Joule Thief

[Koen1]

alright then now I get what you're working on
So that switch thing using a photodiode or whatever, that you were
trying to figure out, that was intended to detect when night falls
and then switch on the JT+LED circuit?
Could you not use the solar panel for that in its entirety?
As in, sun goes down, solar panel no longer produces any output,
circuit detects the drop in output and uses that to switch on
the JT circuit... that kind of thing?

As for the JT battery charger, maybe this link could be usefull:
http://www.instructables.com/id/Joule_Thief_Charger
It's fairly straightforward though, only shows how a nearly dead
battery can be used to charge another battery that has more charge in it.
And the implementation is extremely simple: the guy just
hooks his rechargeable battery to the JT output directly.
So perhaps not quite as usefull an instructable but still, there's
a version of a JT charger idea.

Here's an idea: if we can use "ground radio" signals to charge caps,
we could run a JT battery charger on those, then use the charged battery
to run a JT with LEDs... And maybe you wouldn't even need a solar panel
to make your night light work...
Plus it could recharge while the LEDs are lit up too.
And of course this could be combined with your solar panel to produce
even more output...

@all: do you think it is possible to build a JT circuit that does not use
transistors, with minimal use of resistors? After all, resistors primarily
produce resistance, turning usefull electrical energy into waste heat,
which seems like squandering good electricity.
I ask because I've been wondering if a JT circuit was already possible
before transistors were developed... Sure, I suppose the right electron
valves ("vacuum tubes") with a control grid could, at the right voltages, be used
as a transistor, and I suppose the same could be done with a "magnetic
amplifier"-like transformer with a control coil... And Tesla of course used
spark gaps to produce the "make and break" effect...
But would it be possible to do so without spark gap or space charge zone
through which the voltage pulse must pass, in a solid state version?

Also, and this is another idea that just came to mind, could we not
use an air-core coil to produce a self-oscillating LC type circuit,
and then use this oscillation as the "make and break" for one or more
JTs? And of course if that would be possible, the next step would be
to see if we can't wind the JT coils on the air-core coil itself...
But maybe that's just a silly idea. If so, please point it out.

Regards,
Koen

P.S. The high-pitched soundthat hardly anyone hears is now,
after modifying a third camera circuit, confirmed and must come from
the magnetostrictive effects on the transformer cores. Apparently
some cores of equal size are not of exactly the same composition
and this results in different magnetostrictive properties.

[innovation_station]

koen1

im glad your here you have  great ideas...  cant wait to build them ...


on the light stick ... i was going to have the second transistor switch the aux cells and batteries on and for the cfl unit ...

and use the solar units as is  only to recharge the batteries the unit on the top will control both jt and cfl ...   but there is a problem where the transistor does not supply enough juice for the cam unit to operate ... there is lots of juce in the battery but  the transistor will not supply it

i think all i need to do to compleate that unit is find 1 that will handle the power .... 

or finally get gadgets mod to work  ......  i have not had good luck with his mod as of yet ..  but im sure it is because i do not have the same transformer and switch ..


i have  1 more ring to wind on my 54 10 mm super bright  pannel ...  this thing goes like .5 mile .... 

i used pin spots...  12 deg spred leds...   verry nice security light 

ist!!

[nievesoliveras]

@nievesoliveras

4/34 test results , 2n3904 all 22 awg coils , on ferrite (black ferrite that came from a pc used to be covered by white plastic)1 inch buy .5 inch, With a 1 k resistor 16 volts, tuned with a 20ohms i got 29.2 volts.

Mark

Thank you @mk1 !
I thought that it was going to give more volts.

What do you think on multivibrating the camera transformer as this guy does with his transformer and gets lots of electricity.
He uses 24 volts, we will use 1.5v with the same multivibrating idea.

His video link: http://www.youtube.com/watch?v=t3yvdXIjGSo

Jesus

[innovation_station]

so there you have it !!!


congrats!!

ist!

thank you jesus ..

[jeanna]

@jesus, that is a really cool circuit he has. transistors feeding into each other and also into the transformer.

@All,
Since I was the one who mentioned the multivibrator, I want to comment.

Although it is a good idea, I personally like the mag idea better. It is potentially fussy, but if the 2 magnets can be tuned to resonate, and the saturation of the materials will probably determine that, such that they switch on and off, I think we can get this going without the need for a switching transistor. Maybe the toroid or 2 toroids can supply the switching as well as the transformer-amplification function simultaneously.

I am happy to play with the transistor for now, myself, until I iron out the wrinkles in the designs that are not related to the transistor. Then I will be ready to try to remove it and tune the magnetic aspect of the cores.

I think MK1 is on that already.

@Koen, by the end of your post, I think you answered your own question about the many directions we are going.

I am happy with all the cross-seeding of ideas. I will be glad to return to this on March 2 when I return from my winter camping thing. I am still here til friday and will keep playing right up to the last minute, I think!

Thank you everybody,

jeanna

EDIT:
I was just going back to fetch the drawing from MK1 on that mag amp and I see, way before me, Gadgetmall has  put the astable multivibrator in the mag amp circuit. I was thinking along other times at the time and didn't pay close attention! It is always good to re-read this loong thread!  -j