This is it !

[gmeast]

burnout

When you hook up an inductor to that you should closely monitor your FET to be sure it doesn't over heat and burn out on you.

Yes ... I'm going to slave another comparator to this, and that will sharpen it up. It also gets much better at lower voltages. I know that there are Fets who's gates are 'logic level'. I've never used them and don't know squat about them, their turn-on times, turn-off times, bandwidth, etc. Knowledgeable info' from anyone knowing anything about these critters is appreciated.

Greg

[gmeast]

When you hook up an inductor to that you should closely monitor your FET to be sure it doesn't over heat and burn out on you.

Yes and also, as the coil begins to saturate, it starts to become a 'joule' heater and begins to appear (to the transistor) as a short and then shortly after that "purple haze".

I'll also be running a rail-to-rail final output stage to insure positive turn-on and shot-off. Much heat is inadvertently realized when the device is not adequately shut off 'all the way'.

Also, with the simple PWM I posted, I am able to get an On-time of 1.7usec at 40kHz & on a 7VDC battery (dead 9V) & consumes 13mA and at 13.5VDC power supply it consumes 17mA. It's actually about a 6% duty cycle at that frequency. The rise time is 1.04usec and the fall time is 0.8usec. So I see no reason why this won't work as a PWM for driving a power component for our experiments.

I've added two pots to my rig that allows (very,very nearly) independent control over frequency and pulse-width. This I think will be important at 'tuning time'.

Greg

[cikljamas]

Hi guys !
I just came by to say hallo, and to tell you what i am doing just
right now...Well, i spent a lot of time last few days messing around
with Sucahyos stingo and tonight i made my first little joule thief...
I ll attach one photo with it (schematic and device), but i just saw
on youtube a few videos presenting big joule thief and i must ask you
(especially Crackahcrackah) what do you think about this stuff ?
I believe it has a large potential for charging and lightning cfl and
spare energy of the primary battery. Tomorrow i am going to make
big one, this thing is very, very interesting to me...But i ask you here
about it because i think that maybe we could even take something
from that project and use it in our, or better say your project, what
ever it is going to be at the end of the road ?

[crackahcrackah]

I ll attach one photo with it (schematic and device), but i just saw
on youtube a few videos presenting big joule thief and i must ask you
(especially Crackahcrackah) what do you think about this stuff ?

From looking at the circuit diagram on the 1st page of the respective thread, and about 7 pages deep, he's not isolating the longitudinal waves from the transverse waves after inducing spikes.

Unless I'm mistaken, that's essentially the same circuit he was talking about here: Bob Boyce replication fully disclosed

when he was warning me about sticking my fingers in the water with the lit LED and 200-220Vs running through it. The thing is, when you don't use symmetrical capacitors around the break you are not isolating the current from passing through that section of the circuit. This is not a trivial issue.

We're all dealing with the same inductive spike phenomena, but how it is manipulated after it's creation is just as important as figuring out how to create it and optimize it's occurance. It can be made safer to work with using Tesla's approach, and it can be used in manner's different to conventional, mixed signals (transverse +longitudinal).


If you build the Stingo for testing and learning just understand that those spikes are dangerous. Look at the Tesla 568,176 patent, and all his HF HV patents to see the familiar symmetrical placement of capacitors about the break, and the coupling of the line between the caps to another inductor, functioning as the primary, which excites the secondary, that the load /lights hook up to.

[gmeast]

joule thief

Hi guys !
I just came by to say hallo, and to tell you what i am doing just
right now...Well, i spent a lot of time last few days messing around
with Sucahyos stingo and tonight i made my first little joule thief...
I ll attach one photo with it (schematic and device), but i just saw
on youtube a few videos presenting big joule thief and i must ask you
(especially Crackahcrackah) what do you think about this stuff ?
I believe it has a large potential for charging and lightning cfl and
spare energy of the primary battery. Tomorrow i am going to make
big one, this thing is very, very interesting to me...But i ask you here
about it because i think that maybe we could even take something
from that project and use it in our, or better say your project, what
ever it is going to be at the end of the road ?

Hi cikljamas,

We should use anything that can provide free energy for the average guy ... or anyone else for that matter. Personally, I think that the basic joule thief circuit should be required in any device that's intended to run off of disposable batteries. Can you imagine the waste of energy, the energy thrown away into the trash (every minute) in the form of batteries that can no longer power the device(s) they came from? IT'S CRIMINAL!

On a lighter note, I added in the last Op-Amp on the 2060 Quad to use as a second comparator & buffer stage. I now have independent frequency and pulse-width control (I think I mentioned that already). At 20kHz I have 3.5usec on-time, 2usec rise and fall times for 12 V peak to peak. At 40kHz I have 3.6usec on-time, all other parameters the same. At 5V peak to peak I have 1.9usec on-time, 0.9usec rise and fall times & at 20kHz I have 1.8usec on-time and all other parameters the same. At 50kHz I have the same exact as at 20kHz. At 50kHz, the power consumption is 22.25mA and at 20mHz ir consumes 10.6 mA. At 60kHz, the 12V P-P it consumes 40 mA of juice. What I really need is a logic-level FET ... anyone got some?

Are these numbers fast enough for this effort ... I mean for pulsing the inductors we have (or are going to have/make/buy/whatever)?

Anyway, I'm on the toroid and wire quest, or if anyone can suggest a ready-made one then Great! and thanks. I'd also appreciate a suggestion for the power components.

Greg

[Farmhand]

Hi all, now that i've had a bit of a play with stingo and spark gap, I have a few idea's that i'm gonna let bounce around in my head for a while.

So now i'm going to study hard back through this thread, to try to get my head around things, some of it is not easy to understand so I might be asking a question or two soon.

I have just about got caught up on the most important "other work " so I may be able to put together my CMOS desufator i've been wanting to build. I did mean to do it a week or more ago, as i said then it'll be good practice for me.

It will have frequency and pulse width adjustment and going by the way the inductor i made for it works with Stingo it will be a very powerfull little radiant energy spike maker.

Then if I can get that to work i will be qualified to buid this project. I'll even write myself a Certificate of Qualification.

Just as a passing thought I was wondering.
If I was to use a driven electric motor as a generator to drive a big Tesla Converter/Transformer would I be able to use a three phase motor to a tripple primary transformer or a single phase motor to a quad primary or what? What would be the best way to go I wonder ?

Anyhow Cheers

[Farmhand]

Here's a scope shot from the circuit I just built , it's got adjustable pulse width and adjustable frequency. Seems alright I can make it run and charge a battery on less than 10 Ma.

http://public.bay.livefilestore.com/...206.png?psid=1

What do you guys think of this one. I've had it over 60 Khz too. But my scope is 1970's vintage and very low end. So I cant tell much from it, I'll try to borrow an expensive DMM to measure the Duty cycle.

Cheers

[gmeast]

waveform parameters

Here's a scope shot from the circuit I just built , it's got adjustable pulse width and adjustable frequency. Seems alright I can make it run and charge a battery on less than 10 Ma.

http://public.bay.livefilestore.com/...206.png?psid=1

What do you guys think of this one. I've had it over 60 Khz too. But my scope is 1970's vintage and very low end. So I cant tell much from it, I'll try to borrow an expensive DMM to measure the Duty cycle.

Cheers

The duty cycle looks to be very nearly 50%. Can you post the circuit and a picture of your toroid please?

Thanks,

Greg

[Farmhand]

Here's the circuit it is one of Seamonkeys a desulfator. 50/50 duty cycle are you sure. I'm not using a toroid on it, i'm using an inductor I made specially for it, it is a 10mm cylindrical core, metalic of some kind ,came from a big high end UPS it's wrapped with 3 layers 1mm wire 28 to 30 winds all up.

Here's the inductor.

http://public.bay.livefilestore.com/...003.JPG?psid=1

And the circuit

http://www.energeticforum.com/attach...desulfator.jpg

http://www.energeticforum.com/attach...desulfator.jpg

http://public.bay.livefilestore.com/...002.JPG?psid=1

That scope shot in the post above is when it is charging a battery probe on the inductor side of the recovery diode.

Is there any reason why I can't drive three or four fets with this circuit with a couple of modifications?

[Farmhand]

Oh yeah, I see now the mosfet i'm using is a 36N03 it's a logic level fet. The RDSon is < 17 m ohms


Some other stuff here

td(on) turn-on delay time VDS = 15 V; RL = 0.6 W; VGS = 10 V;
RG = 10 W
- 6 - ns
tr rise time - 10 - ns
td(off) turn-off delay time - 33 - ns
tf fall time - 19 - ns

I was just thinking the 60 Khz statement may be due to a false reading of my DMM because of measuring the wrong way. Sorry. I'll check max frequency later.How should i measure the frequency with a DMM? Where to place the probes ?

Anyway that was practice now i can build another one like yours .

Cheers

[gmeast]

i'm wrong

Here's the circuit it is one of Seamonkeys a desulfator. 50/50 duty cycle are you sure. I'm not using a toroid on it, i'm using an inductor I made specially for it, it is a 10mm cylindrical core, metalic of some kind ,came from a big high end UPS it's wrapped with 3 layers 1mm wire 28 to 30 winds all up.

Here's the inductor.

http://public.bay.livefilestore.com/...003.JPG?psid=1

And the circuit

http://www.energeticforum.com/attach...desulfator.jpg

http://www.energeticforum.com/attach...desulfator.jpg

http://public.bay.livefilestore.com/...002.JPG?psid=1

That scope shot in the post above is when it is charging a battery probe on the inductor side of the recovery diode.

Is there any reason why I can't drive three or four fets with this circuit with a couple of modifications?

I didn't look at it close enough. I thought the oscillation was ringing through the PWM. I now see a more interesting waveform. So what is that? A switch turns 'off' and you get a positive going first spike? then it looks about 15% - 16%. That's better.

[gmeast]

nother one like mine ...

Oh yeah, I see now the mosfet i'm using is a 36N03 it's a logic level fet. The RDSon is < 17 m ohms


Some other stuff here

td(on) turn-on delay time VDS = 15 V; RL = 0.6 W; VGS = 10 V;
RG = 10 W
- 6 - ns
tr rise time - 10 - ns
td(off) turn-off delay time - 33 - ns
tf fall time - 19 - ns

I was just thinking the 60 Khz statement may be due to a false reading of my DMM because of measuring the wrong way. Sorry. I'll check max frequency later.How should i measure the frequency with a DMM? Where to place the probes ?

Anyway that was practice now i can build another one like yours .

Cheers

I don't know ... I'm not that impressed with the rise and fall times. I'm thinking that we may have to use a 'real' Fet gate driver to achieve (or match) the ns turn-off / turn-on times of the Fet.

My Op-Amp project was fun, but I'm not sure it will work well. Does anyone have any 555 screen shots of a PWM THEY made from it. I'm just interested in the output signal of the 555. I downloaded a PDF on MOSFET gate driver circuits and it's way complicated and highly dependent upon Fet design. What the Fet is doing is of no concern right now ... not to me anyway.

[Farmhand]

Yeah i've got the pulse width adjusted out to charge desufate batteries, and the frequency down to 4.5 Khz, I'll take another shot now it's using 320 Ma.

These are at 4.6 Khz using 320 Ma. ATT 1/10v

http://public.bay.livefilestore.com/...001.JPG?psid=1

This one is streched out horazontally.
http://koa0iw.bay.livefilestore.com/...003.JPG?psid=1

Oh it's 4.30 am here. I can sleep during the day.

[Watson]

Supercharged JT

Hi guys !
I just came by to say hallo, and to tell you what i am doing just
right now...Well, i spent a lot of time last few days messing around
with Sucahyos stingo and tonight i made my first little joule thief...
I ll attach one photo with it (schematic and device), but i just saw
on youtube a few videos presenting big joule thief and i must ask you
(especially Crackahcrackah) what do you think about this stuff ?
I believe it has a large potential for charging and lightning cfl and
spare energy of the primary battery. Tomorrow i am going to make
big one, this thing is very, very interesting to me...But i ask you here
about it because i think that maybe we could even take something
from that project and use it in our, or better say your project, what
ever it is going to be at the end of the road ?

That's the circuit that I call my Supercharged JT. It has much greater efficiency than a conventional JT. Search for supercharged joule Thief or high efficiency joule thief and my blog from almost 2 years ago will pop up.

I ordered some transistor sockets from Mouser and i put one into my Supercharged JT. I tried various types of transistors and found that the ones with higher gain do a better job of illuminating the LED. Out of more than a dozen different transistors, I found the 2N3904 put out the least light from the LED. These just don't have the ability to handle several hundred milliamps of peak current. The 2N4401 and BC337-25 are so much better, and put out much more light. Another one that is cheap and easy to find is the Fairchild SS8050, which puts out better than the BC337. You can find them inside toys or buy them from Mouser for less than five cents apiece.

[gmeast]

Pwm

Hi all,

Well I'm not so bummed after all. I just found another Op-Amp with the same wide range of power supply. It has 4 X the slew rate as the one I'm using now. There's one even faster than that, but I think this will do. I wired everything up with a socket so all I have to do is unplug the old one and plug in the new one. I also ordered a logic level power MOSFET with only 1.7/1000 ohms of 'on' resistance ... that's pretty good. I ran out of solder too ... that's coming.

Here's PWM now:
http://02d1852.netsolhost.com/radian...us_control.jpg
The pot-left is the frequency control:1.8kHz to 65kHz
The pot right is pulse-width control: down to 0.8% at some 20kHz and down.

You can see the two probes ... one for PWM out and the other for its inverse signal ... actually just another comparator made from the 4th 'unused' OP-AMP of the quad. The traces track 'exactly' the 2060's published slew rate speck, so that made me optimistic that a peppier OP-AMP will make the difference. Heck, I can get an op amp that will go 230V/usec, but not in a quad package. The 2060 will do 4V/usec, my new quad will do 16V/usec.

Toroid still unresolved. Talk at you more later.