Check Out This Novel Zenaji Battery Installation in WA

Zenaji battery installation

Jason’s garage in Geraldton. Those funky looking black tubes are 4 x 1.93 kWh Zenaji batteries.

Jason Bertelsen isn’t a name in the renewables sector – he’s a customer. However, his story, in particular his decision to put money into Zenaji batteries, intrigued us enough to get in touch for an interview.

Why? Zenaji batteries are relatively unknown in the Australian market. They don’t use lead-acid or conventional lithium-ion chemistries that currently dominate the home energy storage scene. They use Lithium Titanate chemistry that the company claims offers 22,000 charge cycles and fast charging (according to Wikipedia, this comes at the cost of lower energy density).

Jason told us his choice came after years of wanting to add a battery to his solar power system.

“Every time I looked at the problem, there wasn’t a solution I liked,” he said.

About Tesla Powerwall batteries, he said:

“I wasn’t really happy with the lifespan, the guarantee of only a certain amount of cycles. You’ll never really get the return on investment – that was really a factor.”

He was also unhappy with the loads the Tesla batteries could handle, because he wanted something that would be “rock-solid” for anything he throws at it.

“We don’t really use a huge amount of electricity, because there’s only two of us”, he said – “but nobody likes surprises simply because they ran too many appliances at once!”

Another feature of the Zenaji batteries that brought him over the line is their 20-year warranty. With that in mind, and with a 7-year-old 2.4kW solar system that was due for an upgrade, the time was right to make the jump.

“I did more and more research, I had a chat to the company, and I was ready to go.”

The Geraldton, Western Australia resident’s installation was undertaken by local company BTM Energy, and as Jason said on Facebook, the install ran at a leisurely pace and took four months to complete.

As he explained to us, he started by going down the Solar Analytics road, and the long install cycle meant he had the chance to gather data to compare his old system to the new at every stage of the process.

Jason said BTM Energy was the only installer he spoke to that was prepared to work with a battery provider (and chemistry) that was brand-new to them.

“When I showed them the Zenaji batteries, the owner of BTM went and bought some – so they ironed out the kinks of an install at the boss’s house. That made my install run very smoothly!”

“I’ve always wanted to be a bit more energy-independent – I didn’t want to disconnect from the grid entirely. I wanted the feed-in tariffs, and the sharing arrangements.”

His system comprises:

Zenaji Aeon batteries with inverters etc

The Zenaji batteries are installed in a garage, alongside the inverters.

The Selectronic inverter is set up to disconnect from the grid and create a local microgrid in the event Western Power has an outage, and the Fronius inverter is certified to work with the Selectronic.

“When we first got started, we had an idea of what we wanted to do – it wasn’t a step that was ‘needed’ but I wanted to know what was happening, so we put Solar Analytics on, to get readings of what the house was doing before we upgraded.”

That provided useful information, such as a misconfiguration in their solar hot water system.

“It’s great, but we don’t like having cold showers in this house – so we would tend to leave the booster running pretty much all the time. It chews some electricity if it comes on at the wrong time of the day, which it was!”

On many days, the booster would start at 4am or 5am, meaning the water was nice and hot for the shower but at that hour it would be coming either from the grid or the batteries.

To take care of that, BTM Energy installed a relay to control the hot water service.

“When the Selectronic detects there is enough solar coming in, that’s when the hot water starts heating, as soon as it’s the right time of morning. Only occasionally does it need any extra, and that’s restricted to certain times of the day.”

As well as the long warranty, Jason told us he was pleased by the design of the Zenaji batteries, saying they’re “neat and don’t take up huge amounts of space.”

SolarQuotes founder Finn Peacock, a dedicated reader of specifications, wanted to know whether Jason was comfortable that his Zenaji battery warranty would survive Geraldton’s heat.

The warranty is in this PDF, and what attracted Finn’s attention is reductions in warranty life if the Zenaji batteries are cycled on days that fall outside its rated temperature of 5-35℃. For example, if the battery is cycled on a day where the temperature is over 40℃, the warranty reduces by five days.

“These are installed in my garage. It’s all shaded, and the garage doesn’t get too hot, so I wouldn’t think the batteries are going to get too warm. And the Selectronic needs to be indoors.”

And there’s another way to protect the warranty in the case of these batteries: if the ambient temperature is going to be out-of-spec, make sure you don’t cycle the batteries!

If you’re interested in this intriguing Zenaji battery installation you can follow Jason’s journey here.

About Richard Chirgwin

Richard Chirgwin is a journalist with more than 30 years' experience covering a wide range of technology topics, including electronics, telecommunications, computing and science.


  1. Greg Flint says

    Should Jason Bertelsen happen to read this:

    1. The information I’m curious about doesn’t appear on (yet).

    2. Did you decide against RedFlow Zn-Br flow batteries during your assessment? I’ve ordered 3 of them, because:
    (a) no cycle degradation during useful lifetime of 10 years +
    (b) less burnie than high-C Li-ion cells
    (c) the possibility of refurbishment at the end of life.

    I won’t change what I’m doing, but I’m interested in this other alternative that’s new to me, because I’ll be expanding my battery bank over the next couple of years.


    • Hi Greg
      I’ve been following Redflow for years via its stock price. Price has been going down for a few yrs. They originally moved production to Nth America and had problems. They then moved to Thailand. Seem to be concentrating more on off grid telecommunications for their battery use. The thing that finally stopped me going with them was Solarquotes blog on battery testing at the Canberra testing centre. Redflow performed badly. It’s worth looking up.

      • I did read the blog ( several times over a period of 12 months or so, which included a number of updates to the table, before making my choice. The only item I can see where it “performed badly” is roundtrip efficiency of 80% (compared to 90+% for most others).

        Roundtrip efficiency is of low importance to me, whereas non-combustibility and resource-use minimisation are of high importance to me. My PV system already exports 3 times as much energy as I use, and I’ll be increasing exports over the coming years – the household battery bank is part of that. I will also provide ancillary services to the local Darwin grid.

        I reached a point a couple of weeks ago where one of my Li-ion battery-powered devices survived 4 years. Prior to that, I had had 100% failure of Li-ion batteries in ALL devices – usually by swelling of the battery, or leakage of fluid, and once by fire (of the ninety-six $200 Kokam cells in my car, which burnt the car to the ground).

        • Lee McCurtayne says

          The swelling of battles does indicate overcharging, and could be attributed to the charge not being switched off once charged. Not good where lithium is inline.

    • Erik Christiansen says


      I’d favoured Redflow for years, but for the midling max charge/discharge rate.
      The Zenaji is significantly more generous there, allowing a smaller investment in less battery kWh for the same steady power delivery. And LiTi is not so flammable as plain Li-Ion, more like LiFePO4.

      The reported five-fold life-shortening of discharging Zenaji at 40 C is cause for caution, without a doubt. My recollection is that Redflow thrives at that temperature. My workshop is insulated and will be airconditioned, so I’ll have to check whether Zenajis are permitted in a workshop.

      If my budget allowed 3 Redflows, then I wouldn’t worry about power draw, and it’d be a more temperature tolerant robust installation, I figure.

  2. I don’t see much point in having 22,000 cycle capacity. (and it’s only a claim). How can they possibly claim it? The company hasn’t been around that long to justify that.

    That’s 60 years+ at full 100% daily discharge. Most people don’t even live in their homes for that long, you would have to be 20 years old to get one to see its lifecycle by the time you get to 80 and assuming you live in the same house.

    At least with Tesla, they had batteries in their cars for at least 10 years to see how lithium performs before Tesla started offering home battery storage and were confident enough to offer 10 years warranty with unlimited cycling via solar charging (or 37.5MWh if the battery was charged from the grid) with 70% degradation at 10 years.

    My Tesla Powerwall is just a touch over 3 years now. It’s rated capacity is at 13.263kWh (so about 98% of original rated capacity of 13.5kWh at 48% average daily utilisation. So, I estimate I should have about 85% battery capacity at the 10 year mark at the current 3 year figures to date, shall report in 7 years time if this website is still up, or even at the 5 year mark!).

    In regards to safety, it’s highly subjective. Everything is safe to use until something goes wrong. I have not found one incident where a Tesla Powerwall had become unsafe (ie., caught on fire). If there was, I would like to know the story about it . If a product is installed correctly and operated as stated, then there shouldn’t be any issues. Unless there was a manufacturing defect which is bound to happen in all products, after all, humans are not perfect and will never be able to design and manufacture a perfect product.

    My Tesla Powerwall 2 is operating in conditions that ranges from -5C in winter to 48C in summer (this is in the Hawkesbury area 55kms north west of Sydney by the way, not some outback town). No issues yet with it apart from losing comms with Tesla for a couple of hours one time which was remotely fixed.

    The comment: “He was also unhappy with the loads the Tesla batteries could handle, because he wanted something that would be “rock-solid” for anything he throws at it.” How can he possibly know this? He doesn’t have a Tesla Powerwall. A bit presumptuous to make comments like that.

    I don’t understand this logic that would have been of concern with the power limitations of the Tesla Powerwall for this customer. The Selectronic inverter selected is a 5kW battery inverter. How is this different to Tesla’s integrated inverter in the Powerwall? Both are 5kW and this determines your max instantaneous loading. There’s no mention of the Zenaji battery capacity itself. (apart from the specs from which I gather is 1.93kWh per unit and there are 4, so that gives 7.72kWh) which is way less than Tesla’s PW2 at 14kWh. That’s $12,000 for 7.72kWh of storage, not including the battery 5kW inverter? Hmmm, compared to Tesla’s 14kWh battery with an integrated inverter at $12,500 (not installed).

    Judging by the picture, the Zenaji batteries are bigger in all dimensions (height, width, depth) than the Tesla PW2 and that doesn’t even include the battery inverter.

    I have 44 panels at 11.325kW (3 different solar systems installed at different times) but they all work seamlessly with the TPW2 (which I’m really surprised as they are of different brand and vintages from 2011 to 2018). The TPW2 has several times created a microgrid for home (I’ve had 140 hours of blackout over 60 instances, 3 instances were 24hours+ each). The TPW2 never failed and keep the solar systems operating at its max during the day when the grid was down. I even got to share my excess power to my neighbour to help keep their fridge going and being to turn on a kettle.

    Sorry, IMO, I think this exotic setup is oversold/overpriced given the specs when compared to a Tesla PW2. Unless there is some glaring detail I missed that flips the opinion.

    I looked at Sonnenbatterie and was not overly impressed with it. Capacity vs cost vs warranty compared to the Tesla Powerwall 2.

    Looked at Redcell, ditto, too expensive for comparative capacity size, but yes flow batteries have complete 100% capacity after 10 years but not a major decision factor.

    I don’t know of any other integrated battery system that could do better than Tesla’s then/or Sonnen’s.

    • Aside from no cost advantage compared with other battery types, flow batteries have lousy round trip efficiency for domestic application and that significantly hurts ROI.

    • Hi Graham,

      First of all I’m not denying the Tesla Powerwall is cheaper or more compact if capital cost is your main concern this is not the system for you.

      The interest in the warranty is more about the 80% @ 20 years (Zenaji) vs 70% @ 10 years (Tesla) than the 22,000 cycles due to the inherently different chemistry. Granted the company is relatively new but the chemistry is not. Due to the energy density of LTO being less than NMC and LiPO4 it has not had the same market drives from the EV sector in recent years but for stationary applications it is ideal due to it’s longevity especially once the cost comes down further. It’s inherent stable chemistry makes it more suitable for residential applications and also LTO doesn’t have the cobalt issue although i understand that is being engineered out from NMC currently.

      RE the inverters, while both the PW2 and the SP PRO 481 have 5kW continuous rating, the SP PRO has a much larger surge power rating than the Tesla PW2 – 12kW for 30 seconds rather than 7kW surge power on the PW2. The battery obviously needs to supply that power plus losses too but the LTO can deliver high surge currents making it a good match.

      There’s a lot more you can do with the SP PRO such as smart control thorugh relays and additional DC coupled solar which expands your options in special applications or when grid capacity is limited.

  3. So let me get this right:
    – $12k for 4 x Zenaji Aeon batteries (7.7kWh)
    – $6.6k for the Selectronic SP Pro inverter
    – $? for installation

    That’s an installed cost of $19k-$20k for a 7.7kWh storage system. $2.4k-$2.5k per kW.

    Please correct me if I’m wrong, else that’s a certifiably insane price to pay for a home storage system. It would take more than 60 years for me to get such a system to pay back!!

    And the stated criticism over Powerwall’s (5kW) discharge limit being restrictive yet this Zenaji battery system is being controlled by a 5kW Selectronic inverter. How does this make any sense?

    • @Alex

      Doesn’t make sense to me…….

      For that price and a bit more, could have had 2 Powerwalls with 28kWh of storage!!

    • It’s all about the cost over 20 years and having true indefinite backup capable of surge power even during a heat wave – this system could run fully off-grid indefinitely. once you factor in you are only paying for one installation instead of two it makes sense. A PW2 installed will still be ~$13,500.

  4. Why does the above article not include the total system cost?

    A rough estimate is included in one of the above posts in response to the article, but, it is only a rough guess.

    And, with the Solar Analytics component, that, I believe, has an ongoing monthly fee, compared to an inverter such as the Goodwe GW5048D-ES, which includes three of the components above (for a great deal less cost, I believe); the

    A Fronius Primo solar inverter;
    A Selectronic 5kW SP Pro multimode (battery) inverter;
    Solar Analytics monitoring;

    -what is the cost (initial capital outlay plus ongoing costs), compared to the cost of a Goodwe GW5048D-ES inverter?

    And, what advantage would the Zenaji battery system have, over a 12kWh BYD LVS battery system (supposed to be good to up to 50 degrees centigrade)?

    How do the levelised costs per warranted kilowattt hour, compare between the two battery systems?

    • @Brett
      Not only a lack of information on total system cost but barely a system spec mentioned and little to no critical assessment of the claims made.

      What is the actual usable capacity? (~1.9kW/battery)
      Round trip efficiency? (Company claim is 96%)
      Actual charge/discharge rates?

      Unless the article got its information wrong, the estimate I made of the system price won’t be too far off the mark. The batteries cost $12k according to the article. They are available to purchase from various suppliers and prices are ~$3200 each give or take $100, so a bit over $12k for four is about right.

      You can look up the cost of the Selectronic inverter, available from various suppliers for around the $6.5k mark. Add installation costs and we are looking at $19k-$20k for the system.

      The Goodwe inverter you mention has a price around $3k.

      As to $/kWh over lifetime – whose/which lifetime? According to the company, these batteries will outlive me by decades, let alone the inverter/BMS used to control them or even the solar PV used to supply the energy to charge them. I see little point in paying up front for battery longevity one can’t use.

  5. Glad to hear of another happy Zenaji customer! I installed my (at least 24H Solar performed the installation…) system about 15 months ago. It covers the usage on my system during peak power price. I also have 4 batteries, with a Victron charger/inverter, & Solaredge SE5000 inverters with abt 15Kw of panels facing (50%)North & (50%) West.
    I saw these batteries at the Melbourne Smart Energy Expo in 2019, then installed just before Xmas 2019. The batteries are NOT Lithium Ion, but are Lithium Titanate. These batteries are commonly used in Electric buses in China. The advantages of these batteries is that they have an extremely long life & can be charged/discharged multiple times per day if required. Unlike some Lithium Ion, these batteries do not suffer from Thermal runaway.
    As far as Graham says, the life is based on the design of the battery & its industrial usage. It is a projected life, but they should perform for the time stated.
    One comment on the installation picture, my batteries are spaced apart for cooling (as per the spec at the time of installation) at about 150mm spacing.
    My Zenaji batteries have performed flawlessly. I am a happy end user also for this Australian product.
    I am currently considering installing Zenaji on my 2nd phase too!

    regards, Doug

    • @doug, it’s still a lithium-ion battery at its core. It just a modified one using titanate for its anode instead (as a negative electrode instead of a positive electrode used in for example in NMC – Nickel Maganese Cobalt).

      The bulk of it is still lithium and the ions is what make the charge and discharge process functional. Without ions moving, you can’t get the negative and positive flows. And lithium is the source of those ions moving. Hence it’s a lithium-ion battery.

      Whether one writes Li-ion-TO or LiTO or LTO, it is still a lithium-ion battery
      Tesla batteries are mostly NMC (NickelMagnesiumCobalt) or NCA (Lithium NickelCobaltAluminium).

      Lithium-ion describes the ion process. It’s inferred when someone says Lithium-xyz, that it is referring to a Lithium ion battery (Li-ion).

      Take a lead acid battery. It uses ions as well. But we don’t hear Lead-ion (or Pb-ion) batteries? It’s inferred that ionic processes are taking place using lead and water (H2O – to get hydrogen ions). Lead is main source here instead of lithium.

      It’s mainly a marketing term to call lithium batteries – Li-ion. It just sounds fancy when we say the suffix “ion” to give an air of some sort of technical wizardry.

      There are many many flavours of lithium-ion batteries. The additional chemistry after lithium is about the electrode component. Sometimes the word ion is dropped to shorten the chemistry make up. Take LiPoFe – Lithium-iron-phosphate, it is still a Lithium-ion battery. The iron does not replace or substitute the word ion. Technically, should be written as Li-ion-PoFe4.The absence of the word ion does not mean it’s not a lithium ion battery. An ion is a atom/particle/molecule that has a net electrical charge. Lithium is the chemistry for the ion movement. You can substitute lithium for other elements such as lead. The anode and cathode is what attracts/repels the ions. Titanate does not replace the word ion. It’s only describing the electrode chemistry. Lithium is still present and the ions are moving through lithium element. Marketing terms are often confusing when describing lithium-ion based batteries.

      The combination of different chemistries give rise to different characteristics and usually those chemistries determines the best application it’s used for when it comes to money, space and capability. For example a battery on a space station would need to be extremely high density to conserve space as opposed to a battery in a large housing on the ground.

      So, it comes down to application use. Personally, I don’t think Li-ion-TO or LTO is the best choice for home battery storage because the economics is simply too high for payback time (not to be confused with ROI which is wrong used again).

      • In addition…..

        Ion is not an element found in the periodic table. Ion is simply an electrical state. It has no identity as an element. It is derived from the greek word ἰόν meaning “to go”. That is to move from one spot to another. Faraday used this term at a time when it wasn’t fully understood how ions moved when metals were dissolved from one electrode to another which we now know as anodes and cathodes.

        Lithium, Lead, Nickel, Aluminium, Cobalt and others are elements found on the Periodic Table.

        The elements (and in some cases compounds – like Bromide) are describing the battery chemistry.

        It’s just easier to say Lithium-ion or Lithium battery rather than the mouthful with all the elements that make up the battery. But some acronyms like LiFePo4 make it sound fun to say – Life Po 4 or simply Life Po, instead of saying Li Fe Po 4 (like fee-fi-fo-fum in Jack and the Beanstalk). When taking latin shortform words and spoken as English, it gives rise to some english words that exist and gives way to confusion what’s been said. If you say to a layperson, I want a Life Po or Lee ion… they would be thinking what are you talking about – a life boat or a lion?

        Ditto for lead acid battery. We don’t say Pb-H2O battery for most car batteries. We just say a lead acid battery. There still remains an ionic process in the lead acid battery.

        However, we can say that alkaline batteries are not ionic because one can not easily reverse the flow i.e., recharge an akaline battery. It is possible but for very short times as they are prone to rupture. These are chemical processes, not ionic. These batteries are known as primary batteries. Can only be used once. Secondary batteries are known as rechargeable batteries.

        Therein underlies the whole process. Ions allows batteries to be recharged many times and safely. It’s a question what is the best base for it. At this point of time, lithium is the dominant flavour due to many favourable qualities and the ability to mix a number of other elements. Finetuning the lithium battery is where the other chemistries come in to add/improve/change its operating characteristics. Lead is another one, it’s cheap but doesn’t have the energy density of lithium. Bromide is another one. It’s ionic too, well it has to be to be a rechargeable battery. But bromide is a compound, not an element which requires something like zinc – a base element. The cost of chemistry is a deciding factor for its application of a particular battery type.

  6. koen weijand says

    LTO batteries are the way to go. Toshiba invented them and called them SIBC. they are used in industrial robot carts needing 5mins charge a few times a day. the misubishi IMEV EV has a a small series japan only equipped with them.
    volume and weight per kW is less favourable than li-ion, the low voltage is always good for cycle life. now prices are coming down , less than 1usd per Ah, but at lower voltage.

    • @koen… LTO is a lithium-ion battery.

      The other elements/compounds added to lithium changes its chemistry, therefore energy density per kg changes.

      There is no singular Li-ion battery. All batteries using lithium is a lithium-ion battery.

      What should be written is the LTO has a lower kWh/kg value than LNMC or LNCA. They are all lithium-ion. One can not say less favourable than li-ion. The chemistry must be stated. The presence or absence of the word ion makes no difference.

  7. Curious Graham, what’s your background? You seem very knowledgeable, are you in the industry or just do a lot of research? 🙂

    • @Paul

      I have a Electrical Trades Certificate, a Post Trade Industrial Electronics Certificate and an Electrical Engineering Certificate (issued 1989 but these days, it’s considered an Advanced Diploma Electrical Engineering based on the units completed). Almost started my Electrical Engineering Degree but switched to IT and have a Graduate Certificate in Computer Science (Networking).

      I remember very distinctly my Trades teacher back in 1982 talking about clean energy, solar panels and batteries was exotic tech for home energy systems.

      • Lee McCurtayne says

        I am very impressed that Australian product is received so enthusiastically, I would happily pay more for a product made here rather that seeing the money go somewhere else. We all want a successful product locally sourced here with the impetus for that manufacturer to grow and add to manufacturing in my own country. Having the “Cheapest” usually comes back to bite us on the proverbial. You see far more “Cheapest” TVs on the council throw out nights than the quality ones. How many “El Cheapo” solar panels suffer the same indignant end?.

  8. It seems interesting to me that people are wanting to make decisions beyond 5 years and insane that decisions and budgets are projected for 10 years and beyond.

    I don’t have the depth of tech knowledge of the participants in the discussion but there appears to be one huge missing point. The technology involved is in a fast ramp up phase, not even close to plateauing. With endless possibilities, i.e. Capacitor batteries, surely there is a possibility that in say 7 or 8 years time that batteries may have dropped by 30%+ and on top of that perform better. Then all of the other tech that makes up a system may be in the same situation.
    The bottom line is that any system purchased now is statistically unlikely to make financial sense in 5 to 10 years and almost guaranteed not to in 20.

    My decision is that I will most likely buy what makes the most economic sense now, knowing that I will be replacing it before 10 years are up with something totally superior and most likely less cost, so it will likely be a Powerwall or BYD etc.. I just hope some of my system may be able to be reused for something when I update or replace it down the track.

    History proves this – a Wang computer with 8Mb in 1986 cost $25,000 but it made sense for what it did. By 1991 we were buying AT286 computers with a 1000 times the power/memory for I think about $2,000. Are we not in a somewhat simular situation.

    • Lee McCurtayne says

      Sure Tech is it’s own worst enema , but we can sit and wait 20 years and by then we will be on that ramp up for another tech. Then again we can go for the cheapest of the cheap. Just wait till you see all those specimens are out on the council pick up night that just haven’t gone a couple of years andthe supplier has evaporated into another not so distant start up. Ther3 lies the great Aussie problem, we buy the cheapest of the cheap. You only get what you pay for.

  9. koen weijand says

    Moore’s law does not apply to batteries, progress has been historically slow. I’m still using recycleable OPSZ cells, does anyone know what that stands for: Offene Panzer Schiff Zelle , or WW2 Uboot battery.
    back in 2006 I talked to an engineer at the SAFT company (german for juice) , but its a french company making hirel Li-ion batteries for nasa and medical. as soon as the voltage goes over 3.95V the electrodes of a common Li (maganese /cobalt) battery deteriorate quicker. at 3.2V for Lifepo4 it already gets better. at 2.4V for LTO longevity is best up to 40K cycles if done slow <0.5C), still 10K for 10C charge. 20C charge would mean 3 minutes, very attractive , close to petrol filling your car ..

  10. Erik Christiansen says

    Apropos “For example, if the battery is cycled on a day where the temperature is over 40℃, the warranty reduces by five days.”, my reading of the warranty
    at the link is that the 80% warranty reduction cuts in at 35 ℃, not 40.

    That 5 ℃ lower threshold throws in many more warranty reducing days than the next 5 ℃, at least here down south. In this La nina year, I don’t recall a day over 35 ℃ this summer, so far, but if we have 20 in an average year, that reduces 22,000 cycles to 17268, and 30 days would be 15592 cycles actual warranty. (Though it’s the proportionate lifetime reduction which is the effective limit if we only cycle once per day.)

Speak Your Mind


%d bloggers like this: