Ultra Caps?

[ploma1]

Hi... here my 0.02 $ about super capacitors.
1.They are VERY efficient
2. They are very suited to be used for regen as they can be charged incredibly fast (that used to be the problem with the SLA batteries)
3. they have very high amount of recharge cycles.

They would be great on a bike which has to do a lot of stop and go and the regen energy could be stored (and re delivered) in/from the Super caps.

I believe we will see more from them in the near future (if more EV get popular)

cheerios, ploma1

 

[paultrafalgar]

For a Google Tech talk on ultracapacitors see: http://www.youtube.com/watch?v=sudQhWMnrW0
and for an IEEE paper on their application to transport see:
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?tp=&arnumber=4077340&isnumber=4077322
I seem to remember that batteries have a much higher capacity than ultracapacitors so the latter are only useful in combination with batteries rather than to replace them. This may be changing.

 

[lfairban]

Since this is undoubtly the best discussion of Ultracapacitors for use with e-bikes, I want to tell you about an Idea I have been thinking about.

It started when I saw this add: http://cgi.ebay.com/ws/eBayISAPI.dl...s=algo=SI&its=I%2BIA&itu=IA%2BUCI&otn=4&ps=42

on e-bay. I was considering using 10 caps in series with my 24V 300W Currie with 12 Ah SLAs and a 20A throttle controler. That would be 65F @ 27 V for $250. If my math is correct, you could charge this up to 24V and it would deliver the max amps my controler can handle for about 3 second even with the battery disconnected, and only lose 1V. Then it would lose another 1V in the next 3s, and so on.

Now, if you connect it in parallel with the battery and after the Battery Management System, it will still deliver the same performance and advantages as if you had directly connected it to the battery itself. The difference is that when the BMS cuts out the battery, the UltraCap will still power the motor. Depending on how bad the battery is depleted, the battery should come back on in a second or so and keep adding power, until it drops again of corse.

So, my point is that if you had connected the UC directly, it would have added maybe 20% to the range and number of charging cycles by lowering the C rate, that is making the discharges less deep. If you connect it after the BMS, it will seem like you are getting about twice the range because the depleted battery will keep charging the capacitors when the assist is not being used.

For me, that is a considerable portion of the time, because mostly I use the assist for launch at a light in traffic, and for climbing hills. I don't think the caps will help me with the hills, but I would really appreciate not having the battery cut out when I hit the throttle at a traffic light.

 

[TylerDurden]

@lfairban,

for ~$250 you might consider some LFP cells to serve the same purpose and add to your capacity too. The LFP in parallel would handle the peak current draw similar to a UC bank. Charging would need to be separate, but you get more capacity since they are storage cells rather than caps.

Search "hybrid battery" on these boards and see if it sounds reasonable.

 

[lfairban]

Re: TD

That is a good point, and I have been watching LiFePO4 and Li-Poly prices for a battery upgrade. I have also been watching UC prices, and they are not dropping as fast as I anticipated, but they are at a point now where they might be a viable option. One LiFePO4 battery has a cut off at 20V:

http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=4295

where my SLA will go down to about 10V before it cuts out. So, no matter what batteries you use, they are still going to cut out when I try to launch if they are suficiently depleted. I think this is because there is no back voltage at 0rpm, and they drain tons of current until you get going.

I think I was actually trying to solicit comments on the UC strategy. As you say, my next purchase should be better batteries. At some point, it may make sense to use the UCs with a smaller battery rather than using a larger battery without them. One reason to get the UCs first is that I could also use them with LiPO4 batteries when I get them. With SLAs, they are not really worth the price or weight, and would be for the most part experimental, to see if they work as anticipated.

 

[cerewa]

I was thinkin about using A123 cells as a substitute for ultracaps.

A123s paralleled with cheap low-C-rate LiFePO4 seems like not a bad idea to me...

 

[lfairban]

That is not a bad idea.

It would address the high current issue, but those batteries cost alot. They say they are abuse tollerant, but I haven't seen alot written about them. Can they be discharged down to zero volts, or do you need some type of battery management system.

You could get a 5Ah 26V battery pack for about the cost of a 65F UC pack. The nice part about the UCs is that you can completly discharge them and they don't care.

 

[swbluto]

OK, gotcha. I didn't know the farad capacity dropped when run in series. That kind of kills the whole thing from my perspective.

I will just double my AH capacity with more lipos and be happy.

Matt

Maintain capacitance and increase voltage by series-ly going parallel.

 

[recumpence]

That would require a HUGE number of caps.

Waaayyyy too much money.

Matt

 

[swbluto]

That would require a HUGE number of caps.

Waaayyyy too much money.

Matt

Gotcha. I just realized the cost of these capacitors yesterday.

 

[GaryMedia]

It seems that 50-Farad 22Ah ultracaps are in the marketplace now.

http://www.echo-electronics.com/blitzbzcap50050faradamplifierstyledigitalsupercapacitor-p-1076.html
http://www.amazon.com/AROUND-INDUSTRIES-PLCPBT50-Battery-Capacitor/dp/tech-data/B000E3V9FS/ref=de_a_smtd

I couldn't get much info beyond shipping weight (which seems to be about comparable to a LiFePO4 of similar capacity), but these have the innate advantage that they can be manufactured in e-bike friendly form factors, can support 20C discharge rates, can be charged very quickly, and discharged to zero. The highest working voltage I've found in my 6 minutes of research is 21V.

We need to contact some manufacturers.

 

[dirty_d]

same capacity as 2.72 3000F maxwell 2.7V caps
the other one seems like a lead acid battery with a 50F capacitor in parallel.

 

[lostcoyote]

i'll be really curious what EEstor has under their belt.

they're supposedly working on something that can rival LiPOs using capacitors with a barium titanate high dielectric material.... but so far, the company is being real quiet. not sure if it's all hype or not but they have signed an exclusive deal with Zenn motors and have some sort of thing going on with Lockheed/Martin.

in a few more years, there may be some really good stuff happening with ultracaps using exotic composite materials such as perovskite and/or nanostructures.

i'm afraid to say that maxwell ain't cuttin' it here yet.

 

[recumpence]

I used to have a contact at Locheed Martin. But, he moved on. Bummer, I would love to know what is going on (not that he would tell me). :wink:

I think caps are the best energy storage system partly because of the EXTREMELY high number of recharges they are capable of, not to mention fast charge and discharge capability.

Matt

 

[tropmonky]

I will just double my AH capacity with more lipos and be happy.

Matt

This is clearly the best way to go in your situation anyhow... I mean the only time ADDING to a battery cluster with CAPS is when you need more amperage than the batts can handle on their own. Otherwise one should always look to add more standard batts to make up for the amperage need. With the 20AH he won't have any problems with amperage and voltage sag, AND he gets more AH at the same time. You just can't beat that.

Kyle

 

[Lock]

JM Energy Corporation
http://c11asmtg.securesites.net/en/index.html

"Our Lithium Ion Capacitor has about four times higher energy density than a conventional electric double-layer capacitor and has a voltage of 3.8V which is much higher than the 2.5V of conventional electric double-layer capacitors."

14-18 Wh/kg

Represented in the USA by JSR Micro, Inc.:
http://www.jsrmicro.com/pro_othersIonLithCap.html

Sorry. No idea on price.

Availability per the JM site:
October, 2008(plan) Manufacturing facility of 300K cells/year
2009 600K cells/year
2010 1.2 million cells/year
2011 2.4 million cells/year

tks
Lock

 

[lostcoyote]

interesting research...
http://www.gatech.edu/newsroom/release.html?id=1362

 

[317537]

Yes, they would be parrallel with the battery pack. Normally capacitors just charge themselves according to how much current can be sent into them. Theoretically they have an unlimited recharge speed, though in practice they can only pull power from a battery just so fast.

Basically, an ultracapacitor bank would give out huge current for a few seconds to aid the battery with huge current spikes that would be drawn from it. The battery pack would not see such a huge amp draw, thus drastically reducting the amount of voltage sag seen under load.
Matt

Sorry to be a bringer of bad news but the ultracap loses farads when added in series. add the total capacitance up and divide the number of caps and you have the answer.

So 5 X 55 farad caps would equal 11 farad capacitance.

I am working on an idea I got from the zener light buld bms idea of adding a 10 or more caps at 2.5 volts in parralell.

Like this.
Rig up 4X5 watt Zener @ 6.8 v, 10 X 55 farad cap, 4 X 5 watt Zener @ 5.6 onto the battery terminals, the zeners need to be reversed and block all voltage below 12.4 volts if you pull regen the caps will deliver that current as long as the the first zener is open to 6.8 volts (or whats left of the battery when sagging) and the second one is open at 5.6 from what is left of the battery after going through the 6.5 drop and the 2.5 volts potential that the capacitor is holding holds zener 2 open at 5.6.

From what I figure you can get a voltage sag of below on a 12 volt battery of down below 10 volts before the 5.6 volt zener will cut out.

This is all in theory as I havnt got any zeners big enough to cope with regen current but the theory need some development, ive almost got 10 caps ready to try out.

The reason why i chose 10 caps is because I estimate the legs will be lucky to push 4/5 amps at that voltage at most.

I have a led torch I designed some time back and i use ultra caps, it charges with in a blink of an eye and burns for hours. I used 2 X 55 farad caps in series = 27.5

I am dubious to my zener recomedations, so far to say youll need more than 1 amp to deal with regen. Its a concept in the making.

Would a regulator between each cap return the farads?

At 2.5 volts there are plenty of componants that can regultate the greed of anything tied up to regen.

Any voltage higher than 12.7 volts and it gets dumped on to the caps,

 

[recumpence]

The farad capacity drop issue was brought up previously. That is the main reason I dropped this idea.

Thanks anyway.

Matt

 

[PaulM]

Does anyone know why the voltage limit on these ultra-caps is so low?

 

[swbluto]

Does anyone know why the voltage limit on these ultra-caps is so low?

Generally, to increase capacitance with "standard" dielectric material(the insulator between the plates to prevent a short-circuit), the plates are brought closer together which ends up reducing the "safe voltage" as it takes less voltage for the short-circuit to happen. (larger voltages can short-circuit across longer gaps in a given dielectric, including air. Believe or not, but "static electricity" is beyond 1000v which is why it seems to be able to jump across longer distances than the voltages we work at.)

 

[PaulM]

That makes sense, I thought it was going to be more complicated.

 

[swbluto]

That makes sense, I thought it was going to be more complicated.

I'm sure it can be.

But I'm not that knowledgeable.

 

[solarbbq2003]

how about a very very large trailer, with normal caps from junked tvs etc, um maybe not

 

[Link]

how about a very very large trailer, with normal caps from junked tvs etc, um maybe not

LOL

Current-gen supercap = 1/10 - 1/2 the energy density of SLAs.

Normal cap = 1/100 - 1-10,000 the energy density of a supercap.