Thane Heins Perepiteia.

[gyulasun]

Stefan,

The powder Iron core I'm using is good for DC to 1MHz. See the attached pdf document.
My oscillator is running at 14,3 KHz. This is within the cores ability to handle the frequency.

Groundloop.

Hi Groundloop,

You are completely right this core was designed for that frequency range.

My understanding on Thane transformers is to use cores with high permeabilities (mu) for the output coils and cores with low permeabilities for the input coil. (At least there should be a big difference between the two, favoring higher values for the output core in order to keep flux much better inside the toroid and a high (mu) value helps this very much  The core you are using has got a permeability of 75 from the data sheet and this I think an unfortunately low value for this task.  I apologize saying this, I do know one has to work with components at hand from his junk box.

You will hate me when I say to use iron wire formed into toroidal shape of a chosen diameter (preferably isolate the iron wire turns from each other) , this way you could form a toroid core that has sizes of your choice and has permeability about in the range of 1000-1500 at least. And then you could make the output coils on this iron wire toroid core, maybe winding them easier than on the Amidon toroid due to the preferably higher OD and ID sizes...   What do you think?  I suggest this only assuming you do not have at the moment a ferrite toroid core of at least  6-8 cm or higher OD AND with 3000-4000 or higher permeability value... 

I really appreciate your excellent hand-on work on this topic!

rgds, Gyula

[Groundloop]

@Gyula,

Yes I agree with you. How about a Metglass core for output? I have ONE Metglass core with a very high permeabilities (mu).
The problem with using that core is that it currently has a 50mmx20mmx30mm Grade45 Neodym magnet in its middle.
How I can get that magnet out of that core without destroying the core is the big question. (I'm afraid of big Neos.)
Insulating soft Iron wire (painting eg.) is out of the question. Way too much work. But a good idea anyway.

Groundloop.

[Heinstein]

Dear Groundloop,

Try not to get too distracted by your efficiency numbers right now.
Can your data answer the questions?;

?Do fluxes cancel or add??
If they are additive then test #2 should provide a higher output power
than test # 1.

Can a primary be magnetically isolated?
Best test ? short out both secondaries ? does the input current change?

TEST # 0 Open Circuit Test R = infinity

Input
Volts =
Amps =             
Watts =

Output  (both coils in parallel)
Volts =

TEST # 1    R = x

Input
Volts =
Amps =             (should be the same as test # 0)
Watts =             (should be the same as test # 0)

Output
Secondary 1 connected, Secondary 2 open:
Volts across load =
Power =

TEST # 2   R = x

Input
Volts =
Amps =                 (should be the same as test # 0)
Watts =                 (should be the same as test # 0)

Output
Secondary 1 and 2 connected in parallel across load:
Volts across load =
Power =

Thane

[gyulasun]

Well there are iron wire for garden use and such iron wires are insulated with colored pvc layer much like normal copper wires. People dealing with selling flowers could tell you where such insulated iron wire is available.

Re your metglas core, what sizes does it have?  Somehow sliding/slipping the magnet sideways you could remove it...

Gyula

[Heinstein]

Dear Groundloop.

Do you have a 1 k resistor handy?
This was the only load where I could get any encouraging power output.

Thane