Hello Ariovaldo,
 
 thank you for posting a video of your replication.
 
 Unfortunately the focus is not clear when you show the power meter. Can you please write down in a post the power just before resonance and when resonance lights the bulbs.
 
 I'm also trying to understand what happened at the end when only one bulb lights. Do you know what causes this?
 
 Thank you for sharing

Luc

Yeah, the video is bad...I have a regular job and I just have time at night or weekend and my intention is to do some more tests. About the end, when just one light bulb still on is because when you increase or bring down the speed we lose the resonance and the system starts to pulse and the result is voltage spikes. The others lights bulbs just blew out.


Cheers


Ariovaldo

gotoluc (and others interested in QEG audio spectrum):

Thank you Rfacts for taking the time to do the spectrum analysis and post your results.

I think I better understand why they say 400Hz is the resonating frequency. They are speaking of mechanical resonance and not necessarily electrical resonance.

Luc

I will write down everything that you suggested and I will do my best this weekend to test it and put in a table that can be understandable.


Ariovaldo

Thank you very much.

As a suggestion, if you are concerned about the high voltages when the system is in resonance, you can put an oscilloscope (or a multimeter) across only one of the series capacitors.  That will allow you to verify that the system is in resonance with no load and the measured voltage will be lower and will likely not damage your oscilloscope or multimeter.

MileHigh

gotoluc (and others interested in QEG audio spectrum):

Appreciate you sharing the QEG Morocco video that you edited, I'm sharing some audio data from that video.  I fed the audio from your edited video to a PC based real time FFT audio spectrum analyzer.  This allowed me to capture and save the audio spectrum of the QEG as it was ramped up with the light bulbs off, when the light bulbs flashed, and when the light bulbs stayed on.  I left the SA settings at default and have attached the screen captures that I saved.  Most of the screen captures were taken with a span of 0-500 Hz (50 Hz per division) to provide more detail of the most significant part of the audio spectrum.  There are some screen captures with a span of 1KHz, 2KHz and 5KHz to show higher order harmonics. 

There is a frequency at 100 Hz that first appears 2 seconds into the video when the variac knob is first rotated to start the QEG and this frequency is present throughout the QEG run.  This 100 Hz frequency must be the hum from the full wave rectifier which is used to rectify the 50 Hz AC input to power the DC motor, so it serves as a good calibration check.  The light bulbs start to flash when ~400 Hz double peak frequencies appears, I happened to capture it at the 30 second mark in one of the repeating cycles that you edited to extend the flashing light bulb view time.  The ~400 Hz double peak frequencies only appears when the light bulbs flash on.  The light bulbs stay on when the 440 Hz frequency is present along with a 400 Hz frequency, at this point both of these frequencies are constantly displayed until the QEG is ramped down.  Most of the screens have a red marker at the 440 Hz frequency, you can see the frequency the marker is set to at the top of the captured screen.  You can also see an 800 Hz and 880 Hz harmonic frequency on the attached file with the 1KHz span.  Regardless of the QEG outcome it will be very interesting to find out how the QEG output is optimized.

It may be that the optimum output is achieved when the 400 Hz and the 440 Hz frequencies are tuned to match.  Is there enough data here to determine which is the mechanical and which is the electrical resonance frequency?  If mechanical resonance and the electrical resonance (or a harmonic frequency for parametric operation) need to be aligned a real time FFT audio spectrum analyzer like this one may prove to be a very useful tuning tool.  The screen capture with the 5KHz span that I've attached was saved to display the whole instrument control panel so you can see the software application was developed by Fatpigdog Industries.  I purchased the Excalibur 4.06 professional version, it has a very intuitive control panel with a very good set of features and the input audio can be from a file or a microphone - technical support is provided by email.  I'm not associated with them, I just think it's a good product, so for anyone interested in more info it can be found, downloaded, and a registration code purchased at this web site:

http://www.fatpigdog.com/SpectrumAnalyzer/Excalibur.html

To insure that we're using the same reference, this video is the source of the audio:
https://www.youtube.com/watch?v=0ALGxtNZ_2Y

Note: Attached files were converted from .bmp to .jpg to minimize file size.

Nice work. But.... but.... no Linux version! And nonfree.

For us Linux users, there is Audacity.

http://www.youtube.com/watch?v=VZbZa99ocPU

Check this out....


Even I had the "hope" to find out something new, until now, my generator is very inefficient.
I will do some more tests this weekend.



https://www.youtube.com/watch?v=8x0Zpb67Uw4


Cheers
Ariovaldo

Awesome video Ariovaldo!

From my SPICE simulations I have found that this device has some unusual performance behaviors, One of which is that the more resistance you put in the primary circuit up to a certain point which seem to be roughly around 10% of the the total impedance of the primary at the operating frequency, the more power the device will output.  I am using an 11 k load in my spice simulation. 

I have also found that the system will go through surging oscillations if the resistance of the load is tool light, and that if the resistance of the load is too high the system will simply fail to go into resonance.   So, the moral of the story is more load is generally better than less load.  Additionally, adding load resistance will take voltage off the primary windings, so again more load is better than less load as far as helping the system to avoid over voltage transients in the primary windings.

I'm thinking that your generator would easily light up a series sting of 20 100W light bulbs maybe more put in series with the primary.

Cheers,

FB

"It's not so easy to teach a new trick for an old horse", but I will try to make a better one in a couple days. Also, the KW-meter positions was not so good.
Making it short, the power consumption was about 1200 watts and the load was not more than 6 x 60 watts lights bulbs. The output voltage in each coil was about 600 volts, but drops with the load. As I said, I'm just testing and is not my intention to make polemics statements.  I'm just an old guy that has as hobby to build and test stuffs..
Independently of the results, is an interesting approach the way that the system works, having the resonance when reach a certain speed. In this case, was about 2800 rpm, but I can bring it down, increasing the capacitance. Also, it is a danger machine, since the voltage is very high in the excitation (high voltage) coils.


Cheers


Ariovaldo

Voltage generating in the system I think is related to drive frequency by the relation d/dt, that the rate of change in the inductance of the system created by the rotor speed vs time.  slowing down the system will lower the output voltage.

I'm very glad to see you have a flywheel on the motor.  I think this will be an important part of the system.  Ultimately, I would like to see the motor, flywheel and CEG all on one shaft for the best efficiency on the mechanical side of things.

Not shown is the very thing for which it is named.  You cannot have a rat trap be effective if there is no rat.
"Ou es le" Quantum exciter coil   

Very true.

But if you build it like Mr. Tesla did, it shouldn't need one.