harvests momentum to recharge batteries/capacitors

[myzter]

http://www.youtube.com/watch?v=ZwmixtM_Us0&playnext=1&playnext_from=QL

Using a series of kinetic generators, I mounted a device onto a modified mini electric ebike, Iam able to charge a couple cells very quickly.

I figure the mini bike Iam currently building, can double the power generation, the generator would be enclosed in a 7 x 7 x 4 inch enclosure.

The kinetic generator has so far shown to be scalable.

 

[myzter]

I have built a micro ebike which harvests momentum to generate power via the rear wheel, and is technically capable of 100+ watts with zero pedaling!

100% of the time the rear wheel is turning, power is generated, fed to the motor, batteries, or ultra capacitors.
Several wiring combinations are being tested .. which is great fun !!!

To engineer the gen setup from increasing the load or amps draw on the motor
I keep each generator output below 0.50 amp each
this is done by using a resistor as the load. I will probably end up using SP5T switch to utilize the required resistance

With 12.5v series capacitors setup (5x2.5v)
I wired them in on a booster switch, to place the caps in series before the motor after the controller.

I guess I have a question for the ES community...
what would be the preferred energy storage and what about using a battery and capacitor in tandem
example:
Lead acid
LiFePO4 / A123 M1's @ 8ohm
Lipoly
Capacitor / ESR 2.2 ohm (0.8 ohm caps on order)

 

[Momentous]

wow
very interesting myzter!

The new super capacitors can charge just as fast as discharge. So if you have the resistors setup to be switchable, and the voltage range plotted ... probably the limiting factor here would be the capacitance of the caps

 

[liveforphysics]

You can get lower ESR from LiPoly cells, and hold between 30-200 times the energy for a given amount of weight.

I personally use around 1kwhr of battery on a normal trip around town, which is half of my regular LiPo battery pack.

 

[Jeremy Harris]

If I understand this correctly, then what you're trying to do is build a pedal-powered generator with the ability to charge at a rate higher than that you could get from using regeneration, using the high charge current capability of super caps to maximise energy recovery.

The most efficient way to get around the very low load resistance that a fully discharged super cap presents to the generator, would be to use a switched mode constant current charger. This would limit the charge current when the cap was discharged, in the same way as your series resistors, but with much less loss (the resistors soak up a lot of power in limiting the current). Such a device should be fairly easy to build, as you only need current limiting really, rather than a true constant current.

I suppose the obvious question to ask is, would it be worthwhile? As a learning experience, I'm sure it's great. As a viable means of generating enough power to power an ebike, probably no. Even the very best regenerative systems only recover about 8 to 10% of the total energy expended, and only then when the conditions are good for regeneration (lot's of stops and starts, or hilly terrain, at modest speeds).

The bottom line is that you can't recover any of the energy used to overcome aerodynamic drag, tyre rolling resistance or frictional losses in the drive train; that energy is effectively "lost" for ever (in reality it's just warmed up the planet a very tiny bit). The only energy you can recover is that used to accelerate the bike to speed and that used to climb hills, which isn't much when compared to the total energy used, normally.

Jeremy

 

[Momentous]

In your youtube description, you mention you are using 6 generators, and two are not evenly matched.. with a total of 9 possible in your design
I can think of a good question: Are you going to redo your demo with all 9 generators ???

liveforphysics

You can get lower ESR from LiPoly cells, and hold between 30-200 times the energy for a given amount of weight.

I found that Myzter is using a 2.7V cap with 310F @ ESR 2.2
can you provide a link where I can view a lower ohm and higher capacitance LiPoly cell ???

 

[tostino]

He wasn't saying a single cell. He was saying for the same weight, and size you could use lipoly to get better performance than ultracaps.

 

[liveforphysics]

In your youtube description, you mention you are using 6 generators, and two are not evenly matched.. with a total of 9 possible in your design
I can think of a good question: Are you going to redo your demo with all 9 generators ???

liveforphysics

You can get lower ESR from LiPoly cells, and hold between 30-200 times the energy for a given amount of weight.

I found that Myzter is using a 2.7V cap with 310F @ ESR 2.2
can you provide a link where I can view a lower ohm and higher capacitance LiPoly cell ???

http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=9965&Product_Name=ZIPPY_Flightmax_5000mAh_6S1P_40C

These are about 2.5mOhm's per cell, which is nominally at 3.7v. You will find that even running 1P, this provides lower Ri than ultracaps for making any voltage pack. There are even lower Ri cells out there, but you pay a lot more than these cost. You get 6 of them all assembled in series in this little pocket sized pack.

As far as capacity, a 2.7v 310F cap can hold 1129J, or 0.313whr of energy assuming you've got a magic trick to take use of it from a voltage of 2.7 all the way down to zero.

The little 5Ah hobby city cell is 5Ah with a 3.7v nominal voltage, giving it 18.5whr's, or right about ~60times the energy holding ability of the super cap you mentioned, and unlike the supercap, the cell doesn't have to be discharged to 0v to extract all the energy, so your pack stays at a useful voltage for doing work throughout the range of discharge.

 

[myzter]

The caps are 62g each 2.2ohm, and only slightly larger than A123 M1 cell
Electric Double Layer Capacitor (EDLC)
Boostcap® Ultracapacitor 310.0F +20/-0%, 2.7V
L=61.0 mm, OD=33.0mm
-40°C to +65°C
max. peak current , 1 sec 240A
max continuous 30A
power pd 6400 w/kg
energy emax 5.06 w/kg

all ready on order for the new year 2010
650F only .8 ohm / any comparisons out there ???
power max 15,100 w/kg
max continuous 105A
max peak 1 sec 875A
weight .2 kg
58mm X 60mm

got some additional parts ..
all the 9 generators should be working / but not matched / with a video this week...
this last video has just 6 gen, unfortunately 4 were on the fritz
I might add, I burned alot of parts to get here...

this is all fairly new to me ..

New video
[youtube]1P5tmXbta6U[/youtube]

Dec 25 / 09 - found the gen set.. is not-calibrated

 

[liveforphysics]

Ehh, as far as things that compare with ultra caps, it's 5whr/kg compared to 130whr/kg for LiPo, and the LiPo still has a power density of 4,800w/Kg, and it's able to hold that for minutes rather than peter-out to nothing after a couple seconds like an Ultra-cap.

Batteries are all about energy density. When working with decent LiPo, you very quickly reach a point (like in a 2kg battery pack) where if the battery could discharge at 1million watts, or just 10,000w, it really makes no difference in the bikes performance, because you're controller/motor/bike etc can't handle unlimited current rates. Once your pack is able to meet the power demand your controller/motor can handle, then the only part that effects the functionality of the machine is the amount of energy stored in the battery.

For example, lets use the 650F caps, and let's say I add them to my normal 100v 20Ah pack. I need a minimum of 37 of them in series, which makes a 17.65F cap at 100v. This is 87,622 Joules of energy added to the pack, or 24.6whrs. It's going to add 7.4Kg to my pack, and increase the volume of the pack by a large amount as well.
My current draw limit will be no different at all, because it's current limited by the controller. The pack will have another 29.6mOhm resistor paralleled with it, and it was 18mOhm total before, so new pack ESR/Ri would change to be ~11mOhms. This would mean when I'm pulling 100amps off the pack, the voltage would drop from 100v to 98.9v rather than 98.2v, but this ~0.7v higher pack voltage would last <0.5second before settling back down to same 98.2v that it would be if I had no caps there at all.

So, if I decided to do 37 of the fantastic new ultracap you mention, I would add 7.4kg to my pack for an unnoticeable amount of additional energy (24.6whrs), and 0.7v less sag on the pack for <0.5seconds of discharge when drawing 100amps off the pack.

Or, for the same 7.4kg in the battery packs I linked above, you could add a full 1Kwhr of energy, 40kw of additional power ability, and take up a lot less volume as well.

 

[myzter]

Hello .. thats all very well... and keeps me well informed..
when charging a cap or liPoly, in a timed environment...
which can store the generated power the most efficiently ?
/// me.. I want voltage and quick recharges.. only a 5-10 second turbo boost is O.K. with me..
/// this bike has no pedals, planning a ride with minimal batteries makes the ride good fun!

Also Im not saying the caps are the best suited for this design, they probably make the process of charging more difficult..

I've yet to describe the gain in amps

 

[liveforphysics]

The gain in amps is super easy to explain. It's no different than if this were AC power, and you plugged in a 10:1 transformer drawing 1amp at 100v, and then drew 10amps at 10v off of it. You're current went up by 10x, but voltage went down by 10x, so power stays the same. In reality, every time you have a change or transfer of power, you always loose some of it. It goes through a brushed motor, and only 80% is kept, it gets generated back to electricity, and only 60-80% is kept, etc etc. In each step energy is thrown away as heat.

As far as storing power efficiently, those LiPo cells can handle 3-4C charging continuously, and easily 6-10C little bursts of recharge. If you're running say a little 48v 10Ah pack, that would be 2.88kw-4.80kw of power to regen. This is way more than your back tire could handle before it starts sliding. lol

 

[myzter]

OK, so when Im charging the caps, the voltage goes down, and the amps go up...
when my new watts up meter arrives (mid Jan. 2010), I will be able to test test test.. and collect some data showing just how much gain from constant regen with the different energy storage solutions, maybe possible

To charge the caps in a efficient way, I would need several parallel lines with varied Ohm resistance
With batteries, maybe 1-2 parallel lines.. just have to find a couple more meters .... and I'll charge them both

 

[liveforphysics]

To charge the caps efficiently, you will have ZERO resistors. Otherwise this defeats the whole point of having minimal energy loss from the low ESR of the caps...

This is yet another advantage of the battery, as it's voltage stays in a narrow range, maybe 30% change. No dealing with 0v caps acting like shorts issues.

 

[liveforphysics]

I will be able to test test test.. and collect some data showing just how much gain from constant regen with the different energy storage solutions, maybe possible :?

Constant regen would be the same thing as constant energy loss. From a system perspective, it's not a bit different than just hanging a big resistor across the battery terminals to be shorting and wasting power as heat as you ride around.


Also, if you're going to have no pedals on this bike, you're going to need to tow a trailer behind you filled with more mass in ultra-caps than you and the bike weigh together if you want to get anywhere without just scooching your bike along with your legs.

 

[myzter]

yeah,& on.. found the objective

I can dial in the gen output right into the: 0.19a - 0.31a range @ 1.9 - 3.1 watts respectively

By charging a battery of 3 -A123 M1 cells @ 9.74v resting voltage the gen can bring the voltage up 0.02v in a minute

However.. I can parallel the output from the total of 9 generators on-board the bike.

makes me think that maybe with - kinetic, wind, sun, and piezoelectrics the battery draw on a bike could be conserved by 50%
what does anyone think ??

 

[liveforphysics]

The wind and kinetic are only going to be steady losses on the system. The sun is an opportunity to make a gain. The challenge will be if the solar power gains can out weigh the additional weight and wind drag losses of having the array.

 

[myzter]

wind energy can be harvested in only 6 mph speed - the stringed instrument idea from http://www.humdingerwind.com
the body part could be out of 3/8 polycarbonate with an air intake and lined w/Evergreen panels
ya not the most prevalent but THEKPV ride has tight features

 

[Arlo1]

Lukes right. Constent regen will make the drive motor work harder and all you will get is a less eficent bike in the end.

 

[myzter]

Instead..
It seems....
I have utilized both a regen controller and an automatic sensing switch
the resistor I use, controls the pulse width.

8)

 

[monster]

i've no idea what your trying to achieve here but you bike could be adapted to act as a nice hybrid add-on of a car?

 

[liveforphysics]

Lets say you put a 200w wind turbine setup on your bike. Let's say you've got a no expense spared perfectly optimized carbon fiber wind turbine optimized for the speed you will be moving, and it's coupled to a CISRO motor used as a generator to give no cogging and 98-99% mechanical to electrical conversion.

In this best of the best situation, you're only going to be drawing about ~300-350w off your battery to generate that 200w back into the battery. You could save a ton of money, weight, time and expense and just remove the wind setup and parasitic generators and just replace them with a big 200-300w resistor you put across the battery terminals. It would run silently taking up minimal space while accomplishing the same end result.

 

[myzter]

ya ya boring..
more of the big bad resistor across the poor battery terminals again and again..

i have not mentioned anywhere that I need to generate 200watts, the post I made recently I provided result from the watts meter which were in the 2-3 watts range NOT 200watts

The S12 from Busch & Muller

due to a special negative feedback system that cancels out most of the electrical drag, leaving only the mechanical drag.

also the E-WERK

E-WERK provides power of up to 16 W, and all that with idle losses of only 0.03 W at 18 mph!

Or even the LightSpin from Lightspin AG

 

[liveforphysics]

It doesn't matter if we're dealing with 1w or 2,000w.

In the examples you linked, they talk about a 70% efficiency capable generator that can do 16w. This means, let's pretend it has it has zero mechanical losses (which may be insignificant compared to other losses), it means it draws 20.8w of energy off the wheel to supply 16w of electricity. Now put this back into your battery, (lets just pretend 100% battery charge efficiency), then run it through your controller, motor, drivetrain, (lets estimate very high at 80% system efficiency). This means you're drawing 25w off your battery to be putting 16w back into your battery. It's the same thing as throwing resistors across the battery to be draining 9w, except it takes up more space and weight.

 

[Tiberius]

Hi Luke,

I'm very interested in your comparison above between LiPo and supercaps. Sure, if the total energy is a major consideration then a battery is best, and, yes, the varying voltage on the cap can be a problem.

But the varying voltage can be overcome, or even exploited. What about the situation when you only want to store a small amount of energy for a short time, but the rate of storing and delivery is important. I'm thinking of storing energy when you go into a corner and then delivering it back when you accelerate away again. Does the supercap win out then?

Nick