This is the most interesting part of the work, where we will consider various options for the SGP, using fairly simple formulas (1.2-1.5) from the previous section. To do this, we will change the waveform for the generators G1 and G2, the phases and the frequency ratio between them, vary some values of the elements of the circuit 1b, substitute all this in the formulas, and see what we get.

1. Rectangular pulses and the same frequency in G1 and G2
Consider the version of the PMG proposed earlier, where the signal of the generator G1 was rectangular, varying from +1 to -1, in this case, the inductance L was changed by the generator G2 from 1 to 0.2 (Fig. 2b). For ease of viewing, graph 4 has been supplemented with voltage values U, a change in inductance L, and is now shown in Figure 6. We made the resistance R equal to one.

The calculation method is as follows. We obtain the current values for the graph from the formula (1.2), and already proceeding from these values we calculate the power according to the formulas (1.3-1.4), and from here - the increase in efficiency according to the formula (1.5). In this case, the gain is as follows: \(K _ {\eta 2} = 15.6 \).

If the phase of the inductance change is shifted by 180 degrees, as in Figure 2c, then the increase in efficiency will change slightly: \(K _ {\eta 2} = 17.8 \). By changing the inductance not five times, but two, the increase in efficiency will decrease by about 80%: \(K _ {\eta 2} = 9.7 \). If the resistance R is increased to ten, then the increase in efficiency will greatly decrease, up to one: \(K _ {\eta 2} = 1.04 \). You can check this data yourself and enter your own using the following subroutine for MathCAD.

Fig.6. Circuit current - red graph, voltage U (t) - blue, change in inductance L (t) - green

The advantage of this version of the PMG is obvious: the same frequency and even the phase of rectangular oscillations for G1 and G2, which can be achieved, generally speaking, with one generator. The disadvantage is the large mutual influence of these generators on each other, which leads us to a parametric circuit of the first kind and to a subunit increase in efficiency. However, this mutual influence can be partially compensated by using special cores, for example, with magnetic anisotropy.

2. The frequency of G1 and G2 differs twice, the form of oscillations of G1 is sinusoidal
The above example works at the same frequencies in G1 and G2, which does not always give good results in terms of mutual non-influence of these generators on each other. A more interesting, but more complex PMG is obtained if the frequency of the generator G2 is doubled relative to G1. Moreover, the signal form for G1 is sinusoidal (Fig. 7).

Fig.7. Circuit current - red graph, voltage U (t) - blue, change in inductance L (t) - green

The remaining parameters of the circuit are as follows: inductance L changes from 0.3 to 1, and the frequency of G2 is slightly out of phase with respect to G1, the resistance R is 0.5. At the same time, the increase in efficiency in such a scheme can reach values of 4 and higher. There is also a drawback to this approach and lies in the fairly accurate adjustment of the phase shift between G1 and G2, as well as the careful selection of the load resistance. Nevertheless, there is evidence that this approach has allowed some inventors to create highly efficient PMG devices.

There is also a shift in the negative (positive) part of the current graph here, but this is not so obvious compared to the previous example. The displacement is obtained due to the asymmetry of the oscillations themselves relative to zero. You can check this and make your substitutions in the following subroutine for MathCAD.

3. The same frequency in G1 and G2, the form of oscillations in G1 is sinusoidal, the signal G2 is pulse
Perhaps the most interesting version of the PMG is a pulsed one, when the inductance in a parametric circuit changes abruptly and impulsively, in this case, the pulse length is many times less than the oscillation period. There are many sub-options here when you can change the phase between G1 and G2, while simultaneously changing the load resistance and the coefficient of inductance change. You can experiment with such changes in this MathCAD subroutine.

The following graph showsthis option (Fig. 8). The resistance R here is equal to 0.3 and when this value decreases (up to a certain limit) it gives an additional increase in efficiency. The length of the pulse that changes the inductance L is here equal to 2% of the oscillation period, and the coefficient of change is 5. At the same time, the shape of the current in the circuit remains almost sinusoidal, but slides into the positive region, which, with these parameters of the circuit, gives an increase in efficiency by 4 times. It should be noted that this is far from the limit.

Fig.8. Circuit current - red graph, voltage U (t) - blue, change in inductance L (t) - green

The disadvantage of this version of the PMG is the precise phase control between G1 and G2 and a rather sensitive load. But the benefits for researchers are enormous. The most important thing is that the frequencies in G1 and G2 coincide, which means that the master oscillator can be common. It remains to create the correct inductance change and phase shift between the oscillation in G1 and the pulse in G2.

Let us dwell in more detail on the impulse change in inductance L. The simplest and known method for at least 100 years is the spark method, when the ferromagnetic material of the coil core is excited by a pulse current in the arrester circuit. By the way, in the same way it is possible to excite the environment in the case when the core is absent. Also of interest is the method of shaking the coherer by a weak electric field [1], which leads to rather significant changes in its structure and permeability. A slightly less well-known method is mechanical. In this case, the permeability of the core L changes by changing the position of a nearby ferromagnetic material, or a magnet (Adams, Bedini generators). There are more exotic ways to change the inductance, for example like this, when ferrite in an electric field, at certain frequencies, changes its permeability.

There is also a completely unusual way of parametric change in inductance, due to changes in the properties of the core at the moment of impulse. It requires some preparation and patience, because finding such a point, while simultaneously selecting the voltage, pulse front, its length and material, is a rather difficult task. usually ending with rolling into a parametric of the first kind with the corresponding classical efficiency. Another disadvantage of this option is the partial or complete destruction of the core material over time. But this method does not require a second generator at all, and pulses are supplied to the coil from the first (G1). From it, or from the secondary winding, it is removed into the load.

We can only talk about the use of such generators as sources of free energy if we manage to find a relatively low-cost way of changing L with the help of G2. In this case, the power in the load \(P_R \) from formula (1.5) must exceed the sum of the capacities spent on the operation of two generators: \(P_U + P_ {pm} \). In this case, a completely natural question arises: where does the additional energy come from?

There are many answers to this question. Some researchers argue that such energy is taken from the ether, but ether, by its definition, is energy. A more modern approach to this problem assumes that energy is pumped from the environment, for example, from the earth. Indeed, some PMG installations require grounding, and we already know how parametric circuits can receive charges from this huge reservoir. But the author is still inclined towards a fuller use of internal energy of the electron in such systems, when it receives acceleration, or increases its energy by acquiring a higher potential, at the time of changing the inductance of the coil. This assumption is supported by the fact that the current graph is shifted to the positive or negative region, while the coil itself becomes a power source.

From graphs 6-8, one regularity is immediately visible, which consists in the mandatory shift of the current in the parametric circuit relative to the voltage U from the generator G1. This shift should be about 90 degrees and not change much when the inductance L changes. Thus, the first step in tuning the PMG is precisely to achieve such a shift. After that, you can proceed to the second part - connecting and configuring the G2. In the most general case, when these generators work together, it is necessary to achieve a shift in the current waveform to one of the areas: positive or negative. This will mean the achievement of the required effect.

I must say a few words about the core material for the L coil. It is known that even at optimal operating frequencies, ferrites have rather large magnetization reversal losses. In addition, they have completely different characteristics Stoletov curve for different magnetization planes. Often, the initial magnetic fieldTheir value in different planes differs by orders of magnitude. It is more optimal to use cores made of modern amorphous materials, for example, metglass [2-3].

In this work, the fundamental possibility of constructing free energy devices based on second-order PMGs with two generators was shown. These studies were done using classical methods for calculating electrical circuits and compared with working devices. The calculations did not take into account the energy costs for the parametric change in inductance, but, as practice shows, they can be both relatively large and relatively small in comparison with the overall energy of the device. It all depends on the method of such a change, the material of the core and other parameters of the circuit that the researcher chooses.

Subject to relatively low costs for parametric change in inductance, it is possible to achieve rather high rates of increase in efficiency of the second kind, which will cover the energy costs for this change, and even, in part, for the power supplied to the circuit. In the limit, a self-excitation mode is possible, which must be limited by special additional elements to protect the circuit.

As the author sees it, the best option for the PMG can be one in which the same frequencies are observed for the generators G1-G2, and the change in inductance L will be made in pulses (option No. 3). Also, researchers should pay attention to modern metamaterials from which the coil core will be made. It is best if they have anisotropic magnetic properties to reduce the mutual influence of generators G1-G2.

This technique assumes that the device operates on a relatively smooth section Stoletov's curve, which characterizes all ferromagnetic materials. In a real coil, this can also be realized only partially.

The calculation method presented here can be applied to circuits containing parametric capacitance, as well as to other types of electrical circuits with parametric elements.

For the materials of this work, a specialized calculator was developed.