Today I have been trying to produce the flashing lights effect that we see in URFA's videos. I started by taking the Kacher 24V supply from what he calls the synchronisation point 'B' (shown on a previously posted schematic from URFA) via a series connected 148uH choke.  This did not appear to show any effect, so I removed the 470uF electrolytic cap that I had on the output of the PWM / push-pull supply filter, leaving 470nF cap across the output. This started to show a little response as I moved my hand across the antenna and saw the lights varying a little in brightness. It was not until I started altering the frequency of the Kacher with a ferrite rod insert that I saw the maximum effect and this was with the kacher operating at around 950KHz, where the effect peaked. The effect was not as great as seen on URFA's video but the Kacher was having a pronounced effect on the brightness of the bulbs. My grenade coil frequency 1/4 wave is as previously reported - 1.316MHz and PWM is running at 21.9KHz. I don't think that I have so far optimised the value of the series choke for max effect. I suspect that URFA used one of his green high perm toroids for his choke. I have noticed that the amplitude of the Kacher waveform increases a little when the device is connected to my domestic earth. My power supply is isolated from earth.

hi Hoppy,

I am glad you have achieved the required effect on the bulb using kacher by tuning with ferrite rod.

---------------------------------
The next 3 days from Friday i shall commence testing again.

Jeg,

I did find that unless I use 4V7 or 5V6 zeners (base to emitter) I cannot get good operation.

I have bought some at 5V and they are in place. Perhaps my choke is not the appropriate.
I was looking the peaks from mosfets on the choke just behind them. It needs diode before feeding katcher as the thin pulse imposed on 24V, go up and down for 1-2 cycles before it cease down.

Hi Itsu. I had thought of this possibility already and tried disconnecting the scope ground clip
lead earlier but leaving the scope probe tip connected, but the waveform is still similar to the way
it is now. Also the bulb is glowing quite a bit brighter now than on the previous day. Something
has changed from a day ago, but all circuit connections are not changed (it is a simple circuit setup)
and the kacher output waveform has stayed about the same. I don't have an explanation for the change
in output waveform as of yet. Also, moving the 'antenna' winding around only varies the output waveform
voltage by about 10V to 15V peak or so, so that is not it either. 
All the best....

P.S. Itsu, if you get the chance at some point, if you can show a scope shot with scope probes
at same locations with only kacher driver on and a 10W to 25W bulb or so across the grenade with earth ground connected,
I would be interested to see what you have for waveforms. My blue scope probe was 7 or 8 cm or so away from
the top winding of the tesla coil.

P.P.S. Itsu, you are right that the bulb should not light at only a few volts RMS. I guess the likely explanation
for my different waveform a day ago may well have been a poor connection or incorrect connection of my scope probe.
The waveform I am getting now should be the right waveform.

Hi Void,

here the screenshot of that setup,

yellow is across the 25W / 220V bulb,  one side grounded
blue    is 5cm away from the antenna

Only the Kacher is running at 25V

Regards itsu

Jeg,

I have found that any toroidal choke around 100uH or higher will work. The choke will prevent the Kacher pulling too higher pulse current and diverting current from the push-pull / PWM. If this choke is shorted, the load bulbs will dim noticeably. Below is a scope shot of the waveform taken on the cathode of the series diode to ground after the choke. The critical bit is getting the Kacher frequency tuned for the maximum effect. In my case its around 950KHz. Where have I seen that frequency mentioned in past posts?

Hoppy this choke doesn't have to be common mode? How you prevent the HV return? I think your series diodes to act better if anode at 24V clean dc instead of ground. They will send back to the source anything below 24V.

In my case its around 950KHz. Where have I seen that frequency mentioned in past posts?

Hi Hoppy. I think some people have reported a grenade resonance point around there.
All the best...

Hi Hoppy. I think some people have reported a grenade resonance point around there.
All the best...

Yes, thanks Void, you have jogged my memory. Interestingly, we do not see the effect demonstrated by URFA in any of Ruslan's videos.

Jeg,

I have not removed the 1000uF cap shown after the Kacher common-mode choke (not fitted on my device) However, I found it important to remove the electrolytic cap after the push-pull filter choke and replace with a non-polarised cap around 1uF. This has the effect of reducing the amplitude of the leading edge of the push-pull pulse waveform and maximising bulb brightness. This electrolytic cap (470uF in my case) severely attenuates bulb brightness on my setup.

Thanks a lot Hoppy. I will proceed the same. I already dismounted katcher for surgery.   

   John.K1:
   I always have had the problem with one or the other fet always getting hotter than the other one. Especially when lighting 5 to 7, 100 watt bulbs.
  I think that some or maybe even most of the guys will notice this overheating as well, when trying to light some REAL loads.
As yet, none of us has made a circuit that can light 1000 to 2000 watts, as Akula/Ruslan have shown possible. Nor are they really outputting that high a load, as all their bulbs will dim somewhat, as each additional bulb is added to the circuit. That is probably why they don't show what a similar bulb does when connected to the grid, and placed next to their circuit, to show the lumin comparison. 

  I also thought that the free end of the kacher antenna may not be the best way to go, and tried to place my Joule Ringer core directly on top of the yoke, and fed the HV output from that ringer core to light a CFL bulb, (to test the HV output) which was also then connected to directly to the grenade. But, no real additional effect was observed, with or without the CFL lighting, and also connected as well, which was lighting pretty brightly.