Bucking Coil Inverter

[voltan]

hi mate. i think blocking osc. is where the magic can be found. i'm new to chris sykes stuff, but i like it a lot. from what i've read,elsewhere,it sounds a lot like when you light bulbs with what textbooks describe as an open circuit that is radiant energy and it means you have standing waves and the connection point/s is/are a current node.you are seeing cold electricity.a challenging process generally initiated with abrupt d.c switching events on the primary side, but being an assymetrical transformer, it could work the same or better when driven with sine waves. i read that chris is leaning towards ac sine input, but your blocking osc is delivering the goods anyway. see how far you go with that first. simple but effective is often the best way. if i got most of that right,then with a 3.5 v input your achievement is quite incredible. replicators out there are playing with thousands of volts to try and manifest the cold stuff. it's not hard to make an led glow off 1 wire with an av plug,but your putting out a fair bit of light there. if your results are replicatable, this could be the start of something big.
a variable resistor or trim pot to base would help to optomise the performance, unless you have already zeroed in on 220 ohms using one or the other.
like sputins suggested,after you have optimized it, make it permanent and presentable and maybe keep it in a display case or something.if it turns out to be a breakthrough project it should be kept intact and preserved. too many people have taken parts out of working projects and later they can't get it to work like before. you will build bigger and better ones and who knows, it could be a piece of history one day.

[voltan]

ok.with a r and d budget of 85 cents the best i could do was 6 ferrite rings iv'e had for about 15 yrs.18mm od, 10mm id and 6mm wide,so only 36mm long, and they are held together reasonably well with just heatshrink.because there are 5 joins in the core i rubbed the faces flat with 320 wet and dry sandpaper on a perfectly flat surface,using plenty of water and turning the rings as i rubbed them in a circular motion, to try to maximize contact area between the rings.i don't know if it helps or not tho. i wound the secondaries with awg 34. then 2 half lengths of heatshrink over them. then about 20 turns bifilar of awg 27 over 1 secondary with tape to secure it, for adjustment or removal.to get the output voltage up for 240v led globes it has a turn ratio something like 1:8 between the driven primary winding, 1 of the 2 strands, and the sec it's over. i found an early model 240v led globe. it lights to about half of the brightness compared to mains power,when i adjust R to about 18 ohms. my light duty power supply sags to 1.5 v, so pic 2 was with 1.5v, .5 amp input.and the mje340 is only rated to 500 ma, so both need upgrading, but i think it should go well then. i will get a new generation led globe from bunnings next week as well.it works the same on 3 wires,but drops off with 2 wires,1 from each sec.
also no gap between the secondary coils on this one. not sure if any of these design details are critical. maybe we can establish some guidelines after there have been more replications and results posted.
i haven't tried it,but connecting the mains input of a switch-mode adapter,with the right output spec, to the output of these things might be an easy way to close the loop.
cheers

[voltan]

hi all. there's a very informative article on the net called 'the spin field ferrite sleeve antenna' by graham maynard. intergoogle it.to get to the good part scroll down about halfway until you see 'the ferrite anomaly solved' in red,then read to the final notes.his take on the benefits of hollow ferrites,the localized magnetic fields around the current carrying coils on a ferrite rod or tube,etc in combination with chris sykes pdf, could be the key to understanding and explaining skywatchers extraordinary results.
cheers.

[Wistiti]

hey Sky.
Are you still here?

[SkyWatcher]

Hi wistiti, yes I'm still here, I'm studying h-bridge circuits and may build one of those to drive the partner coil setup.
Or run a speaker wire over from my stereo and record some different frequency sinewaves on a cd and try that to drive it.
Only thing about the large stereo setup, is i'd have to use a killiwatt meter to see amp draw changes and input/output.
Another option for square wave, is to use an rc speed control, though that has frequency limits, but high power ability.
I hope emjunkie comes to his senses and is able to ignore peoples words, oh the ego.
peace love light

[voltan]

hi sky. something like this might be worth a try. the bifilar primary flip flops n-s-n-s, providing a virtual ac input.both transistors are npn and the diodes help to protect the base emitter junctions from transients.
*after i changed the bifilar winding on my attempt, a 1 transistor version, it would only go with no diode,but the 2 transistor scheme might need the diodes to run,not sure. so i guess try it with and without.
cheers.

[SkyWatcher]

Hi voltan, thanks for the circuit, though that's the one i'm using now, just without the diodes.
That circuit is good, though some other versions of it, are more adjustable.
Also, i don't think it can use only one primary wire with that flip flop circuit.
I'm going to try using some small pc speakers output i have laying about, then i can use a pc program or linux in my case, to sweep through different frequencies.
The pc speaker output can power a single coil at true sine wave.
Though the rated speaker load impedance will have an impact on output I'm sure.
peace love light

[Zardox]

SkyWatcher - Off topic but Linux is the only way to go. I can't believe that people are still useing Microsoft.

[Wistiti]

Hi Skywatcher and all!
Yes Sky, am agree with you, the ego is something people have to free off...
Good idea the speaker output!
i have made a test this morning with my tv yoke ferrite and old Sony cd player i have it is write 24w output on it. I cut the wire that come from the player that goes to the speaker and connect to a single wire primary coil wound over one of the 2 partner coil. I drive it with a tone generator app i have on my phone.
I see 2 things:
1-best output is obtain with the second way to conect the partner coil shown by Em junky(The end of the first connect to the end of the second and the output is the start of each.)
2-There is a big difference with the frequency... My best output is around 220hz.
I try sine, square and sawtooth wave. For me until now, the square wave give best result.
To be continued!
ciao!

[voltan]

based on wistiti's early results, something like this wave generator kit from jaycar, on the input of a car amp that can handle 2 ohms on the output side might go well. both items on 12 volts.
square or triangle waves from 100-20,000 hz,probably adjustable limits.not sure if the kit outputs a signal that swings plus and minus.it may need additional biasing to drive the amp both ways.
nice work wistiti.
cheers

[SkyWatcher]

Hi folks, i hooked up one speaker output to the primary inducer coil.
Placed bifilar in series as one coil and it measures from 3-4 ohms.
Which works out fairly well, since the speaker is rated at 4 ohms or measured bewteen 3-4 ohms.
It liked 200 hertz the best, in square and sine wave, used audacity in linux mint to create the tones.
Best efficiency i could tweak with a 110 ohm load, was around 57 percent at almost couple hundred milliwats input at 5 volts, usb powered speakers.
So as Chris sykes has said, it needs a good magnetic field to get it working efficiently.
As the flip flop circuit so far has managed 87 percent efficiency, though i have yet to tweak that setup.
peace love light

[Wistiti]

Hi Sky!
I am agree with you must have a more power input or review your coil winding to acheive a better magnetic field... If your primary is stiil bifilar try just one wire... It may help. Tonight i have seen also there a sweet spot on the gapping that put a bit more on the output with no more on the input.
The type of load to seems to mather too...
Am still learning!
Ciao!

[voltan]

hi all. i rewound my primaries with about 80 turns of awg 34, so the turns ratio is more like 1:2 now. with the old lamp the best output was with the sec's connected like in pic 1.
then i got a new 4 watt led lamp. i couldn't get it to light until i tried a voltage doubler, maybe due to the low turns ratio and the lamps rated 220-240v. then i found the output scheme in pic 2 worked the best.
*i have since found the additional diode up the top improves the efficiency a bit more.
i might run a ground wire out the window and try that on it. still trying different connections on this.
cheers.

[Wistiti]

Good work voltan!
the way you connect your partned output coil is the way i have more output.
Try to play with the gap of your ferrite bed betwen the 2 partned coil output.
Ciao!

[voltan]

thanks wistiti. my sec's aren't adjustable, and no gap, but it sounds like a very good feature to incorporate in these replications. same goes for sliding primaries.
cheers