Akula eternal lantern 4

[d3x0r]

The gap does matter; I had some of the glue leftover from where the space was on my core, so I wasn't actually getting it closed... removed the glue, and a closed core does not do the same thing.


On akula's (an prior times I saw this' the collector rises before the base does down... so it had looked like the collector was the first...
on my video, the base is falling first for sure....though in prior experiments I have seen that it appeared the collector was first to react...


Although; akula's scope is on the other side of the cap/resistor from the base... so maybe the base is falling first (the capacitance is satisfied by current from the emitter...) but then it recharges very fast

[itsu]

... so looking at details of akula's board... he has the power supply with a long wire soldered to a large-ish capacitor...
Hmm schematic must be wrong... the pulldown(1k) and variable(0-.22n) cap must be on the other side of the 10ohm&10n to the base...
*sigh* and there's 2 transistor things that aren't represented...

I don't think that would make any sense to have the  pulldown(1k) and variable(0-.22n) cap at the other side of the 10ohm&10n.
The fysical location on the print does not confirm this, also the pulldown (1K) would be needed there right at the base when using an IRF (MOSFET)
which he mentions he would start using when he found the ferroresonance of the core.

The "2 transistor things" are in the 2e diagram in post #1 of this thread:

Regards Itsu

[d3x0r]

I don't think that would make any sense to have the  pulldown(1k) and variable(0-.22n) cap at the other side of the 10ohm&10n.
The fysical location on the print does not confirm this, also the pulldown (1K) would be needed there right at the base when using an IRF (MOSFET)
which he mentions he would start using when he found the ferroresonance of the core.

The "2 transistor things" are in the 2e diagram in post #1 of this thread:

Regards Itsu

I concur that the pulldown should be on the mosfet side...
But; the physical layout of the board... the variable capacitor is not near the mosfet... and there's a resistor in parallel with it...


I'll check the other schematic; have been working from a simplified version apparently... but with the PNP/NPN part of the gate driver there too... would a pulldown even be needed?

[d3x0r]

Overlaying the two schematics and fading between them... other than slight alignment differences of a name and a couple traces... this is the only difference

[Void]

The 'flyback mode' or 'flyback effect' can be seen when just observing the primary with no secondary.
Basically all this mode is, is putting a small gap in the ferrite toroid core to increase the reluctance
of the ferrite core a bit. The result of doing this (adding the small gap) is that the back pulse created by the coil when the driving
pulse is quickly switched off climbs up to a higher voltage peak value. I am not sure of the actual physics of
why this works, but that is the end result of introducing the small gap in the ferrite core.

In CRT TV's, they need to have a  high voltage, and putting the small gap in the flyback core allows them to get a higher
voltage on the back pulse with less secondary windings. Having the gap gives a narrower pulse but a higher voltage
peak, so the area under the curve for the back pulse probably stays about the same, but maybe the core with no
gap is a bit more efficient.

I have attached two scope shots with my scope probe going from the Drain to the Source (ground), taken with the exact
same coil and the same pulse driver settings, but the first scope shot is with no gap in the core, and the second scope
shot is with about a 2mm gap in the core. There is a maximum gap width you can have before the voltage peaks start to fall off again.
I have 3VDC on the coil that is being pulsed, and the pulse width is about 20uS.

You can see that with the gap, the back pulse on the Drain climbs a lot higher in voltage, but it is a narrower pulse.
That is really all there is to the 'flyback effect'. The gap just causes the back pulse from the coil to go to a higher
peak voltage value, but the pulse is narrower.

There may be other differences in behavior of the ferrite core when using the gap as well, and you can see a bit of a
difference in the resulting waveform on the Drain of my FET when there is no gap, and when there is a gap.
In my particular circuit setup, with no gap, after the back pulse is done the Drain is close to 0V and then slowing
climbs back up to 3V. With the gap, the Drain actually goes a bit negative for a little over 2 microseconds, and then shoots
up very quickly to a bit over 3 V (i.e. there is some overshoot), and then drops back to 3V after a bit more. The number of bounces
you see after the initial back pulse on the Drain or Collector completes may just be due to small capacitances in the circuit in
conjunction with the coil giving some resonant bounces, and may not be of too much significance at all in regards to the
'ferromagnetic resonance' which Akula is talking about, but I am not certain.