Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

  I need to know which mosfets, zeners, caps, diodes, resistors are really the best to use, for our particular purposes. To make new Mazilli circuit with at least 12v, or even higher higher voltages.
  Maybe, Geo has some new suggestions, especially for the feed-back path.
  If there are any proven induction oscillator Mazilli circuits that are yielding the output that would make for a good replication? Please let me (us) know.
I'm be making a new order for some components to be brought to me from the U.S.
                                                             NickZ

[Hoppy]

  I need to know which mosfets, zeners, caps, diodes, resistors are really the best to use, for our particular purposes. To make new Mazilli circuit with at least 12v, or even higher higher voltages.
  Maybe, Geo has some new suggestions, especially for the feed-back path.
  If there are any proven induction oscillator Mazilli circuits that are yielding the output that would make for a good replication? Please let me (us) know.
I'm be making a new order for some components to be brought to me from the U.S.
                                                             NickZ

Nick,

I have already given you a link to a Mazzilli circuit and component list which will give good results. See here. I suggest you also get hold of some 50W, 12V halogen lamps and connectors for a current limiter. These can be wired in parallel for higher current loads or in series for higher voltage supplies. Geo uses a lot of low value capacitors in parallel, so instead of the 0.68uF cap specified, you could try a number of 0.1uF caps. The linked document does not specify the power rating of the zener diodes but I have already suggested that you go for a 5w rating. A suitable inductor can be made by fully winding a large yellow coded ferrite toroid (which you may already have) with 1 layer of 1mm dia magnet wire.

[SeaMonkey]

The Mazilli Oscillator is a novel circuit which
is just too simple in its construction to be
reliable and/or versatile.

Economy of parts in the Gate Drive may
work for some time in the short term, but
the MosFets will be unnecessarily stressed
which will lead to their early failure.

For maximum reliability Gate Drive Chips
offer the best solution.  The circuit will be
somewhat more complex as a consequence
but the additional "good design" will assure
maximum efficiency and reliability.

In order for the circuit to function over a wide
range of input voltages some additional
safety features must be included in the design
as well.

Simple circuits can be fun to play with but for
serious work to be done with MosFets they
must be properly and adequately driven.

 

[Farmhand]

Seamonkey would you suggest an oscillator chip to provide the two out of phase signals such as a SG3525/SG3524 - TL494 - CD4047 or such thing ? Then a driver chip for each mosfet like a TC4420 driver chip or maybe a dual driver chip ? If so those are pretty much what I use for inverters and stuff up to several hundred kHz. I usually use a 12 or 15 volt voltage regulator at the input to the signal section of the circuit and allow the mosfets the switch the applied voltage in full. If someone already has a preferred single signal oscillator chip they could maybe use a flip flop to create the two signal trains from one. I think it could be done with as little as two CD4011 chips and a CD4013 flip flop, one CD4011 for the oscillator, and one CD4011 or CD4001 with the four logic gates made into a pair of paralleled gates to drive the mosfets. CD chips are very inexpensive a CD4001 costs about 30 to 50 cents. Such a circuit would consume a very small amount of power. These days I would probably use a picaxe or arduino for the signal source so that extra function can be utilized easily. The frequency and/or PW can be altered using the analogue to digital inputs and 5K pots for voltage dividers. The picaxe 08M2 is a small 8 pin microprocessor which is very useful for such things but would require an additional 5 volt regulator.

If 5 Watt zeners are required at the gates of the mosfets on the mazzili circuit to protect them that says a lot of wasted power to me, but what do I know. I try to drive regular mosfets with a 15 volt supply and logic level mosfets I usually use a 10 volt supply. I use an electrolytic capacitor after the regulators. There is a lot to gain by studying the proper ways to drive a mosfet, they become very reliable.

The picaxe solar charge controller project I mad was almost unbreakable, I can momentarily reverse connect it and get away with it. And the switching is awesomely clean.

I basically use the picaxe circuit I described to control this boost converter. It's a workhorse and I can use it to charge batteries, it can deal with 5 amps input at 12 volts. The coil cores are the limiting factor they get warm.
http://www.youtube.com/watch?v=eLnNgfaha10

I can post the code I wrote which is very basic, if anyone wants it. The max voltage and max current can be varied now I made a modification.

Here I use the Converter to test the solar battery conditioner circuit operation at different input power levels and adjust it.
http://www.youtube.com/watch?v=j7FTJ7yLeDY


Cheers

Edited inverter to converter. my bad sorry

[SeaMonkey]

Good question FarmHand.  An Oscillator Chip is always an
option to provide the two complementary drive signals
for the two MOSFETs;  but then again, a free-running design
which sets its own operating frequency under varying loads
is good too.  Both styles can be made very versatile with
sufficient improvisation.

A micro-controlled circuit would be the absolute best
"Cadillac" sort of circuit which would be nearly impervious
to (or adversely unaffected by) loading with some good code.

It's a good goal to work up to through the various implementations
of the progressively more capable versions of the circuit.

Simple is good for starters - more advanced techniques will assure
the very best performance and reliability.