Kapanadze Cousin - DALLY FREE ENERGY

[Black_Bird]

@verpies

First, thanks for the help.
Second, you are right about the neon. From the picture it seems the resistor value is 1.5k ( brown, green and red), so current across neon should be about (150-62)/1500=59 mA, which will take up to 8.8Watts, high, but not unexpected.

[Hoppy]

@itsu

You have problem with ferrite and probably wiith saturating КТ926А.

Problem can be solved on several ways.  Use high speed current driver to put more currnet in base of transistor. Or use another 74AC00 as buffer between Nano-Pulser and base of КТ926А. Connect all input in parallel and wire it to output of Nano-Pulser. Output pins also in parallel and to base of Transistor.
Or use configuration of two pnp transistors like in NMOS driver posted earlier on this  thread. Also you can use push-pull configration PNP-NPN like level shifter.
Bases connect together and on the output of Nano-Pulser, collector of NPN +12v, collector on PNP -12V, both emitters together on resistor 10-20 ohm, other end of resistor on base of КТ926А.

Ferrite transformer need to be rewinded. I think you have very big ferrite ring and thin wire. For ferrite 7х4х2 mm µ=3000 use 0,35mm wire diameter. 

Wind primary like on the pic below. Wind several test windings - for example 10 , measuire L and calculate permeability of ferrite core. It must be known to determine inductance of unsaturated and saturated core and his relation ratio. Relation ratio is need to determine the frequency response of transformer.
Ideally you need Vector Network analyser but this equipment is very expensive and is tool for professionals in RF design. Verpies have right. Transformers on different cores are the most complex radiotechnical components in electronics.

Сергей В,

With respect, why should Itsu or anyone need to change Dally's design completely as you are suggesting. Dally's design either works as claimed or does not. We are now in a better position thanks to your help and others, to replicate very closely to Dally's build. The only thing I see Itsu needing to do, is reduce the size of his toroid to the specified dimensions on the schematic. It is now clear that if Dally had a direct drive between pulser chip and transistor as per his schematic, then the transistor would not saturate and the diode would not avalanche. It also appears that he omitted the 20R resistor and 220uH choke. The task now is to see if the omission of these components makes any difference.

We must not loose sight of the purpose of this exercise; that is to replicate and hopefully validate Dally's claim. Failing to validate his first design will not help to encourage anyone to build his new design.

Regards
Hoppy

[verpies]

Winding direction notation is easily drawn incorrectly.

Maybe by newbes but Belkin is a Pro and it is not like him to make that mistake.

I assume that the Belkin circuit is driven from a transistor switched from a fast forward edged pulse with sufficient current to saturate the transistor. Under this condition, I can envisage that the diode might avalanche.

With 150V power supply and 2:1 turn ratio, the Tr1 will generate 300V, at most - that's not enough to avalanche the diode at the secondary.

However, even if it does, I do not see how this can have such an enormous impact on the main coil as to produce the overall result as claimed by Dally.

Well, that's the $64000 Question!  Up until that point everything happens according to classic physics and electronics.
If it stays conventional then this will be a dud and if something exotic happens in that big main coil then we will be happy.

Also, we need to know the construction of the Belkin transformer - is it a toroid or was this just assumed by Dally?

I think that this is a standard russian toroidal core K7*4*2 M3000HM, as stated below this Belkin's schematic diagram.

[verpies]

From the picture it seems the resistor value is 1.5k ( brown, green and red), so current across neon should be about (150-62)/1500=59 mA, which will take up to 8.8Watts, high, but not unexpected.

Сергей sees gray, Black_Bird sees green and Verpies sees black, go figure 
Fortunately the doubt is only about the second color band which means that the resistor is either 180Ω, 150Ω, 100Ω  ...not enough deviation to matter.

[Сергей В.]

Yes have two, both burned. He still waiting new Digital Oscillograph pack delivery from China !!