Joule Thief

[kooler]

xee2
i forgot they had diodes in them
makes it more interesting.. to why it gives the ion breeze  with pos
ah i might cut the side off and bypass the diodes and see what happens...

[jeanna]

It works just fine BUT it does cut down on the light output total and I don't like that.  anyway, for what it is worth, here it is.

Thanks Bill.
That was well done.
Well, some people buy the unfrosted 100w filament light bulbs and hang them from the ceiling with no shade. It is definitely brighter, but it glares, and as I mentioned,  it is a funny thing that it doesn't always light up the room any better.

I am grateful to you for doing that test.

=======
@All,

I finished my lamp and darn... this is so strange.

I went to a lot of trouble to get the battery-rheostat-switch in line with the joule thief under the fixture.
I did all that so I could use the switch.
I even fed the secondary wires through the little holes around the bulb screw place.
I managed to get the thing so it went out when the switch was thrown, but it continues to drain from the battery.
It might be a faulty switch, but I bet it is because it is doing a joule thief radiation shuffle, or some other kind of dance.

So, it looks sort of like the other globe lamp except it has a rheostat.
The light is the 19 led LoA light so it has fewer bulbs than the 24 in that one, and it is beautifully smooth in the light it gives off.
I mean there are no bright light spots coming through the glass bulb.
Those strawhats are very nice in that way.
But, it is not really any brighter.

AND,
I measured the rheostat and 0.6v is going through when the light switch is off. !!!!! 

I guess I will need to make a magnet switch like Bill's, if I keep this, but I think it is not working.
Perhaps the thing to do is to make a growlight tube lamp out of it since the extra high voltage will be put to good use.
So, this was a frustrating, but a learning day.

-----
I have another question I am asking myself.
And, I will ask all of you.

We all know that we must put our leds in series because the led with less resistance will glow really bright for a while but it will eventually kill it because too much amperage is going through it.
OK, but that is for a normal dc circuit.

This HV has negligible amps, so I am wondering if the HV excitation of the phosphors makes that [keep it in series]  consideration unnecessary.
It sure would be nice, especially since most of those cheap Chinese flashlights which are overdriven use leds in parallel, and the modification is easy to just add a joule thief and remove 3 batteries.

I am wondering if anybody else has made any of these mods with the jt secondary, and can share some experience.
[gadget?, ist?, MK1?...anyone?]

You know, if it works, we could solder up say 20 or 24 leds in parallel and wind only 10 turns on a secondary and it will be done!
If it works...

thank you,

jeanna

[Magluvin]

Something I know of LEDs, if you have 2 in parallel after a limit resistor, usually only one will light due to barrier voltage differences in each. LEDs also have an overdrive effect when pulsed. Like circuits for laser diodes, I believe 10khz or so, the chip gets super excited then return to normal brightness after the excited stage during the pulse.
If you have ever shined an led (clear lens) on a wall or paper, you can see the chip and even the anode wire, tiny.
If you increase the pulse rate from low to high, you can see each side of the chip get bright 1 at a time, till eventually they all get excited and very bright. Soo the jt's are most likely taking advantage of this excited state and this can account for the brightness that you get from them, even at low currents.
But if you want them to be in parallel, use some really low ohm resistors for each in order to to let each be able to reach barrier voltage before another kills it for the rest as in when they are parallel.
Mags

[innovation_station]

Thanks Bill.
That was well done.
Well, some people buy the unfrosted 100w filament light bulbs and hang them from the ceiling with no shade. It is definitely brighter, but it glares, and as I mentioned,  it is a funny thing that it doesn't always light up the room any better.

I am grateful to you for doing that test.   { me 2 im gratefull for ALL you have done ! }

I am wondering if anybody else has made any of these mods with the jt secondary, and can share some experience.
[gadget?, ist?, MK1?...anyone?]

You know, if it works, we could solder up say 20 or 24 leds in parallel and wind only 10 turns on a secondary and it will be done!
If it works...

thank you,

jeanna

yes j

i can say i have driven 6 leds with no diffrance .. from 4 turn secondary and i have had 2 ...  i need to use a weeker battery ...  or there too bright and my transistor still gets too hot .. this is why i choose the CANS..  and cuz i have not used them b4 ..

that was a 4401 npn   

so in short if i want to run 10 leds from 1 coil ...  and you want a 30 led bunch maybe you need 5 turns ..  and 3 coils   never tryed it just about all my leds are burnt out .. need to some how get a few hundred thousand more ... lol

as long as all coils are the same .. it dont care ...  and DONT TAP BACK SIDE!

here is a pic

w

[jeanna]

Thanks magluvin.

But that is way too much work for me or anybody, which is why I asked.
However,  the rheostat or even a fixed resistor at the pos of the battery does reduce the voltage coming from the battery, so that aspect could be controlled som without too much soldering.

btw,
I am regularly using about 20k-50khz with my secondary circuits.
It was easy to make the one in series parallel.
Do you really think this applies here? Or, maybe does the high freq/hivoltage  just excite the whole array sort of like corona, and not really get into the led itself.

Thank you. This gives me something to add to my thoughts on it ,

jeanna