Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

   All:
   I also tried to use my signal generator to inject pulses onto the grenade coil, through a 2 turn coil onto the grenade's 168t coil. Which was grounded on one side to my 37.5 meter ground line. Then tried to match the same or harmonic frequency on my Kacher by cutting it up like a salad into pieces shorter and shorter to try to match the same frequency. That led to little to no results for me.
   Now I don't try to match to any fixed nor set frequency, I just try to see where I get the best sync and output at the bulb(s). Similar to what Geo was doing, as he had no scope, nor much else to work with. Yet, he has obtained the best results so far, of any of us to date. And was why I was kind of hard on him for not keeping in touch, and sharing his results with us. Be they good or bad results.
  Also keep in mind that each added bulb will change the running frequencies of the whole circuit, as can the types of ground lines, etz...

   I only trust what I can see with my own eyes on the self runners type videos, and I even take that as possibly not what is really going on, at all. But, it's the best approach that I can come up with, up to now. Along with our house detective, Hoppy's advice.
   
   Although I see no comparison between my same grenade/Kacher circuit running from the Mazilli driver, as compared to what the TL494/UC4420 can do at the bulbs, up to now. Yet, there is no OU in either case, much less self running.
However, this will not stop me from continuing with this project, I'll just have to keep learning and testing, as I still have much left to try.
  I thank you all for your cooperation and interest in this thread. It's good to see some of the old timers back, as well.
                                                                                                                                                                           NickZ
 

   

[sparkmen]

hi all , trying again to post , my previous post is still waiting for moderator to appove (?)
anyway , here comes ahgain with only one  attachement
I've made the circuit as attached and when using "earth" connection to mentioned point , battery draw is almost 4 time less
voltmeter is across 0.16ohm /2w resistor. I can sense the difference in consumption by touchin the resistor, getting hot.
voltage across res without earth connection is 130 to 160mv( 0.8-1A) and when I connect the earthing is 35-50 mv.
not equiped with osciloscope to see the wave or freq, duty cycle is at minimum.
maybe someone can give me some clues on it?

[lost_bro]

See Itsu's video of a similar circuit below:
https://www.youtube.com/watch?v=wgtnWuuCMDI


^{NOTE: This is not a FE circuit..}

Good day Verpies

Interesting that you used the *timing cap*, Pin#6 & 7 on the CD4046 as your *master_timer* for the TL494 *slave*.  Nice work!
I used a Chinese made ($1.50USD) SG3525 w/ LM358 pcb for the *slave*.  The Chinese pcb already has the totem pole output transistors in place and an onboard LM358 for over-voltage feedback control, and also has input pins for soft_start and over_current protection.

I originally tried to use Pin# 3, (sync pin) for the *slave* connection to the *master*.   It seemed that the maximum _Capture & Lock_ bandwidth was only about 3kHz with the sync pin as input.  Since that was NOT sufficient bandwidth for the application, I had to find another viable route.

I finally used Pin # 5  (timing cap) input as the *slave* connection, as you did on your schematic with the TL494. 
Using Pin #5, I was able to obtain a variable bandwidth over 10kHz for _Capture & Lock_, both being the same ( I used Pin #13, Type 2 phase comparator) on the CD4046BE.

I used Pin # 4 (output Pin) on the CD4046 as the *master_timer* but it was necessary to use a _frequency_doubler_ between the *master* & *slave* to make it work.

Instead of a *Passive_lag_loop_filter* for VCO input, I set up a LF412CN as a  *2nd_order_active_loop_filter* on the VCO input, the response time and dynamic range is excellent, I could see a visible difference over the first order lag loop filter I had tested before.

Here's a quick video of the final prototype up and running,  the scope only shows the complementary push pull outputs of the SG3525.  I am manually changing the Frequency of the SG that the PLL is set up to track.  The dynamic range *capture & lock*  is over 10 kHz in these tests, but it can be expanded by changing values of the resistors on the CD4046BE.  When the PLL reaches it's *capture & lock* limit, the SG3525 push pull output no longer responds to the change in SG frequency input.

https://youtu.be/PE2M8lHR3Eo

Actually for a cheap Chinese pcb, you can't beat the features for the price.  The trick is how to _connect_ it to the PLL.

.................
the note in the photo below is *incorrect*, I should show pin#5 for the *slave* connection and NOT pin #3.
.................

take care, peace
lost_bro

[NickZ]

  After the PLL 10 video, Itsu gave up on that schematic and discontinued the use of that type of circuit. He then went to the commercial PWM TL494, which is now selling on ebay for $6.  As he found that the PLL did nor work keeping up with the varying frequency changes, when changing loads.
  However, I find the NON PLL video 1, much more interesting and useful for my purposes, and so, I will also buy one of those cheap and effective PWM circuit boards, as well. They are now only $6 including shipping. I wish that I had gone that route, months ago.
  Itsu's video:
  https://www.youtube.com/watch?v=1Tgkg1XrDlk

   ADMIN!!!  Please fix this thread.
                  Sorry guys. If someone knows how to get their attention, and have them take care of what I did here, please do so.
 

   Ebay, TL494 circuit board:
    http://www.ebay.com/itm/TL494-KA7500-DC-9V-25V-Driver-Board-DC-DC-Converter-Inverter-Boost-Module-PWM/181449170972?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%26asc%3D41451%26meid%3Da21e8f78e99c4541832e9a49304da384%26pid%3D100005%26rk%3D4%26rkt%3D6%26mehot%3Dpp%26sd%3D232115117269

[itsu]

Nick,

the PLL used in the PLL series of video's was Olegs designed PLL which turned out to be not a real PLL system.
Therefor i stopped using it as it did not do what i was wanting it to do (locking).

So offline with verpies we worked on a PLL system which should be able to do what i want it do do and the result can be seen
in the above by verpies linked diagram.

It is still in breadboard stage and i am putting it on a PCB for doing some more tests, so hopefully i will be able to "lock" the yoke / inductor combination





How about your MOSFETs?   Did you change the primary coils for resistors?   Does it (TL494) work properly then?.


Itsu