Kapanadze Cousin - DALLY FREE ENERGY

[Dave45]

You can limit the current with a 7812, radio shack carries them

http://www.researchcell.com/electronics/7812-pin-and-circuit-diagram/

Another current limiter might be a better option

[NickZ]

  Dave45:
  That may be a good idea.

   I have not seen a transistor version of any circuit lighting up several hundred watts of incandescent bulbs yet. They do make Led bulbs light well, up to a certain point, usually less than a 100 watts though.

  If the induction cooker circuit can light lets say 2000 watts of load, then maybe we need to buy some of those, instead. Or some ready made mosfet driver circuits that can adjust for duty cycle as well as frequency, like what Akula has done.
But, his first device also overheated, and was not perfect. Maybe his later design have been improved by now.
  I do love simple circuits, and was the reason that I went for Geo's round core in the first place. I still think that the yoke can outdo the smaller flyback core, as Akula has shown recently.
  Anyways,  Hoppy, yes, I'll try to find all the right components, including those bigger Zeners. Have you got the Mazzili going yet?  Could use your help on this, as well, to show us all how it's done.

  I believe every word that Geo has told us about the circuits, and everything else, and I'm hoping to make a useful working circuit, soon. 
And, I'm also hoping to see him come up with a looped circuit that can light many bulbs with no overheating, 24/7.  That illusive loop is what we are all waiting to see.

[SeaMonkey]

What I see is that when using transistors for the Royer circuits, even the faster switching transistors, they can't do what the Fets can do. As, I can't light a single higher wattage incandescent bulb, (not even a 25 watt) on the transistor circuit.  Compared to when using the Royer with the IRF 620 or IRF 634, on which I could light two 50 watt incandescent bulbs at least partially on my yoke core when using just a 12v, 4.5 aH battery.  So, it looks like Geo is right, when saying that Fets are the way to go.

Transistors of the Bipolar variety are rather "tricky" when
used in power switching circuits and can be quite a bit more
finicky than the newer MosFets.

The Beta or Current Gain of a transistor is variable;  not
only from one transistor of the same type to another,
but also in operation.  The more Collector Current the
application demands, the lower the Beta or Current
Gain of the transistor will be.  At high current levels
the gain of a typical power transistor may be as little
as 3.

This means that in order to assure the switching
transistor is driven into "hard saturation" for minimum
internal resistance and maximum efficiency it must
have sufficient Base Current to drive it for the entire
duration of the transistor "ON" time.

An oscilloscope is a necessity for fine tuning the base
drive to a power transistor in a test setup.  The Collector
to Emitter voltage will decrease to some minimum value
when the transistor is saturated and delivering to the
test circuit the desired level of collector current when
it is provided with adequate base drive current.  For a
collector current of 10 Amperes the needed base drive
may be as much as 3.3+ Amperes.

Unless the switching transistors are driven hard enough
to attain hard saturation the power output of the circuit
will be disappointing and the transistors will run much
hotter than normal.

There are new bipolar transistor types which are sold as
"Low Vce Sat" which will exhibit a very low voltage from
Collector to Emitter when saturated at a high current
level;  some as low as 0.3 Volts which makes them
competitive with MosFets for certain applications.

But they must have the right amount of Base Drive to
assure the Low Vce Sat.

[NickZ]

   Sea Monkey:
   Always good to hear from you...  It's midnight here now, but I just had to answer your post.
   As it's been the overheating issue of the core/resistors/transistors/chokes, etz...when higher current or voltage is provided to the base or gates. Which also has been the ongoing problem up to now on practically all of these devices. Which no one has resolved, at least that I'm aware of. Other that duty cycle, frequency controls, core wire turn counts, and such.
  So, the question is how to provide sufficient gate/base current to saturate the transistor, with out adding to the pile of fried components.   A little magic is involved here... 
  Do you have a solution???  As simply increasing the current to the gate or bases has not shown a reliable long lasting output, or results, so far. And adding more load as Igor did when connecting the 2000 watt bulb, to his induction hob crt resulted in a less than satisfactory outcome.

  However, my main concern is how to loop these circuits, not just how to make them more efficient, or providing the highest gain. Although sufficient gain must also be factored in.

[Hoppy]

Hoppy, yes, I'll try to find all the right components, including those bigger Zeners. Have you got the Mazzili going yet?  Could use your help on this, as well, to show us all how it's done.

Nick,

I have had several Mazzilli's up and running. I think I have already guided you sufficiently on how to achieve a reliable working Mazzilli. However, to clarify, use the circuit here and make sure you fit the 15V / 5W rated zener diodes and provide current limiting in the form of a halogen bulb for simplicity. Ideally invest in a current limiting bench PSU if you intend to do a lot of electronics work. If you get the build right and current limit your supply, you will succeed in getting a reliably working Mazzilli.