Joulethief SEC exciter and variants

[woopy]

Hi Xee2

thank's for proposal

i just tested it now

and without the diodes i only get a small noisy plasma arc for the same amp draw

and with the diode it melts really well youpp ???

I think it is time for you to get a go on this circuit

thank's for input

Laurent

[quantumuppercut]

Hi Xee2

thank's for proposal

i just tested it now

and without the diodes i only get a small noisy plasma arc for the same amp draw

and with the diode it melts really well youpp ???

Interesting. The way I see it is that more charges is being stored where the diodes junction meet thus giving more juice on discharge. It's like discharging 2000V 1 Farad vs 2000V .01 Farad. It would be less discharge frequent with the diodes if I'm correct. It might be better for oscillation because I think the diode configuration acts as a variable capacitor. I think it was Mike who taught me that.

[jimboot]

@Laurent are you using the HV diodes in yours? I have no diodes in mine yet. Added another 3 transistors though I couldnt work out where they went from the vid.

[woopy]

Hi jim

yes you need the diodes. I think 2 microwave owen diodes should work (not tried ) I did not try lower voltage diode, but feel free to try it.

I retry this morning and for sure without the diodes you only get a dizzling blue plasma arc and the steel do not melt.

I look at the melting point while melting and the best melting is when there is a small distance (about 1 mm ) between the 2 pieces of metall (the croco and the steel wire) and inside this small space there is i yello-orange arc.

good luck

Hi Quantum

Can you elaborate please or make a link to Mike theory, it seems to be interesting to explain this melting, and perhaps try to recover this power to do something more interesting than melting steel

Thank's

Laurent

[quantumuppercut]

Hi Quantum

Can you elaborate please or make a link to Mike theory, it seems to be interesting to explain this melting, and perhaps try to recover this power to do something more interesting than melting steel

Thank's

Laurent

Hi Laurent,

Mike just said you can put two diodes (opposite) to make a varicap. I just interpreted his wording into my own thinking. I believe there are certain diode that designed for varicap but I think every diode has some kind of parasitic effect as a varicap. I have to ask expert on this one.

This is just a possible explanation I think of. The reason I think it this way because if I connect a load straight to high voltage, it does not give heat, but when you build up charges and discharge, it gives heat. I think both process is the same energy. One possible scenario is that capacitance allow (waveform?) to build up. Of course this is all my speculation, you just have to see what makes sense and throw out what doesn't make sense. Thanks for experimenting.

QU

[boguslaw]

woopy

I see you are one step from ...... BIG THING

[boguslaw]

Here is a related question I have, because I'm working on something similar (improving inverter design) but woopy has the simplest circuit I believe.

How can I compute the wire diameter for required amps flowing and frequency ?
Of course I can find tables with values of wire diameter for 50Hz but that's not correct for higher frequency.

woopy, you should measure two things to get ou : frequency and voltage where you have spark. Then it's just a one single change and eventually modifying the amount of turns on output or input side. But the frequency might be still too much for running it from transistor .

There is one additional problem : careful overcharge protection or it can blow.


Congratulations man !

[boguslaw]

Isn't that Don Smith design ?


Please tell me one thing. Is we could have "lenz free" transformer then what would be the simplest way to extract max power from it ???!!!

Answer please ? I need confirmation if we think the same path....

[woopy]

Hi Quantum

thank's for input

at all replicators

Haargghhh i was experimenting the current draw wave on the scope at different voltage during the melting and suddenly i heard the really bad noise of the arcing inside one of the secondary winding and finish.
The coil now work only at very low voltage. So i have to rewind the secondary for the third time

This time i will try 0.5 mm with plastic insulation , I hope that i will get good and reliable result. I probably will not reach to wind 500 turns.

The arcing in the winding seems to be an issue in that system, so chose correctly your wire and good insulation.

OK back to bench


Ouups i was disturbed by my phone and i forgot to post. But in the mean time before dewinding, i decided to see where was the arcing originate from.
So i put more and more volts and suddenly the noise became very standard for plasma arc and i see the arc at about 10 turn from the end of a secondary.

So i dewound 10 turns on each sec and everything is OK for now oouufff

I don't know how long it will last but i decided not to go over 10 volts with this winding, but at 10 volts it melts very well

Hi Boguslaw

if i am near something, so everybody is also,
simply wind a torroid exactly as Sergdo (at OU forum) shows in his video if you type "Sergdo" in youtube you will find his channel with a lot of very interesting videos, and especially one where he lights a filament lamp with 17 watts input and seemingly 2 of those toroid and a very bizarre transformer.

Than apply the circuit from Xee2 some post ago (the second circuit which is how i have conclude after having seen the Sergdo video) et voila.

hope this helps

Laurent

[Bob Smith]

Hi All,
For what it's worth, this circuit did not work for me on a shorted single strand 4 inch diameter coil of soft steel wire. 22 winds primary, 44 (2 layers of 22) winds secondary. Tried 3 or 4 different FETs and variable resistors. Will try solid toroid next. Does anyone know if it has to be ferrite, or will powdered iron work as well?
Thanks,
Bob

[boguslaw]

Hi All,
For what it's worth, this circuit did not work for me on a shorted single strand 4 inch diameter coil of soft steel wire. 22 winds primary, 44 (2 layers of 22) winds secondary. Tried 3 or 4 different FETs and variable resistors. Will try solid toroid next. Does anyone know if it has to be ferrite, or will powdered iron work as well?
Thanks,
Bob

I'm quite sure only ferrite would work or you should have to add capacitor which would be tricky OR you have some special powdered cores , newest are made for high frequencies also.

[woopy]

Hi all

Yep this system is really much toooo sensitiv to arcing.

It happen 2 more times tonight with arcing and i dewinded the arcing and it get a step further and arcing and so on no chance with my enameled copper wire for my setup and long run testing. But what is positiv is that the 4 parallelled transistor seems to accept the work almost cold ??

Ok tomorrow i will rewind everything with really goog insulation wire youp

The most important thing here is that ( as per Sergdo thread ) i could get this steel melting at arround 10 watts,

if you are not puzzled with this result i think we simply have to travel to hell

but i prefer my way

just for info i was able to make a scope trace of the 3watt LED lamp but this time you see the image of the current wave. i think this is reliable as it was made before the destructive arcing in the secondary.

hope this helps

Laurent


th

[Bob Smith]

@ Boguslaw - Thank you for your response. I was hoping to get a kind of super kick going within the wire coil (it's actually rebar wire from Home Depot), but it was not to be. Oh well, now I know Will try to start winding tonight.

My sense from Sergdo's diagram is that the primary coils are wound counter-clockwise, and the 2 secondaries are clockwise. I don't know if this makes a difference, or what others have done; will try this on the toroid as well.

A bientot.
Bob

[xee2]

Updated woopy circuit diagram. I think this is correct.

[woopy]

Hi Xee2

thank's for the beautifull shematic

Just some corrections

The winding seems to be of atmost importance

so on the right winding (joule thief part) the down red output must go to the down blue output. On your design the red down output goes to the upper blue output( See at Sergdo picture ).

Than the resistor to base. I used a 1k 0.5 watt resistor but it become really hot, so i changed with a 5 watt 680 ohm that i have, and it is much better. I would propose to install a 1k variable resistor of 3 or more watt.

Finally i use 6 diodes per side (not 3 ) and there are 1N5408 (rated 3 A each ) i have not tried the 1n4007 (rated 1 A)

Finally as sayd above i used 0.4 mm enameled copper wire for the 500 turns. But the insulation of enameled copper seems to be too weak to sustain the inside current/voltage, and they are arcing after some minutes and destroy the winding.
So i will rewind today the secondaries with a plastic insulated copper wire and i am almost sure that i will not have the space for 500 turns. I will keep you imformed.

The toroid i use is a ferrite 60 mm outer diameter covered with blue epoxy coating, i don't know the material inside the toroid. Sergdo seems to have also smaller toroid, And Jimboot use a smaller one and get results also.

Hope this helps

But for all replicators
look very closely to the winding direction at the picture posted by Sergdo (simply copy the picture exactly, without thinking of confusing clockwise and counterclockwise and get lost and scratch your head each time you manipulate your toroid, after some training it is easy you will see )

Good luck at all

Laurent