Quantum Energy Generator (QEG) Open Sourced (by HopeGirl)

[F_Brown]

I think you mean increase here.

I'm pretty sure I mean decrease.  This touches upon something that has been buggering me for a while now.  "Over-Unity" as well as "Under-Unity" are quite misleading buzz words.  With all factors about a device taken into account any proper efficiency analysis should show that the device operates exactly at unity.  If the COP deviates either way from unity, then the analysis is flawed; something is missing.  That is to say in a proper analysis all input sources are identified and quantified, and all output sinks are identified and quantified.

So my view on the core losses is that for the same input power, core losses would have reduced the power available to be dissipated by the DC resistance of both the primary winding and the load before the power even got that far, because the power has to go through the core first in order to get there.  Just as easily one could say for the same amount of power dissipated in the primary winding and load with core losses added on top, then the input power would have to be adjusted upward to accommodate that increase in the total output power. 

I think you have worked out the magnetic field strength.  If we also know the materials, then we know the conductivity and can compute the eddy currents and resulting Lorentz forces.It could also be that the basic behavior is right but that the numerical approximations are not accurate enough.  At least for purposes of diagnosing the problem adding series and shunt resistances as you have already started to do could provide a good deal of insight.

Yes, the damping circuitry is a band-aid and will affect the accuracy of the resulting simulation.  Ultimately, I would like to run the simulation without any "stabilizing" additions.

The FEMM analyses took into account the the properties of the M19 core material, although FEMM lacks the facilities to calculate eddy currents created by motion.  So some other application would have to be used to determine the magnitude of those rotor currents and drag forces they create.   

What I missed was to collect data on the saturation current level for the primary winding at various rotor angles and include that data in the arbitrary inductor model.  That's going to require another round of FEMM analyses and the creation of a third look up table to properly implement.

Also, I noticed I really should have run the FEMM analyses for the primary inductance and rotor torque out to primary current of 10 amperes instead of just 4, in order to have sufficient headroom for transient current spikes.  So for now it's one step back...


Pmgr:

Have you considered that the entire secondary of the QEG may be a "Red Herring" created by Timothy Thrapp, in order to misdirect people about where power may be successfully extracted from the device.

[F_Brown]

Well, I did a few more FEMM analyses and found that the saturation current for the QEG primary only varies from 4.5A when the rotor is aligned with a pole to 6.5A when the rotor is aligned in between poles.  So, I set the saturation current level in the arbitrary inductor model for the median value of 5.5A, and all the simulation problems went away.

I was able to remove all the "stabilizing" additions.  The QEG model 2.1 now runs 10 times faster, which is about the same speed the 1.0 version ran at, has exponential build up of current and voltage, has a roughly sinusoidal wave form in the primary, has asymmetrical cogging torque.  I think my tables are working as intended now.

More tests under way...

And I do still need to generate the saturation current vs rotor angle data and create a third table for that.   

I also slowed down the rotor drive to see what would happen

***

Date: Tue May 13 17:19:51 2014
Total elapsed time: 42.525 seconds.

RPM 6,000 Load 1k5 Tank 200 Hz

torque: AVG(v(nm))=-21.6878 FROM 0.4 TO 0.9
load_rms_volts: RMS(v(load))=3476.13 FROM 0.4 TO 0.9
load_rms_amps: RMS(i(load))=2.31742 FROM 0.4 TO 0.9
load_avg_va: AVG(v(load)*i(load))=8055.65 FROM 0.4 TO 0.9
load_avg_volts: AVG(v(load))=0.19909 FROM 0.4 TO 0.9
load_avg_amps: AVG(i(load))=0.000132727 FROM 0.4 TO 0.9
coil_avg_va: AVG(84*i(l1)**2)=451.117 FROM 0.4 TO 0.9

13,677 watts input
8,507 watt output
0.624 COP

[F_Brown]

Ari,

I tried to match your setup, with 30 ohm load, and about 900 watts draw at the motor.  This is what I got. 
I think it's in the neighborhood aye?

Date: Tue May 13 18:38:54 2014
Total elapsed time: 36.147 seconds.

Load 30 Ohms (in series with primary)
Rotor speed 100 RPS
Tank Resonance 200Hz

torque: AVG(v(nm))=-1.2249 FROM 0.3 TO 0.8
load_rms_volts: RMS(v(load))=73.9814 FROM 0.3 TO 0.8
load_rms_amps: RMS(i(load))=2.46605 FROM 0.3 TO 0.8
load_avg_va: AVG(v(load)*i(load))=182.441 FROM 0.3 TO 0.8
load_avg_volts: AVG(v(load))=0.00585677 FROM 0.3 TO 0.8
load_avg_amps: AVG(i(load))=0.000195224 FROM 0.3 TO 0.8
coil_avg_va: AVG(84*i(l1)**2)=510.836 FROM 0.3 TO 0.8

769.7 Watts input to rotor shaft
182.4 Watts dissipated in 30 ohm load, COP 0.237
693.2 Watts dissipated in load and primary winding, COP 0.901

769.7 watts provided by a 85% efficient motor means motor
will draw 905 watts from mains.

Here is image of voltage and current waveforms across the load:

[Farmhand]

I'm pretty sure I mean decrease.  This touches upon something that has been buggering me for a while now.  "Over-Unity" as well as "Under-Unity" are quite misleading buzz words.  With all factors about a device taken into account any proper efficiency analysis should show that the device operates exactly at unity.  If the COP deviates either way from unity, then the analysis is flawed; something is missing.  That is to say in a proper analysis all input sources are identified and quantified, and all output sinks are identified and quantified.

You're quite right Mr Brown, if all input and outputs + losses are considered everything runs at unity. This only leaves one logical definition for over unity and that is anything with a C.O.P. of over 1.0 is over unity, this can be done as C.O.P. is a function of our input to actual output.

An example of very high C.O.P. devices are wind turbines and solar panels.

An example of a self runner is a solar system with storm shutters or solar position tracking where the shutters or the tracking motor is powered by energy collected by the solar panels themselves. Or a solar powered vehicle even. Self runners already exist.

I believe the holy grail is a machine that can self run 24/7 and output excess power by utilizing an environmental energy input so that the fickle sun and wind become obsolete.

However I don't think the QEG has any such facility for any free environmental energy to be input. Although there are other possibilities but not many people think of the actual mechanism of harnessing extra energy from a given source and set out to do so, most are hell bent on replicating shaky claims in the hope they will be the one to show over unity and save the world.

I wish all the thousands of minds would think in their own way about how energy can be utilized from the environment.

Some see lighting LED's for extended periods as useful, I wish them well but I don't agree, considering our environment is so rich in man made electrical noise LEDs can be lit in a lot of funky ways but OU it isn't and not very useful. If I run a bought from the shop 300 Watt inverter from one 12 volt battery with another next to it the unused battery can excite a coil and light an LED from one battery pole due to the HF noise. Ala Stiffler ect.

When my electric fence was in good shape and had no grass shorts I de energised it and scoped the open wires and ground, I was able to see three or more different frequencies on there. I can pick up the electric fence pulses on my scope from over 50 meters away by scoping the ground and a coil. I'm thinking of a way to calibrate the scope trace when the fence is working well without shorts from branches or breaks, so that I can diagnose the fence's operation from my experiment hut, check for shorts. 

..

[MarkE]

COP and efficiency coefficients generally refer to the useful output divided by the expended input.  Output that is in the form of unused heat is usually considered loss.  For the same sunlight, a solar thermal collector where heat is the useful output is around 75% efficient delivering that heat to the heat transfer fluid, while PV where electricity is the useful output is constrained to about 20% for single layer modules.  For a machine like the QEG the idea is to be a generator.  Anything less than self-running makes it a complicated and noisy heater.