWh version – will deliver an economic return. They estimate at an IRR (internal rate of return) of 9 per cent. That represents a pay-back of about six years. If they are right, then that means that mass-market adoption ain’t so far away as some would believe, and incumbent utilities might wish.

So, how do they get there? This table will help illustrate, along with the explanation is below.

tesla storage ubs

The price is the big ticket item in the equation. UBS notes that there has been a big debate about the the difference between the price of the battery – US$3,000 – and the installed price quoted, closer to US$7000.

But it says that while that may be the case in the US, where rooftop PV is vastly more expensive to install than in Australia, it believes the difference between the battery price and the installed price would not be that high in Australia.

The analysis also assumes that the battery can be fully utilized 7KWh of gross energy every day and that the solar system is large enough to charge the battery and still provide the behind the meter power. It also assumes that the US$ online price can be directly converted to A$ and will be the price in Australia.

The UBS analysts say they are confident that inverters can be obtained for around $1100. They have used a “Powador” model in their example, priced at  A$1025. The installed cost at $5,175.

As for how it relates to the market, UBS takes the example of the Ausgrid area retail peak electricity price of $0.51 Kwh and the price paid for selling electricity to the grid at  $0.06/kWh.

It assumes the system is 89% efficient, and it assumes 4 hours of installation labour @$100 an hour.

“On this basis we conclude, ignoring tax, that the system can provide an 11% IRR which is better than the home loan rate and about a 6 year payback.” It notes that the battery will more likely suit larger detached houses with higher than average consumption and larger than average solar systems.

Now, some people may find the UBS price estimates as on the optimistic side. But even if labour costs and balance of system costs are higher, and push the total installed cost to around $6,300, then  consumers with a solar system already installed will still get an IRR about equal to the home loan rate.

(See also a later story, Morgan Stanley sees 2.4 million Australian households with battery storage, which also includes a forecast of 6-year paybacks in some states).

Comments

90 responses to “UBS: Tesla Powerwall can deliver 6-year payback in Australia”

  1. Warwick Avatar
    Warwick

    Yep, definitely on the optimistic side and not the best research by UBS…If you’re in the Ausgrid area, the flat rate is about 25c/kWh with Energy Australia or the other retailers. Whilst the full calculations are not revealed, you should not compare the difference between 51c and 6c…as you are taking the two most optimistic values i.e. the cost of energy on a contract is about half the 51c quoted here and the value from a solar system can be more about the offset of retail consumption at 25c or so rather than a measly 6c for export. i.e. you’re better to self consume when the retail price is above the feed in tariff, so adding storage adds no value at those times.
    Storage is a good idea, but the economics are not as attractive as these figures first suggest…

    1. Ken Dyer Avatar
      Ken Dyer

      Warwick, the figures may appear a little rubbery but that is because the costs for batteries are dropping even faster now that the Tesla Powerwall has been released, so the economics become more attractive. If anything UBS has been conservative.
      http://theconversation.com/battery-costs-drop-even-faster-as-electric-car-sales-continue-to-rise-39780

      1. Rockne O'Bannon Avatar
        Rockne O’Bannon

        Conservative? That is stretching it. The assumptions are the problem. I find three or four assumptions to be a lot less than conservative.

        I am an ideal customer. I am not sold.
        You know, if the size is not matched really closely to what you are generating AND using pretty much all the time, these numbers fall apart pretty quickly.
        Even granting that no system is ideal, the question becomes: why would I want a battery with x + y capacity when I only need x?

    2. Aaron S Avatar
      Aaron S

      I agree with Warwick, price of power is more like 25c/kWh so it will take twice as lone for pay back

    3. Rockne O'Bannon Avatar
      Rockne O’Bannon

      Thanks Warwick. I tell you, I can’t get the numbers to work even for JAPAN, so I have been moving around the internet seeing what people are doing. Kudos to people on this site who are taking the time to do the math. That is surprisingly rare.

      You seem like you have considered the real world aspects well.

      I expect to have surplus after the FIT in Japan ends in several years, but I don’t know what the other costs and conditions will be then. Even so, if one cannot be assured of using storage at pretty near its capacity all the time (it is expensive), then the expenditure never quite earns the expected return.

      Keep in mind one other scenario. If you are generating 15 kW during the day,and have a bunch left over, and you consume most of your grid electricity after the sun goes down, the battery system could be useful even when the retail price is higher than the FIT.
      Also keep in mind that a lot of people designed their systems to maximize the power generated, not to match their own consumption. Therefore it is likely that they will be generating excess.
      The huge price differential between your cost of generation and the amount you pay is what drives the returns from storage. Therefore, unless you have absolutely free solar (no FIT) and really high grid rates, this will not pay off.

      1. Warwick Avatar
        Warwick

        Rockne, it’s good to see that you recognise my critique of the article is about getting the numbers right rather than being against the technology (which I think has great potential). Perhaps the easiest way of explaining the value of storage is the two opportunities it presents: 1) Time shift energy consumption from the grid’s peak charge times to off-peak where there is a significant difference in price 2) Store excess PV energy where the energy consumed later is at a higher value if consumed from the grid than the FiT offered on exports.

        I’m not sure whether the UBS report was done by some interns or possibly deliberately chose the extreme values to make the economics more attractive, as it looks suspiciously like they’ve chosen a controlled load scenario (i.e.usually for off-peak hot-water) where the 2-8pm price is around 51c and the off-peak price (10pm to 7am) is around 11c. PV alone possibly makes good sense to produce during the peak and offset consumption from peak charges but why would you bother storing excess PV energy during the day (i.e. you can offset shoulder at about 20c/kWh or off-peak at 11c) against the FiT of 6c so for a gain of only 5 to 15c/kWh after losses. If they wish to choose the controlled tariff load, then the likely best scenario would be to ignore PV completely (if you use a lot of peak power) and charge up at 10c/kWh in the off-peak to displace peak power at 51c/kWh, therby saving 41c/kWh before losses…

        They need to clean up their analysis and detail their assumptions so people can assess its worth. Storage is coming but not quite as quickly as this research suggests…

        1. Rockne O'Bannon Avatar
          Rockne O’Bannon

          I am also an “iconoclast”. I am perfectly willing to buy or not buy based on the numbers, not the gee whiz factor.

          Right on the whole first paragraph. And I have offpeak rates too, so I am certainly considering both propositions. I have thought about the “problem” for years.

          As a point that might interest you, many analyses on the internet are glowing, but they are done by Morgan Stanley, which is a major client and lender to Tesla. I think they have an equity stake too. It is pretty bad. There are no disclaimers. You are right to be wary of estimates.

          Regarding your second paragraph, You seem to be noting that the UBS people are not differentiating between the two value propositions. And they should
          1. For people with a FIT going from “some high number” to “effectively zero” all that “extra” solar power could be stored to offset the electrical demand up until 10 pm. For me, this is the big deal because 10 kWh should just about cover it.
          Under absolutely ideal conditions, a 10 kWh could give me 3 dollars American every day. In the real world, more like 2.
          2. For people with a high differential between peak and off peak, there is also a benefit, which is the difference between rates times the capacity of the battery.
          Under ideal conditions, a 10 kWh battery could give me about 2 dollars a day. In the real world, maybe 1.5. Maybe a lot less because, just as day follows night, I can’t discharge my battery during the day and charge it too! haha.

          If you could to BOTH 1 and 2 above, it has to be noted that you are doing two cycles a day and you will burn your battery out in 6 years!

          So we come to the same conclusion. This battery seems like a great idea for somebody, but for whom? Elon Musk obviously, but beyond that?

          1. ChrisEcoSouth Avatar
            ChrisEcoSouth

            To add to the mix, the system “should” momentarily discharge the battery during the day to cancel out import spikes, whether of short or longer duration. The battery would then recover by charging once export starts again, minutes or hours later. How much or little this “light cycling” affects the battery lifetime, I don’t know.

          2. Rockne O'Bannon Avatar
            Rockne O’Bannon

            Ah! More worms for the can! So this is not something someone controls with timers and switches? Yikes. I smell major inefficiencies and unintended consequences.

            I am finding the comments very enjoyable. Nobody but nobody bothers to really think about the cycling and costs and contracts involved.

            Finding “profit” on the other side of this maze is not easy. I am sure most people will just throw their money on the table and trust to the gods. I fear that is likely to have the expected results.

  2. Michael Tournier Avatar
    Michael Tournier

    Just to Clarify a few things with this article.
    i am 100% behind battery storage and the Tesla model is a fantastic little product, for ther poeple want to save a few dollar on thier power bills.
    Essentially there are a few flaws in the design, when it come to the beliefs of OFFGRID for as little as $3500USD.
    From the specs that i have read essentially the system is capable of a 2000w max load capacity, in real life terms this is the power used to cover 1 Aircon or 1 electric kettle, your evening house lighting with the fridge running, but anything else will need to come from the grid. Oven, Hot plates will be grid.
    This article also points to the Powador inverter, this is a grid tied inverter not a hybrid, so it does not have the capacity to charge batteries, i have found that most Hybrid inverters are coming in at around $2500USD.
    I have also been told that the utilities in Australia will not allow the systems to be charged on an off peak Tariff, they will have to be charged from Solar or the peak tarrif, if someone can correct this statement, i would be happy to hear it.
    This means in common terms that if you use 4kwh of power from the batteries due to efficiency losses it takes 5kwh of grid power to charge to full capacity again.
    $$$save $1:20 but cost $1:50 to charge.
    Ideally if you can charge from the solar and buy from the grid the necessary power that the system does not cover, then it becomes viable.
    Most systems will need a minimum 4000w load capacity, and in real life terms that is still too low.
    Utilies require an AS 4777 compliant inverter to have the grid changover to be less than 6000w at any given tranfer.
    IN real life suburbia most housholds will reach this load at any given meal time, these are Peak demand times.
    To be realistic and even come close to “OFF GRID”
    you will need:
    3 x tesla 7kw @ $4850AUD to give the 6000w load capacity needed
    1 x Solar Edge @ $2500AUD
    Solar panels @ $5000
    Installation @ $2000
    Total price $24,000
    This is with no local material cost% mark ups.
    Dont get me wrong i Love Tesla and would buy one myself Car and Powerwall with a carport car charger, because, “wouldnt you be the cool kid on the block”, its like owning an Apple TV, Iphone, computer, and tablet, they just work seamlessly and lets face it they are cool!!
    Thank You Tesla you have bought the New Iphone of power stations, and have bought power storage to the world.

    1. juxx0r Avatar
      juxx0r

      It has a built in Charger, called a DC/DC converter.
      It can’t charge from the grid, it charges only from DC, from your solar.
      It does 3.3kW peak, so 3 would be 9.9kW.
      The article was about On Grid, not Off Grid.

      1. Chris Fraser Avatar
        Chris Fraser

        Could somebody attach a DC Bridge/Transformer to charge the battery from the grid ? Just a little night time energy from a nearby wind source to top up the battery ?

    2. neroden Avatar
      neroden

      Cook when the sun is out?

    3. Coley Avatar
      Coley

      Thanks for that, it’s the most understandable comment up to now, at least for us readers of ” maths for dummies”;)

    4. Richard Pyro Lim Avatar
      Richard Pyro Lim

      How does it compare now that the Aussie dollar has fallen to just $0.70 to the us dollar?

  3. ChrisEcoSouth Avatar
    ChrisEcoSouth

    These are a lot of assumptions. I thought $US3500 was supposed to be wholesale? I have not heard confirmation that the Powerwall actually has the internal charger/inverter necessary? If it does, then why is an external inverter being quoted at all? If it doesn’t have a charger inverter internally (like other brands do/will do), then I guess the price of the Kaco is a “stand-in” price for the external charger/inverter, as the Kaco by itself will not fulfil that role at all. In which case that price may well be too low an estimate. Furthermore, installation costs go up with the number of external components to be ‘wired up’, as they all have to meet rigorous (Australian) standards for LV enclosure etc. $400 for installation, cable runs etc is definitely on the low side, IMHO.

    1. juxx0r Avatar
      juxx0r

      The Tesla Powerwall has an internal DC/DC converter; this is the charger. The powador is for inverting to 240V.

      As for the connections, Solar to Powerwall, powerwall to inverter, job done. It’s only 1 more connection.

      1. ChrisEcoSouth Avatar
        ChrisEcoSouth

        If the customer has a PV system already, then why purchase another grid-inverter?
        The Pwall must be placed out of the weather – that may add 10’s of metres of conduited cabling, with associated isolators and enclosures.

        1. juxx0r Avatar
          juxx0r

          You’re right, don’t purchase another inverter, knock another grand off the price.

          1. ChrisEcoSouth Avatar
            ChrisEcoSouth

            Integrated charger/inverter is certainly the way to go, but the Pwall is still short on details, and costings as the discussion here.
            I’m keenly aware of the costs, and want to reduce them, as we are selling these sort of systems now. This is why I am saying details of the Pwall are missing or not clear.

          2. Lawrence Coomber Avatar
            Lawrence Coomber

            Chris the tendency to integrate other power products into inverters is a mistake. Of course it is a “value add” option for the on-grid inverter manufacturers but this strategy will backfire, and is only viable because on grid technology has a silent partner assuming all of the risk for guaranteeing customer supply continuity [the grid].

            The best inverter/converter designs ensure they do only one function [conversion] and do it efficiently and robustly. Endurance and long term functionality is the converter’s primary role. It should not be over-burdened with peripheral functions.

            Imagine the scenario when the failure of a component within an inverter ancillary module [charge controller or monitoring unit] for example; brings the inverter to its knees and shuts it down because of an inconsequential component failure.

            Ancillary power product modules are best left to 3rd party technology developers which can then be integrated to the system as required by the system designer. Also, competition between 3rd party developers brings about technology improvements; increases competitiveness, and drives down costs.

            I manufacture off grid inverters using these key points foremost in our designs and the results tell the story. The same logic should be employed by on grid manufacturers.

            Lawrence Coomber

        2. Robert Haylar Avatar
          Robert Haylar

          I think it can be installed outdoors, but a new inverter will be necessary. The battery’s price is purposefully low to attract attention. The real cost is buried in the forced purchase of an inverter and installation. The battery will only operate with specific inverters that are part of the package. I think UBS have missed that. The inverter must be ‘software compatible’ with the battery’s management system.

          Edit
          I understand that Fronious and Solar Edge have converters, said to work with the Powerwall.
          Details of the Powerwall are very vague, but there is this from their website.
          “Powerwall consists of Tesla’s lithium-ion battery pack, liquid thermal control system and software that receives dispatch commands from a solar inverter.”

          1. ChrisEcoSouth Avatar
            ChrisEcoSouth

            To be sure PW details are sketchy, but the details that are there seem to match up with other companies’ approaches to the same thing. The electronics required for any such battery is a DC-DC converter, which will conveniently step down from array voltages of (say) 250V to 450V, and step up to a nominal 360V for the (existing) grid-inverter.
            The present model Fronius and Solar Edge systems do not connect direct to 48V, so filling in the gap says the PW will need a DC-DC converter.
            Batteries can be placed outdoors, but like everything else placed in Australia, it must withstand temperatures ranging between 35 degrees and 42 degrees Celsius for up to 2 weeks at a time – and I mean the temperature does not drop below that even at night!! This is outside the temperature range requirements for warranty of most (if not all) manufacturers of any Lithium technology.
            Another hidden installation cost consideration.

          2. Robert Haylar Avatar
            Robert Haylar

            I see, thanks. That assumes a nominally 48V battery? I was thinking the battery is nominally 400V, because that is what is in Tesla’s car. The 10kWhr unit uses the same cell as the car, the other uses the same format cell, different chemistry. The DC/DC unit would then be a simpler MPPT controller, rather than a full up/down converter. The battery’s BMS would then control both units That seemed to me be the cheapest option for them. It’s perhaps too early to calculate costs, when there is so much missing information.

            As for temperature, you are right. The specification is ‘operates from -20C to +43C’ but no Lithium cell will retain all of its characteristics at room temperature over that range.
            Varying cycle life, capacity and efficiency over that range,

          3. ChrisEcoSouth Avatar
            ChrisEcoSouth

            Yes – if 400V then would have to be DC step-down/step-up from array (with specialised MPPT), as arrays may vary from 200V to 600V(!); and output direct to inverter as you say, and the control has some sort of limiter (if not split-mix with array pass-through) to control over-discharge.

          4. Catalytic Engineering Avatar

            Existing Tesla Model S battery packs (85 kWh model) appear to be 16 modules in series. Estimate about 5 kWh per module. Each module has 6 subsections in series. Each subsection has 74 cells in parallel.

            So for a Li-ion cell voltage of 3.5 – 4.3 VDC (crudely speaking) you get a subsection voltage of also 3.5 – 4.3 VDC, and a module voltage of 21VDC – 26V.

            If the Powerwall is two modules in series you get 42VDC-52VDC which is nice as it meets certain American safety requirements, and it gives about 10 kWh DC. Total ~900 cells.

            Could be a totally reworked design with 100 subsections in series (to give 350-430VDC), and 9 cells in parallel per subsection, but this would be a radically different design from the existing Model S technology.

            http://www.teslamotorsclub.com/showthread.php/34934-Pics-Info-Inside-the-battery-pack

    2. Jacob Avatar
      Jacob

      U$3000 is the final retail price of the 7kwh Powerwall.

  4. Jo Avatar
    Jo

    I think the calculation is a good start while I agree with Warwick that some of the assumptions may be too optimistic.

    I hope that we will also find in this series of articles a calculator like the ATA Sunulator that can calculate for the full year with half hourly interval data for electricity demand. This should give a much clearer picture of the benefits of a battery.

    In addition I would like to encourage all authors to NOT use the payback time as a description of the financial viability of the investment in a system. This is misleading as it makes sure that solar and other investments in renewable energy or energy saving cannot be compared to normal investments. But investing in these things is like any other investment (and may give a warm feeling in addition for doing the right thing). Or has anybody heard about the payback time of a bank account, a term deposit or a share portfolio?

    Using the IRR, as done by UBS, is a much better way of describing the financial outcome in ‘investor terms’. Some people may have difficulties in understanding what the IRR is. However it is actually quite simple. It is exactly the same as what we know as ‘comparison rate’ in home loans, with the only difference, that the investor is the ‘bank’.
    Or in other terms: the IRR is the interest rate that a bank account had to have in order to provide the same financial outcome as the investment that provided the IRR.
    So a comparison rate (IRR) of 9% is a brilliant result for an investment if you compare it to an investment in a bank account or a term deposit!
    ..

    .

    1. juxx0r Avatar
      juxx0r

      Also note that UBS use a 2.5% increase per year in the electricity price.
      And i think the assumptions are a bit optimistic:

      1) i don’t think it’s reasonable to be able to recharge every day of the year to 100%.
      2) It’s not possible for everyone to use their 7kWh during the peak times.

      1. proud kiwi Avatar
        proud kiwi

        Well yes the UBS can be wrong for instance the WA cost of electricity I know we’re not connected to the National Grid has gone up 4.5% this year and will probably continue above the rate of inflation for the foreseeable future making the CBA even more compelling.

      2. Rockne O'Bannon Avatar
        Rockne O’Bannon

        juxxor
        Thank you for using common sense in addition to the numbers. That assumption of “full capacity/ full utilization” should raise alarm bells to anyone who actually has a solar array.

        Nobody should forget that if you are not charging it up all the way, you can’t use it all the way.
        And if you don’t use the power in it, it will be left over in the battery the next day, keeping you from getting the full capacity the next day too.

        I am ecstatic that people are doing the math. But let’s all remember we are in the real world people.

        1. juxx0r Avatar
          juxx0r

          Turns out after i replicated UBS’s numbers, they were only using 5kWh/day

          1. Rockne O'Bannon Avatar
            Rockne O’Bannon

            Yikes.
            You know, one would think with all the hype and excitement that this Powerwall thing would be a slam dunk, but it just isn’t. It seems like so many parameters have to be just right, or the return is very marginal or negative. The model results are very sensitive to parameters, not robust.

            Truth be told, Australia seems to be closest to a “good market” for this because all those FITs are ending, but it just amounts to throwing money into batteries to spite the utilities. Or is it a hedge against higher rates, maybe? Proud kiwi seems to cite that as a benefit.

            I will list the assumptions that pretty well have to be right on for the Powerwall to be a good deal.
            1. Very low cost solar generation. (way below grid parity)
            2. High rates for grid electricity
            3. Full charging every day.
            4. Full discharging every day
            Now how many people make it this far? Not many. And then there are factors that are beyond an owner’s control
            5. Installation and ancillary costs below 7000 (really?)
            6. 4000 cycle lifetime (seems pretty high).
            7. Finance costs
            8. Fees from the utiility (they will have no response?)

            What I have found is that working backward is a good way to evaluate this. At what price would I be willing to buy such a system to get a 10 year payback/breakeven? My number is 7665 US dollars for a complete financed installed 7 kWh 100% efficient system, but I live in Japan. And that is AFTER the FIT period finishes, assuming zero FIT.

            One last thing. As far as increasing electricity rates, one should probably note that even in Japan, since the earthquake, and I live in the quake zone, rates have not gone up 2.5% per year. Even in our extreme case, it has been more like 2% per year. To be entirely fair, if one includes fuel and renewable surcharges, it is more like 4% per year, but those go down as well as up.

          2. juxx0r Avatar
            juxx0r

            1. How is 4.8c/kWh?
            2. 45-50c/kWh, and increasing from 1st July by 4.5%
            3. UBS’s numbers showed you didn’t need this and modelled only 5kWh/day
            4. See point 3.
            5. Doubling the price for installation?
            6. Elon said probably double that
            7. Finance costs are irrelevant when they’re half your return, i.e. 4.5% homeloan rates compared to 9% IRR
            8. Fees from the utility are the problem here, this is part of the solution.

          3. Rockne O'Bannon Avatar
            Rockne O’Bannon

            1,2 That is an amazing differential. I have to admit that if I had a big bunch of solar panels on my roof in Australia, these numbers alone would make me look for something.
            3,4 You absolutely need it. It is a no-brainer. With any high fixed cost, you have to maximize throughput. Buying 7 kW of capacity and then using only 4 kW per day is not going to get you a return on anything but the installation. Also, 3 and 4 are separate. If you charge and don’t use, you lose. If you don’t charge, you can’t use. They are separate and independent risks/inefficiencies.
            5. I have heard installation in Aus is cheap. So “shrug.” No way I could do that in Japan. That is for sure.
            6. Elon says? Haha. Why not just assume unlimited cycles and be done with it? But the guarantee is 10 years, if I recall correctly. I guess if you had a battery that would seriously last forever, then we are wasting time…we could get our money back just by selling someone else this perpetual motion machine!
            7. FInance costs are never irrelevant. They are a riskless cost vs. a risky benefit. Last time I was in Aus, the casinos and banks were in separate buildings.
            8. Well, just for instance, California has threatened before to cut people off the grid if they use offpeak power to charge batteries for on peak use. Batteries are supposed to be used for renewable power only, or PG and E is under no obligation to give any benefits whatsoever.
            It is very romantic to “stick it to the man” by wanting to go off grid, but it is a real life pain. Taxes, fees, etc. are what we pay to participate in society, and if the goal is non-participation, then why even bother with all this numerical mumbo-jumbo?

            You know, getting back to 1,2, all I can say is wow. I hope I never find myself in that situation. I am also pretty sure I will never find myself in that situation. If Tesla cannot make it there, they won’t make it anywhere.

          4. juxx0r Avatar
            juxx0r

            3,4, you absolutely don’t need to because UBS gets you 9% IRR with only 5kWh/day.
            8, All this thing is that UBS has modelled is a UPS, there are currently no regulations on UPS’s, why would that change?

            So UBS gets you 9% on 5kWh, on 7kWh it’s worth 17%, on 5 and 4.5% Escalation which is closer to the truth round here you get 10.7%, with 7kWh and 4.5% you get 18.8% IRR

          5. ChrisEcoSouth Avatar
            ChrisEcoSouth

            UPS – the definition gets murky here. We can define it however we want, but utilities are the arbiter. They can choose to classify it differently if they want to.

          6. Rockne O'Bannon Avatar
            Rockne O’Bannon

            And for whatever reason they want to. Safety. Equity. Think of the children, etc.
            You guys know how this works in Aus better than I do, but it seems like one either trusts the utilities or does not trust them. I don’t hear too much of the former here.

          7. Rockne O'Bannon Avatar
            Rockne O’Bannon

            We disagree.
            I am interested in 3,4 because, in my climate, there is a considerable amount of fluctuation in the yield from my solar. I might reach 100% today, but one day last week, I got only about 20% of the rated capacity. Then one has to consider that the “surplus” from your solar has to be adequate to meet your demand and THEN to charge your battery.
            To me, if you don’t/can’t charge, you miss all benefits for the day. If you don’t/can’t discharge, you miss those benefits too.

            Capacity matters. Smaller systems are more likely to be intensively utilized, but they will also not capture as much benefit. Excess capacity is an albatross 300 days out of the year, giving marginal benefit the other 65.

            So you and UBS don’t care. OK. Fine. I see it as a point of great sensitivity and great risk of raising inefficiencies.

            Regarding 8, “Why would that change…” Look, I would never mean to imply that Australian utilities screw their customers. I am sorry if I gave you that impression.
            Where I live, I have 10 and 20 year contracts. Ironclad. I guess people in Oz invested on a contract of 5 years or so with utilities? Here, we are looking at falling prices for electricity. Slowly at first, but if nuclear power plants come back online, rates might really drop.
            So this “assumption of escalation” driving the analysis is a suspect situation for me. 4.5% per year? That is realistic? That sounds catastrophic, and I am a guy who lived through a 9.0 earthquake. With coal and gas prices tanking, people in Australia think they will be paying DOUBLE for electricity in 15 years? Given enough fear, people will buy anything, but… really?

          8. juxx0r Avatar
            juxx0r

            We are paying double what we paid 15 years ago.

            The increases the few years before were more like 7.5%

          9. Rockne O'Bannon Avatar
            Rockne O’Bannon

            Respect.
            If you expect that to continue, don’t even bother with the analysis. It is a no-brainer. Buy the thing and hang on tight.

            In truly high inflation scenarios, bean counting goes out the window. In my lifetime, I have seen 20% inflation rates (where a 20% investment gets you zero real return) and negative inflation rates (where holding cash is better than riskless). If you expect your dollars to be worth that little in the future, go spend em.

            Good luck. I thought this was a technological or cost/benefit issue. If it is an hyper-inflation hedge issue, this discussion is meaningless.

            You made my day. I am going to enjoy Mad Max a lot more. I didn’t know it was a documentary. (kidding)

          10. juxx0r Avatar
            juxx0r

            It’s a badly managed grid issue, failure to recognise change issue, failure to innovate issue, Diesel power is good issue, the people have no other options issue.

            Stick around here and enjoy the show.

            This powerwall isn’t even the real game. If the 100kWh units that tesla sell for $250/kWh were used to take whole blocks off the grid, we’re talking ANOTHER 30% cheaper. That’s solar at 5c/kWh, storage at 10c/kWh, Gas backup for 20c/kWh and the grid is still charging at it’s cheapest 25c/kWh plus plus.

            So long and thanks for all the pineapples.

          11. Rockne O'Bannon Avatar
            Rockne O’Bannon

            Holy smokes. Given reasonably low fixed costs and high expected inflation and grid collapse….

            So… this is not a consumer product, it is infrastructural reform. Certainly, Australia is out there exploring new frontiers. Very popcorn worthy happenings afoot.

          12. Catalytic Engineering Avatar

            See some analysis here on timing (and capacity) of charging and discharging – it’s a model developed in Germany for household consumption.

            http://loadprofilegenerator.de/energy-arbitrage-potenial-of-a-7-kwh-battery-for-an-average-german-household/

          13. Rockne O'Bannon Avatar
            Rockne O’Bannon

            Thanks for that. The algorithm in paragraph 4 at that site really says it all.

            I have done my own simulations, and that is always what I come up with. In a slightly more complicated scheme I have, I use “rough weather predictions for the next day” which improves things somewhat. Any rough and ready forecast will give some indication of how much sunlight is coming down the pipe.

            However,
            Whether a person can actually hold to that algorithm and switch things around properly with sufficient discipline is questionable.
            Whether a reliable automatic system can be contrived, and whether regulations by a utility will allow that is questionable.
            etc.

            This is a weird time. Ratepayers are fighting a turf battle with utilities. The latter might fight tooth and nail to prevent turnkey solutions for consumers.

          14. Catalytic Engineering Avatar

            I was interested in these numbers as I had done something similar in the German case (much smaller split between feed-in-tariff and purchased pricing, but I assumed 365 cycles / day and 7 kWh DC out)

            How did you replicate the UBS numbers? I could do it if I used the following additional assumptions (following your comments)

            2.5% price escalator per year
            365 cycles per year (not 400)
            4.3 kWh AC discharged per cycle
            0.45 cents per kWh profit (0.51 minus 0.06)

            The 4.3 kWh AC discharged could be thought of as follows:

            4.85 kWh DC out of the solar panels and into the battery
            4.46 kWh DC coming out of the battery after 92% DC-DC round trip battery and power electronics efficiency
            4.3 kWh AC after one-way inverter effficiency of 97%.

            I agree with your finding that the cycling energy is much lower than 7 kWh but maybe we’re making a different assumption from UBS somewhere else?

          15. juxx0r Avatar
            juxx0r

            I think that they just scaled it down to get a real world average.
            The numbers look even better when you use the 4.5% escalation that was just announced round these parts for this years price rise.

    2. ChrisEcoSouth Avatar
      ChrisEcoSouth

      I’d love to use IRR to sell these, but at the end of a bank investment term (for arguments sake 15 years), you get to withdraw your principal, but at the end of the 15-year life of the Pwall, you get to throw it in the bin (depending on scrap value). I’d really like this analysis to work better, though!

      1. juxx0r Avatar
        juxx0r

        At the end of the 9% IRR you’ve already withdrawn your principal.

        So you’ve made 9%, you’ve got your money back along the way and you’ve still got a product good for 80%. Surely that beats a bank?

        1. ChrisEcoSouth Avatar
          ChrisEcoSouth

          I ran some calcs so I can see the IRR includes the capital now, thanks. But we are still assuming the PWall is an all-in-one box with bi-directional inverter, and has usable capacity after 15 years (I’ve seen optimistic numbers on batteries before). Margins have not been added for the product! That alone skews the result. I’m making it out closer to break-even at 15years with 2.5% IRR/bankrate, which is still nice, but not exactly setting the world alight.

          1. juxx0r Avatar
            juxx0r

            I can match their numbers exactly by using only 5kWh and a 2.5% escalation rate over only 10 years.

            I don’t know who is assuming a bidirectional inverter, certainly UBS isn’t otherwise they’d use more than 5kWh a day.

            It’s warrantied for 10 years, it returns twice what your homeloan costs you, who cares if it falls over in year 11?

            Incidentally there’s more information on the tesla website now:

            http://www.teslamotors.com/powerwall

          2. ChrisEcoSouth Avatar
            ChrisEcoSouth

            Ta – I checked the website: Its the DC/DC converter that is bidirectional, and presumably can ramp both its inputting and outputting up and down. If so, it will connect to almost any existing grid inverter. Being $US3k retail makes the numbers work more as you say; overseas shipping, and additional isolator etc, will drive that up further. Its also still assuming the PW can be placed next to the existing inverter – most won’t be able to, due to direct sun at that spot.
            Website still foggy – talks of power in a power-cut, and “inverter not included”, but doesn’t describe anything further. Looks like the power-cut capability will require a Tesla-dedicated appliance inverter added on – this will be restricted by the rate of discharge limitation on the PW.

          3. Miles Harding Avatar
            Miles Harding

            The DC-DC converter would explain a few things like
            the limited power handling capacity and
            how they protect the battery from external silliness.

            We may need to wait until the product condenses from it’s current vaporous state and we can get some technical documentation.

      2. Jo Avatar
        Jo

        This is not true!
        at the end of the 15 years you do indeed get no money back. But you have received higher payments during the 15 years to compensate for that.

        1. ChrisEcoSouth Avatar
          ChrisEcoSouth

          Yes indeed. Through more discussion here, I have rescinded my misunderstanding of IRR. Apologies.

    3. Giles Avatar

      We have already published a calculator here – where you can vary your assumptions. Download it from here.

      https://reneweconomy.wpengine.com/2015/what-the-tesla-powerwall-battery-means-for-households-61055

      1. Jo Avatar
        Jo

        This calculator is a good start.But it is a calculator for just one (average) day of the year.

        1. Craig Memery Avatar
          Craig Memery

          Jo, we’ll be releasing a battery solar version of Sunulator soon. Email if you’d like a go with the beta version. Cheers, Craig

    4. Rockne O'Bannon Avatar
      Rockne O’Bannon

      Although I would not use the payback for most investments, it is interesting here, in my opinion, because of that little cycle number: 4000.

      If you are charging and discharging this thing every single day, you are going to be done with it in 11 years. As far as I am concerned, if I don’t get my money back before then, the investment is a loser. I would have to put more money in to get all of my initial money out.

      I will say frankly that I have been pricing similar systems for several years now, and even with free and clear solar power, I can’t find one that is worth it. Charging a car with a 10 kWh or 30 kWh capacity makes more sense.

    5. Rockne O'Bannon Avatar
      Rockne O’Bannon

      Brilliant? A bank account or term deposit has less risk, don’t you think?

      Finance professors will tell you that IRR falls apart if you go from positive cash flows to negative during the lifetime of an investment. NPV would be better AND it would let you incorporate capital costs better than IRR.

      Payback or breakeven is not as useful as NPV, but it is robust, and let’s face it, with this apparatus, if something goes wrong, then wanting to know “when I get my money back” becomes the big question.

      1. Jo Avatar
        Jo

        I don’t think that a term deposit has less risk than investment in a battery if good warranties cover all risks. Actually a long term bank deposit has quite a high inflation risk. If inflation goes up, the income from a bank deposit does not rise. However the savings from an investment in renewable energy will improve with inflation because the savings will increase if the mains electricity cost rises with inflation.

        No finance professor has ever told me that IRR falls apart. I don’t even understand what ‘falls apart’ means in this context.Numbers do not fall apart.
        IRR does incorporate capital cost. Again I do not understand what ‘better’ means.

        As I described before ‘the if something goes wrong’ risk needs to be covered by warranties.

        .

        1. Rockne O'Bannon Avatar
          Rockne O’Bannon

          You cannot use IRR if you have a negative cash flow beyond the first investment. So if there is some failure not covered by warranty of any component, and that failure is sufficient to turn a cash flow negative during the period you are examining, then IRR will give a spurious result. I am sorry if a finance professor never told you that. I think there is a way to adjust it, but you have to use shorter periods to do it.

          I am decidedly NOT criticizing your result. But if an inverter only lasts five years, say, and if there were no warranty, the IRR result would have to be tossed.

          I use NPV if I am serious about evaluation. It solves all the problems. It accommodates all cash flow structures. One can also use it to compare one plan vs. your next best alternative investment. You can incorporate expected inflation, different durations of investment, negative cash flows in intervening periods, etc. .

          “I don’t think that a term deposit has less risk than investment in a battery if good warranties cover all risks.” Most people would disagree that a term deposit in an Australian bank => warranty from Tesla. If you or anyone wants to put money in the Bank of Tesla, go ahead. Keep in mind that this company does not even use GAAP accounting. If the Australian banking system and its insurers are less reliable than that, you are entirely justified. Many many American companies have recently refused to honor legitimate claims of gift cards, even in bankruptcy courts. It’s a risk.

          You give the case of inflation rising. OK. Sure. But one must also concede that if the price for electricity falls or even if regulations change, your Tesla battery will be less useful to you than a time deposit. That is how risk works. Personally, I price solar and storage systems as options, not as bonds. Options can be “out of the money” just like renewables can be. They therefore DEMAND a higher risk premium than bonds. “Free money” does not exist.

          Which is the other problem with IRR. Nobody uses it to compare investments that are not “like” investments because they entail different risks and assumptions, as in this case. Even given different portfolios or risk profiles or taxation, people will rationally make different choices. A 9% IRR with certain risks might not be preferred by someone who is willing to forego the returns and risks entirely. In fact, most Australians will make the latter choice.

          That’s it. I won’t change your mind, I know, but that is how I see it.

          1. juxx0r Avatar
            juxx0r

            Tesla does use GAAP and also non GAAP but because GAAP doesn’t fit the current phase of it’s business well enough to accurately describe what is going on, they quote both. You can’t have a go at them for that.

            In Europe you can get negative interest rates, which is literally free money, or from the other side your term deposit costs you.

            I’d take the 9% and i’d bet my left nut that i’d end up with more than that thanks to the management of our utilities.

          2. Rockne O'Bannon Avatar
            Rockne O’Bannon

            I’m not having a go at anyone. If someone says to me that a certificate of deposit at a bank is “less than or equal to” a warranty from Tesla, I am going to call that dumb. Being an American, I know how things come and go. You know, JC Penney and Sears, names which you have probably not heard of, used to be iconic, and now they don’t pay their bills. GM almost went toes up less than a decade ago. Enron had everybody fooled. If we are going to call Tesla a company and not some manifestation of Jesus Christ, we have to be realistic. OK. I have to be realistic. Others can go on believing whatever they want. Hope springs eternal.

            As a matter of fact, I have lived with negative interest rates. And no, it is not free money because you still need to borrow it. That is controlled by people who will charge you a fee. I will say again that there is no such thing as “free money.”

            Based on our other exchanges, and given such high expectations of inflation, I would probably buy a Tesla battery for use in Oz even if it did not give a 9% IRR. As I said above, I have no particular problem with that result per se.

            I am in the market for storage. I am keeping my eyes open. Looks like this might be a good thing for Aussie conditions. The math does not work for me yet, but it might someday.

          3. Jo Avatar
            Jo

            I cannot see a mathematical reason why an occasional negative cash flow would create a ‘ spurious’ (=false?) result for IRR calculation.

            Besides NPV and IRR are intrinsically linked: IRR is the interest rate (in NPV terms = discount rate ) at which the NPV of an investment is zero.
            So instead of using the IRR formula in Excel, one could also use the NPV formula and use the solver to find the discount rate for a NPV of zero. Would that be spurious as well?

            The main reason why I recommend using the comparison rate rather than NPV is that most people would not have a clue what NPV is, but they most likely know about the concept of comparison rate (=IRR) in home loans. The formula for the comparison rate is even written into the law of NSW: ( http://www.austlii.edu.au/au/legis/cth/consol_reg/nccpr2010486/s100.html ).

            This concept can be easily understood: In simple terms the comparison rate for an investment is the interest rate of a term deposit that with the same initial investment and duration creates the same financial result (NPV) as the investment.

            I agree that investments should not only be judged by IRR or NPV but also by their risk profile and also bey their liquidity. But that is not in contradiction to my point that we should make investments in renewable energy comparable to other investments instead of avoiding the comparison by just using the payback time, which some ‘specialist; even call ROI.

            Coming back to the Tesla power wall. It will have a warranty of 10 years which will cover most likely everything including the battery and a DC/DC inverter. So unless the company goes broke there financial risk by a defect inverter after 5 years.

          4. Rockne O'Bannon Avatar
            Rockne O’Bannon

            OK Jo. Have at it. This is not rocket science. “The concept can be easily understood…” by having a look at Wikipedia.

            Here is a casually found link to an explanation that is actually biased to our side. It seems to be comprehensive, explains WHY many finance professors advise not to use IRR, but this guy is biased in favor of it.

            http://members.tripod.com/~Ray_Martin/DCF/nr7aa003.html

            If you want a CFO/accountant to explain it to you, then here:

            http://www.cfo.com/printable/article.cfm/3304945

            And Wikipedia offers this. Please scroll down the the PROBLEMS USING IRR. There are…. several.

            http://en.wikipedia.org/wiki/Internal_rate_of_return

            As I said earlier, this is standard stuff. If you have never had a finance professor advise you of these problems, then I guess you are not hanging out in the right places. I did problem set after problem set designed to give the wrong answer if IRR was used. One of the things I like least about irr is that even after you are all done, you have to start all over again with different projects, durations, etc. If you use NPV, you can simply recalculate and compare. Or calculate separately and compare.

            But all that aside, it now appears to me that if electricity rates are going up by leaps and bounds in Oz, all of this “bean counting” is not very useful anyway. The results get swamped by inflationary expectations. 9% IRR? If rates go up as they have in western Australia the last 10 years, a 0% IRR would still be a winner.

            You know, I like to watch old movies, and I saw this exchange in one of them just today. I had to grin.

            Guy 1: This check is as good as gold.

            Guy 2: Gold is as good as gold. Nothing else is.

            See, it is common knowledge and has been for at least a century. And I think that applies just as well to any promise made by any vendor, including Tesla. A term deposit has numerous public safeguards and insurance backing it. Saying … let me see… your words were…. “I don’t think that a term deposit has less risk than investment in a battery if good warranties cover all risks.”… well, that is already heavily qualified as it is because I have never seen a warranty cover ALL RISKS, even inflation and deflation. But even so, it is a tall statement to make. I have cited numerous cases where large companies have defaulted on warranties, but I believe that no US depositor has lost either principal or interest from an FDIC insured bank account or financial instrument, ever. EVAR.

            Maybe you intended hyperbole. Maybe Australia has a worse banking system than Thailand. Maybe you expect hyperinflation. Sure. But the statement seemed really out there, so I said what I felt needed saying.

          5. Jo Avatar
            Jo

            Interesting, that you provided three links that prove my point. The Wikipedia statement “The IRR is the annual interest rate of the fixed rate account (like a somewhat idealized savings account) which, when subjected to the same deposits and withdrawals as the actual investment, has the same ending balance as the actual investment.” is pretty much the same as the explanation that I provided in my former post.

            My discussion has not been about major investments to deal with ones lives savings but just about improving comparability of investments in renewable energy.

            But haggling over NPV versus IRR is missing the point. My initial post was arguing against the ‘habit’ of using payback time as an investment description for renewable energy projects.

            Payback time misses to describe the gains after the break even and in its simple form it does not take the time value of money into account. Using payback time as the sole measure (as it is often done in Australia) keeps renewable energy projects incomparable to other investments.

          6. Rockne O'Bannon Avatar
            Rockne O’Bannon

            First of all. Back up. I told you I did not question your present analysis results, but told you of problems with IRR. I explained payback as a quick and dirty method, but said I defaulted to NPV instead of using IRR, which has problems. You got huffy and told me you have never heard of problems with IRR. If you have never heard of these… If no finance professor has told you of them, then you need to learn some more. So I sent the links, which are excellent.

            And you claim they prove you right? They prove your ignorance, JO, flat out.

            The Wikipedia article explains, what… 7? 9? common weaknesses and criticisms of IRR? I certainly proved my point. It took less than five minutes to find all three links by the way. The first link above was a conciliatory softball for you. The second link above shows how ridiculous people who trust IRR truly are. The third link above is simply common knowledge.I think each one is very good. The last two are excellent, with very clear explanations of IRR weaknesses.

            Couldn’t take it gracefully, though. They prove you right? Allright JO, here are some quotes you might have missed. If you want to use non-rigorous, substandard, misleading methods, go ahead, but see what CFO.com says. They call IRR analysis foolish and risky. Are you a CFO or accountant JO? Do these quotes prove you “right” when you say you have never heard of problems with IRR?

            “Maybe finance managers just enjoy living on the edge. What else would explain their weakness for using the internal rate of return (IRR) to assess capital projects? ”

            and

            “For decades (this was written in 2004!), finance textbooks and academics have warned that typical IRR calculations build in reinvestment assumptions that make bad projects look better and good ones look great. Yet as recently as 1999, academic research found that three-quarters of CFOs always or almost always use IRR when evaluating capital projects.”

            and

            “Our own research underlined this proclivity to risky behavior (of using IRR at all). In an informal survey of 30 executives at corporations, hedge funds, and venture capital firms, we found only 6 who were fully aware of IRR’s most critical deficiencies. ”

            And the Wikipedia article, although mentioning numerous weaknesses of IRR, particularly mentions one I brought up.

            “In the case of positive cash flows followed by negative ones and then by positive ones (for example, + + − − − +) the IRR may have multiple values.”

            I am going to stick to payback to sniff at a bone, and I won’t chew on it until I do NPV.

            I consider IRR to be a waste of time. Not worth the data entry.

  5. Brad Sherman Avatar
    Brad Sherman

    Can the more knowledgeable readers laugh test this for me.

    From what I’ve seen advertised I could install a 5kW PV system for $10k without subsidy. My existing 5kW system produces more than my house consumes (I average 11 kWh per day). Put in 3 x 10 kWh Powerwall systems (let’s call it $12k) and I’m guessing I have enough current to run my oven and other appliances and can weather 2 days with no sun.

    To my way of thinking this seems close to a viable off-grid system (maybe I need another 20 kWh backup as a hedge). My current Actew power costs me $0.22/kWh (100% green power) and my connection charge adds another $0.082 / kWh at my consumption level.

    If I was starting from a clear block with no grid connection, I’m guessing I should deduct, say, $5k just to run the wire to the street and connect to the grid plus whatever the meter costs. I’m going to call the net cost of PV + 30 kWh storage as $22k – 5k = $17k. If it lasts 10 years I make that $0.42 / kWh. This compares to my current rate of $0.30/kWh. On current trends my Actew power will cost $0.38/kWh in another 10 years.

    If subsidies for the PV are $5k or connecting a block to the mains costs $10k rather than $5k then the cost of the PV + Powerwall comes to $0.30, i.e. the same as I’m paying today for green power. Imagine an entire new suburb going 100% PV + PowerWall with no need for the government to run the grid out to it. Presumably this would open up more options regarding future development opportunities.

    I’m not really interested in IRR calculations (but as an investor in other domains I appreciate them) because I place a high value on self-sufficiency and reducing my vulnerability to the whims of the market. I’m keen to disrupt the status quo so I’d use the logic outlined above plus the feel good factor of offering the finger to the ‘man’ to justify off-grid systems in Canberra today.

    Assuming I haven’t made an outrageous error in my assumptions here.

    1. David Osmond Avatar
      David Osmond

      Hi Brad, just a couple of comments. If you wanted to go off-grid, you’d need to work with the 7kWh Powerwall which is designed for daily cycling, rather than the 10kWh version.

      Secondly, I don’t know where you live, but here in Canberra, my worst week of the year for solar generation is about 1/4 my average generation. So if the same applied to you, for your worst week, your 5kW PV system may generate on average only about 5 kWh/day, so over the week you would have about a 42 kWh deficit (assuming your winter consumption also averaged 11 kWh/day). And being winter, its extremely likely the next week might also be significantly under average.

      To reduce the amount of required storage, you’d probably want to add some extra PV, and probably also get a back-up generator, that you will hopefully only have to use very occasionally during the worst weeks of winter.

      1. Rockne O'Bannon Avatar
        Rockne O’Bannon

        My worst days go to about zero. And that can last for several days. But, well… that is Japan so….

  6. Catalytic Engineering Avatar

    Interesting discussion here!

    We did a similar calculation for Germany but were much more conservative than UBS. We assumed a full 7 kWh (DC) of cycling per day, US$ 0.31/kWh avoided purchased electricity price, and US$0.12/kWh sell back to the grid. It just about earns 2% IRR.

    One concern I have with the UBS analysis is the baseline case. The way their model is structured, any excess electricity from the solar panel is stored in the battery, so that you don’t sell it to the grid at A$0.06/KWh. But if you could consume the solar energy yourself (in the case where you don’t buy the battery), you would effectively avoid buying from the grid at A$0.51/kWh.

    In other words, that 5 kWh DC is only profit if it exceeds the home consumption you would have had at the time the solar power was generated. i.e. this assumes is a rather oversized solar system.

    Conversely, you have to consume all the 5 kWh energy you stored in the battery, otherwise you don’t “earn” that avoided cost either. What’s average Australian energy usage in kWh/day?

    Bruce

    [edit – misread the price]

    1. Catalytic Engineering Avatar

      (It’s actually lower than 2% if you subtract the baseline case, as UBS did)

      Here’s a link to our article

      http://www.catalyticengineering.com/the-berlin-powerwall/

      1. Robert Haylar Avatar
        Robert Haylar

        I think you may be the author of ‘Ten Top Facts about Tesla’s Powerwall’?
        If so, why do you think the battery is two car modules, rather than a 400V cell stack? I would have expected the necessary DC/DC conversion to be too expensive.
        Efficiency will be worse than indicated if you are correct, because the advertised figure will be contrived to be peak efficiency.

        You are doing a good job. It’s refreshing to see that Australians are more skeptical than Tesla’s Elon Musk fan club.

        I don’t like this rather second-rate battery, nor its intended purpose, which is to use homeowner’s solar energy to redirect money that should rightfully go to the homeowner or the grid, into Tesla’s pocket.
        UBS are being quite dishonest in their claims.

        1. Catalytic Engineering Avatar

          Yes, that’s us.

          A 400V string of batteries would be very sensitive to single point cell failures. Having more cells in parallel would be more robust (you would lose only a small amount of power and energy and could keep running). And using automotive packs would be a sensible way of leveraging existing product development to achieve the price point.

          The Powerwall spec sheet mentions it has a DC-DC converter, which should be able to handle boosting from ~50VDC on the battery side up to ~350VDC on the inverter side.

          The 92% round-trip DC-DC efficiency figure appears to include both those power electronics and the battery chemistry itself, but you’re right that we don’t know what the efficiency is at low power, or end of discharge, etc.

          Really looking forward to more real-world data …

          1. Robert Haylar Avatar
            Robert Haylar

            Yes, from Tesla’s point of view, using what they have makes sense, but the boost converter would operate at a 7:1 ratio, meaning large input currents, so compromised efficiency. Like the choice of cell itself, the product suits Tesla more than it suits the customer. Not sure that single-point failure is relevant. If one cell fails, the others see increased demand.

            Like you, I would like to see one taken apart and tested.

          2. ChrisEcoSouth Avatar
            ChrisEcoSouth

            Larger input currents may well tend to be limited by the converter itself, which in turn is selected for that limit, which saves the lithium material from being over-current itself, thus reducing lifetime stress. I’m seeing this on other manufacturers offerings, and means that this category is primarily for grid-FiT arbitrage, and not driving appliances independently. Ideally, lower internal resistance lithium-tech (ie runs cooler) would need to be used for stand-alone applications.

          3. Robert Haylar Avatar
            Robert Haylar

            No matter how the cells are arranged (50V, 400V) each cell will see the same current for any given output.
            There are more parallel cells in a 50V battery, but each parallel cell string sees 8 times the current of a 400V battery.
            From that point of view, cells are equally stressed in each case.
            Conversion from 50V>400VDC>Inverter is not efficient.

            When adding a battery to an existing 400V PV/inverter system, then the conversion in the battery makes sense, but not if there is a choice of 50V?

          4. Robert Haylar Avatar
            Robert Haylar

            I meant that at 50V there is 8 times the current of 400V, but at 50V there will be 8 parallel cells/strings to share the output current.

  7. John Leslie Thirgood Avatar
    John Leslie Thirgood

    Tasmanian power rates just under 25 cents kw hr lights and power then . 17 cents kw hr hot water and heating.and off peak near 12 cents kw (Domestic rates) Basing the rate of power at 51 cents for peak periods only in other states where applicable distorts the numbers dramatically. However I can say that a 5 kw system in Tassie with typical 68% feed-in rate at 6 cents basically halves the annual value , therefore if the Powerwall utilizes all the feed in then it could add $1000 annually in returns. remembering that the average tassie bill for T31 sits between $1200 & some $1600 for power consumption. Hot water and Heating power is on separate metering and not currently able to use solar generation. one would need to forgo the cheaper rate tariff to then combine all to the dearer t31 25 cent rate ! , to be able to take advantage of the solar /tesla system. Then consider that in summer DEC one makes some 12% of annual solar generation and again in Jan, then near 60% of the years solar gain in summer half of year, June July , Aug typically sees under 15% of annual solar generation combined, yet we typically use peak power in the worst solar months. so weighing this into the figures one can not store the summer excess for 4 ~6 Months later thus feed in will still occur and if one swaps hot water and/ or heating as well to the higher tariff so as to take advantage of your own generation, unfortunately in winter one will be paying some 30% more on shortfalls from storage that must then kill off any advantage.
    However in QLD say running a.c. will assist the Tesla battery option. but most would wait until the 10 ~15 year 40c ~60 cents feed in contracts run out until even thinking about batteries.
    So lots to consider here, current energy consumption that is valid to offset from storage, seasonal use variations, current feed in contract rates etc.

  8. Rockne O'Bannon Avatar
    Rockne O’Bannon

    General mea culpa to everyone.
    I have been nitpicking on the calculations, but it was shown to me that ratepayers in Australia are facing rapidly increasing rates. I looked around and found that rates in Western Australia have gone up 77% since 2008 (!).

    Certainly, if people expect this to continue, or if they expect even half such increases in the next 10 years, the analysis is almost meaningless, i.e. a waste of time. Given those expectations of inflation and the costs shown in the article, buying the battery system is a no-brainer.

    An IRR of 9% or even 1% is trivial compared to inflation of 50-70%. A loan today will plummet in value over ten years. The savings on the electric bill will increase every year.

    And that drives the bottom line.

  9. 573v3 Avatar
    573v3

    Where is the story? Tesla’s Powerwall is not novel. Ergon in Cairns is testing battery systems, and i understand they have several different models in comparison…none of which are Tesla.

    Who saw the ABC’s 7.30 Report last night?

    There are several lithium battery systems already available in Australia… That’s the real story, it is happening in Australia already without Tesla… though i am not surprised by how much hype has been purchased. It is a pity it has been wasted on something that is not available and will not be available until 2017, 2016 if you’re lucky.

    If the Tesla battery AS4777 compliant? I do not think so and If it is not then it cannot be connected to the grid.

    Tesla’s indicated costs are mis-representative. Tesla says that they have the 7kWh battery for daily cycling and the 10kWh as a back-up. Does this mean that the two are intended to be used in tandem? Also, (sorry, but i am no electrician) if the US grid is 110v and V=IR & P=VI, then is the kWh’s available halved for 240v infrastructure?

    Good Lithium batteries have a Depth of Discharge of about 80% and design life of about 6000 cycles… or nearly 20 years. Much better than lead-acid in both respects. So a 10kWh system would have 8kWh available

    To install battery back-up to the grid, the electrician needs to be both DC & AC certified. There are not so many of these and they seem to be quite busy already installing grid-connect solar systems.

    Apart from a suitable inverter, what of the control systems? How does the system determine when to use/store Solar or battery or grid? I think you will find Tesla’s Powerwall is not much more than a lithium battery. It reminds of a large cordless drill battery…no more.

    I have seen the BYD system available from EnergyHQ. That was shown last night on the ABC’s 7.30 Report… well a glimpse at least. The BYD is a 10kW LiFePO4 (Lithium Ferrous Phosphate) 8kWh usable battery with 3.3kW inverter and all the control systems and electronics fully integrated into a unit the size of a medium sized fridge. This system knows how to simultaneously draw from the battery and grid at the same time, if the load exceeds 3.3kW. BYD’s system is a fairly simple install. The price of the BYD is more like AU$15,000. Nothing like 11% IRR… but then, neither is Tesla’s.

    Cost/benefit of a solar system is determined by how much can be utilised and/or sold. I’m in Queensland, so will use local data for this… a $10,000 5kW solar system is designed to return about 184MW during it’s warranty life-cycle.

    $10k / 184k = $0.055 / kWh, without STCs. That is five and a half cents per kilowatt hour. Of course the full value of this is dependent upon how much solar power can be utilised. Sufficient to say it’s much cheaper than grid power already. Solar Grid Parity? Get used to it!

    Why not do a story of what is available in Australia… now. AS4777 grid compliant. Tell us how they are working, what they’ve cost and how much benefit can be achieved? Go ask Ergon in Queensland, or IT Power in Canberra…

  10. Steve Avatar
    Steve

    $400 install, you can at least double or even triple that, for a start it weighs in at around 100 kg, it requires an electrical CES, electrical inspection, probably a check by the distributor, and so on, no one has seen one in the flesh yet and not even a spec sheet or data sheet available, the Tesla shop in Melbourne is a stand at a shopping center, the 1800 number has already been changed or disconnected, try getting info on the battery it is impossible, this is a start but this is not going to be the end game, and it will leave a lot of people disappointed, I have been in this industry longer than most and the hype factor for this battery is like a drug, people are scrambling to get a piece of it and it doesn’t even exist yet, wait and see peeps, don’t get caught up in the BS, we simply have no hard facts on this battery yet and the fact we cant even get a spec sheet is a big worry.

  11. Rockne O'Bannon Avatar
    Rockne O’Bannon

    Saw mention of this in the Australian media. Maybe this needs to be added into calculations?
    “What might end up happening is that regulators will demand that customers leaving the grid pay some kind of disconnection fee (e.g. $1,000) as compensation to the other users.”

    I wonder if people will pay it and leave or wind up grudgingly remaining hooked up to the grid to pay the base fee every month.

  12. Sam Avatar
    Sam

    Is there a link to the original UBS report?

  13. Kurt Alexander Avatar
    Kurt Alexander

    One thing I have not seen below is the facts of what we are currently getting.
    I am currently paying 25.378 cents per kWh for tariff 11 (Domestic use).
    I am currently paying 18.454 cents per kWh for tariff 33 (Off Peak use).
    I currently get 44.000 vents per kWh for my solar feed back to the grid.
    So in my mind, if I am storing this and not getting the 44 cents per kWh, only to be able to use it later, then I am only getting at the most 25.378 cents per kWh for my solar, instead of what I currently get at 44 cents per kWh.
    Why would I do that just to be able to store 7 – 10 Kwh?? I am actually losing 18 to 25 cents per kWh by storing it.

  14. warren carr Avatar
    warren carr

    Zealand
    Everybody so far has missed 2 points about the Powerwall.
    It has a standard 10 year warranty,BUT THE WARRANTY CAN BE EXTENDED TO 20 YEARS.
    Yes this will cost extra and the cost is not been published yet that i can find.
    The second point is a game changer in that the Powerwall comes with software installed that lets the owner of the system sell,trade the excess stored power on the spot power market.
    Called Reposit Power.
    So instead of getting 0.6 cents Kw you can sell your power for up to $13 Kw plus in time of high grid demand.
    This system is already being trialed in some parts of Australia by home system owners using the Magellan Power Res1 storage unit.
    This unit is at the moment expensive.
    So it is impossible to calculate the payback time with this new variable thrown in.
    But like mobile phones this is a game changer.

  15. Oz_Mark Avatar
    Oz_Mark

    What utter nonsense – it assumes the existence of a large PV (whose IRR it destroys) and then assumes 51c/KWh at night? – only numerically or financially illierate could fall for it (many of the other assumptions are optimistic too – you’ll be lucky to get 85% charging efficiency and 6.5KWh out of it)

  16. Lawrence Coomber Avatar
    Lawrence Coomber

    Like all “wow factor” commercial products, the least accurate assessments about its commercial viability and realistic practical usefulness will be those comments offered by anybody with a direct or indirect commercial interest in the product. At best, commentary will be vague and void of corroborated factual statistics. The Tesla Powerwall fits neatly into this category of product.

    Any grid connected energy product should be visualised as operating in a joint business partnership with “the grid electricity supplier” who is also the senior and most dependable partner in the overall power plant. The “grid electricity supplier” is the one who effectively assumes all of the risk and responsibility to ensure supply to the customer.

    This distorted scenario, presents as yet unknown financial and performance complications and risks to the three stakeholders involved in the power plant; “the system owner”; the “Tesla Power Wall system provider”; and “the grid electricity supplier”. The inevitable risks inherent with these triumvirate business arrangements are well known by experienced renewable energy engineers with a keen understanding of the business imperatives that each stakeholder is both responsible for, and beholden to [now and into the future].

    Battery storage systems in on grid scenarios [energy plant business partnerships] are not immune from negative business outcomes, easily predictable in advance – but surprisingly not yet a point of discussion in most technical forums. There is much more analysis yet to be done before this technology format can be considered useful and practical. Off grid systems are an entirely different animal though.

    Lawrence Coomber

Get up to 3 quotes from pre-vetted solar (and battery) installers.