Joule Thief

[TheNOP]

@AbbaRue
thanks for the core making technique.
i will probably try it eventually.

[innovation_station]

@All
Before trying any new core materials try running them across a magnet a few times and then see if they remain magnetized.
If they remain magnetized then you don't want to use them as a core material. 
If they don't attract a magnet at all then using them is probably no different  then using an air core. 

I went down to the beach on the weekend, still ice on the shoreline here in Canada
but some Canadians were out in shorts and short sleeve shirts after all it was 52 degrees out. 
Anyway, I brought a magnet with me wrapped in a plastic bag, and I moved it through the sand to collect some magnetite.
I collected about a pound of it in the short time I was there.
I used a hard drive  magnet which may be strong but it doesn't have a lot of surface area.
The next time I go I plan on bringing one of my 4 inch diameter ring magnets.
They sell them at hardware stores mounted in a steel base with a 1/4 inch hole in it, to mount a coat hanger hook.
I hope to collect a lot more magnetite next time.

You can mix the magnetite powder with fiberglass resin and hardener and pour it into a mold to make any shape core you want.
They call this a "Dirt Cheap"    way to make transformer cores. 
Any patch of sand should yield some magnetite if you run a magnet over it.
They call it "Riverbed Magnetite" and experts say it makes the best core material of all natural sources.
Google the subject to find out how much resin to magnetite to mix.
You don't need a lot of resin and it is quite cheap, a liter of it should make dozens of cores if not hundreds.
For a mold you could use wax and carve the shape in it, then remelt it to make a new mold if it gets damaged.
I'm not sure if the resin will get hot enough to melt the wax while curing though. Never tried it.
Or you could use some type of plaster or clay material. Drywall mud is quite cheap.
I would build a large mold with many cores in it so I only need to mix the resin and magnetite once.

Addendum: Here is a link on this forum on making magnetite cores. What an awsome forum we have here!
http://www.overunity.com/index.php?topic=3318.380

i think this is the most valuable info in this whole topic .... 

thank you sir...!!

i have NANO MAGNITTE... 

ist

but it was not cheep and was not purchised for this style unit ..... 

CREATIVITY HOLDS NO LIMITS 

check this out 

http://www.youtube.com/watch?v=UCfrmYZ2zko

[jadaro2600]

@jeanna,

I'm having no luck getting a third wire to pickup properly on one of my custom duos ...on top of this I've only been able to get so much out of them when it comes to voltage and salvation of current.

SO, I'm going to switch up and try some coil tricks to see if I can't get the spikes to work a little better.

[jeanna]

@jeanna,

I'm having no luck getting a third wire to pickup properly on one of my custom duos ...on top of this I've only been able to get so much out of them when it comes to voltage and salvation of current.

SO, I'm going to switch up and try some coil tricks to see if I can't get the spikes to work a little better.

OK jadaro and Koen and Altraz and anybody else having trouble,
My first question is have you made a joule thief in the standard way?

If not, please try doing that. I will go through it now:

Using some tele wire wind 2 wires close together making 11 turns of them. This gives you 22 turns in all.

Then, because you used 2 different color wires, and you can tell them apart, connect the beginning of one to the end of the other. Just twist them with their exposed copper ends for now.

Put this twisted-together end into the red+ rail of the breadboard.

Then put one other end into the line that is at the end of the base resistor,
then the other end at the collector of the transistor.

Use a 2N2222 or 2N3904 transistor
1k ohm resistor at the transistor base.

Make sure this base resistor is right on the base and that the other end of the resistor can go to the toroid. I notice there are some circuits around that use this resistor at the other side of the toroid. Do not do that for now, OK?

====
EDIT: add a jumper to the black rail from the emitter of the transistor.
I made a better description. If there is still time I will rewrite it here. I just made the changes that make it better, though. I will be back to find that better description later.
====

After you check it all and the battery connection, add a led between the C and E of the transistor. If it does not light turn it around.

Only now can you play with the secondary.
But, first, remove the led that is in the regular jt spot.

I found that 5 turns guaranteed a light from the secondary. (I mean even on a bad toroid)


I hope this helps.
---

I also have been having trouble with one of my breadboards. The foil must be torn away too much. Also, one of my leds doesn't want to light much and it can make the parallel array turn off. This never happened in January. Also, my expensive double ultrabright is looking yellowish. I am not getting a bright white color no matter what. I have no idea why. My components have not been connected to anything for 4 weeks.


@jadaro, I know you are using a different circuit design. There are too many interesting variables going on here to make more than one change at a time makes the whole process too difficult. And besides IT may be the trouble.

@all,
OK you just heard about the trouble I am having with components and breadboard.

I just made a start on a MK2 or 3.

Before I started, I wanted to doublecheck the toroids that I have made before just to have a beginning set of numbers.
I chose one of the nice early ones. It is too small for a lot of secondary, but it is made just like the one I just described and has 20 turns secondary made with red 30 gauge mag wire from RS.

I was checking for amps so the volts and brightness are a little down too. I used a 9,5 ohm resistor at the red rail of the battery.
Here is what it showed.

.27v,  28mA no light anywhere, just the transistor is on.
.26v  27mA led in the jt place
.16v  16.8mA 2 lights from secondary of 20 turns.

"MKHybrid"=made like standard jt but with MK1 numbers of turns on same 'filter' toroid
.20v  21mA no led
.20v  21mA led in jt place
.10v  10.5mA superbright looking yellow. Regular leds no light???

Another filter toroid wound with 6 bifilar and 8 turns secondary
.23v  24mA no light
.23v  24mA led in the jt place
.19v  20mA led on secondary

The new one I am testing
MK2 4 turns each at 180 degrees
.25v  26mA no light
.25v led in jt place
.21v  22mA led on a quickly wound 8 turn secondary.

Don't you think it is interesting that on ALL of these, the volts and amp draw go down when I use the secondary to light the led's?
--

@MK I remember reading your reason to separate the bifilar into 2 parts and separate them by 180 degrees. So far, I have not had better results by doing that and I am wondering if you will describe what happened that made you decide to split the bifilar turns?

thank you,

jeanna

[AbbaRue]

@jeanna
I live in Sarnia, ON. which is at the bottom tip of Lake Huron I'm about 60 Miles from Detroit, MI.
We had a nice mild week here, Wed. we reach 65 F.

I don't know if regular wood glue or silicon would work. Some people have used epoxy glue but it can be costly.
The best resin I know of is fiberglass resin.  You can buy it at any auto supply shop, it's used for auto body repair.
You mix the resin with the magnetite really well and then you add the hardener stir it some more and pour it into the mold.
The mixture will be a thick paste so you should get some latex gloves so you can push the mix into the mold with your fingers.
Something like playdough, or window plaster.
It cures in a few hours, and then you have your toroidal core.
The ratio is 70% magnetite to 30% resin. 
Nice thing is you can experiment with odd shaped cores to see what results you get.
Maybe even some type of 3D core like in your illustrations.