@ All:

I found this over on hackaday.com.  It is a pcb board project without the board.  I think this idea might help us get the Fuji circuit as small as possible.  If anyone is interested, I can come back and add the link but it is over on their site.  ***EDIT*** Link:http://hackaday.com/2011/03/31/pcbs-without-any-substrate/

@ Jesus:

Thanks.  That looks to be very interesting.

Bill

@ All:

I found this over on hackaday.com.  It is a pcb board project without the board.  I think this idea might help us get the Fuji circuit as small as possible.  If anyone is interested, I can come back and add the link but it is over on their site.  ***EDIT*** Link:http://hackaday.com/2011/03/31/pcbs-without-any-substrate/

@ Jesus:

Thanks.  That looks to be very interesting.

Bill

Very cool!!!

@Jesus

I think what you are trying to do is good in running a CFL and to charge the battery at the same time, but the problem will be to allow the CFL to put out sufficant brightness and have enough power avaliable to charge the battery.

You could have the CFL switching on and off so fast that it wouldn't be noticed by the eyes (saves power also) and when the transformer turns off capture the EMP and feed it back to the battery.

First of all, you would notice the CFL will work longer, then refineing the circuit should bring success, the goal being making it to work all the time.

Reading the last 2 pages of postings has been jogging my memory where some people managed to recharge a battery while running a small electric motor, they modified the armature winding contacts (snapped off the contact tag where the wire was soldered to the copper segment) so at every 3rd pole there was a small spark which automatically sent energy back into the battery. During this moment in time there was no current being drawn out of the battery and momentum carried the armature to the next segment so it was being kicked at every revolution, which is frequency.
Th faster the armature rotates per second, the more energy is avaliable, eventually there is more energy to charge the battery so it all keeps running.
 
You could do the same thing by using a 555 IC as a vibrator to switch two seperate transistors, either on to off in a flip flop fashion, let the CFL be switched on by TR1, when TR1 is off, harvest the BEMF from the transformer through TR2 to the battery.

Gary's idea of a ringer circuit should work as well.

I could be wrong here, but I think John Bedini drew up a circuit where one part of the circuit needed to be off or the battery couldn't be charged.

You mentioning here about only charging the battery by touching the battery with the diode working a split second has got me thinking about it all.

jim

Jim:

I like your idea of the pulses.  To me, pulses are always good.  I was just reading on another topic of the high dielectric strength of water and had maybe a crazy idea?  What if I were to take some of my clear plastic tubing and form it into a ring, fill it with water, and seal it?  And then, yes you guessed it, wind it as a toroid?  A water filled toroid.  At first glance, one might think that water would not have any magnetic properties but then, neither does air and folks use air cores all of the time.  One might even try filing it up with a ferro-fluid just to see what happens.

I don't know....just thinking out loud over here.  When I get a chance, I will give it a go.  If it works, then anyone, anywhere could make their own toroids and even vary the properties easily.

Bill

I think this is a excellent idea because ferro-fluids  material is composed also of oil , since oil and water don't mix the ferro fluid should react like a super fluid .

As long as the water is not ice , air should also be removed .

Btw , nice to see you all here .

You are still surviving down there in the land of Oz , Electric Jim .

Mark

@Jesus

I think what you are trying to do is good in running a CFL and to charge the battery at the same time, but the problem will be to allow the CFL to put out sufficant brightness and have enough power avaliable to charge the battery.

You are right!

You could have the CFL switching on and off so fast that it wouldn't be noticed by the eyes (saves power also) and when the transformer turns off capture the EMP and feed it back to the battery.

First of all, you would notice the CFL will work longer, then refineing the circuit should bring success, the goal being making it to work all the time.

I dont know how to make the cfl to switch faster with the original circuit.

Reading the last 2 pages of postings has been jogging my memory where some people managed to recharge a battery while running a small electric motor, they modified the armature winding contacts (snapped off the contact tag where the wire was soldered to the copper segment) so at every 3rd pole there was a small spark which automatically sent energy back into the battery. During this moment in time there was no current being drawn out of the battery and momentum carried the armature to the next segment so it was being kicked at every revolution, which is frequency.
Th faster the armature rotates per second, the more energy is avaliable, eventually there is more energy to charge the battery so it all keeps running.

I also read about that a few years back but have not implemented it.

You could do the same thing by using a 555 IC as a vibrator to switch two seperate transistors, either on to off in a flip flop fashion, let the CFL be switched on by TR1, when TR1 is off, harvest the BEMF from the transformer through TR2 to the battery.

Can you post a schematic?

Gary's idea of a ringer circuit should work as well.

I could be wrong here, but I think John Bedini drew up a circuit where one part of the circuit needed to be off or the battery couldn't be charged.

You mentioning here about only charging the battery by touching the battery with the diode working a split second has got me thinking about it all.

jim

I will find the circuit that converts 1.5v to 12v and post it here, pointing out where the sudden connection circuit is needed. (a switcher that connects the diode output for a fraction of a second once in a while).

@nivesoliveras and others

The flip flop sends a high (which is a positive voltage) from it's output, which can be used to switch on a transistor. After a preset time, it will then cycle off, the on-off times in the circuit are generally fixed, but it is possible to have it adjustable for the mark space ratios (the ON times and OFF times) to be made longer or shorter.

Here is a link to a site that gives a very good explanation for those who don't know what a Flip Flop thingie is all about. It has a PCB that has been populated, there are very few parts involved and this one is very basic, it has no way to control the times when the LED is on or Off, these have been are fixed by the designer via the chosen resistor and the capacitors. They are chosen because they give reliable starting of the circuit which can be tricky sometimes

The circuit can also be used as the main section to drive a iron core transformer to make a UPS, but that's for another day
 
http://electropart.info/schematic-diagrams/led-flip-flop-project.html
   
To alter the times that the LEDs are On and Off, you need to increase or decrease the value of the resistors or the capacitors separately, or you can change all of them.
You cannot alter the 470R resistors as these are required as a current limiter for the LEDs, but you can change the values of the 10K ohm resistors, I would change the 10K resistors to a 1000 ohm resistor and insert a 10k pot between the top of the resistor and the positive voltage supply rail. The 1k ohm resistor serves as a buffer between the transistors base so it won't be fed too higher a voltage.
You could also place a 10k ohm pot across the electrolytics, setting the wiper to the high or low side will either make the LED be on for shorter or longer times.
You could instead use a BC548 or a 22222 etc etc transistor.

The circuit as seen switches ON and OFF the two leds alternatively, what we need to do is use the circuit to feed pulses to a larger transistor (BD139) to switch on and off two circuits, the first one being the supply powering the CFL battery, the other switching the charging circuit.

So if the CFL can be made to turn on and off faster than we can see it happening it will give the impression it is always ON, during the OFF times the back EMF from the collapsing magnetic fields in the coil can be used to recharge the battery driving the CFL, this should be able to extend the time that the CFL can be driven by the battery.

One thing to remember though, this circuit is designed to work with 9v DC, we need to make it work with 1.5 volt DC.
That's the challenge.

Suggestions,
The LEDs could be removed, and a wire link be inserted in it's place (takes care of the 3v that is necessary to make the LED fire up). Without the LEDs it should be possible to make this circuit perform at a much lower voltage, the transistor turn on at .6 of a volt, use geraniums to bring this lower to .2 volt if necessary.
Make sure the transistor chosen can handle hi spikes though.

Take the Hi at the base of each transistor (leave the transistor in circuit) by soldering a wire from it to the base of another transistor, or solder a wire from the collector to go to the base of another transistor (BD139)

Feed the 1.5 volt through the collector of the BD139 and take the switched Hi (1.5v) out to drive the CFL

Do bacilly the same with the 2nd transistor from the flip flop that drives another BD139 whose Collector is soldered to the output coil and the emitter is soldered to the Pos of the 1.5 volt battery supply.

I haven't done this as I haven't got the bits here to see if it works, but it may or may not work, only one way to find out is to give it a go.

Testing the flip flop
A scope (CRO) connect the earth clip to the Negative side of the battery, the tip to the base of the small transistor/s will show you if the Hi pluses are on the output, connect the probe to the BD139 emitters and you should see the same pluses there also, that tells you the FLIP FLOP is working and the BD139's are firing (switching) OK, and you would be able to see the mark space ratios as well.

jim
     

A small modification to my last circuit.

I have now shown the POTs which are placed "across" the electrolytics, adjusting the settings to the left will make the capacitor take longer to discharge, moving the wiper to the right will shorten the times of discharge.

Setting either wiper differently will allow you to switch on and off the BD139s at what ever times you wat them to be on and off for.

jim