LED Flash light ..Pulsed LED, underdamped oscillations

[Someone Electro]

its not a good practice to conect LEDs directly

 

[williB]

Ok i am in the process of winding new coils ..
i'll let ya know how things turn out..

 

[williB]

well i have finished it ,sort of ..
it very well might be the worlds largest shaker generator..
it is presently getting a little over 300mA .. i've maxed out my little analog meter , (250 mA max).. lol
i have increased the current output by eliminating the dead spots in the output wave..

 

[Someone Electro]

how good dose it charge an cap?

 

[williB]

it only takes several seconds to charge my 5700 uF cap array up to 30 V..

 

[audioguru]

Why did you use so many turns on your coils that the voltage is way too high? After shaking for several seconds the 5700uF cap charged to 30V will power an LED through a 1k current-limiting resistor for only about 10 to 15 seconds. You need to use a constant current source or PWM to keep the LED from dimming as the cap discharges, but 30V is too high for PWM since the peak LED current will be too high.

 

[williB]

Why did you use so many turns on your coils that the voltage is way too high? After shaking for several seconds the 5700uF cap charged to 30V will power an LED through a 1k current-limiting resistor for only about 10 to 15 seconds. You need to use a constant current source or PWM to keep the LED from dimming as the cap discharges, but 30V is too high for PWM since the peak LED current will be too high.

it is not the number of turns ,per se , that increased the voltage ,in so little time,as much as the transformer i used, but without the increase in the number of coils , the transformer wouldnt have helped..
in fact i used a 2K resistor ..
i will be using PWM and a constant corrent source..

 

[williB]

conserving energy

C2 is initially charged to say 30 V..
C1 is 0V..
R1 is 2K ..
LED is a white ~ 3 V forward voltage drop..
when switch is thrown C1 charges up to C2 voltage minus the LED diode drop..
But the beauty of this configuration is C1 still has about 27 V in it..
the problem is C1 only took about a second to charge up ..
but the energy is still in the system..and the white LED had a respectable 4 mA going through it..
am i just delaying the inevitable?
or can i use the energy stored in C1 to repower the LED later on..somehow??

 

[Someone Electro]

I like that design!

 

[williB]

when i put a 3V lithium battery in place of C1 (same polarity as C1, plus at top ) the led stayed lit for about 7 mins.
only using about 1mA current..
how about using a rechargable in place of the lithium battery or again a super cap. ..??

 

[audioguru]

Only 4mA through an LED for a flashlight?
Those very dim solar garden lights that don't light-up anything are about 4mA, aren't they?

The LED flashlights at the store use 24 ultra-bright LEDs operating at 20mA each. for a total current of nearly 1/2A, from 3 or 4 D cells.
That's what is called a flashlight. What's yours called? He, he. :lol: :lol:

 

[williB]

Well , when their flash light batteries burn out mine will still be going..
ps got a link to that light mentioned??

 

[Someone Electro]

You cod put in a seting of brightnes.

Soyou can have high brightnes and les on time and low brightnes and more on time

 

[williB]

You cod put in a seting of brightnes.

Soyou can have high brightnes and les on time and low brightnes and more on time

exactly ,
now i have to brush up on my pic PWM

 

[Someone Electro]

You can also use ADC to make it in an constant curent surce. and an nother ADC to conect an potencimeter to set the brightnes

 

[williB]

this is how i was able to get over 300mA ..

scope is set @ 5V /div 20mS /div..and scope is connected directly across the bridge rectifier and a high value resistor to cut down on ripple from the diodes..and without the transformer connected..

Out of 180 mS , two passes , 140 mS or so , of it , is a non zero value..

 

[audioguru]

I see a peak of 15V, an average of about 6V and hardly any current into your high-value resistor. How did you connect a current meter to measure 300mA?

 

[williB]

why do i bother..sigh

 

[audioguru]

Hi Willi,
Don't give up!
Maybe you could use 3 or 4 shakers as pistons in an engine, driven from a crankshaft. Then use one of your small windmills to spin it and it should create enough electrical energy to light a few LEDs brightly.

The reciprocating shakers would waste mechanical power, so maybe a smoothly rotating generator is best. :lol:

 

[williB]

i ordered (10) .22F super caps last night . At 5.5V they could hold 33J energy..
Below is a charging curve for one of the 1F caps i believe..so the actual charging time will be less per cap but i have 10 so i figure 30 or 40 seconds of shaking should charge them ..