This page has been robot translated, sorry for typos if any. Original content here.

§ 15 Life and death of method No. 13. (unwinding generator)


The ghost of a freebie roams the Internet! Method No. 13, is also known under the names "Winding Generator", "Reactive Power Generator (?!)", "Electric Heating", and the like. A distinctive feature of this method of stopping (rewinding) electric meters is that it does not require changes in wiring diagrams, grounding, and generally interference with the existing power supply circuit. The emphasis is always on what acts on electronic meters. Usually this circuit is called - "Method No. __. Electronic." It’s even hard to come up with a more legendary and discussed method of rewinding counters on the network. It is like communism - it seems here just a little bit and here it is happiness. Detailed information (microcontroller circuitry and firmware) costs $ 11 (although some irresponsible individuals sell it for $ 2), but I found it on the network for free. However, do not run after your electronic and cash money and grab onto the soldering iron.

Scheme, theory and embodiment.

Let's get acquainted with the content of the sent description, the text of the original description is in italics, I will highlight my comments in bold.

Theory and principle of operation of the circuit: In the first quarter of the mains voltage period, energy is consumed from the network, that is, capacitor C1 is charged, but charged through transistor switches A and D which are controlled by high-frequency pulses, that is, energy for charging is consumed by high-frequency pulses. It is known that counters including electronic because they contain an induction current sensor with a magnetic circuit with a limited frequency conductivity (let me disagree about electronic ones - a low-resistance resistance can be used as a shunt for measuring current, which cannot have a limited frequency conductivity in principle. But a coil can also be used, but with a very small inductance, so the error introduced to it is very small.
For a more complete explanation, I will provide excerpts from the standard factory instructions for the electronic meter:
Analog signals taken from a precision current transformer, in the core of which there is no iron , and a resistive voltage divider are fed to the inputs of the microcontroller. The microcontroller converts the analog signals to digital, multiplies them and calculates the average power P (t) every second. Energy consumption is determined by integrating P (t). "
Counters electronic single - phase active energy GEM Operating Instructions

For an electronic counter, the magnetic field generated by the current coil is completely unimportant. Rather, it is not limited frequency conductivity that occurs, but the sampling frequency of the electronic counter. That is, if an electronic meter measures current at a frequency of 100 times per second, and if the load is consumed at a frequency of 200 times per second, then theoretically every second pulse will be unaccounted for and energy will be taken into account only half. And now again, look at the factory instructions:

2. Technical characteristics of the electronic counter GEM .

Accuracy class 1.0 or 2.0 (GOST 30207)
Rated voltage, Un 100V; 120V; 127B; 220V 230V
Rated (maximum) current In 10 (60) A - class 1.0; 10 (100) A - Cl. 2.0
Rated frequency 50 Hz or 60 Hz
Sensitivity 0.004In
Operating temperature range –20 0C to +55 0C
Power consumed by the voltage circuit <0.75W; <1VA
Power consumed by the current circuit <0.05 VA
Counter constant 4000 or 2000 pulse / kWh
Max. number of tariff zones 1 or 2

the sampling frequency of the counter (clock frequency) at a frequency of 60 Hz should be at least 120 Hz (this is one of the theories of higher mathematics, as I recall, I know, but I know for sure that for accurate reproduction of an analog signal the sampling frequency should be double), but most likely it is even higher and the load should be consumed with a frequency of 240 Hz, and it should be synchronized with the internal frequency of the counter, which is hardly feasible in practice.) , and induction, because In addition to the magnetic, they also contain the mechanical part of the measuring system, and have a very large negative error during the passage of rf current. (here, to some extent, one can agree that the induction meter does contain an inertial aluminum disk. But another point arises. In order for the savings to be measured, the frequency must be high enough. And this will strongly clog the household electrical network with harmonics (frequencies different from the network frequency), which will endanger other devices that are not designed to work with high frequencies (different from the frequency of 50 Hz). Moreover, the pulses arising in the network will be very powerful (otherwise there will be no underestimation effect.) Of course, as a rule, at the input of household devices there are usually filters, but they are definitely not designed to filter currents with a power of 1.5-2 kW. Therefore, such filters will surely fly, and after them the devices themselves. And what we have? We’re saving 10 bucks on electricity and burning a TV for 100, and it’s quite possible along with the neighbor’s (the network is shared). The question is, do we need such savings? Although of course you decide. And running a little ahead. Housing wiring also has some kind of capacity, and the question immediately arises - if the frequency is high enough, and this should be so, will all these high-frequency pulses be smoothed out with the wiring capacity. As a result, it may be that the capacitor for a series of high-frequency key openings simply will not charge and it will have nothing to give to the network.) It remains in the second quarter of the period to discharge the capacitor into the network without any impulses, through the same keys. Similarly, the second half-cycle through the other shoulder of keys C and B. I wonder why everything is so complicated? So, for example: Consumed 2 kW, the meter took into account 0.5 W, ideally gave 2 kW, the meter took into account -2 kW. The result of the period is that the induction meter spins back at a speed of -1.5 kW, and the electronic one costs up to 1.5 kW. Where did these numbers come from? Let us leave it to the conscience of the author, but even with working schemes they will certainly be much less.
So already at the stage of theoretical analysis, we found out that this will not work on electronic meters, it remains to find out how efficient the circuit is for induction ones. Although to turn them back there are a million ways much easier and more efficient.

fig. 1 Signal diagram.

The purpose of the circuit elements:
VD1-4, DA-1 in Figure 2. Microcircuit power.
VD5.6, R5.6.7 in Figure 2. A pulse shaper synchronously with a 50 Hz network.
VD 2, R 5 in Fig. 3. rectifier, module power supply.
VD3, C1 in Fig. 3 stabilizer.
VT1 in Figure 3 is a key element.
The pulse frequency f = 1.0 ... 3.0 kHz.
VHS 3-4 pulse shaper output.

General scheme: Fig. 2.

Details: VD1-4 - diode assembly KTs 402B; VD5, VD6-D226. Or analogues 1N4007
C1-20..40mkf x 400 V (you can use both electrolyte and non-electrolyte)
C2, C3-47mkf 12v; C4-22rF.
DA1-78LO5 or KREN5A (5V) or LM7805.
VT1, VT2-KT315.
R1, R2, R3, R4-1.1kom; R5-1com. All 0.5W
Tr-r 1-220v, III-7v, II-12v. low power
Quartz - 4MHz.
Modules A, B, C, D are identical and are assembled as follows:

Fig. 3.
Details: VD1-D243, VD2 - D226; VD3-KS156A.
C1-20mkf. 12v
DA1 - PC120 (optocoupler).
VT1- KT809 (400V, 3A) on a radiator (for all together) 100x150x50mm
VT2- KT315
R1-10kom, 0.5watt
R2-5.1 ohm (multi-watt approximately 10 watts)
R3, R4, -30kom; R5-20kom, R6-1.1kom. 0.5watt
The remaining resistances are 1 W.

I wonder how the transistor VT1 will behave when a negative voltage is applied to its collector relative to the emitter, even more interesting how will VT2 react to this? Will it turn into a jumper? And I wonder why in the R1VD1 chain such a solid resistor? Is this chain can save from something?

Microcircuit D1 is a simple microcontroller that runs according to the program recorded in its memory (in accordance with the schedule for turning on the keys Fig. 1). Programming is carried out through connector X1.
The firmware is tuned to 2 kHz pulses and a duty cycle of 50 \ 50.
These parameters can be changed before compilation.
For programming, copy to notepad and save with the .HEX extension
I did not cite the firmware and source code of the program because the circuit is still inoperable, however, if people ask, I will post them.
In the absence of a programmer or controller, the control circuit can be assembled according to other principles, including logic elements. The winding power, with C1 = 20 μf, is approximately 1 kW. It is interesting by what formula the capacitor energy was calculated, but oh well, more on that later. Increasing the capacity increases the power, but other VT1 transistors are needed. Do not forget about fuses. When setting up, it is better to use C1 = 5uF, not an electrolyte. Yes, the author was modest, in general, electrolytes are certainly cheaper and smaller, but they will not withstand a long constant charge-discharge for a long time and there will be a small BUK.

It so happened that neither the positron nor the megaphase began to answer my questions, and the searches for like-minded people on the megaphase forum and in the positron complaints book led to my IP being blocked. Therefore, I decided to break the task into smaller ones and simulate fragments of the circuit in Micro-Cap. Look at the figure, I extremely simplified the circuit (I threw out the zener diodes and added a second power source, since the optocoupler is closed, I didn’t draw it either) and now I can even see with a naked eye that a kayak will come to the transistor VT2. Why is that so? Yes, everything is very simple - when a negative potential is applied to the collector of the transistor, the K-B junction plays the role of a diode, moreover, with a cathode on the collector.

Why do transistors burn?

( Note by elremont. To confirm that this is all true, I will give excerpts from the forums in which this problem was discussed. I do not write the addresses of the forums, I think this is not so important.

Posted by: PRO (prokrs [dog] {17-04-2005 20:10}
This circuit does not work !!!
More precisely, it works only in terms of burning transistors!
Positron is a scam!

Actually, you can only save 100 watts!
These are my real experiments!
Author: victor (victor1964 [dog] {06-12-2004 08:47}
The scheme is not working. Who wants to burn transistors can experiment !!!
Author: sasa (nefeld74 [dog] {05-05-2005 14:56}
I ordered from POZITRONA (method 13 and extension cord) and they do not work.
Guys who can help me advise the work scheme.
Thanks to everyone.

I did not contact the authors of the posts, but I have no reason not to trust them.)

But as it turned out, these are just flowers ... Let's look at the currents flowing in the circuit. When the capacitor is charged in the first quarter of the period, the current flows through the following circuits: from pin 1 of key D to pin 2 of key D, then through a capacitor to pin 2 of key A and finally to pin 1 of key A. Isn't it strange? Apparently, according to the idea of ​​the positron (megaphase), the current should change its direction after the capacitor or the keys conduct the current equally in both directions.

Look at figure 5, doesn’t resemble anything? But they didn’t guess, this is a circuit of a single-phase voltage inverter (a voltage inverter is called a DC-to-AC converter). Guess which pin the current source (capacitor) is connected to. Again they didn’t guess, the current source is connected to the AB terminals, and we remove the change from the CD terminals ... continue?


Many people ask whether the principle of operation is correct - indeed, induction meters have a large negative error under RF load, but the principle is false otherwise. With electronic meters, this trick does not work (maybe I just didn’t succeed).

And again elremont :

And now I will give excerpts from the forum where professional electricians discussed this problem:

Method number 13 works, although I myself have not tried it. There are several drawbacks - transistors quickly burn out and a rather strong interference goes to the network !, so the telly does not show well (among neighbors too). The neighbors will begin to figure out what is happening.
Hello colleagues! Now there will be a revelation! METHOD 13 Positron does not work and cannot work in principle! It has already been 2 months since we bought this method (11 bucks is a penny), as a result, we got the following results: the diagram is correct there, the firmware of the microcircuit is also clear, the oscilloscope shows that everything is fine. BUT!!! Any attempt to connect the finished device to the network fails. A powerful transistor burns out immediately. Everything was tried: they interchanged the modules, changed the face values ​​of the parts, modified the device, but the result is the same. In addition, there are experts in this matter (not in the CIS) who have explained why this will not work. In addition, if anyone succeeds in assembling this method, such interference will go into the network that it becomes impossible to use the unit in the apartment. I have another scheme that will really work, but the cost of its parts reaches 500 US dollars. In addition to selling, collecting it is unprofitable. RESULT: Positron is a freeloader who trades in unverified schemes. QUESTION: there is such a block the size of charging for a mobile phone, when it is plugged into the socket, it slows down or completely stops the meter (single-phase), while a small, inconspicuous bug should be installed in the meter. We have a real block, we need a bug scheme. Somebody knows? PROPOSAL: who are interested, I will send for FREE the entire positron scheme of method 13. I am waiting in exchange for any interesting information on theft. We can communicate on this topic personally, write to me at . And for general reference, at the same time I will try to open a new topic. Write!
I've been led by the idea No. 13 of positron suckers , etc. No, I didn’t buy the circuit, I developed mine, on another microprocessor, on different keys. He wrote a software program that implements the principle, picked up the circuit, picked up the circuit, checked the waveforms of the signals on the oscilloscope, everything is OK - the capacitors charge, the keys open when necessary, ONLY ONE PROBLEM - THE COUNTER DOES NOT SPIN BACK, EVEN DOES NOT BRAIN! I tried to change the frequency of 2-10 KHz , checked everything again , everything worked, spent a lot of time developing a circuit, developing an algorithm for the “machine”, writing and debugging a program, assembling and debugging a circuit, IT DOES NOT WORK AT ALL ... ... Twists and twists, EVEN DO NOT BRAKE, AS IT IS NOT AT ALL .... It's all about the physics of processes. There is little hope that I have some tricky counter: ECHO co-197 (Kharkov), which I doubt very much (((((((((I look at this thing - like that old woman with a broken trough ..... Show me I’ll give money for the happy owner of working method No. 13 ... Maybe I also sell a 100% "working" scheme for $ 1000

so I once seduced this idea. also sat like "an old woman with a broken trough" it all depends on how to relate to this fact, or forget it as an annoying curiosity, or like replenishing a piggy bank of invaluable experience - and replenish the ranks of the "beaten" - for each of which, as you know, two are "unbroken" give

-A guest-
About the method number 13.
This is similar to another attempt to create a perpetual motion machine in which something is created from nothing.
To test this method, a circuit implementing it was assembled, and it was revealed that this method does not work. The counter disk just stands still. And when you connect other consumers, it begins to rotate where it should be - FORWARD, as if nothing had happened.
The fact is that the energy from the network in the first half of the half-cycle goes into the charge of the capacitor, and in the second it returns back to the network (with the exception of some losses). And it does not depend on whether the pulses charge the capacitor or continuously for half a half period.
I will try to explain this with a simple example.
If we scoop a bucket of water from the well and pour it back, what will change? Nothing.
And now, if we collect water in a bucket in portions (pulsed charge mode) and pour it back?
We will only expend our efforts on lifting water from the well and pouring it back.
In the case of the circuit, this is the purchase of parts, its assembly. And spending money on the purchase of the scheme itself, from which there is no sense.

Although, if you assemble and adjust it, you will undoubtedly gain experience in electrical engineering and electronics.
If you are interested, write to I will send for a moderate fee.
it's all right. we draw energy from the network, then we give it back in the same amount. that is, minus losses due to processes, heating, and so on. BUT !!!!!! the idea then is the error of the measuring device at a high frequency !!!!! I personally think that this air conditioner, which is charging, smooths out our high-frequency pulses. it is charged not from 0 volts at each pulse, but from the previous point of charge.
in general, the idea of ​​a counter error at a "high" frequency seems deliberately false to me - specially launched "to the masses" to lead the wrong way - after all, the idea of ​​an "electronic" winder could not be born out of nowhere (there is no smoke without fire).
I think the essence is the following — it is known that when a direct current is passed through the counter current coil, it “lies” - the fact is that the CHANGE in the magnetic flux in the core of the current coil does not come from a sinusoid zero, but a little later (depending on the value of the direct current) )
our task is to “magnetize” the core of the current coil. as?
1.-connect before and after the battery - does not fit - you need to get into the counter.
2.-magnetize until there is no current in the network - this happens regularly - 100 times per second. when zero in a sinusoid, we have a circuit — a phase in the socket — a shunt coil — an automatic circuit breaker — RCD —... a secondary winding of the power transformer —... RCD — a circuit breaker zero in the socket — in total, a circuit without voltage with a resistance of several Ohms-- in this moment is discharged into the Conder socket (300 V 10000 mKf) through 100 A key-core
the current coil will “fly away” into saturation and turn into a permanent magnet - and if further current flows through this coil, the core’s magnetic field CHANGE will not further recharge our partially discharged conder and proceed to the negative half-wave of the sine wave. this is just a theory born from looking at 100A keys
a guest
The counter, in fact, has a negative error at an increased frequency.
If you turn on the counter according to the STANDARD scheme, connect it to the low-frequency generator and increase the frequency from 50 Hz, then with increasing frequency the disk slows down and stops at 560 ... 600 Hz (counter type CO-2, for other types the frequency may be different). The voltage was sinusoidal and rectangular, without a constant component. In both cases, the effect is the same.
A pulse charge (interruption of the charge current) alone does not produce anything. The average value of the current through the capacitor is the same as without pulses. And little depends on the frequency. This is verified.
In general, if the capacitor is charged from a separate source, and discharged into the network, the counter starts to rotate backward. But where to get this separate source, and is it not better to immediately supply the load from it, because he must develop the same power that we want to rewind. At the same time, we become a source of food for our neighbors.
The idea with magnetization is likely to work, too, but from a separate source: a battery, for example. Significant power is required to charge a 10000 μf Conder to 300V in less than a quarter of a period.
a guest
Even if there is a working circuit, it most likely works with a separate current source. Its power should be not at all big.
In order for the meter to unwind at a speed of 2 kW / h, it is necessary to pass a current of 9A through the current winding, while the voltage on it is only 2.7V.
The power is 24W, and if you unwind 1 kW / hour, then only 6 watts. The difficulty lies in how to transfer it to the current winding through a network in which there is a voltage of 220V.

These are the pies ... If, even after what you read above, you have an itch to spend 11 bucks, buy a better book on electrical engineering. There will definitely be more benefits.