Supercapacitors vs batteries. Capacity.

[jonescg]

Some sleuthing on the viability of supercapacitors:
http://forums.aeva.asn.au/viewtopic.php?p=66819#p66819

The Arvio Supercapacitor is simply a lithium ion battery in disguise...

 

[Sunder]

Considering the Toshiba SCIBs have ~50c rate both directions, and cycle life more akin to capacitors than batteries, makes me wonder what kind of application needs super capacitors over these kinds of cells.

You can raise their voltage to anything you want without loss if capacity, they are smaller and lighter, cheaper, narrower voltage range, etc...

 

[Buk___]

Some sleuthing on the viability of supercapacitors:
http://forums.aeva.asn.au/viewtopic.php?p=66819#p66819

The Arvio Supercapacitor is simply a lithium ion battery in disguise...

What happens if you interleave SCs and Li-ion cells in series?

 

[jonescg]

You end up with a Li-ion battery that takes up too much space and cost more than it should.

You would almost be better off putting a 3.5 kWh LiPo pack in parallel with a 7 kWh LiFePO4 pack. There will be some current dumping between the two, but it would have plenty of power and useful capacity.

 

[Buk___]

You end up with a Li-ion battery that takes up too much space and cost more than it should.

You would almost be better off putting a 3.5 kWh LiPo pack in parallel with a 7 kWh LiFePO4 pack. There will be some current dumping between the two, but it would have plenty of power and useful capacity.

It was just a thought experiment

I meant like this:

47.6A is 2C for the 7 Li-ions; and the SCs are separated, so (I think) they don't combine as capacitors.

The SCs charge pretty much instantly, but what happens to the Li-ions? For the few seconds the SCs are charging? And afterwards once the charge current is removed?

And once fully charged, the SCs give you instant, high amps for a few seconds and the Li-Ions give you capacity?

 

[billvon]

Okay. Let's do the math with the Maxell 3.2V 3.3Ah SCs from above.

To get 25V you need 8 in series. 3.3Ah translates to 3712F.

Not sure where you got that.

In general the best way to approach capacitive discharge is F=(A*S)/V. So if you want 320 amps for 5 seconds, dropping from 25 to 8 volts, you need 94 farads. To get 94 farads at 25 volts you'd series 10 Maxwell BCAP350 parts (2.7V, 350F) for strings of 35 farads - then put 3 in parallel. That's a total of 30 BCAPs at about $14 ea, for a total of $420.

Note that that is a constant current discharge, not a constant power discharge. So actual runtime would be longer, since you probably don't need 320 amps at 25 volts (that's 8kW or 11 horsepower). In addition, while starter currents can easily peak at 320 amps, they don't stay there for very long (which is why those 8-gauge wires you have to your starter don't melt.)

 

[Buk___]

Okay. Let's do the math with the Maxell 3.2V 3.3Ah SCs from above.

To get 25V you need 8 in series. 3.3Ah translates to 3712F.

Not sure where you got that.

In general the best way to approach capacitive discharge is F=(A*S)/V.

F = 3.3Ah * 3600 / 3.2 = 3712.5

 

[billvon]

What happens if you interleave SCs and Li-ion cells in series?

You get pretty massive imbalance. Putting them in parallel can give you lower ESR and more surge capacity, and can help if you are mixing short, very high current discharges with longer steady discharges (i.e. you have a rail gun on your bike.)

But for most ebikes you get more performance improvement by just adding more battery in parallel rather than a supercap.

 

[amberwolf]

the SCs are separated, so (I think) they don't combine as capacitors.

They're still in series in the circuit, so their capacitances are still combined the same way.

The SCs charge pretty much instantly, but what happens to the Li-ions? For the few seconds the SCs are charging? And afterwards once the charge current is removed?

They are in series with the SC so the current thru them is the same as the SC.

And once fully charged, the SCs give you instant, high amps for a few seconds and the Li-Ions give you capacity?

Not in series. In series, the lowest capacity unit is the limiting factor.

In parallel, that would be true. (whcih is why JonesCG suggested a small LiIon and large LiFePO4 pack in parallel).

IN the series case, the SC current is limited by the LiIon resistance (if it's higher than the SC), and in any case the LiIon would experience that super high current, which might be bad for them if they're not designed for it (some are, some aren't).

 

[amberwolf]

But for most ebikes you get more performance improvement by just adding more battery in parallel rather than a supercap.

Yes. There are a lot of supercapacitor and ultracapacitor threads discussing the particulars of this (it's almost as common an idea as putting a generator on your front wheel and a motor on the back, and cruising down the road forever on the free energy but at least the SC / UC works, if impractically large).

 

[Buk___]

But for most ebikes you get more performance improvement by just adding more battery in parallel rather than a supercap.

If you read the OP of this thread, you'll see that I wasn't looking to use SCs (for an e-bike or anything else),

And back up there I also calculated that it would cost $6000 and 54kg to replace my £200/2kg Li-Ion pack (and that was apparently a way underestimate).

So, once again, the question/diagram was just a thought experiment.

 

[spinningmagnets]

There are several youtubes showing a super-capacitor bank starting a car. The most common example is a pack of six Maxwell 350-Farad 2.7V supercsps, for a max of 16V. (Almost identical in size to a common D-cell), also Amperics five 3.0V for 15V max...

The main benefit is that...in extremely cold sub-zero weather, the common lead-acid batteries sag badly.

One youtuber finally did a test that I was curious about. He removed the fuel pump relay so the 4-cylinder engine wouldn't start. He then began turning over the engine, and it spun for about 30 seconds before the 12V super-cap bank died.

The stock lead-acid battery would struggle in cold weather, but...in warm weather it would spin the starter until the starter brushes fried...a very long time....

Edit: another youtuber tested the amps when starting a 4-cylinder. There was a split-second peak of approximately 200A to get the engine spinning, and a continuous 100A to keep it going...

 

[Hillhater]

Some sleuthing on the viability of supercapacitors:
http://forums.aeva.asn.au/viewtopic.php?p=66819#p66819

The Arvio Supercapacitor is simply a lithium ion battery in disguise...

Are they refering to this ?...
https://arvioshop.com.au/supercapacitor
It certainly breaks all the rules for supercaps.

 

[jonescg]

Are they refering to this ?...
https://arvioshop.com.au/supercapacitor
It certainly breaks all the rules for supercaps.
[/quote]

Yes, it looks like you could fit a conventional Li-ion battery inside the box with plenty of space for supercapacitors for window dressing. The Li-ion battery would provide the necessary power and storage while the caps are there to help with the marketing.

 

[fechter]

Are they refering to this ?...
https://arvioshop.com.au/supercapacitor
It certainly breaks all the rules for supercaps.

Does almost sound too good to be true. In most cases, that means it isn't.
Specifications:

Nominal Voltge: 48VDC
Voltage Range: 44VDC-54VDC
Capacity: 3550Wh
Maximum Charge Rate (0%-100%SOC): 100A
Maximum Discharge Rate (100%-0% SOC): 100A
Maximum Charging Voltage: 54VDC
Internal Resistance:≤3mΩ
DC to DC roundtrip efficiency (@100A):99.1%
Operating temperature: -30°C to 85°C
Galvanic Isolation:1500V
Projected Cycle Life3: 1,000,000
Projected Calendar Life1,3: supercap cell: 45 years, module control electronics 10-15 years
Shelf Life2:10 years
Warehousing: can be stored at any SOC without effecting cycle life

3550 Wh is pretty respectable. It does not say how large or heavy the unit is. Seems like it would have to use a DC-DC converter to maintain useful output voltage, but they claim 99.1% round trip efficiency, which does not seem possible with any DC-DC converters I've seen.

 

[jonescg]

3550 Wh is pretty respectable. It does not say how large or heavy the unit is. Seems like it would have to use a DC-DC converter to maintain useful output voltage, but they claim 99.1% round trip efficiency, which does not seem possible with any DC-DC converters I've seen.

75 kg and about 63 litres volume for the 3.5 kWh version.

They describe an "Embedded Li-ion battery for charge retention circuit" but deny that it's a Li-ion battery. It could be a lithium ion capacitor, which is a hybrid between a capacitor and a Li-ion battery cell - low energy density but high cycle life. Obviously they are in the sale phase, so information is hard to come by. But if there is a capacitor capable of the quoted figures I'd love to know more about it in the literature. There are a few installed in Off-grid installations in Australia and some cycle testing going on in-house. Will be very interesting to see how it pans out.

If you triple the size of a conventional Li-ion battery so that you cycle to 1/3 depth (or less) you will effectively quadruple the cycle life. One must consider the cost / benefit of course...

 

[Hillhater]

So, whar you are suggesting is that at 75kg, there could easily be 10-15 KWh of lithium cells (LTO ?) having an easy life cycling only 20% DOD, and using caps for "surge" loads (if necessary) to give the performance quoted .
Much more likely than the huge improvement in super cap performance implied by their own sales blurb.
Someone will tear one down soon to reveal the truth, after all, they have been on the market for over a year.

 

[liveforphysics]

Strange they would be 100amp rated in our out when that's just over 1C for the device, making it in the realm of decade old laptop cell performance but with half the energy density.

 

[Punx0r]

Unless there really are caps in there and the 100A limit is due to a dc-dc converter? Regardless, it all sounds a bit fishy.

New supercap tech targeting 10X the energy density of current supercaps: http://www.wired.co.uk/article/superdielectrics-supercapacitor-electric-car-battery

 

[Hillhater]

New supercap tech targeting 10X the energy density of current supercaps: http://www.wired.co.uk/article/superdielectrics-supercapacitor-electric-car-battery

That sounds much like a Robert Murry Smith device ?...
http://revolution-green.com/1-megafarad-supercapacitor-robert-murray-smith/

 

[TheBeastie]

I have a friend is is redoing an oldish campervan to live in for the foreseeable future.
For his battery pack to power all his stuff hes going with 3.6KWh lead-acid battery. which I think was going to be around $400 range.
I said why not go 18650 lithium, I worked out from nkon.nl it would be about $1100 for about 3.3KWh but of course its going to have around 3-4 times more total cycles and be lighter and require less space.

But for him its all about the upfront cost rather than the long-term cost as we all can see if you're going to be using it long term then you actually save money with the lithium because you're going to be able to use it 3-4 times as long.

And this is where OUR bias sets in vs Lead Acid.. And this is where I argue its the EXACT same thing with supercapacitors. If you can overcome the voltage drop problem and the charging problem and the space problem (which you can in a hybrid or pure setup for a campervan at least) then why wouldn't you want a "battery pack" that can do 1million cycles, as we know that is going to last much longer.
Because if a hybrid that does the heavy discharge work from the capacitor should save the batteries lifecycles significantly.

We all know that high discharge kills Lithium longevity/lifecycles, even on that other forum the guy forum was saying "lipo only has 200cycles" but we all know thats only true if you beat the shit out of it in heavy draw and its the exact same thing with 18650s as this chart shows with a modern 20700 cell.

For me its been frustrating argument because I don't mind people saying "I don't really care about the environment, I don't care about saving money in the long run, I just want convenience, a simple lithium/SLA battery is just far easier to manage etc." Because again, how can you compare even 2000 cycles vs 500,000 cycles to 1million cycles?
And I FULLY accept that, I am the exact same type of person, we pretty much all are, but I never ever seen anyone admit that, and I don't understand why that is so hard.

https://youtu.be/CzqmWgBSrHE?t=10m55s
Just watching that video on that all in one 3.55KWh supercap, looks very impressive and promising.
As he says at this point, hes been charging and discharging it about every 50minutes for months now and there hasn't been any degrading in total capacity.
[youtube]CzqmWgBSrHE[/youtube]

 

[Hillhater]

.......If you can overcome the voltage drop problem and the charging problem and the space problem (which you can in a hybrid setup for a campervan at least) then why wouldn't you want a "battery pack" that can do 1million cycles, as we know that is going to last much longer.
Because if a hybrid that does the heavy discharge work from the capacitor should save the batteries lifecycles significantly.

Can you overcome the voltage drop issue effectively ?
..or the space problem ..( especially in an RV ) ? ...10-20 times bigger than lithium !
..or the charge balance/BMS issues ?
Or the cost issue ?
There is a reason no one uses S'Caps for storage..?
SCaps are not well suited to capacity storage (ok for power !). And lithiums biggest drawback ..cost...is constantly deminishing.
I would have thought the practical solution for an RV storage battery is a repurposed salvaged commercial EV pack, ot at least the cells from one.
Typically, for that $1100 you could get a ready built, 5.3 kWh , 24v, Tesla module, that would only need to cycle 50% DOD to give more capacity than the 3.6kWh of lead, and have a cycle life that outlasts the RV or the owner !
Similar capacity Volt or leaf Modules would be even cheaper.

 

[spinningmagnets]

The Tesla Model-S has a 300-mile range. What if...part of the battery was swapped-out for a super-cap bank?

My son has a Chevy Volt, and he lives 8 miles from work. The perfect candidate for an electric car. Imagine the super-cap bank would only power a Tesla Model-S for 10 miles. BUT...it would charge up in two minutes. It would still have a 200-mile lithium battery on standby @ 4.0V per cell. If he charged at home and at work, the lithium would never get cycled at all...super caps are supposedly able to last millions of cycles. MILLIONS.

I'm not saying that all EVs and hybrids should do this, I'm saying it's an interesting option that I personally would like to have. The latest ultracaps use nano-graphene-aerogel (insert buzzword here) to greatly increase the surface area of the collectors (and range per volume), but...they are not made from exotic materials. Dirt cheap.

As for the half of the power pack that uses a power-dense chemical battery, solid-state batteries (no liquid electrolyte) will likely reduce the size of that part, and also sodium-Ion based battery chemistry will shrink the battery even more. Maybe small enough that there will be more room for a larger Supercap bank?

The smaller the 200-mile battery is, the bigger the supercap bank can be. Imagine a 50-mile supercap bank that charges in 10 minutes?

 

[TheBeastie]

.......If you can overcome the voltage drop problem and the charging problem and the space problem (which you can in a hybrid setup for a campervan at least) then why wouldn't you want a "battery pack" that can do 1million cycles, as we know that is going to last much longer.
Because if a hybrid that does the heavy discharge work from the capacitor should save the batteries lifecycles significantly.

Can you overcome the voltage drop issue effectively ?
..or the space problem ..( especially in an RV ) ? ...10-20 times bigger than lithium !
..or the charge balance/BMS issues ?
Or the cost issue ?
There is a reason no one uses S'Caps for storage..?
SCaps are not well suited to capacity storage (ok for power !). And lithiums biggest drawback ..cost...is constantly deminishing.
I would have thought the practical solution for an RV storage battery is a repurposed salvaged commercial EV pack, ot at least the cells from one.
Typically, for that $1100 you could get a ready built, 5.3 kWh , 24v, Tesla module, that would only need to cycle 50% DOD to give more capacity than the 3.6kWh of lead, and have a cycle life that outlasts the RV or the owner !
Similar capacity Volt or leaf Modules would be even cheaper.

Sorry, I always add to my posts but as I later, after I added and watched that supercap video demonstration, browsing this other forum talking about it all http://forums.aeva.asn.au/viewtopic.php?f=17&t=5486&start=25#p66874 , this is what I have always had in the back of my mind.
They have taken the raw supercap capability and added some electronic smarts to do all the hard work, and I think this is what everyones had in the back of their mind is why can't someone make a truly nextgen BMS to handle the voltage drop and charge problem of supercaps.

https://youtu.be/CzqmWgBSrHE

Just like how we have hated unreliable Chinese transformer based spotwelders that are heavy/costly to ship and don't last long, we have all felt someone could make a smarter low cost designed spotwelder with no heavy transformer thats just smarter and better, sure enough there are now two Arduino mosfet based spot-welders from folks smarter then us like the Kweld and the one just called "Arduino spot welder", both of these are from folks in Germany, as far as I know.

It's the same thing we supercaps, this is what I imagined someone would do eventually, use modern electronic smarts to BMS the crap out of supercaps.

Sure their are size problems of supercaps, but some applications can be quite acceptable for a pure supercap, that Sirius 3.55KWh super cap could fit in a campervan easily.
The other half of supercaps usefulness in the market place right now is just hybridization of lithium for short burst heavy draw like ebikes/evs. Most of the heavy draw from batteries is from speeding up from a stop which by far the biggest battery life cycle killer, if we can marry the supercap to do that workload with electronic smarts then surely the battery pack can last 10,000 cycles much more easily.

 

[Hillhater]

.....that Sirius 3.55KWh super cap could fit in a campervan easily....

Easily ?...maybe, but it weighs 75kg and costs $4k + ..and is rumoured to be mostly Lithium cells anyway !
3.6kWh of most any other battery would weigh less and certainly cost much less .

.

The Tesla Model-S has a 300-mile range. What if...part of the battery was swapped-out for a super-cap bank?

.... The latest ultracaps use nano-graphene-aerogel (insert buzzword here) to greatly increase the surface area of the collectors (and range per volume), but...they are not made from exotic materials. Dirt cheap.

..... Imagine a 50-mile supercap bank that charges in 10 minutes?

SM,.. Have i missed some significant new developments in S'Caps performance and price ?
Just how big and heavy would a 400v SCap pack have to be to power a Tesla 50 miles ? (15-20 kWh equivalent )
And exactly how "dirt cheap" would that pack be.
Further, exactly what technical advantage is it bringing to the game ?

PS. Note:-
... you dont need to imagine a 50 mile charge in 10 mins ...
A tesla supercharger can charge 20kWh in 10 mins already.