Will Super Capacitor Energy Storage Save The World?
Note from Finn: People have asked me if I would recommend buying a ‘super capacitor’ over a conventional battery. My short answer is no. Personally I think the technology is too new to consider it in a residential application. As Ronald says at the end of the post: “…they are a new thing and sometimes new things have problems that aren’t apparent at first.” Buyer beware.
Australian company Arvio is now selling a supercapacitor energy storage module. If you go to their site you can see the Kilowatt Labs Sirius module that stores 3.55 kilowatt-hours of electricity on sale for $4,950 including GST – here’s how it looks:
But GST may not be an issue because at the Smart Energy Conference in Sydney last month, Peter Wilson of Arvio said it would only be sold to trade, which means only to installers and wholesalers. So if you want to buy one retail you may have to look elsewhere.
The supercapacitor cells are manufactured in China and the module is assembled in Dubai. Arvio says the supercapacitor modules can be used as drop-in replacements for batteries and will work with any battery or hybrid inverter1. Three inverter brands specifically mentioned by name were Selectronic, SolaX, and Victron.
The modules have a full 10 year warranty that promises they will suffer no decline in capacity in that time and allows for unlimited cycles. This is far superior to the warranty of any battery on the market that I know of.
The Arvio Independence Day
As well as individual Kilowatt Labs Sirius modules, Arvio can provide an all-in-one supercapacitor storage system for use on or off-grid. They call it the Arvio Independence Day. In the United States, Independence Day is either about making tea in a harbour or aliens blowing up the White House. I’m not sure which. But here it is a large metal cabinet containing a Selectronic inverter, a charge controller, monitoring equipment, and generally two Kilowatt Labs Sirius supercapacitor modules for 7.1 kilowatt-hours of storage. More storage modules can be added if desired or you can have just one.
This is the Arvio Independence Day.
It’s capable of providing 7 kilowatts of continuous power and can provide more for brief periods. If there is a blackout it will take 20 milliseconds to switch over to off-grid use. It can be located indoors or outdoors, so presumably the final version won’t have the holes in the top you can see the picture above. The whole system comes with a full 10 year, unlimited cycles warranty with no decrease in storage capacity. It covers everything including the inverter, charge controller, supercapacitors, and the cabinet and its connections and electronics.
It’s not cheap.
With 7.1 kilowatt-hours of supercapacitor storage it will set back a homeowner roughly $25,000 including GST. But one of its features is if you move you can take it with you, so you don’t have to leave your investment behind.
This is not the Arvio Independence Day.
Advantages Of Supercapacitors
Supercapacitors have a number of advantages over batteries when it comes to energy storage. These include extremely rapid charge and discharge times and the ability to be cycled a vast number of times without degrading. Arvio says the Kilowatt Labs Sirius supercapacitor module has the following characteristics:
- Can be discharged 100% without problem.
- Fully charges or discharges in 30 minutes.
- Has an operating temperature from -30 to 85 degrees Celsius.
- 96% round trip efficiency from DC power into the module to DC power out.
- No chance of thermal runaway, which is a technical term for, “Oh my God! The battery’s set itself on fire!”
- A 10 year warranty that covers 100% of the module’s original capacity with unlimited cycles.
The warranty is far better than any battery I know of. It’s impressive even for a supercapacitor, which can be expected to fall to around 80% of its original capacity after 10 years of heavy use. But as heavy use for supercapacitors involves almost constant charging and discharging, which is not how home battery storage is used, Arvio may be confident there will be no deterioration in capacity over 10 years.
The claim that it fully charges or discharges in 30 minutes is interesting as its tech specs below say its maximum continuous charge and discharge rate is 100 amperes and its voltage range is 44 to 54 volts. This means it should take over 45 minutes, but the tech specs also say it can be charged or discharged at 300 amperes for short periods, so that may account for it. This is similar to a battery rather than the extremely rapid charging and discharging I would expect from a typical supercapacitor.
Technical Specifications
Arvio sent me a sheet of technical specifications for the Sirius energy storage module and I have taken a screenshot of most of it and shoved it below:
The self discharge rate of 5% over 25 days is excellent as I’d expect a normal supercapacitor to lose about half its charge over that time.
High Specific Energy For A Supercapacitor
With 3.55 kilowatt-hours of energy storage and a weight of 75 kilograms, the Kilowatt Labs Sirius supercapacitor module has a specific energy2 of 47 watt-hours per kilogram. But if we just look at the cells, Arvio states they have an energy density of 70 watt-hours per kilogram. This is more than 10 times better than a typical supercapacitor. Arvio says this is possible because their supercapacitors use graphene, which is a sheet of carbon one atom thick. The graphene is all scrunched up inside the layers of the supercapacitor, giving a huge area for storing charge.
I decided to check if the energy per kilogram Arvio claims is possible for a graphene supercapacitor and soon found this paper published in 2010. Here’s it’s abstract:
And here is an article written on it. According to the paper, the claimed energy density is definitely possible and they managed it 8 years ago, which is not an unusual amount of time for technology to get from the lab to the marketplace. Further improvements have been made since then as this paper from October last year says graphene supercapacitors were made that stored 148.75 watt-hours per kilogram. But note the abstract to this paper does not state this specific energy was obtained at room temperature, so it may not be as good as it seems.
Australia has also been working on graphene supercapacitors at Swinburne University in Melbourne.
This is apparently what scrunched up graphene looks like when magnified a hell of a lot. (Image Source: Physicsworld.)
While Arvio’s stated 46 watt-hours of energy per kilogram is excellent for a supercapacitor, it’s still not much better than a typical lead-acid battery and a module of lithium battery cells can store more than 3 times as much per kilogram. But when it comes to stationary energy storage weight and size are less of a concern because most homes can spare the required space and have strong enough floors.
Kilowatt Labs Supercapacitors – Poor Energy Efficiency When Fast Charged
Arvio have put out some videos on the supercapacitor and I found this one surprising:
They charge a 3 watt-hour supercapacitor cell in about 3 minutes and 40 seconds and say they have lost one watt-hour as heat. That means its charging efficiency is only about 75% which I would personally consider to be horrible for a supercapacitor. If it is the same when discharged, its round trip efficiency would be around 56%, which is terrible. I would expect a typical supercapacitor to have a round-trip efficiency of perhaps 95% if charged and discharged over one minute.
Unfortunately this poor efficiency isn’t only the case when it is rapidly charged in under 4 minutes. In this video they charge a cell over 15 minutes and I’ve taken a screenshot of their results at the end:
This means they needed 3.7 watt-hours to charge a 3 watt-hour supercapacitor cell. That’s a one way charge efficiency of 81%. Actually, it’s slightly worse than that as they didn’t wait for it to stop charging, which supercapacitors do by themselves when they are full. If discharge is similar we are looking at a round trip efficiency of around 66%.
Because the supercapacitor performs so poorly when charged over 15 minutes, I didn’t understand why Arvio gives its DC to DC round trip efficiency as 96%. So I sent Paul Wilson of Arvio an email asking about the RTE, or Round Trip Efficiency, and he replied:
“The 96% RTE on the unit itself is an actual test that we have done when running the turnkey unit at about a 1 hour charge and 1 hour discharge rate. This is more load than most applications would require. Very little heat is generated in this cycling test and the average over 24 hours was about 96%. We expect that we can get this to about 97% soon with some recent upgrades just introduced.”
So it seems that while the cells don’t act like I’d expect supercapacitors to when fast charged, they do appear to act like some lithium batteries when charged or discharged over an hour.
If we put the estimated round trip efficiencies along with Arvio’s figure on a graph you can see the large jump in efficiency when charged over an hour3:
As one hour is 4 times longer than 15 minutes, which is close to 4 times longer than 3 minutes 40 seconds, it allows for a good comparison between time periods.
Determining battery round trip efficiency by charging for an hour and discharging for an hour is normal procedure. And when used for home energy storage, the supercapacitors will often be charged at an average of less than this rate, though it will depend on the size of the solar system and household electricity consumption.
But no matter how Arvio determined their 96% figure, what they definitely need to do is change the 99% figure they give on the page where they are selling the Kilowatt Labs Sirius module. That is not correct according to according to Arvio’s own testing:
Misspelling efficiency is not considered a defence.
Kilowatt Labs
As the supercapacitor is assembled from Chinese-made cells by Kilowatt Labs in Dubai, I decided to check out the Kilowatt Labs website to see what they had to say about the supercapacitor module or just capacitor module, as it says on the front:
They had some interesting things to say in the second half of this video from 2:43 onward:
https://vimeo.com/152428341
- It can be charged in 30 seconds, which as we’ve seen above, would not be a good idea, especially with the huge amount of waste heat that would be created.
- It has a round trip efficiency of 99.1%.
- Its operating temperature is -25 to 80 degrees which is slightly different from the figure Arvio gives.
It made me wonder if they were even talking about the same supercapacitor module. It was difficult to tell because nowhere on the Kilowatt Labs site is there a picture of it. Fortunately, I was able to find Kilowatt Labs videos showing the supercapacitor module. Here’s a screenshot from one called, “The making of Kilowatt Labs Supercap Module,” that clearly shows the same labeling as the module Arvio is selling.
According to the video, it’s apparently made by having all the pieces magically fly into place.
It Is A Pseudobattery
I have mentioned the cells don’t charge how typical supercapacitors do and how the module’s efficiency when charged over an hour is similar to some lithium batteries. Overall its performance characteristics are much closer to a battery than a supercapacitor and, after giving it some thought, I realized this is to be expected.
The basic difference between batteries and supercapacitors is supposed to be that batteries store energy as a chemical reaction and supercapacitors store it as a static charge. But this isn’t exactly true. Batteries can store charge like supercapacitors and supercapacitors can store energy in chemical reactions. Both battery and supercapacitor cells are faced with similar bottlenecks getting energy in or out. Once supercapacitors start storing energy on structures similar in scale to the molecules that store energy in batteries their performance is going to become similar to that of batteries. The graphene that’s said to be used in these super capacitors is a sheet of carbon a single atom thick, so the scale is pretty small.
Because of their battery-like performance, some people have suggested that what Arvio has is actually a battery and not a supercapacitor, but I’ll extend Arvio the courtesy of assuming they aren’t dumb enough to buy a box of batteries and mistake them for supercapacitors.
Kilowatt Labs, in their video above, refers to their supercapacitors as a pseudobattery and that is as good a name as any. It’s not a battery but it behaves like one as far as the user is concerned, except it is a hell of a lot more durable. Or at least we hope it is a hell of a lot more durable, since that’s its main selling point.
Future Developments
Provided it performs as promised, Arvio’s offering is an impressive piece of technology. While it is pricey at the moment, its 10 year unlimited cycle warranty is bloody good. I’d say it’s a safe bet the technology will improve and the price will fall in the future. Ideally they will give batteries a run for their money when it comes to stationary energy storage. They may even improve in specific energy enough to become suitable for use in electric vehicles. But even if they don’t, stationary supercapacitors can be used to charge electric cars with stored solar energy overnight. It’s normally not economically practical to do this with home-batteries because it the cost of degrading the battery is generally not worth it compared to charging with off-peak electricity, but with supercapacitors this should not be a problem.
Another interesting application Arvio is investigating is using home supercapacitors to provide ancillary services to the grid that help keep it stable. Due to its apparent ability to supply these services continuously without degrading, if households are paid at the same rate for ancillary services as large generators, then Paul Wilson of Arvio says the supercapacitor modules will rapidly pay for themselves.
But I personally wouldn’t be in a big hurry to grab a supercapacitor module or two just at the moment. This is because they are a new thing and sometimes new things have problems that aren’t apparent at first. Personally, I would wait until the results of independent tests are published and I would also check out the reports of early adopters. Arvio has said they have had no problem with getting pre-orders, so there will soon be a considerable number of people out there with them. I am definitely looking forward to hearing how they perform.
Footnotes
- Also called multimode inverters ↩
- Specific energy is simply a fancy term that means energy per kilogram. Sometimes people use the term energy density for this, but that’s actually energy per liter. ↩
- Arvio’s figure is for an entire module rather than a single cell which should drag its efficiency down ↩
About Ronald Brakels
RSS - Posts
There has been some discussion about this Storage device and not all of it has been good.
https://forums.whirlpool.net.au/forum-replies.cfm?t=2707005&p=1
http://forums.aeva.asn.au/viewtopic.php?f=17&t=5486
See my article for non-engineers entitled “How to tell a Capacitor from a Battery” at
http://forums.aeva.asn.au/viewtopic.php?p=67687#p67687
Kilowatt Labs “supercapacitor” fire, before and after photos here:
http://forums.aeva.asn.au/viewtopic.php?p=69713#p69713
Pictures are NOT the same “before and after”, something fishy.
1) look at wiring config of far left box “before and after”
2) look at SIZE of second box from left “before and after.
3) the far left box sizes are DIFFERENT.
Other discrepancies in the photos too.
DONT BELIVE IT!!!!!
HMMMMMMMMMMMMMMMMMMMMMMM
All the boxes and wiring configs look the same to me:
https://www.dropbox.com/s/tzfx47khd7vho7n/sirius.jpg?dl=0
The plastic trunking with the BMV has melted off entirely, but that is to be expected.
Still, strange there is nothing more to this story, nor any comments from manufacturers or suppliers.
You have overlayed two different pictures from a person standing in two different locations.
Of course the images don’t line up!
Hi Andrew. I believe Guy’s point is that they overlay well enough to show the mistake that J made. J was counting from the left and failed to notice that the after shot was a wider angle and included a box that was outside the before shot, while the “box” that was second from the left in the before shot was actually a piece of plastic duct which has vaporised in the after shot.
I have one of these units that I purchased to evaluate.
My view was that if it worked out well enough I would purchase more.
While I had it in service, I was using it to boost my hot water system using it’s stored energy and the solar panels whenever there was excess energy.
At my place if the sun is out, this was usually around 11:00am.
Since my 1000Ah lead gel batteries started to fail (5 years) I have replaced them with a couple of BYD 13.8KWh batteries and an LG Resu10 (not currently in service)
While my system is now up and running, time is killing me for reconfiguring my system to accommodate the supercap and the Resu10 into their roles.
To note, there is no charge profile using an SMA sunny island inverter that will adequately charge the supercap.
I currently have a pair of sunny islands running the place and an Selectronic 482 (currently idle) that I will stick on the resu10 (noted the 5KW limit on the resu10) but am wondering what would be the best option for an inverter for the supercap?
I could use a Midnite classsic 200 to charge it but still will require an inverter capable of running the HWS which has a 4.2KW element.
Suggestions?
You’ve seen the photos. Why wouldn’t you treat it like an unexploded bomb, and take it back to whoever sold it to you? Clearly it burns exactly like a lithium ion battery, and clearly its BMS (if it has one) failed to protect it.
We installed 2 of these supercap batteries in 2019 in our off grid system and have had zero problems sofar (5 years). We have drained them to zero multiple times with no problems. Turn of all appliances and they restart automatically or push the restart button.
With prices similar to lithium ion and the 10 year swap out warranty, I think it is a no-brainer what to go for.
I’ve had a couple of Sirius supercap units pencilled in, in place of the Redflow, for just over a month, now. The major annoyance with the ZnBr battery was that the charge rate limitation prevented its direct connection between solar charger and battery inverter in a DC-coupled system. The mandatory regular 100% discharge also necessitated either a back-up battery or assured sunshine during the recovery.
There are buses in China running on supercaps, charging quickly through a pop-up pantograph, then zooming off for 20 km or so, using 1 kWh/km net, without overhead wires, according to videos on the tube. (Except at the charging points, e.g. terminal stop.) Regenerative braking helps there, as the supercaps can take it all.
I was concerned about how much of the energy stored on a capacitor (super or otherwise) is retrievable by an inverter, as it needs an input voltage in the general vicinity of its design voltage in order to operate efficiently. But reflecting on the fact that the energy stored on a capacitor is 1/2CV^2, we can expect that 3/4 of the stored energy has been withdrawn before the voltage drops by half. I’ll have to check the Selectronic inverter to see how low it’ll go. At low SOC, a capacitor has insufficient voltage to be useful, so the 100% discharge tolerance is of limited practical use, I expect.
The cabinet holes (if this is the final version) will have to be there to cool the equipment. A screen would be fine indoors… Ventilation fans would be needed if the equipment is to last long in this configuration outdoors. Nice article!
The LFP batteries on this side of the pond are about a thousand dollars US per usable KWH.
With the price per kWh of $4950/3.55 =$1394.37/kWh, compared to the (incresed) price for the Tesla Powerwall 2 of (about) $12500/13.5kWh = $925.93/kWh, I think perhaps, the Tesla Powerwall 2 might be a better investment, especially, given that the Tesla Powerwalls are (apparently, maybe) available to the retail public.
But, I think the even better investment, given the names involved, and, what they may represent, is the idea of the “aliens blowing up the White House”, especially, while the unknown life forms that have currently taken over the White House, are inside it. It would simply be yet another case of “USA friendly fire” – alien creatures killing their own kind.
Does the $25,000 figure include the whole box and all its contents i.e. Selectronic SP Pro etc? or is that just for the storage elements?
That’s everything.
When I completed the Tafe Stand Alone Course, we had a general discussion on combining a super capacitor and batteries. I would appreciate your thoughts. Essentially the super capacitor takes the peak load when a device starts, which prevents the battery suffering the shock load. Once the shock decreases the battery is able to supply the load whilst the super capacitor re-charges. 🙂
Combining a battery and a supercapacitor has been done and appears to work as expected. And example is the CSIRO Ultrabattery which combines a lead-acid battery and supercapacitor and is being sold in India for home use:
https://www.csiro.au/en/Research/EF/Areas/Energy-storage/UltraBattery
And I guess the Nissan Leaf electric car is kind of another example as it has a supercapacitor which is used to handle loads from its regenerative braking system.
Adam above is being diplomatic. I’ll be less so. If you read the thread over on aeva.asn.au, it’s clear that there are no supercapacitors in the Sirius box – it uses lithium titanate batteries. Every test that Arvio have posted on youtube shows that this is what they are. Arvio seem to be collectively unable to accept that, but it is clearly the case.
What is not known, is what kind of quality these cells are. Good LTO batteries might be a reasonable investment, although cycle life is temperature dependent. Cheap LTO cells – well who knows?
This means that the Kilowatt Labs claims for the performance of this device are completely false – especially those relating to temperature range, cycle life and charge time. I don’t suggest that Arvio knew this when they began their dealings with Kilowatt Labs, but given the interaction over at aeva, they must have doubts by now.
Berno, could you please share with us the basis of your inference that the units contain lithium titanate batteries? (Verifiable information would help a lot.)
The storage cycling video I have watched shows repeated half hour charge, half hour discharge of full capacity. It seemed to me that they had set out to check the datasheet usage spec of 1 million charge/discharge cycles, to the extent feasible. I am not aware of any other technology than supercapacitors, which offer 3.55 or 7.1 kWh capacity and such a cycle life.
That note 3 in the datasheet adds “Addional terms and condions, including a limited warranty, will apply at the time of purchase.” indicates to me that the claim has a technical basis, i.e. practical limits.
As for the hyperbole of the 99.1% RTE in the brochure, the datasheet admits:
“DC to DC Roundtrip efficiency (@100A): 99.1% (at caps) (tested to 96% at terminals)” Add that to the 1 million cycles, and the module doesn’t look much like any electrochemical battery yet seen.
I suggest you grab a cuppa and have a read of http://forums.aeva.asn.au/viewtopic.php?f=17&t=5486
Things are certainly heating up (excuse the pun) in the energy storage market – and not just in Labs in more. Bring it on!
Hi Ronald,
Have you got your hands on one of the units to independently test the claims of Arvio and Kilowatt Labs?
No, I have not. And I was unable to find any independent test results online. Hopefully the results of independent tests will be published soon so there will be no doubts about what they are.
We received our first phone call from a customer who has one of the Arvio units and was looking for advice on reconfiguring his Plasmatronics PL60 regulator.
So they are out there and available for paying customers, but haven’t yet seen someone with the means to do independent testing.
It’s a shame that customers are buying based on the claims before the independent confirmation.
Ronald, you could perhaps pick one up here:
https://solarbatteriesonline.com.au/product/sirius-energy-storage-capacitor-module-3-55kwh-48v/
Jason at solarbatteriesonline is enthusiastic, and mentions energy density improvements coming through from the labs, and lower self discharge figures as well.
That would be an excellent way to discover what is in there. The 3 milliohm internal resistance and very high charge/discharge capability do point to at least a supercapacitor front end. The 54v max voltage is equal to 20 * 2.7v supercaps in series. (Then bunches of that string in parallel to make up the capacity.
Now, 3.55 kWh is 12.78 Megajoule. At 54v, and 1/CV^2 for the energy on a capacitor, that’s around 8400 farad. But when 20 * 3000 farad supercaps (which will only take 2.7v each) are put in series, the string capacitance is 3000/20 = 150 farad total. Thats exactly 56 strings in parallel, for 1120 supercaps, at $4.18 per supercap, if the box and electronics are free. That’s crazy cheap for supercaps. (But expensive for the storage capacity.)
I’m waiting on the building surveyor’s autograph on my building application, but hope to have a roof up before the end of the year. For the off-grid system I want DC coupling. The Redflow won’t take the PV array output current, and one can’t rely on the inverter always taking lots of it. The high charge rate of the Sirius module is therefore enormously attractive.
A photo of the innards would be a scoop of major proportions, either confirming or dissipating the cloud of uninformed volubility whirling in that other pool.
Considering coals that BYD was raked over for their 10 year warranty in their review by solar quotes, have you see the warranty statement for these units?
https://www.solarquotes.com.au/blog/byd-brings-b-box-battery-battleground/
The warranty I’ve seen is this:
https://www.solarquotes.com.au/blog/wp-content/uploads/2018/04/kilowatt-labs-warranty-V2.0-2018.pdf
And if that is what is being given to Australian consumers then I believe there are some serious issues with it for sure.
Sorry Ronald, these devices are NOT supercapacitors. No supercapacitor or hybrid, or psuedo, or lithium supercacpacitor exists which can achieve the energy density that this device claims to achieve.
Until the cells are submitted for independent testing to prove what they are, the device is a fraud and should not be sold.
Ronald Brakels wrote:
“Because of their battery-like performance, some people have suggested that what Arvio has is actually a battery and not a supercapacitor, but I’ll extend Arvio the courtesy of assuming they aren’t dumb enough to buy a box of batteries and mistake them for supercapacitors.”
The thing is, Arvio didn’t have to be dumb, because the batteries do not look like batteries, they look like capacitors, and they have nothing written on them! But you can find devices that look exactly the same, being sold on Alibaba as 1.3 Ah lithium titanate (LTO) battery cells. And the tests Arvio have posted, show that their voltage/current/time behaviour is exactly like LTO battery cells.
Erik Christiansen wrote:
“A photo of the innards would be a scoop of major proportions, either confirming or dissipating the cloud of uninformed volubility whirling in that other pool.”
Arvio have posted videos showing the innards. You can see them here: http://forums.aeva.asn.au/viewtopic.php?p=67082#p67082
Erik Christiansen wrote:
“The storage cycling video I have watched shows repeated half hour charge, half hour discharge of full capacity.”
If you ignore what Paul Wilson is saying in that video, and instead look carefully at the data on the screen, it shows, in two different ways, that it is only cycling the battery by 57% each time (between about 33% SoC and 90% SoC). And it is doing it at a rate of 0.75C. This is extremely gentle treatment and provides no evidence whatsover for the claimed million cycles between 0% and 100%. You can see more details here: http://forums.aeva.asn.au/viewtopic.php?p=67188#p67188
Ronald Brakels writes:
“Further improvements have been made since then as this paper from October last year says graphene supercapacitors were made that stored 148.75 watt-hours per kilogram.”
Scientific papers quote watt hours per kilogram of /active material/. You can divide that by about 7 to obtain the watt hours per kilogram for a complete cell, as this also contains aluminium and copper current collectors, plastic separators, liquid electrolyte, a case and terminals.
Also, as you can imagine from that image of the crumpled graphene¸ it might not weigh much, but it takes up a huge amount of space compared to batteries, which can use the full volume of their material even though it is packed/stacked solidly. There isn’t anywhere near enough volume in the Kilowatt Labs case to fit enough graphene supercapacitors to store the amount of energy that it stores, even if they were the most perfect graphene capacitors that are theoretically possible, and cost was no object.
Ronald,
If this was ment to be a serious , credible , product review,…you have done your reputation harm.
You knew there was much controversy over the claims being made by Kilowatt labs and Arvio , yet you did little to clarify of check those claims or even ask relavent questions.
How can you review a product without even physically seeing it, let alone performing suitable tests ?
Richard,
Which parts of the post do you disagree with?
Please be specific.
Finn
The bit I disagree with is calling it a review without ever seeing, touching or testing.
https://www.solarquotes.com.au/blog/arvio-supercapacitor-battery-review
When I am looking for a review of a product, I expect the person to have one or have at least used the product.
So you don’t like the URL?
Not just the URL, but also the premise that this is a product review. If you google Arvio Supercapacitor review, the top hit is this site, with the subject and blurb as follows:
“Arvio Super Capacitor Battery Review: Interesting but Raises Questions.
Arvio’s Kilowatt Labs Sirius Supercapacitor, now selling in Australia, has the best warranty of any battery we’ve seen here. Read this in depth review of its claimed advantages over regular solar batteries.”
I disagree that this qualifies as an in-depth Review of the Arvio Super Capacitor Battery and it’s advantages over solar batteries. I’d suggest instead:
“Arvio’s Energy Storage: The Next Big Thing or Just Big Claims?”.
Then include a section at least acknowledging the now very detailed criticism and uncertainty around the spec sheet claims. Specifically, an independent test of the 80-85 degree operating temperature which would definitively reveal this to be a revolutionary new technology.
Ronald Brakels wrote:
“Kilowatt Labs, in their video above, refers to their supercapacitors as a pseudobattery and that is as good a name as any.”
Have another look at that video. They don’t use the term “pseudobattery” at all. Instead they combine 4 terms, some of which are contradictory: “double layer”, “pseudocap”, “non-chemical” and “battery”. The terms “double-layer”, “pseudocap” and “battery” describe 3 different types of electrode. Capacitors and batteries have two electrodes each. Even if we grant that they may be using the term “battery” very loosely, to mean “something that can store electrical energy directly”, it is a fact that all supercaps have one double-layer electrode and one pseudocap electrode, and a pseudocap electrode is most definitely “chemical”. Hence all supercaps are chemical. Only the double-layer electrode is non-chemical.
Pseudobattery would be a very poor name for the Kilowatt Labs device, since it is a battery pure and simple. “How do we know?”, I hear you ask.
Arvio have done the world a great service by showing us what the cells look like and by publishing various voltage/current/time tests that they have performed on individual cells. The latter show unequivocally that they have no significant capacitance (less than 1 farad) and that they look and behave exactly like lithium titanate cells.
Ronald Brakels wrote:
“It’s not a battery but it behaves like one as far as the user is concerned, except it is a hell of a lot more durable. Or at least we hope it is a hell of a lot more durable, since that’s its main selling point.”
And there’s the rub. Since we now know, thanks to Arvio, that they are batteries and not supercapacitors, we know their claims of 1,000,000 cycles and 45 years, and operation at 85 °C, are completely false. Mind you, those claims would be extreme even if they were supercapacitors.
I think we need some independent review of these Sirius Capacitor Module units.
My interest is perked but is perked by these claims and that has to be a warning sign.
If no independent test or certification of the unit exists then we should probably stay away. This article is also making way too many assumptions to be any kind of comfort.
Great article Ronald. You achieved your goal. You got people talking and demanding the facts without exposing yourself to any litigation from the company. I liked it.
There’s been an Interesting change of heart by one of the posters above over on the aeva thread …
I can’t wait for the consumer lines to open, with the second generation versions of these things. Hopefully they will price-match the LiPo market and drive it under.
One thing to remember is that the supercapacitors are all individually controlled by circuits. They use PWM controls, and are subject to PWM losses as well as the individual circuits “flaws”. That is why they behave more like a battery with a crappy PWM charge controller on them.
I am sure each cell, minus the circuitry, will deliver the 99%, but the circuits are not giving that at the terminals. Purposeful misleading marketing. Though, in all honesty, maybe the issue with the specific unit that you tested, was an issue of someone trying to force-feed the unit, and damaging it a little in the process. Though I suspect it is more of a “play on words”, like the whole “weights-per watts”, which always leaves-out the “required hardware controllers”, like Tesla does. (Add the price of the other required hardware and Tesla’s cost per watt is not the same. But still, much better.)
At the end of the day, unless they reduce the cost per watt, it will never replace a similar battery in a car. No matter how good it is. It will still be a novelty, as LiPo was, and now is not. It still needs more “tweaking”.
Hopefully the world will switch to 24v or 48v DC systems. Then we will all see some major gains. Using these lower voltages is killing advancement. They demand such high amps and controlling systems that just hold everything back. People are not messing around with wiring anymore. Time to stop nerfing our devices in fear that someone will grab an exposed wire. We even have safety features for that, which are cheaper and easier to manage than on 12v systems.
I have one of these on order.
Additionally I have a spare array made up of 20 panels into a Sunnyboy inverter and a spare Sunny Island inverter.
Would like to configure my existing static loads to be driven by this module.
If someone is interested in assisting with configuring of a monitoring solution and making it available to the web,
I would be happy to host the results.
My place is off grid on the northern beaches, Sydney.
Andrew
There appears to be some basic Physics that is being overlooked. It is this:
The Energy stored in a capacitor ;
Wc= 0.5 C^2 (1)
The total charge stored in the capacitor is then
Q=CV (2)
That charge came from a source (battery, solar panel) and the work done by that source is:
Ws=QV (3)
Put (2) into (3) and you get:
W= CV (4)
This is exactly double what is stored in the capacitor !
This is a well klown result from Physics. It does not matter how you control the charge process, the energy stored will always be half of the work done by the source that was used to charge the capacitor.
Hence if you claim that you can get 99% round trip efficiency, you are not charging a capacitor. That is where batteries, despite all their downsides, are more efficient!
Hans,
First, a little typo; the energy on a capacitor should be 0.5CV^2, in your equation (1) But after that, the question of half the energy remains to be answered. There are pageson the internet, showing that the other half of CV^2 is dissipated in the charging circuit, but they use a primitive resistor as the only voltage adaptation between the supply and the capacitor. I.e. they define that the losses will occur, by their incredibly inefficient circuit design.
In contrast, PV inverters test at up to 96% efficiency, and they briefly store all their energy on input capacitors. The switchmode power supplies in our computers and microwave ovens may test at closer to 90% than 96%, but they also transfer a heap of energy via an inductor for brief storage on a capacitor, this time on the output. They do not magically lose half the energy.
My impression is that some have forgotten their calculus, or lost view of the fact that the energy on a capacitor is the _integral_ of charge accumulation at an increasing voltage. When a capacitor is charged at constant current, the voltage ramps up at a constant slope. The graph of voltage over charge is a straight line, with the slope equal to the reciprocal of the capacitance.
The energy stored is the area under the QV graph, and we know that the area of a triangle is half that of the rectangle, i.e 0.5QV. (The integral is just a fancy way of calculating the area under the graph, more useful when the line is all over the shop, due to varying charge rate.)
With a DC-DC converter following the ramping capacitor voltage (Instead of a crude lossy resistor) we input e.g. 1/.90 *0.5QV joules if the converter is 90% efficient, and 0.5QV joules is what’s stored, i.e only 10% loss in the converter.
With a good converter, the physics then show that if we also had a thundering big capacitor behind it, then the two together would make a _more_ efficient energy store than an electrochemical battery.
All we need now is for Arvio to send one of their batteries to join in the government testing of serious battery candidates already under way, as described on the other Solarquotes blog. If the voltage does not go down to zero volts, or shoot rapidly to a battery-like voltage on charging, the we are permiitted to query the product’s description.
Sorry, I omitted the connection between the formula equivalents:
Energy on a capacitor = 0.5CV^2 = 0.5QV = 0.5 Q^2/C
Now paragraphs 1 & 3 above work together a bit more cohesively, hopefully.
As someone recently looking at super capacitors to use as backup power in my sailboat, ive seen that most caps arent made with Graphene. What you show in the picture above is activated carbon, not Graphene.
Why not set up a facility in Australia rather than importing these things from China? Talk about a missed opportunity.
Supercapacitors require 20 times the volume to store the same energy as a lithium ion battery. Using graphene in place of activated carbon doesn’t change the volume requirement, it only increases the conductivity and hence the rate at which such capacitors can be charged and discharged. But as we have seen with the Arvio/Kilowatt Labs device, which is in fact a lithium titanate battery, such batteries are already capable of charging and discharging much faster than is required for either home storage or electric vehicles.
The far higher volume of supercapacitors compared to batteries is inevitable, because capacitance is inherently a /surface/ phenomenon, while battery storage uses the entire /volume/ of the electrodes.
Dave – I think you’re probably right about the Arvio batts, but using nano tech like graphene can increase the specific energy density – you just can’t reduce the volume that much. At least the links below which are probably the world leaders in curved graphene are not making outrageous claims.
Take a look at these links:
https://www.youtube.com/watch?v=KQ2Eo6wl5r0
https://www.skeletontech.com/ultracapacitor-technology
Thanks Mondo. I had already seen those before I wrote the above. I believe my 20 times figure was based on the skeleton supercaps. I think they manage to squeeze 3200 F into the standard “coke-can” size whereas activated carbon only gives 3000 F.
Thanks ERik, I did have a typo (the V) was left off!; it should indeed be 0.5CV^2.
But the rest is correct and the power supply doing the charging has to supply twice the energy stored. It does not matter whether you use a resistor to limit the current or some other current control system, the energy source has to supply twice the energy stored in the capacitor. THe way one charges the capacitor will determine whether the other half of the energy is dissipated as heat, sound, light (spark) etc buy does not impact on the energy that is required to charge the capacitor.
I put this on the blog because it seemed to be overlooked by others. I am not trying to run down super-capacitors, quite to the contrary. They have many advantages over batteries as exemplified in the blogs and elsewhere. However, this is one disadvantage. Actually, I am looking to construct a super capacitor bank to augment my off-grid system.
magic pudding
The Sirius super capacitor is a magic pudding. First have a look at the Kilowatt Lab’s patent US20180076644, it describe a composite battery comprising of battery (lithium) and supercap array in tandem. Read the patent description “each of the batteries is electrically coupled to a corresponding super capacitor”. Other company making similar product will call it simply ultra battery. The only difference between them is that Sirus have somehow more sophisticate control circuit. Second just having a look at the weight of the Sirus magic pudding and in front of your eyes it turn suddenly in to a bullshit. The most technologically advanced Maxwell super capacitors have 7.6Wh/kg so 3.5kwh Sirus will be have minimum 467kg, not to mention a word “chinse”. This is a halve tonne unit not a 60-70kg as clamed by Kilowatt Lab’s. To have a first in the world affordable supercap storage device we need at last another decade.
Patience please.
Mark
Mark – not that your wrong about anything else but Maxwell are not the most “technologically advanced” – the just manufacture the best available individual capacitors. The most technologically advanced ultracaps would be Skeleton. But even they only claim about 4x the energy density of the rest.
https://www.skeletontech.com/all-products
They tend to produce modules used at the front end of regen braking systems and other high demand applications (like the Chinese buses) often backed by a lithium ion battery.
See the nano Nouvelle patents, Sunshine coast Qld
Dear Ronald,
I was suspicious about the 96% round trip efficiency.
The eqn: Energy = 0.5 x C x V^2 applies to capacitors,
yet the energy is supplied as QV. 50% dissipates as heat and if you charge fast enough as EM radiation. Typical foundational physics.
There seems no way they can have those results as a capacitor, it must be a battery and the subsequent claims typical of super capacitors does not apply.
Since the last discussion was in July, I went looking for more information.
At this site there’s a video of the test:
https://powerforum.co.za/topic/871-super-capacitor-batteries/page/5/
https://www.youtube.com/watch?time_continue=661&v=tD7MXTfumJs
This test showed a turn-around efficiency of 93%.
Something about cells made in China and assembled in Dubai that makes me feel unsettled.
Maybe that is just me?
January 19, 2020
kudos regarding Ronald Brakel’s SolarQuotes piece on the $4900, 3.55 kilowatt hour, 75 kg Sirius Energy Storage Module advertised by Arvio of Australia – – very interesting reading! However Arvio’s spec weight of 75 kilogram and price tag for the SESM seem much too low – – could these be double checked?
A 3000farad, 2.7 maximum volt capacitor cell stores 3.037 watt hr of energy so if the SESM stores 3.55 kilowatt hr it must contain 1169 such cells, regardless of their interconnection ( ;for the module to be rated at 48 VDC eighteen would be connected in series and then 65 of such series would be connected in parallel). Shunbin of China offers such 2.7volt, 3000f cells for $25 apiece (and Maxwell for $59) and it is difficult to see how Arvio’s SESM containing 1169 such cells can be priced at about 5 times less than 1169 x $20 = $23380. Also Maxwell’s 2.7 volt, 3000f cell weighs 1.124 pound so 1169 of them weigh 1314 lbs.How on earth can Arvio’s 1169 2.7volt,3000farad cell SESM only weigh 75 kg = 165 pounds, eight times less?
It would be of real significance if Arvio’s SESM specs weight, volume, and energy storage are indeed as they advertise. A Tesla automobile requires about 90 kilowatt hour of energy in its battery pack or 25 Sirius 3.55 kilowatt hr modules. But if 25 Sirius modules actually weigh 25 x 1314 lb = 32850 lb instead of 25 x 75kg = 4125 lbs then a capacitor powered Tesla sedan (or more realistically a pickup) is seemingly a long way off instead of fairly close.
Three years ago we took the plunge and tried these “batteries” out. 10 year warranty and an Australian distributor made us want to give them a go. Within a few months one of the modules died and was replaced quickly so felt the support was good. 3 years on the units only charge to 70% meaning we need to start the generator almost every morning. We are not new to battery life. We had hassle free running with lead acid for 8 years before one of the batteries died. Our installer, a generous and very supportive local fellow is trying to get at ratified refund from Arvio since they have been ignoring him for months regarding our problem (started at the beginning of 2021). He is worried too as we are not the only installation he has done. My recommendation: Avoid this product and this supplier.
My Arvio “supercapacitor” died around the 2 year mark.
Sent it to them for a warranty repair.
Arvio stated “No Warranty” as they claimed that the supercap pack had
water damage. (it looks like an overcharging issue as the BMS didn’t work correctly and there was no water anywhere near the unit)
When I got the pack back from them (shipping to and from them at my cost, Sydney to Melbourne and back), The Supercap pack had not been opened so I fail to see how they determined water damage as the cause.
Being an electrical engineer, looking at the design from an engineering perspective, the build was very agricultural. I noted that IC’s weren’t plugged in properly and there was a processor with a bent leg just touching to make contact as well.
I did take pictures at the time too.
I could have pursued it with fair trading but felt it a waste of time.
Hi Finn, you are being too kind. The Kilowatt Sirus unit does not use super-capacitors, it simply uses ‘Lithium LTO’ batteries – a technology that has been around longer than Lithium LFP (used in the Powerwall 2). Further, as you have touched on, their performance specifications are somewhat ‘fanciful’ to put it kindly – as many reviews have pointed out. So lets just call it what is – another Battery. True super capacitors do exist and simply are not suitable in this area of use.
Andrew
After my Kilowatt Labs supercap leaked and failed, I pulled it apart and have been using the cells for other purposes.
One thing that has entertained me greatly is my grandson’s toy car.
You know the ones that 3 and 4 year olds get to drive around in.
Well the one we had, the lead gel battery died so I repurposed a couple (3) of the “supercap” cells into it and charged it up.
That thing screams around and lasts forever!
I built a charge circuit for it and a couple of times a year they charge it when it is sounding a bit tired but has never missed a beat.