Author Topic: Flywheel power (Read 3128 times)

willib · « **on:** March 19, 2007, 04:57:23 PM »

i have been using a flywheel on an excercise bike lately

i have two 1 farad caps connected in parallel for two Farads total capacitance

they are connected to the alt output between the load battery and the alt.

how would one rate the power from a flywheel?

what i do is pedal fast and let the flywheel charge the batts for over a minute.

The starting current into two NiMH batts (in series at 2.73V rest voltage)is 597 mA and after a minute it is still above 250 mA

i have used a 12V taillight to limit the current but later i will try PWM to adjust the current into the batteries , {less loss from lighting/heating the bulb)

]

in the graph above the current was sampled in one second intervals .

i clicked 'start ' too late so i had to pedal a bit more , hence the rise in current at the beginning

it starts to decline at 10 seconds into the sample session.

and lasts for 114 seconds

i think by 100 seconds the flywheel had stopped or was spinning very slowly at that point, and the caps were discharging the last of their power into the batteries.

willib · « **Reply #1 on:** March 19, 2007, 03:16:25 PM »

this next one is with the PWM set to a low value of on time vs. off time

hence the low starting current.

the rotors stopped spinning around 100 seconds the rest of the curve is the caps discharging into the batteries.

you can see the batteries are still taking in 2mA after 400 seconds , thats over six minutes

willib · « **Reply #2 on:** March 19, 2007, 03:44:27 PM »

This one is with the pwm almost full on

It starts out with the current a little more than 3.5 Amps

at that rate of charge the Flywheel and caps only lasted about 30 seconds

i thought it was interesting

s4w2099 · « **Reply #3 on:** March 19, 2007, 07:34:54 PM »

teh area underneath the curve should be the amps/seconds. It would be nice to calculate it to know the amount of energy that the flywheel has at certain RPM. Assuming the alternator and diodes loss.

willib · « **Reply #4 on:** March 19, 2007, 07:44:25 PM »

i made a video of the minigen II in action , and uploaded it on YouTube, its processing , ATM ...

the motorcycle hub ,spins true as anything , but the minigen II's connection to it is not perfect , so it wobbles a little , thats why i made the support mount as flexable as possible..

the flywheel isnt going particularly fast in the video , i hope it comes out ok .

it just gives you a better idea on how it works.

willib · « **Reply #5 on:** March 19, 2007, 09:11:28 PM »

i couldnt wait for youtube so i uploaded it to photobucket

here is the link

http://smg.photobucket.com/albums/v200/harsh/?action=view&current=Video2.flv

tell me if it works for you?

s4w2099 · « **Reply #6 on:** March 19, 2007, 09:30:49 PM »

its working nice for me. Have you tried to calculate the total power of the flywheel with your data?

willib · « **Reply #7 on:** March 20, 2007, 08:09:12 PM »

this is part of the data file i have to work with , there are 115 lines in the whole data file.

each sample is at one second intervals

i have isolated the decimal with a pascal program, 490, 485 ect..

if i could add them up , all i would have to do is divide them by 1000 before dividing them by 115 to get Amps per second..

but i am having trouble

it all comes down to how the numbers are represented in the program ,

490 is actually three seperate integers 4 , 9, and 0

i cant find a procedure or function combine 4 , 9 ,0 to represent four hundred ninety...

09:09:44 AM 0.490A

09:09:45 AM 0.485A

09:09:46 AM 0.480A

09:09:47 AM 0.476A

09:09:48 AM 0.477A

09:09:49 AM 0.499A

09:09:50 AM 0.538A

09:09:51 AM 0.559A

09:09:52 AM 0.590A

09:09:53 AM 0.597A

09:09:54 AM 0.595A

09:09:55 AM 0.591A

09:09:56 AM 0.585A

09:09:57 AM 0.580A

so thats where i am at the moment , i could go 4 times 10^2 + 9 times 10^1 + 0 times 10^0

wow that would work!!

s4w2099 · « **Reply #8 on:** March 20, 2007, 08:47:47 PM »

nutty :-P

willib · « **Reply #9 on:** March 20, 2007, 09:03:12 PM »

hey i got it to work

for the first data file i got 30.0 , not sure of the units?

i get 33212 when i add all 106 three digit integers (not 115 because in the beginning i was still pedaling )

so if i divide that by 1106 , it should be correct?

thats 1000 because i essentially multiplied it by 1000 when i stripped off the decimal point. , and 106 because there are 106 samples

so the area under the curve is 30 Amp seconds?

it getting late , ..

willib · « **Reply #10 on:** March 20, 2007, 09:14:35 PM »

let me correct that 33212/ 1000 = 33.212

33.212/ 106 = 0.3133

sorry , did i mention i was tired.

RP · « **Reply #11 on:** March 21, 2007, 01:18:58 PM »

Willib,

212/106 = 0.3133 is the average current
212 IS the amp seconds. If you'd had 1 amp for all 106 seconds then you'd have 106 amp seconds.

Dividing this by 3600 = 0.0092 amp hours.

s4w2099 · « **Reply #12 on:** March 21, 2007, 05:59:27 PM »

Ok, I got almost the same results. Check this out:

17.5 + 12.5 + 5 + 2.5 + 1.25 + 0.63 = 39.38 a/s

39.38(amps per second)/60 = 0.66 a/m

0.66(amps per minute)/60 = 0.0109Ah (amps hour)

So you got there a nice battery.

willib · « **Reply #13 on:** March 21, 2007, 07:39:39 PM »

wow , and that was with a different data set!

very neat .

RP · « **Reply #14 on:** March 21, 2007, 07:48:35 PM »

Remember, it's amp seconds and amp hours (Amps X Hours), not amps per hour or amps/hour. I'm really not trying to be picky, I just don't want anyone to wonder why the math won't work if you divide.

s4w2099 · « **Reply #15 on:** March 21, 2007, 08:36:45 PM »

I think PWM will give you greater power IF the powersorce (propeller) is being stalled by the alternator. In this case apparently the alternator will not stall the flywheel so maybe thats why we have the same results from different datasets.

It must be fun to try to add very high quality bearings and see how much energy you can extract from the wheel when friction is minimal, (Fr->0).

willib · « **Reply #16 on:** March 21, 2007, 08:56:14 PM »

so is amp-seconds a useful unit of measure

for most of the test the battery volts stayed pretty much constant i think...

lets say that the battery voltage was constant for the whole test , that would give us watt-seconds a unit of energy stored in the flywheel

http://www.digitaldutch.com/unitconverter/

to be honest i dont remember the battery voltage at the time , but i was charging two NiMH batteries

if we use 3V as a voltage , that would give us 33.212 * 3 = 99.635 or 100 watt-seconds ..

the alt is definatly able to stall the flywheel, without question..

it is the load i was using at the time that was not using that much kinetic energy from the flywheel

earlier i was using the AC output directly to heat a nichrome wire , very hard to pedal and the flywheel would stop in seconds without more pedaling , not as much fun as charging a battery

s4w2099 · « **Reply #17 on:** March 22, 2007, 08:25:44 AM »

I guess so. Amps second times battery voltage would give you watts second. The same math that we use for regular batteries and alternators should apply to this situation too.

willib · « **Reply #18 on:** March 24, 2007, 05:42:50 AM »

The bearings that came with it are pretty nice

"I think PWM will give you greater power IF the powersorce (propeller) is being stalled by the alternator. In this case apparently the alternator will not stall the flywheel so maybe thats why we have the same results from different datasets."

the reason the different data sets were the same , is because i got the flywheel up to the same initial speed in both tests. as best i could

the amount of voltage rise , during charge up , was dependent on which batteries i was charging at the time.

the batteries that rose the least were my new Duracell NiMH 2650mA ( around $10 at staples) for four.

i've also noticed that when a rechargable (NiMH) is really low their resistance seems to be higher. and seems to go down as they charge up

i've also beefed up the wires between the alt and the rectifiers and the ones between the rectifiers and the two one Farad caps and the ones comming from the caps

i was getting tired of the old wires melting , the new wire is a multistrand 12AWG 300V , with gasoline and oil resistant insulation.

s4w2099 · « **Reply #19 on:** March 24, 2007, 08:51:08 AM »

I dont quite get it then. So if the power curve is the same what would be the advantage of the PWM for this application?

willib · « **Reply #20 on:** March 24, 2007, 11:24:25 AM »

This particular application was charging NiMH batteries

the pwm limited the current into the batteries

enabling me to get the flywheel , and the alt, really moving without swamping the batteries with excess current, thus using more of the flywheels momentum .

s4w2099 · « **Reply #21 on:** March 24, 2007, 04:15:03 PM »

Ohh I see. Thanx man that clarifies everything.

:-D

willib · « **Reply #22 on:** March 24, 2007, 09:53:20 PM »

sorry but i was in a hurry last time , so i thought i would get to the point.

i actually did not have an answer after reading your reply the first time.

then i got on the bike and started charging six NiMH batteries at once , two parallel strings of three in series , without pwm , and i got reminded why i used pwm.

without pwm the alt rpm is to slow and hard to pedal

with pwm the current is adjustable leading to faster alt rpm ,and more flywheel action

it feels right when there is a little resistance on the pedals , not too much , or worse , too little .

News:

Author Topic: Flywheel power (Read 3128 times)