Li-ion alternative

[lizardmech]

I have been looking around at recent super capacitor products that have been released fairly recently, it would be possible to build a 200Wh pack, weight is around 4kg cost about $1400 USD. 200Wh doesn't sound very appealing vs 1000Wh you find on high end EUC however the capacitor has some unique advantages. Very high discharge rates 30A continuous 80A pulsed, charge rate is the same as discharge, you would likely find regenerative braking is much more effective and charge time would be 15 minutes or less. Additionally they don't burn like li-ions and would survive 10,000+ cycles.

Do you think people would be interested in them? Price would likely come down a bit if enough EUC packs were being made with them.

[Hunka Hunka Burning Love]

What would the size be like? When I think of capacitors I think large cylindrical bodies that take up lots of space.  Could you fit them in the same space as lithium ion battery packs?  I'm not sure that people would go for 200 wh for a lot more money.

EDIT:  How about a hybrid lithium ion / supercapacitor pack?  Would there be any advantages of having half the pack as capacitors to help keep costs down?

[lizardmech]

Size isn't too bad, these new ones recently released are smaller. It would take up similar space to 16S2P with 18650 cells. Using them as a buffer might work as well, I asked the manufacturer if they have a recommended way to use them like that. I'm not sure if it's a just a matter of connecting them in parallel and the capacitors will naturally take most of the load due to lower resistance or if you need complex controllers to regulate the two.

[esaj]

Interesting, I looked at some supercaps earlier, but usually there were at least one of these limitations that prevented the use as high(er) voltage battery replacement:  too low maximum voltage, too high ESR for serious discharge/charge, too heavy, too large... and the big banks could cost quite a bit (hundreds or thousands of euros).

[Chriull]

4 hours ago, lizardmech said:

I have been looking around at recent super capacitor products that have been released fairly recently, it would be possible to build a 200Wh pack, weight is around 4kg cost about $1400 USD. 200Wh doesn't sound very appealing vs 1000Wh you find on high end EUC however the capacitor has some unique advantages. Very high discharge rates 30A continuous 80A pulsed, charge rate is the same as discharge, you would likely find regenerative braking is much more effective and charge time would be 15 minutes or less. Additionally they don't burn like li-ions and would survive 10,000+ cycles.

As far as i know with supercapacitors the voltage is proportional to the charge. Not like with LiIon cells between ~3-~4V but from zero to max (as it is with normal capacitors, too: U=Q/C with Q=Q0+I*t for a constant current I)

So for driving the motor a (high) power DC/DC converter is needed to get the voltage from the supercapacitor in the neede range of ~55-67V.

(Or other way round: If with the max voltage of 67V the supercapacitor would be charged to 200Wh and can be used down to a voltage of 55V only 200Wh/67V*(67V-55V)=35Wh could be used.)

For regenerative breaking a (high) power DC/CC converter is needed. The breaking force of the wheel is proportional to the current flowing. So the firmware has to controll the current of the DC/CC converter which charges the supercapacitor. (This could maybe be realised with the motor coils and the H-Bridge itself, if proberly driven by the firmware?)

So two converters are needed for max powers of ~3kW. If over the whole range they have some efficiency of ~85-95%  ~5-15% are burned as heat. Imho by now not practical for wheels.

1 hour ago, lizardmech said:

Size isn't too bad, these new ones recently released are smaller. It would take up similar space to 16S2P with 18650 cells. Using them as a buffer might work as well, I asked the manufacturer if they have a recommended way to use them like that. I'm not sure if it's a just a matter of connecting them in parallel and the capacitors will naturally take most of the load due to lower resistance or if you need complex controllers to regulate the two.

Imho the prob will be (again) the needed power electronics. 

Besides the above mentioned DC/DC converter to get the voltage up for motor driving and the DC/CC converter for regenerative breaking additionaly:

- a second DC/DC converter which charges the battery pack once the Supercapacitor gets over a certain charge.

- another DC/CC converter from the battery to the supercapacitor to keep him charged while "normal" driving.

Imho by now something for electric vehicles with more than one wheel, where all this could be transported and effectively cooled...

[lizardmech]

These cells behave more like batteries but technically it's not a chemical reaction so they are classed as capacitors not batteries. Voltage range is 2.5v-4v and you just use them with conventional BMS systems. They look similar to Lifepo4 performance but with enormous charge rates and without the fire and chemical dangers of li-ion and lifepo4. I'm not certain what is inside these ones beyond them being inert. Most of the supercapacitor startups have been using graphene and carbon nanotubes, so I assume it's probably those doped with some secret compound or something similar.

I have to double check but the charge/discharge rates were symmetrical from how I read the datasheet, assuming you could pull 1KW from a power socket you could charge a 200Wh pack in about 12 minutes. I'm not certain how much range you would gain from regen if it can cope with 80A pulses and store them you would possibly see much longer ranges than with a 200Wh battery. I'm not certain how much li-ion EUCs manage to retain vs just converting to heat.

[Keith]

The parameter I haven't seen discussed on this thread so far is self discharge or leakage. Capacitors are not exactly known for their ability to hold charge indefinitely.  There is going to be a huge difference in usability if they (say) lose 10% per hour over if they lose 10% per week. 

Either way using them as the temporary storage for an energy recovery system during braking does sound like a promising line of research - anything that improves braking performance has got to be a step in the right direction.

[lizardmech]

On the data sheet they sent me it says 0.1-0.3mA leakage over 72 hours. I might have made mistakes in my math but 0.0003A X 3.8V = 0.00114 watts (0.00114 x 72) * 18 cells would mean 1.4 Wh lost over 72 hours on a 200Wh pack.