LED Flash light ..Pulsed LED, underdamped oscillations

[williB]

LED's use very little current for the amount of light they produce , ie not much is wasted as heat..
pulsed LED's use even less..
i am thinking of a circuit that would use the least amount of current from a battery ..and still keep the LED bright..
by using pulsed current and an underdamped LC circuit , i am wondering if we could just get a couple of oscillations out of it, the battery would last a lot longer..
applications : LED flash lights

 

[audioguru]

Hi Willi,
My 6V Ultra-Bright Chaser project is blinding, has speed and brighness contols and its 4 AA cells seem to last forever. Its circuit with Cmos gates and oscillators (PWM for brightness control) use such a small supply current that there isn't an on-off switch. Since the LEDs are chasing around in a circle, they make my projected shadow move around too. The project is here: https://www.electronics-lab.com/projects/games/004/index.html

 

[Someone Electro]

Nice and simple thingy.

Got the use of those AOL CDs too.

 

[Roboticinfo]

reply

Or you could use a hard drive disk for a even better result.

 

[williB]

hmmm
without R9 your battery would last even longer..no ?

 

[audioguru]

Hi Willi,
Sure the battery would lst longer because without the current-limiting resistor R9, then IC2 would burn-out and the LEDs wouldn't light anymore! :lol:

IC2's max continuous output current rating is 25mA. Even though I have the outputs pulsing and not continuous, I always design things conservatively.
R9 also absorbs the difference in voltage when the battery is about 6.4V when new, and when it is 4V when it needs to be replaced. At 4V my chasers are still pretty bright. :lol:

 

[williB]

thats sort of what i ment ..but what i really ment was ,couldnt you control the amt of current flow with just a transistor ?
also i am thinking of just using two LEDs for the flash light ( torch light )
the input currentwould be limited to 100 mA for around 60 mS per pass..
the photo is the open circuit voltage 5V/div @ 10mS/div of my shaker tube output..
when i hook it to a analog meter and * shake * , i can keep it around 100 mA ..
so you see why i am trying to squeeze as much from my driving circuit as possible..
so i was thinking of using pulse width modulation and a LC oscillator ( underdamped) !! ? :lol:

 

[audioguru]

Hi Willi,
I don't think you can get a few volts at 100mA out of a shaker.
I tried an expensive LED shaker flashlight in the store. I shook it until my arms nearly fell off and it charged its battery or super-capacitor. I turned it on and it produced a few mA of light for a couple of minutes. It is useless as a flashlight. :lol:

 

[williB]

ok uncle scrooge,
i did not say that my shaker was store bought , i made it myself..
granted that as you increase the voltage required , your current goes down.
but as i said , when i hooked it up to an amalog meter i was able to keep the needle at 100 mA , with vigorous shaking i was able to keep it at 110 mA..

 

[audioguru]

A current meter is a dead short to the voltage. A red LED needs about 2V at 20mA, plus more voltage for a current-limiting resistor in case you shake it too fast. Can your shaker provide it? Maybe you should have a full wave rectifier bridge across the coil to double the blinking rate and protect the LED from reverse voltage. :lol:

 

[williB]

i had a full wave bridge across all four coils yes..
as you know ,that is the only way to get the meter to register anything..
now , i was also thinking of using FETs instead of diodes , because of the voltage drop required to get anything out of the diode bridge..
I just ran another test and am getting a little over a volt (scope reading ) when i hook the diode bridge output to the Ammeter , which by vigorous shaking just reported just under 125mA..

 

[audioguru]

I don't know why you have 4 coils, one is usually enough. I don't know where you measure 1V which isn't enough for an LED.

Just put away the meter and connect an LED. Are the flashes bright enough? Add a big filter capacitor and see if the light is smooth enough.

 

[williB]

the one volt (scope reading) is across the ammeter ..
four coils are four times better than one

 

[audioguru]

Your ammeter has such a high resistance that it has a high error, especially when measuring current in a circuit with a low supply voltage.
My ammeter has a resistance of 0.03 ohms and therefore with 1A it has a loss of only 0.03V. At 100mA, its loss is only 3mV. It also measures AC current with the same small loss. Of course my meter is a very accurate DMM. :lol:

 

[williB]

i never said my analog meter was a good one :lol:
it is a piece of crap :lol:
my DMM says that the analog meter has 3.4 ohms resistance..
the reason i was using the analog meter was because the DMM does not change as uniformly as a analog meter does..

 

[williB]

this is the output from the shaker with the output of the bridge rectifier connected to my crappy ammeter..
the pic header is supposed to say 1 Volt /div @ 20 mS /div

 

[williB]

i just connected shaker to Xformer to diode bridge to 5700uF worth of caps.
and was able to charge caps to 30 V !!

time ,at constant shake, to produce a given voltage

V     Time (seconds)
5     5
10   9
15   14
20   19
25   28
30   41

at .5CV^2
gives 5.13J energy @ 30 V

 

[williB]

the transformer is a small mains to lower voltage step-down transformer hooked in reverse..
with a 4:1 or similar transformer i'm sure i wouldent loose so much current in the process

 

[audioguru]

Hi Willi,
You shouldn't step-up the voltage so high to power an LED.

When connected to a 30V supply with a 1k current-limiting resistor, a 2V LED dissipates 56mW and the resistor wastes 784mW.

You can shake your generator for 41 seconds so it charges your 5700uF cap to 30V. If you connect a 2V LED with a 1k current-limiting resistor to it, it will light brightly for a moment then fade to nothing in 10 to 15 seconds. Not a good flashlight.

 

[williB]

I know ....
that is why i am trying to extend the time it stays lit.
i figure 20 min. of light for a half min of shaking would be a worthwhile flashlight..??