Kapanadze Cousin - DALLY FREE ENERGY

[Dog-One]

Very inspirational Nick.  I find myself agreeing with you 100%.

Also slowly beginning to understand bits of pieces of what Nelson has achieved.
Hopefully good things will come of it.

[TinselKoala]

More on my 494 driver circuit, which is basically the Stalker/verpies circuit with a few mods.

I changed the Snubbers to 56R 2W + 20nF 1kV, and I put 0.1 uF poly film caps across pins 1 and 4 (supply and ground) of the 4420 driver chips. I also put MUR1560 high current ultrafast diodes in series between the Drains of the mosfets and the Load. And finally another 1 uF poly film cap between + rail and Source ( - rail) near the mosfet sockets.

All of this keeps most of the spikes from kicking back into the mosfets, so they now don't get excessively hot even on their small heatsinks. The Snubbers do get hot from absorbing and dissipating the spikes so I know they are doing their job.

This morning I managed to destroy the NE-2 I was using for a test load on the output of the small potcore transformer. I have never seen a neon fail like this before: there was a bright red-purple glow then "psst" and the light went out. On post-mortem exam, I saw that one of the electrodes in the neon actually melted enough to bend over and contact the glass envelope, which then punctured and cracked. This was at around 40 kHz, with power to the Drain-Load-Source circult supplied by an external 12 volt battery, and low Dead Time setting on the driver board. Mosfets and Snubbers just warm.

[Bat1Robin2]

Can anyone comment on how wide the pulse width for the tesla/katcher drive is typically set at to get a decent hv field to light a nearby florescent. I started with 500 ns pulse it would not light anything, i then increased the pulse width to around 4 us, and it does pretty good but not the best.

[NickZ]

   Thanks for the update TK, as well as posting your video links.

   Although Stalker has been at this for a while now, he does not show the results of most of his drivers tests once connected to the rest of the device, nor, if his latest devices will even self run, or not. So, I think that it would be best to take all these latest tests with a grain of salt, until further proof of their working are known and confirmed. Which is what I'm up to now.

   In Stalkers latest diagram he shows no jumper wire between the 28t yoke secondary and the 3t coil. It is also showing the 28t coil reversed from the yoke primary coil directions.  I see NO earth ground wire connected to the grenade/yoke coils.

   So, TK as you are seeing now, it's not easy to get the TL494 circuit and fet drivers to work without heating of the mosfets, and still produce a clean signal without interference or overheating.

   I am running the Stalker's normally used snubbers system, which are the 10k resistor to the gates, and two facing 18v zeners, 1k resistors along with the MUR1560 diodes from source to drains. If that snubber set up works for him it should also work for us, as well.
  However, I'm now using the Pinoytech set up for the TL494 circuit, as shown below. Which is a mod of the TopRuslan TL system. So, I've combined the Stalker, TopRuslan, and PinoyTech set ups, all in one.
 
   I'm also replacing all my filter capacitors, and voltage regulators, as most of them got damaged when I reversed the input polarities by accident. And I'm now also using the 104J caps for most of the filter caps, as well. As I see that it's best to use the exact same components as in the diagrams, when possible. And am also separating the inputs to the TL494, and fet drivers, and will not be shared inputs as I had done at one time, previously.

[TinselKoala]

Nick, the Snubbers I am talking about aren't the Gate protection zeners or associated resistors. They are a 10 or 20 nF ceramic high voltage capacitor in series with a low value, 22R or 56R resistor, from Drain to Source. This system dissipates the backspikes from inductive loads and prevents them from heating up the Mosfet body diodes, a major source of heating. You might consider trying something like that on your system. Also I must note, once again, that the schematic you posted is using IRF3205 mosfets, which have very low on-state resistance and thus will not heat up much in normal operation. I'm using them myself.
(Just checked with my local supplier, he is out of stock of IRFP260n and wants almost four dollars each for them... time to make another internet order I guess.)