Does Battery Storage Help Or Hurt The Environment?

an eco-warrior dreaming about a powerwall

If you want to save the environment and have limited funds, then buying batteries should not be a priority.

Millions of Australians have installed rooftop solar to help the environment and save money.

Many are hoping home battery storage will soon let them do the same.

The cost of home energy storage is falling rapidly and it may not be long before it pays for itself.  But at its current price it will not save money, even under favourable circumstances, when used on-grid.  Despite this, there have been hundreds, or possibly thousands, of people who have installed home battery systems because they want to help the environment.

Have these people  made a horrible mistake?

In Australia, with one main exception, on-grid home energy storage does not reduce greenhouse gas emissions but instead contributes to them.  While battery storage is likely to reduce emissions in the future, systems installed now are almost certain to only contribute to global warming over their entire lifespan.  Early adopters would have helped rather than harm the environment if they had instead spent their money on installing larger solar systems or on improving energy efficiency.

Manufacturing And Using Batteries Results In CO2 Emissions

Home battery storage contributes to increased emissions in two ways.  Firstly, manufacturing batteries results in CO2 being released.  And secondly, home storage results in clean solar energy being lost because no battery is 100% efficient at storing electricity.  More electrical energy has to be put in than can be taken out.  If that energy had been sent directly into the grid it would have decreased fossil fuel generation.

Embodied Energy Of Batteries

The world is still heavily dependent on fossil fuels and almost everything you buy has a significant carbon footprint from CO2 emitted during its production and transportation.  Batteries are no exception. Emissions from manufacturing will decline as the world uses more renewable energy and recycled material, but currently making batteries  releases considerable CO2. This is often referred to as a battery’s embodied energy.

How much CO2 is released manufacturing a typical battery system is difficult to determine, as it depends on the type of battery, the materials used, how they were produced, how far they were transported, and how much recycled material was used.

One very simple rule of thumb is that the average manufactured good, containing an average amount of recycled material, will have a carbon footprint of around 2.  That is, for every kilogram of an item’s weight, 2 kilograms of CO2 were emitted producing it.  This is only an approximation, but is surprisingly accurate for a wide range of manufactured goods.

However, for one of the most common type of new batteries, lithium-ion, the carbon footprint may be much higher.  One study found the carbon footprint to be 173kg per kilowatt-hour of storage.  For the battery pack they looked at, 18kg of CO2 were emitted per kilogram of weight.

That study was from 2013.  Because battery technology is advancing so quickly I expect the carbon footprint of lithium-ion batteries to be considerably lower now.  But because I don’t have any solid information, I will simply guess it has fallen by 13% to give a nice, round number: 150kg of CO2 per kilowatt-hour of lithium-ion battery storage. Although I suspect the decrease would actually be greater than this.

Powerwall, LG Chem RESU, And ZCell Carbon Footprints

The most famous home battery system is the Tesla Powerwall which weighs 100kg.  The less well known, but well regarded, LG Chem Resu weighs 60kg.  They are both lithium-ion battery systems with 6.4 kilowatt-hours of storage.  Because the LG Chem RESU weighs less, presumably less emissions were released in creating it, but for the sake of simplicity I will assume each kilowatt-hour of lithium-ion storage releases 150kg of CO2 for a total of 960kg each.

A battery that uses a completely different chemistry is a zinc bromide flow battery.  The Australian designed Redflow ZCell is of this type.  Its storage capacity is 10 kilowatt-hours and with its 50kg enclosure weighs 290kg.  Because I have no information on the carbon footprint of these these sorts of batteries I will simply use the rule of thumb that emissions from manufacture equal 2kg of CO2 per kg of weight.  Since around 100kg of the ZCell’s weight is distilled water which has a low carbon footprint, that means around 39kg of CO2 would be released for each kilowatt-hour of storage or 390kg in total.

Just estimating the CO2 emissions per kilowatt-hour of battery systems isn’t very useful as the total amount of electricity they can store over their lifespans can very considerably.  A better measurement would be CO2 emissions from manufacturing per stored kilowatt-hour over their lifetime.

Fortunately, it is not difficult to determine how many kilowatt-hours a battery system is warrantied to store if it is cycled once per day, as most are likely to be. For the Tesla Powerwall, going by its old warranty, it comes to 16,000 kilowatt-hours. For the LG Chem RESU it is 17,192 kilowatt-hours. And for the Redflow ZCell it is 30,000 kilowatt-hours.

But that is only what they are warrantied for.  Their actual total output should be better as they are unlikely to die as soon as their warranty is up.  I would say a reasonable estimate for the average lifetime stored energy of these systems would be 120% of what they are warrantied for.  I hope this estimate is low, but battery systems do degrade and can’t be expected to operate too far beyond what their manufacturers promise.

So by taking the rough estimate of emissions resulting from their manufacture and dividing it by the estimate of their lifetime storage we get the following:

  • Tesla Powerwall: 50 grams of CO2 per kilowatt-hour
  • LG Chem Resu: 47 grams of CO2 per kilowatt-hour
  • Redflow ZCell: 11 grams of CO2 per kilowatt-hour

Battery emissions from manufacture.

A total of 50 g/kWh for the Tesla Powerwall is not a huge amount, so if you installed one on-grid thinking it would help the environment you shouldn’t beat yourself up over it.  You can wait until I’ve explained about the emissions from charge and discharge losses before you beat yourself up.

Every Time You Charge And Discharge Your Batteries You’ll Lose Clean Solar Energy

Rooftop solar without battery storage sends any surplus solar electricity into the grid. A home with batteries can instead store it to use in the evening. Each kilowatt-hour of electricity sent into the grid or discharged from a battery will reduce electricity generated from fossil fuels by around 1 kilowatt-hour. But because energy is lost when charging and discharging batteries more than 1 kilowatt-hour needs to be put into them to get 1 kilowatt-hour out. This means storing using stored solar electricity will reduce emissions by less than if it was instead sent into the grid.

Round Trip Efficiency

When you charge and discharge a battery, energy is lost. The percentage of electrical energy that can be taken out compared to what is put in is called the round trip efficiency. The figures for the 3 systems are:

  • Tesla Powerwall 92.5%
  • LG Chem RESU 95%
  • Redflow ZCell 80%

Roundtrip Efficiency

The efficiency of batteries can decrease over time.  Just how great the decline will be is difficult to know because manufacturers rarely mention this information. Because of the uncertainty I will be kind to the batteries and not attempt to adjust their efficiency for the effects of ageing.

AC Coupling Causes Additional Losses

Batteries can be charged directly from DC power generated by solar panels. This is called DC coupling and is the most energy efficient way to install batteries. It requires a hybrid inverter compatible with the batteries and these currently cost more than standard grid connect inverters. Solar panels with micro inverters cannot be DC coupled.

Many grid connect battery systems are installed using AC coupling, which requires a separate battery-inverter.  This can be an easy way to add battery storage to a rooftop solar system, but the battery inverter results in additional energy being lost.

A quality battery-inverter such as the Sunny Boy Storage can have an average efficiency of 96.5%. Not all battery-inverters are this efficient, but I will use this figure.  It applies both when batteries are charged and when they are discharged.

So if the battery systems are AC coupled and the battery-inverter efficiency is applied to both charging and discharging batteries, the overall round trip efficiency of the 3 battery systems will be:

  • Tesla Powerwall 86.1%
  • LG Chem RESU 88.5%
  • Redflow ZCell 74.5%
AC coupled roundtrip efficiency (%)

AC coupled roundtrip efficiency (%)

Solar Power Displaces Fossil Fuel Generation

Wherever you are in Australia, supplying solar energy to the grid will reduce fossil fuel generation. Even in clean, green South Australia where wind power sometimes generates more power than the state can use, the excess is exported and reduces Victoria’s fossil fuel usage.

Fossil fuel generation releases the equivalent of approximately 950 grams of CO2 into the atmosphere per kilowatt-hour. So if a battery system has a round trip efficiency of 90%, each kilowatt-hour of stored electricity a household uses will result in the emission of 95 grams of CO2.

When considering the round trip efficiencies of the battery systems alone, their emissions from charge and discharge losses will be:

  • Tesla Powerwall 71 grams of CO2 per stored kilowatt-hour
  • LG Chem RESU 48 grams of CO2 per stored kilowatt-hour
  • Redflow ZCell 190 grams of CO2 per stored kilowatt-hour

When AC coupled with 96.5% efficient battery-inverters emissions will be:

  • Tesla Powerwall 132 grams of CO2 per stored kilowatt-hour
  • LG Chem RESU 109 grams of CO2 per stored kilowatt-hour
  • Redflow ZCell 242 grams of CO2 per stored kilowatt-hour

Putting It All Together

Adding the emissions from the battery system’s manufacture, the emissions per stored kilowatt-hour when AC coupled with 96.5% efficient battery-inverters will be:
  • Tesla Powerwall 182 grams of CO2 per stored kilowatt-hour
  • LG Chem RESU 156 grams of CO2 per stored kilowatt-hour
  • Redflow ZCell 253 grams of CO2 per stored kilowatt-hour

Over their expected lifetimes these AC coupled battery systems will emit:

  • Tesla Powerwall 3.49 tonnes of CO2
  • LG Chem RESU 3.22 tonnes of CO2
  • Redflow ZCell 9.11 tonnes of CO2
Total CO2 Emissions

Batteries Can Reduce Emissions But Don’t

Home battery storage can reduce emissions in 3 main ways:

  • providing ancillary services
  • improving the average efficiency of grid generation
  • storing surplus renewable energy

Only the first now applies in Australia.  In the future batteries storing clean renewable energy may allow fossil fuels use to be eliminated, but right now they do not.

Providing Ancillary Services Can Reduce Emissions

Home batteries can cut emissions by reducing the amount of fossil fuels used to provide ancillary services that keep the grid stable. These services include maintaining power at its correct frequency and providing spinning reserve.  But when batteries are used it is called non-spinning reserve. If your batteries start spinning then something is going drastically wrong.

By storing surplus solar energy and drawing on it when needed, household battery storage can reduce the need for ancillary services by smoothing out the home’s grid electricity consumption.  But this will have a very small effect on emissions from ancillary services. The main requirement for ancillary services is when the largest generator on a grid suddenly trips, rather than coping with variation in grid demand.

Home Battery Storage Can Provide Ancillary Services Directly

In order to effectively provide ancillary services home batteries need an internet connection and software that allows them to supply electricity or store it, as required. It is possible to be paid for providing these services, but currently the pay is considerably less than what fossil fuel generators get for doing the same job. As only a very small portion of the electricity stored in home batteries is ever likely to be used to provide ancillary services in normal use, even when able to provide these services, home batteries are not likely to provide nearly enough ancillary services to offset the emissions they cause.

Batteries Can Improve The Efficiency Of Grid Generation

By storing solar electricity during the day or off-peak grid electricity, battery storage can reduce or even eliminate the need to use peak generators during times of high demand and this can potentially reduce emissions. But in Australia the large majority of our electricity is produced from coal and our peak gas generators tend to have lower emissions per kilowatt-hour than our coal generators. This means battery storage in Australia could increase the average emissions per kilowatt-hour of grid electricity, as a smoother demand allows more of it to be generated from the lumbering coal plants.

Batteries Can Eliminate Fossil Fuels From Electricity Generation

With enough renewable generating capacity battery storage could allow fossil fuel generation to be eliminated. Excess renewable electricity could be stored in batteries for use when there wasn’t sufficient solar, wind, or other renewable power to meet electricity demand.

But Australia is nowhere near the point where battery storage can help us eliminate burning coal and natural gas. Once we reach the point where more renewable energy production regularly exceeds demand then battery storage will be extremely helpful in cutting emissions. Australia would need to generate around 50% or more of its electricity from renewables for us to reach this point.  Unfortunately that is still years away for Australia, despite our wealth of renewable resources.

Export Limited Solar With Batteries Can Reduce Emissions

While Australia as a whole is a long way from producing so much renewable power it will go to waste without some form of storage, there are homes in this situation. Some homes are limited in how much power they can export to the grid. These export limited households are most common in rural areas, especially in Queensland.

Because it is possible for them to produce more electricity from rooftop solar than they can use themselves or export, if these households install battery storage it can definitely reduce emissions by storing clean energy during the day that otherwise would have gone to waste.

Assuming around 50 grams of CO2 emissions from battery manufacture, each kilowatt-hour of stored electricity used in the evening, that otherwise would have been wasted, will reduce emissions by around 900 grams. This is the one clear example where on-grid battery storage can reduce emissions in Australia.

Battery Storage Probably Doesn’t Need Your Support To Succeed

While most on-grid battery storage won’t save money or cut emissions now, some people may be happy to buy it for the warm feeling they get from knowing they have contributed to the development of a technology that could greatly cut emissions in the future.

But at the current rate of price declines in battery systems it won’t be long before battery storage pays for itself even though it won’t reduce emissions. Once it reaches that point huge numbers of people are likely to install it because it will save them money and this will result in increased production and further cost declines. As a result, battery storage is not likely to need your help.

There Are Better Ways To Help The Environment

Rather than spending money on batteries which will lose you money and increase emissions, if you want to help the environment I would recommend spending money in a way that will cut emissions and save you money.

One way is to install additional solar panels, or perhaps a completely new solar system, in addition to one that is already on your roof. If installing more solar isn’t practical, if you are wealthy enough you could consider helping family members get solar or donate a system to charity.

Other money saving steps that can reduce emissions include improving the thermal envelope of your home, getting a heat pump or solar hot water system, installing LED lighting, and buying an electric bicycle and using it instead of driving.

My Recommendation

People buy batteries for lots of reasons. If you want batteries because you love new technology, or because your solar exports are limited by your local network, or because you really want a white shield with ‘TESLA’ written on it in your garage, then go for it.

But if you want to save the environment or save money, then I have a duty to tell you that battery storage increases greenhouse gas emissions and will cost you money instead of saving it.  If you want to help the environment or save money there are much better ways to spend your money.

If you don’t buy batteries for on-grid storage that is not going to halt their development. Plenty of battery systems will still be bought by cashed up early adopters and for off grid systems.  And battery development is occurring at a frenetic pace to supply the growing electric vehicle industry.

If you do buy batteries then please consider a system with a high round trip efficiency and install enough solar panels or make enough efficiency improvements to compensate for the extra emissions. Hopefully before your first system reaches the end of its life battery storage will no longer increase emissions, but instead reduce them by enabling almost 100% renewable generation across Australia.


Updated on 6th of September 2016 to decrease the efficiency of using AC coupled battery storage to closer match reality. 

About Ronald Brakels

Many years ago now, Ronald Brakels was born in Toowoomba. He first rose to international prominence when his township took up a collection to send him to Japan, which was the furthest they could manage with the money they raised. He became passionately interested in environmental matters upon his return to Australia when the local Mayor met him at the airport and explained it was far too dangerous for him to return to Toowoomba on account of climate change and mutant attack goats. Ronald then moved to a property in the Adelaide Hills where he now lives with his horse, Tonto 23.


  1. Very good summation.
    As an old hippie and owner of Positronic Solar, I spend a lot of time each day talking people out of batteries for the reasons you’ve given.
    Besides, if you believe the battery brochures I have a bridge I’d like to sell you.

  2. Jack Wallace says:

    Good to see someone actually ‘doing the arithmetic’, even if some of the assumptions are merely ‘opinion’.
    A general observation:- If the idea is setting a benchmark then, rather than comparing the variables, an unrelated reference would be more valid:- eg a lead/acid battery-bank of similar storage size.
    In fact, I’m surprised lead-acid batteries weren’t included in the evaluation. (along with a $-cost comparison as a level-playing-field benchmark.)

    Further, I’d suggest that “It requires a hybrid inverter compatible with the batteries and these currently cost more than standard grid connect inverters.” once again makes the point that a stand-alone (lead-acid) solar-system would hold it’s own in any across-the-board ‘cost-efficient’ (in resource AND dollar) comparison.

  3. Jack Wallace says:

    ps….. On finding themselves surrounded by hostile Injuns the Lone Ranger asks he sidekick “What do we do now, Tonto?”
    …and the arithmetically-derived response was “What do you me ‘we’ Paleface?”

    I was just a bit curious about how anyone would come to name a horse Tonto 23. Is he the 23rd Tonto…or is it perhaps the horse’s Facebook page designation?

    • Ronald Brakels says:

      I was going to tell you some horse jokes, Jack, but then I thought people might think I was being serious and be horrified and so I decided against it. So please just pretend I told you some great jokes and you laughed yourself hoarse.

  4. Howard Patrick says:

    Five years ago I paid $11k for a 3Kw PV array. I could now get nearly 10Kw for the same outlay.

    Predictions are predictions but do you think the price of battery storage could fall by a similar about over the next five years?

    The are so many thinks happening in the battery space, like magnesium batteries with perhaps twice the storage density, but would be interested in your wise thoughts.

    • Ronald Brakels says:

      Apparently storage has fallen by over 10% a year for the past three years. If it falls by 10% a year for the next 5 years then it will only cost 60% what it does now. That’s only the cost of battery storage and not installation, but I’m sure that will fall as well as installers learn by doing and as manufacturers tweak their systems to make them both smarter and easier to install. So provided regulatory hurdles aren’t placed in their way, I think it is likely home batteries will pay for themselves, at least in favorable situations, by then.

      And if a large Chinese manufacturer, making use of that nation’s large and expanding lithium-ion production capacity, decided to win market share by selling units at marginal cost, then the day household battery storage will pay for itself in Australia could come quite soon.

    • Jack Wallace says:

      Y’need to shop around. Seven years ago I had a 2kw system installed for $3860 ~ by an officially- recognised provider because I wanted the FiT, which was offered @66cpkwh, contracted until 2024.
      I bought my first panels 2nd-hand in 1981 and paid $13.80 PER WATT*.

      Whjilst shopping around you’ll get all sorts of offers from all sorts of people.
      While websites like this are useful, the decision should, in these days of endless opportunity and competition, be based on ironclad, enforceable warranties. Everything else is of ‘tier-2’ consideration.

      After making many detailed enquiries over weeks I chose a private installer (now a company) who was prepared to give me his personal details ~ including address and photos of his own solar-system ~ and to take personal responsibility for a warranty: an ‘industry-standard’ one rewritten to our mutual satisfaction.
      On that basis I was prepared to accept ‘unknown’ (therefore cheap!) panels ~ which have performed admirably ~ and an inverter that ‘experts’ everywhere were shitcanning fiercely. (and therefore also cheap).
      Such a demonstration of goodwill from both sides is beyond price.

      I’ve had two problems along the way: a minor one with a circuit-breaker and a serious one with the inverter. Both were remedied immediately (despite the 200 km distance to travel) without question or cost to me.

      The bottom line?
      It’s your money, Ralph; if you can’t get the deal you want take it elsewhere.
      The right people are out there.

  5. Another good article.

    To what does “In Australia, with one main exception, …” refer?

    • Ronald Brakels says:

      It refers to export limited solar installations where perfectly good solar power goes to waste because production exceeds the combination of what the home consumes and can export.

  6. Jonathan Maddox says:

    All true, but I find that the discussion misses one very important point: additional storage will soon be *required* to allow low-cost renewable energy, most of which is intermittent, to replace fossil-fuel-burning power stations.

    While storing daylight power for use at night *right now* might not reduce gross network-wide emissions, we’re already very close to the point where that will be true (witness occasional negative midday spot prices on sunny days … coal power stations effectively paying customers to take their power). Batteries are getting cheaper, and cheaper does mean reduced embedded energy and reduced emissions in manufacture and delivery, all else being equal.

    Buying batteries pushes down the learning curve. “The Electric Power Research Institute (EPRI) reviewed a variety of data to find that lithium-ion batteries drop in price by 15% per doubling of volume.”

    Volume is happening, and it only happens when people buy today’s batteries. It doesn’t happen based on the batteries you might buy tomorrow when they’ve already become cheaper!

    • Last May, a well known Silicon Valley salesman launched a home battery.
      The battery costs more than promised. It does not last as long as promised. It does not provide backup, as promised. It is of neither econonmic or environmental benefit, as promised. It’s that simple.

  7. Jeremy Lawrence says:

    I think this article is wrong. To help the environment we need to accelerate the transition to a 100% renewable electric energy economy. Solar supporters aren’t fighting a technical or cost battle, but a persuasion battle. Batteries are good for the environment because they persuade the public and politicians that the grid can manage very high levels of solar penetration. The more batteries we have on the grid, the weaker the argument of the fossil fuel apologists. Every battery we connect to the grid is a nail in the coffin for the fossil fuel industry. The Germans subsidised solar and that created the market for high-volume manufacturing in China. Australian home owners can deliver the clearest possible pro-environment message to battery vendors, policiticians and grid regulators by simply buying quality batteries.

  8. Guangming Meng says:

    Hi Ronald, interesting article, but have you properly considered that not all kilowatt-hours are made equal? All that excess solar power in the middle of the day getting exported and eventually dissipating into the dirt, unloved and unwanted. All that excess power doesn’t have the slightest impact on peak demand, but contributes to an oversupply which distorts the electricity market and reduces the incentive to invest in renewables. Isn’t it better to store excess power in a battery and release it when it is needed, i.e. during peak time? With enough of these systems, surely peak demand will reduce and the need for baseload power will be revealed for the myth it is?

    I thoroughly enjoy your articles but I think you’ve missed a trick on this occasion.

    • Ronald Brakels says:

      Looking at wholesale electricity prices for the National Electricity Market at noon yesterday I see the lowest was in Tasmania at 5.9 cents a kilowatt-hour and on the mainland it was 6.6 cents in Queensland. So we are still a long way from solar electricity being curtailed because it exceeds demand. On several occasions combined wind and rooftop solar production has exceeded demand during the day in South Australia, but the surplus was exported to Victoria.

      Solar electricity lowering the cost of electricity during the day does not result in a market distortion. It results in a market being a market. And I think lower wholesale electricity prices during the day is marvelous because it is horrible for the economics of coal power. It also acts as a price signal encouraging storage, but currently electricity arbitrage using batteries does not pay for itself. This could change in the future.

      As for solar pushing down electricity prices during the day and discouraging the building of more solar capacity, the solution to that is a carbon price equal to the cost of removing CO2 from the atmosphere. Perhaps $100 a tonne, although estimates vary. This is because what we really want is not for solar to be deployed, but to not die.

      Fortunately the economics of rooftop solar is not affected as strongly by declining wholesale electricity prices as utility scale solar. Of course, once a region starts to regularly curtail renewable energy production because supply exceeds demand, then energy storage can definitely cut fossil fuel emissions.

  9. Jack Wallace says:

    Speaking as someone who’s been using (and building production machinery) alternative power since 1981, I say again that NO form of STORED solar (or wind) power has the capacity to produce energy nearly as cheaply as (in any terms, including environmental, as the harnessing of tidal power.

    eg. each day the Great Southern Ocean fills and empties Port Phillip Bay (Victoria) FOUR times ~ producing ENORMOUS amounts of energy in both directions. Anyone who’s ever watched The Rip at full flow can’t help by be amazed at the very obvious force of the tidal flow….FORTY CUBIC KILOMETERS at a time.
    (Not to mention the other 100,000 km. of other coastline we have to exploit.
    Turbines in long pipes across both peninsulas could not only produce, consistently, more power than we could possibly use, but could be simply regulated by the adjustment of a stop-valve.

    In terms of solar (or wind) power which needs to be stored the solution is keeping the storage to a bare minimum (lights, computers, tv, etc.) and using mechanical power short-term to run the heavy users like microwaves, dishwashers, etc.
    The main consumers (fridge/freezer/ etc. can best be run, directly, full-on all day long by enough (cheap) dedicated solar-panels/wind-turbines.
    The details have been posted previously. Diddling about with sophisticated, relatively minor options is another terminology for “Pissing into the wind”

    • Peter Herklots says:

      How I agree with you over tidal power.

      Unfortunately construction would take more than three year, so, because the span of political parties in power initiating the development will not be interested.

      That is why a new Snowy Development will never happen again.
      Similarly pumped storage potential is also ignored, which is a pity as water from East of the divide could be pumped up and then discharged West into the Murry Darling Basin. Two birds, one stone.

    • Peter Seligman says:

      I can only agree that I too thought that an ENORMOUS amount of energy was flowing through The Rip. But like the late David MacKay of Sustainable Energy – without the hot air, I prefer numbers over adjectives. So I calculated it. First of all the area of the bay is about 2000 km2 but the average tidal range is only around 0.75 metre giving 1.5 km3 not 40. For the practical reason of not being able to dam the heads, the extractable power is about 40MW, or about 10 watts per Melbournian. At the risk of appearing to be spruiking my book Australian Sustainable Energy – by the numbers, see page 17 for more on this and other locations

    • Unfortunately a few years back in the west some idiot proposed a 22 hour per day tidal station at Derby (up north, big tides). The design was so poor (e.g. taking out just over 160 hectares of mangrove swamps to provide buffering and still not getting 24 hour power out of it) I can still remember it – arguably the worst design for any technology I have ever seen. Ever.

      The reason I bring it up – it has probably killed all tidal power projects in Oz for the next 20 years or so. Even the current ocean trial around the corner here (Rockingham) is only trying to harness wave power.

      On a cynical day I think the design was deliberately rigged to fail, and it was fascinating to see the usual “mindlessly green aficionados” support it. No one with an IQ “above a fence post” (Tom Waits) could have regarded that design as more than an April Fool’s joke – and I include a number of hard core greenies amongst the “entirely correct naysayers”.

      Having said that – high tide places in WAs north would be eminently suitable given even a mediocre design – if anyone had the guts to try to get approval by the current lot of federal politicians. Unfortunately, I doubt they would succeed due to the “long term is three years rule” we are cursed / blessed with.

  10. Interesting article Ronald, but I don’t follow your calculations.
    For the Powerall, the battery and inverter combined efficiency is 88.8%. To provide 1kWh of AC energy delivered to the load, solar DC input must be 1126Wh. Then, 950g could be offset.

    If that same DC solar were sent to the grid, inverter conversion at 96% would reduce it to 1080Wh, equating to 1027g offset.
    Therefore, the battery’s inefficiency reduces the potential offset by 1027g- 950g = 77g/kWh.
    I don’t see how you arrive at 116g?

    • Ronald Brakels says:

      Athomas, whether solar electricity is sent directly into the grid or used to charge AC coupled batteries it will suffer a loss when power from the solar panels is inverted from DC to AC. Since that loss is unavoidable in both situations it can be ignored since I am just considering extra emissions that result from adding battery storage.

      Looking at my results for emissions per kilowatt-hour stored I see… I see that I have made a mistake. I have not properly accounted for losses from AC coupling. This doesn’t change my conclusions, it just means emissions that result from adding battery storage at this point in time are a little worse. I will update the article. Thank you for getting me to think about what I had written.

  11. Do you have any thoughts on the process, cost or emissions at the end of life of a battery. I assume at minimum the battery needs to removed and put in landfill or Tesla talks about recycling batteries at the gigafactory. Won’t that add more emissions, or is that included in your figures.

    • Ronald Brakels says:

      Mol, just as lead-acid batteries are recycled today, I expect lithium-ion batteries will be recycled in the future. This will reduce their carbon footprint by reducing the need to mine and refine new materials. But I didn’t try to take this into account. One problem with trying to evaluate its effect is it is difficult to know how carbon intense manufacturing will be in the future. Fingers crossed its emissions fall very rapidly.

  12. thank you what an interesting read, thought provoking

  13. Ian Gillies says:

    Thank you for your exceptionally well-presented guide through the subtleties of Solar PV, battery, and grid interactions

    I have just been cold-called by the installer of my solar PV system, asking me if I am now interested in adding batteries. At first I had a rush of blood, but my Googling lead to me to your article, which has had the effect of a very bracing shower.

    My “use case” is this. I live in a two-story unit in Melbourne. Summers were growing unbearable, so we installed air-conditioning, rooftop solar, and considerably upgraded roof insulation. The air-conditioner was chosen to be controllable remotely. Similarly, I enabled our dishwasher and washing machine to be activated remotely. The intention was to make the sun supply as much power as possible for those energy-hungry devices.

    There were some gotchas though, new ones revealed by your article. In the case of the air-conditioner I had hoped that, having cooled the well-insulated house during the day, we would be able to turn the unit off latish in an afternoon and remain comfortable during the night. However, during the evenings of days from 35C to 43C and beyond, the heat soon invades. My hopes turned to batteries as the solution to that.

    However, if I understand you correctly, purely in terms of CO2 emissions, I would have had the opposite effect of what I intended. I would be better off giving 100% of my power to the grid during the day, helping to drive down the minimum base-load demand and associated high emission rate, and then drawing up to ~88% of that out of the grid during the evening peak when average emission rates are lower. Had I hoarded the power, I would have access to that ~88%, but forgone an opportunity to offset emissions. Not only that, I would have contributed to smoothing out of circadian demand and helped drive up the level that dirty base-load systems could maintain.

    Furthermore, your article seems to me to imply even a down-side my strategy for ensuring our dishwasher and washing machines operate during periods of high solar generation, despite no batteries being involved. There is a strong financial incentive to do that, since I was not an early adopter. I do better to grab my own free power first and forgo my piddling FIT, rather than get all the FIT and then draw the power out of the grid later at full retail rates. However, in terms of being green, that seems to be reversed. Once again, I would do better to feed all my day-time power to the grid and turn on those devices during peak periods.

    Getting back to the air conditioning, despite the foregoing, it still seems to me that with batteries added to my solar PV system I can cool my house on hot summer evenings and achieve several things. Leaving aside embodied energy considerations etc, I get that evening’s energy far cheaper than I could by drawing from the grid, I don’t actually cause extra CO2 emissions, and I disrupt the major grid supplier’s business model.

    One of the difficulties I have with your analysis is the addition of emissions associated with battery manufacture to the loss of saved emissions relative to the potential of the associated PV system alone. That is not quite the same thing as a system that actually releases that quantity of CO2. Even if the combination of on-grid PV plus battery cannot do as efficiently as PV alone, it can still be a net downward driver of emissions. If the result is better for individuals, then many more people may jump aboard the bandwagon. Increasing the number of systems, even at the cost of efficiency per system, may still be the quickest path to overall sustainability.

    Any comment would appreciated and weighed very carefully. (I will not horse-laugh.)

    • Ronald Brakels says:

      Ian, while exporting solar electricity into the grid during the day and then using appliances such as dishwashers and clothes washers in the evening will hurt coal generators, it is probably not an economically efficient way to go about reducing fossil fuel emissions if it is costing you money. My guess is you would be able to do far more good for the environment if you ran your appliances during the day off solar power and then used the money you save to invest in things such as expanding your rooftop solar capacity, buying a system for a relative, or investing in energy efficiency such as a heat pump hot water system or more efficient appliances.

      With regard to running your air conditioning off batteries, this will result in more emissions than if the solar electricity had been sent into the grid, but these emissions could be offset by installing an extra solar panel or investing in other energy efficiency. So if using a battery saved you money it wouldn’t be too difficult to use the money saved to offset its effects. But, looking at the total cost, battery storage can’t save money at the moment. We are getting closer, but we’re not there yet.

      Getting a battery system now could help lower their costs and bring closer the point they will be of environmental benefit in the future, but we know for certain that expanding rooftop solar and improving energy efficiency will have environmental benefits now and will save people money, so I see them as the better choice. The electric car industry alone will ensure batteries will continue to improve.

  14. So now we’re calculating a very loose g/kwh based on pretty much pure speculation (especially with the redflow graphs), but fair enough, it will take CO2 to produce a battery, so of course the value is not 0.

    But if you’re using a battery and not using the grid for 15 years, you are responsible for less carbon usage overall.

    Sure, the grid is running anyway producing CO2, but that’s like with the energy efficiency argument – so what if you’re on LED lighting and the grid is running 10 coal power plants at full steam anyway. If you’re feeding in solar power at the middle of the day when power is not at its peak, and the power plants are running anyway, how are you saving any CO2 at all?

    By these arguments, we shouldn’t do anything.

    If I take my house off grid AND make it efficient I will be using less CO2 personally, and we need to prompt more people to do this to force power plants to shut down for lack of use.

    THAT is how you’d end up saving CO2 really. What do you think?

    • Yes, it’s difficult to know if home batteries could be effective, even hypothetically. I have tried to work that out, but costs and benefits are nebulous, so can’t be certainly dismissed, or proven.

      One thing for sure, though, is that battery manufacture, transport and installation will produce considerable CO2e. If batteries were rapidly deployed in quantity, those emissions would be added over a short time. Offset would take 10 years or more, but may never eventuate, leaving 100kg of waste.

      It’s a gamble at best, and the benefits, should there be any, are small. The time, money, materials would beter be spent on aggregated or grid systems, which can be designed, and engineered using durable batteries, where appropriate.

      The present proposal is to install consummer grade equipment, to whomever, while simultaneoulsy promising self-consumption, grid-stability, reduced emissions and back-up. Marketing sells batteries, where the benefits are simply assumed to exist. That’s how I see it; an expensive risk, with limited prospects for reward.

  15. Peter Seligman says:

    First class article in every respect Ronald! You’ve said it all.
    I notice in your calculation of CO2 emitted through battery manufacture, that you used the rule of thumb of 2 kg of CO2 per kg of an item. I used a similar metric in my book Australian Sustainable Energy – by the numbers. However I used a figure of 0.5 kg/$ of cost. The rationale is that this is Australian emissions divided by our GDP. On that basis, a $10,000 system equates to 5 tonnes of CO2. This then includes the battery manufacture, inverter, installation and profit.
    I use my method for the following reason. A Rolex watch may only weigh a 100 grams but its huge price supports a company of executives, advertisers, jetsetters, plus the emissions due to the mining of the precious metals. A $20 watch of the same weight is produced by a company that produces maybe 100 – 1000 times as many watches so their impact is commensurately lower. Maybe the best metric would be to use a linear combination of both weight and cost. Are we onto something here?

    • Ronald Brakels says:

      Half a kilogram of CO2 per dollar of GDP still holds up extremely well.

      I certainly agree the less one spends for a good or service the less one is likely to contribute to emissions. (It’s good environmental justification for me to be a cheapskate.) But I don’t think a combination of CO2 per kilogram and CO2 per dollar would improve estimates of emissions, as they are really measuring different things. One is an estimate of emissions resulting from manufacture, while the other is average emissions per unit claim on economic activity. I think they can work as complimentary measures, but not combined.

      I’ve found a PDF of your book and I will have a look.

  16. Hans Sieker says:


    I applaud you, and your commenters, for the detailed analysis and discussion on this topic. Keep it up. Being a recent purchaser of a rural property (currently a city dweller), a property with ample solar potential and through the foresight of the current owner the infrastructure required already being in place, I have been searching for the ‘truth’. Your article, and your commenters, have come the closest I have come across to exploring the “total environmental costs” of solar (and for that matter other) systems.

    So much food for thought. Putting economics aside (at this date still not insignificant), I have always struggled with the end-to-end environmental cost of “renewable energy” systems – be they domestic solar, hybrid or battery cars or wind generators.

    Being an engineer, always asking the detailed questions, putting aside the “I’ll look good to the neighbours with all these panels and batteries”, the science, supplemented by your excellent summary of the issues to be considered, has (at the moment at least) left me with the current conclusion “not convinced to proceed”.

    many thanks Hans.

    • Lawrence Coomber says:


      Don’t give up investing in your own knowledge and education on the subject Hans.

      Off Grid is a mature global technology; and more than capable of fully satisfying your modest rural property requirements reliably and without fuss 24/7, and cost effectively. But like all subjects of course, the devil is in the detail.

      The link below is not specifically about your comments Hans, but will at least add to your understanding of one critical system design imperative that would always feature in best practice DC Coupled Off-Grid designs by RE contractors operating in this space, who have a proven track record over time to have a genuine commitment to delivering both product and system warranties and system performance guarantees, as well as meaningful ‘life of system’ customer support strategies.

      PS. Don’t worry about the neighbours Hans – instead embrace them and create an Off-Grid (Micro-Grid) and sell them discounted electricity; or better still, share the system cost together. Load diversity between 2 adjoining users (or more) brings considerable efficiencies into play regarding system sizing and design overall. There are several business model formats in place for rural (and domestic) Off-Grid (Micro-Grid) (highly efficient) strategies.

      Lawrence Coomber

  17. Unfortunately this is just another article giving one point of view of the effect of batteries in a domestic solar energy situation, while ignoring most others. If we were to apply and believe the exact same calculations to the domestic motor car, none of us would buy one, because they damage the environment more than almost any other modern appliance. Maybe get rid of the batteries in those motor cars and reintroduce crank handles and magneto ignition. Did you laugh?

    In modern communities our every day health and safety depend to a great extent to the provision of the grid systems that provide potable water, central sewerage treatment, electrical power and in some places reticulated domestic gas. We all know that in isolated rural locations families can only try to emulate the convenience and health benefits of those grid systems by storing their own water and installing filtration, installing septic tanks to treat waste water and generating electricity by some means to power necessary modern home appliances and lighting. Bottled gas, bulk diesel fuel and batteries are also common necessities to store energy for use on site. Most of that rural infrastructure is much more expensive to install and maintain than the grid services available to a suburban home. I have been there, done that.

    I have 15 Kl of rain water storage on my suburban home because I like vegetable gardening and the idea of collecting free water from the sky for use when the garden needs water. But if I compare the cost of that water per kilolitre and include the tanks the pump and the reticulation to 5 taps, with the price per Kl of town water over 20 years, I am out of pocket. We all know there is no water feed-in tariff. But I still do it! For me the same reasons go for solar generated electricity battery storage. I can collect free electricity, why not store the excess for use when the house needs electricity. As for me liking a polluting motor car that has lead acid storage battery and will never ever pay for itself, like most people I am willing to pay for the convenience of having a car today with all its environmentally bad points By and by, I also would not connect domestic gas to my house in lieu of solar electricity to heat water or cook. Burning LPG every day in a house produces much, much more CO2 and CO than the net effect of daily storing solar energy in a lithium battery.

    • Peter Seligman says:

      You appear not to have understood the gist of the article Neil. Yes – the car has come with an environmental cost but has provided a significant benefit over the horse and cart. Ronald said that domestic grid-connected batteries come at a significant environmental cost, but with no benefit, except in in isolated instances in QLD. And he indicated that even being a heroic early adopter is less useful than just putting on more PV panels.

      • Ah so! Do you agree that my rain water system comes at a significant environmental cost for no benefit? All that plastic tank and plastic pipe manufacturing came at a cost. The pump used energy. My benefit is zero in dollar terms and it is less convenient than using town water on the gardens. Is that all there is to it? Using Ron’s argument we shouldn’t install rain water tanks in the suburbs?
        What about installing 5 Kw of PV panels on your house to generate 7,300 Kwh per annum but exporting it all to the grid at 6 cents/Kwh ($440)? Is that of benefit to the environment and to whom? What about the other end of the stick? Use all of the 7,300 Kwh in your home and save 26 cents per Kwh for your benefit ($1900)? In the average working home you can’t use much of that 7,300 Kwh during daylight hours. But you could use it all with the help of battery storage. Just like the rain water, the solar energy is free but do we today in a developed society have to leave all of the collection and distribution to the State. I don’t think so. As home owners, we should if we can, be willing to accept a reasonable cost of being able to provide on-site as much of the water and electrical energy we use for our daily enjoyment and wellbeing? If a significant percentage of home owners do that, then the capacity of water mains and electrical transmission lines don’t need to grow so fast. Isn’t that an environmental benefit not mentioned in the analysis?

        • Peter Seligman says:

          Glad you mentioned water Neil. There is a great program by the ATA called Tankulator which helps you calculate how much water you save with a certain size of tank, under specified conditions. I found that with a 2000 litre tank connected to laundry, toilets and garden taps we could save 51 kl/year but a 5000 litre tank only raised this to 56 kl/year. So we opted to 2000 litre and used the rest for other environmental projects. With the solar PV we got 5 kW, which as you say, can generate around 7,300 kWh/year. Later we got a further 5 kW, generating another 7,300 kWh/year, totalling 14.6 kWh/year for us and other consumers. However had I spent the money on batteries, we would lose around 300 kWh/year in round-trip (in)efficiency. So only 7,000 kWh/year for us and other consumers. By the way the FIT in Victoria is about to go up to 11.3 c/kWh on 1 July.

          • Wow, we are flat out finding the NE/NW facing roof area on our house for 5.6Kw of panels. Fortunately it also does double duty collecting rain water. Benefits come in many forms. Because we water gardens in the mornings, the rain water pump runs on solar power. One of the things we grow for 6 months of each year are very thirsty snake beans. In that time we eat about $150 worth at market prices. Their main plant food comes from our compost bin. Just one example of us reducing water and electricity loads on the public utilities.
            I agree that a healthy FIT is very effective at countering the argument for battery storage as part of a PV power system. Much better than the nebulous environmental damage argument of making lithium batteries. Being an early adopter here in SE QLD we are getting 50 cents/Kwh FIT until 2028. But our system limit is 5Kws. We don’t need batteries because that level of FIT does the storage in dollars so much better that you could ever hope to do with storage in Kwh using any kind of battery today or in the future. We’ll have to keep you posted on what we do in 2019.

  18. Thanks for the analysis.
    Storing solar in the (electric) hot water system seems a far better option than batteries.

    It makes sense to disconnect from the gas grid too if getting a bigger pv system.

  19. Peter Seligman says:

    Re your first point – the electric HWS idea would only make sense if you are using a heat pump. With an old-fashioned resistive heater, you will find that in winter on many days you won’t have enough excess PV to heat the water. Then you will be taking elec out of the grid to heat water – wasteful and expensive. You might also have the dilemma as to whether to do this in the day to maximise self-use, with the risk of paying day rates for some water heating, or doing it at night if you expect it to be a cloudy day following. It’s a gamble. And if you end up with cold water – you’re the bad guy!
    Re the second point – yes if you are going for a heat pump.

  20. Hi Ronald,

    In this article you say ‘Each kilowatt-hour of electricity sent into the grid or discharged from a battery will reduce electricity generated from fossil fuels by around 1 kilowatt-hour’. I’m interested if there is any Australian research or detailed analysis that shows this?
    My assumption has been that if coal generation is relatively slow to ramp-up and down then it can’t react to the relatively fast changing (and somewhat unpredictable) rate of domestic solar generation and therefore not all solar feed-in can actually offset generation.
    Similarly, large commercial renewable generation (like wind in South Aust) presumably doesn’t ramp-down when domestic solar spikes during a day and therefore some portion of domestic solar feed-in is effectively going to waste. I have no idea what size that portion is though.
    Are you aware of any analyses in this space?

    • Ronald Brakels says:

      Hi Andrew.

      Our electricity sector spends a lot of energy making electrons wiggle back and forth. Variation in demand introduces some energy losses. For example, gas turbines have to be warmed up before they are used otherwise they can fly apart in a variety of interesting ways. Operating a coal power plant at 60% of capacity (generally their minimum in Australia) instead of 100% can cause its efficiency to fall by a couple of percentage points. But these losses aren’t very large and they have always existed. Demand for electricity has always gone up and down and grids have had to deal with it. The introduction of variable generation such as wind and solar can increase these losses, but they don’t increase the already small losses by a great deal. One reason why is because wind and solar power are so reliable and predictable.

      While one the output of one home rooftop solar system may not seem very predictable on a day with scattered cloud, over an area the size of a town, rooftop solar production is. As we can directly observe clouds from space and with radar, it makes solar output very predictable. At the sites of wind farms and the heights of rotors wind speed does not change rapidly and is also very predictable.

      There have been times in South Australia when combined wind and rooftop solar output has exceeded by state demand and the ability to export surplus electricity to Victoria. This does not happen often. It happened around noon a few days ago, but did not last long and only occurred because the state government now requires two large gas generators be in operation at all times.

      Anyway, renewable generation will normally result in a reduction in fossil fuel generation of an equal or almost equal amount. I can’t think of any studies on this off the top of my head, but you may be able to find some if you poke around on the internet.

Speak Your Mind