Giantkiller moves forward into Full Heterodyning.

[otto]

Hello all,

@point99

show us your coils.

Otto

[Feynman]

@poynt

In all seriousness, I will pretend for a moment you are not trolling and try to explain...   as Tesla and others have stated, the key to understanding radiant energy is in frequencies and resonance, among other things. 

If you only knew the magnificence of the 3, 6 and 9, then you would have a key to the universe."
-Nikola Tesla

I think what we hope to do with the research into digital oscillator control is to provide an open-source framework in which experiments can be conducted systematically, using the scientific method.  Maybe these chips cannot necessarily drive a TPU, but they can provide data on its resonance properties by doing frequency "scanning".  Maybe this helps, maybe it doesn't, but there is only one way to find out.

That is, send different frequencies into a TPU, see how it responds.  Plot the data and upload it.  Or try different wave types (square vs sine) and see how the unit responds. How do different designs affect the frequency response? Can we find resonant peaks? Can we see patterns in the internal resonance?  Can we exploit these properties with digital oscillators or must we contruct analog Colpitts oscillators in order to have clamped feedback?  (ala PLL).  If so, why? There are alot of question which have not been answered, and I personally hope to post some basic research which has not been conducted.  I will leave the advanced TPU theory, design, historical research to those who are qualified, but I think it's important to be able to answer some basic scientific questions.  (ie. what sort of inductive coupling between coils? what is COP? what are the output waveforms, and how to they depend on input frequency and waveform? is frequency response linear or nonlinear? and so on)

On top of this the oscillators GK, Jon and I are building can be used for other things, such as pulse charging batteries, or generating HHO.  But it's more of an open-source computer-controlled digital oscillator framework for doing basic research, automating data collection, and so on.


@all

And speaking of which, I've got some results on the LTC6904 square wave oscillator.  I scanned using the proper control packets, and the error was very much improved.  The error is linear +-2% up to 500khz, after which it becomes highly non-linear all the way to 30Mhz. But still, we are talking +-5% tops, which is not too bad and I am sure can be improved.  I think the next step is to add a high speed buffer IC on the output because the LTC6904 is not designed to drive things directly.  For now, here is the graph of LTC6904 square wave output showing the error distribution.  This was done via computer control on a Linux workstation, using serial over USB as described previously. I am using Victor VC3165 Frequency Counter to measure the oscillator chip output.



This data is from an actual trial I just conducted.  Single Dataset, the error is calculated by dividing the measured frequency by the "expected" frequency.  So 1.0 is zero error, 1.02 is +2% error, 0.97 is -3% error , etc.

And here is a pic of the setup

[poynt99]

Hello all,

@point99

show us your coils.

Otto

We've seen yours and they don't do much do they?

[aleks]

The on time is a bad reference. The shutoff would be a better place to look. This lets the ringing start. But I chose to use square waves. That lets me control the pressure on the electron in 2 directions. I see your point on the sawtooth. Apply slow pressure against the electron then let is snap back. The energy comes back into the device instead of exiting at high speed at the start of the pressure.

With sawtooth you can tune to the resonance correctly. With square wave you'll always have to detune (phase shift) the next 'on' transient in a hope to 'shut off' the counteracting 'off' transient. Beside that sawtooth can be both positive and negative (with only 'on' transient or only 'off' transient) - you decide which one is best suited for you.

[otto]

Hello all,

@point99

at least I made a lot of tests and builded the ECD TPU. So you have no coils. A little advice: just look at the ECD schematic and read the pdf. Maybe even you can something learn. Look at the wires lenghts. You may  "discover" something new.

I would never say that somebedy like dfro is lost until I can prove it. I would never dare to say an idea is crap until I tried it. For myself I can say that this man is right. Its on you to build coils like Mannix showed us and then to follow dfros 2 posts and PROVE that dfro is lost. Until then I have nothing to say to you.

Otto