Kapanadze Cousin - DALLY FREE ENERGY

[AlienGrey]

I thought this article might help but since some of you
appear to have invented a way of using a capacitor
that appears to be able to do the same job,
I would suggest if that is so perhaps you should patent it !
However, i know of no such technique reading this should help.


What is a Silicon Transient Voltage
Suppressor and how does it work?
Application Note
Vishay General Semiconductor
www.vishay.com
Document Number: 88436
256
Revision: 02-Feb-07
By Bruce Hartwig
Senior Automotive Applications Engineer
Transient Voltage Suppressors (TVS's) are devices used to
protect vulnerable circuits from electrical overstress such as
that   caused   by   electrostatic   discharge,   inductive   load
switching and induced lightning.
Within the TVS, damaging
voltage  spikes  are  limited  by  clamping  or  avalanche  action
of a rugged silicon pn junction which reduces the amplitude
of the transient to a nondestructive level.
In  a  circuit,  the  TVS  should  be  "invisible"  until  a  transient
appears.  Electrical  parameters  such  as  breakdown  voltage
(V
BR
),   standby   (leakage)   current   (I
D
),   and   capacitance
should have no effect on normal circuit performance.
The  TVS  breakdown  voltage  is  usually  10  %  above  the
reverse standoff voltage (V
R
), which approximates the circuit
operating  voltage  to  limit  standby  current  and  to  allow  for
variations  in  V
BR
  caused  by  the  temp
erature  coefficient  of
the TVS. When a transient occurs, the TVS clamps instantly
to limit the spike voltage to a safe level, called the clamping
voltage (V
C
), while conducting potentially damaging current
away from the protected component.
Figure 1. Transients of Several Thousand Volts can be "clamped" to
a Safe Level by the TVS
Figure 2. Transient Current is Diverted to Ground Through TVS;
the Voltage Seen by the Protected Load is Limited to
the Clamping Voltage Level of the TVS
TVS's  are  designed,  specified  and  tested  for  transient
voltage  protection,  while  a  Zener  diode  is  designed  and
specified for voltage regulation.
For transient protection, the
designer's choice is a TVS.
The surge power and surge current capability of the TVS are
proportional  to  its  junction  area.  Surge  ratings  for  silicon
TVS  families  are  normally  specified  in  kilowatts  of  peak
pulse  power  (P
PP
)  during  a  given  waveform.  Early  devices
were  specified  with  a  10/1000  μs  waveform  (10  μs  rise  to
peak  and  1000  μs  exponential  decay  to  one  half  peak),
while  more  recent  product  introductions  are  rated  for  an
8/20ìs  test  waveform.  Power  ratings  range  from  5  kW  for
10/1000  μs,  down  to  400  W  for  8/20  μs.  This  power  is
derived from the product of the peak voltage across the TVS
and the peak current conducted through the device.
Packaging  covers  a  broad  spectrum  according  the  need.
Discrete  axial  leaded  components  are  available  in  peak
pulse  power  ratings  of  400  W,  500  W,  600  W,  1.5  kW  and
5  kW.  The  higher  power  devic

[Hoppy]

Nick,

Try to get hold of one or two 12V / 50W auto headlamp bulb, which will give you a more realistic limited current draw than your 24V/18W bulb which will limit current to well under 1 Amp at both 24V and 12V.

[itsu]

AG,

i tried a TVS (54V) across the drain / source, but it started smoking after a few seconds, even when putting 2 in series (108V) they smoke quickly.
I guess the spike (400V) is to high / broad to be tamed by a TVS.   I did tame the spike though.

Then i tried an additional diode across the installed RC snubber resistor creating an RCD snubber, but allthough it caused the spike to be lowered to about 330V, it
started now frying the resistor in the snubber.   
Guess i need a bigger (2W) snubber resistor (180 Ohm in my case).

Finally i used on top of the earlier calculated RC snubber (to tame the ringing) the snubber setup from Oleg, see picture below.
It reduces the spike to about 150V.
It creates some extra ringing, but that is quite minimal.
I had 1 MOSFET starting to oscillate sometimes, but was able to tackle that with using these small ferrite beads on the MOSFETs gate leads.

This Oleg snubber consists of 2x UF4007 diodes VD6 and VD7, a 1uF capacitor C15 and a 1K resistor R26 (1W in my case).

Perhaps it could be improved upon by using a high quality ceramic cap, and/or schottky diodes.
Nothing gets hot quickly now.

Perhaps i need to install this Oleg snubber first, then calculate a snubber for the ringing.

The ringing RC snubber is directly across the MOSFETs drain / source,  the Oleg spike snubber is across the Yoke primaries.


Itsu

[AlienGrey]

AG,

i tried a TVS (54V) across the drain / source, but it started smoking after a few seconds, even when putting 2 in series (108V) they smoke quickly.
I guess the spike (400V) is to high / broad to be tamed by a TVS.   I did tame the spike though.

Then i tried an additional diode across the installed RC snubber resistor creating an RCD snubber, but allthough it caused the spike to be lowered to about 330V, it
started now frying the resistor in the snubber.   
Guess i need a bigger (2W) snubber resistor (180 Ohm in my case).

Finally i used on top of the earlier calculated RC snubber (to tame the ringing) the snubber setup from Oleg, see picture below.
It reduces the spike to about 150V.
It creates some extra ringing, but that is quite minimal.
I had 1 MOSFET starting to oscillate sometimes, but was able to tackle that with using these small ferrite beads on the MOSFETs gate leads.

This Oleg snubber consists of 2x UF4007 diodes VD6 and VD7, a 1uF capacitor C15 and a 1K resistor R26 (1W in my case).

Perhaps it could be improved upon by using a high quality ceramic cap, and/or schottky diodes.
Nothing gets hot quickly now.

Perhaps i need to install this Oleg snubber first, then calculate a snubber for the ringing.

The ringing RC snubber is directly across the MOSFETs drain / source,  the Oleg spike snubber is across the Yoke primaries.


Itsu

yes that's the type of circuit I'm using I have IRF 150' since day one although I did try a FAST IGBT that worked too with over shoots. Now I have IRF150 I don't get any over shoots or heating at all but my yoke is a toroid wound bi-filar as I said with 8+8 turns the dead zone if filled with sine waves (harmonics). The Russians used a barium based yoke, I don't think many will have this it's the key as and its all done in the coils as far as I can see.

[Hoppy]

The Russians used a barium based yoke, I don't think many will have this it's the key as and its all done in the coils as far as I can see.

Yes, If these devices are genuine, then I too believe this is probably the key. I mentioned this along time back but have never been able to source a supply. Apparently some of the vintage 60's loudspeaker magnets and radio rod aerials had barium content.This is a primary reason why I stopped experimenting. Like Nick. I'm also concerned that the modus operandi might be NMR.