EA003341B1 - Method for excitation of an electric arc and devices therefor - Google Patents

Abstract

1. A method for excitation of an electric arc in which prior to supplying an excitation pulse voltage to an arc interval an output of the electric arc power source is disconnected for the time of increasing of short-circuit current to the level of stable arc current, characterized in that after reaching the level of stable arc current by short-circuit current a forcing voltage is supplied with an amplification voltage having the same polarity and being of a greater amplitude after which a delayed excitation voltage pulse of an opposite polarity relative to the electric arc power source is transmitted in the arc interval, the rated value of said voltage exceeds the above-mentioned amplification voltage and has a shorter flank. 2. The method of claim 1, characterized in that the time delay of the excitation voltage pulse is ranged between 3 to 300 nsec. 3. The method of claim 1, characterized in that the ratio of the time delay of the excitation voltage pulse to the duration of its shorter flank is 1 -30. 4. The method of claim 1, characterized in that the forcing voltage is formed by transforming the power source energy from its circuit being short-circuited. 5. The method of claim 1, characterized in that the excitation voltage pulse is formed by transforming the power source energy from its circuit being short-circuited. 6. A device for excitation of an electric arc, comprising an excitation voltage pulse source, a first and a second arc interval clamps, an electric arc source, parallel to which a blocking condenser and a key with a threshold element are connected, characterized in that the device is provided with an internal amplification voltage source, a blocking inductance and a long "cable-type" element, said internal amplification voltage source is connected in accord to the electric arc power source, an excitation voltage pulse source is connected to the electric arc power source by its one clamp via a blocking inductance and by other clamp directly, and via a long "cable-type" element excitation voltage pulse source terminals are connected to the first and second arc interval clamps. 7. The device of claim 6, characterized in that the internal amplification voltage source is designed in the form of series-connected key element, resistance and inductance element sand a voltage capacitance source. 8. A key with a threshold element, comprising a capacitor, a saturating magnetic circuit with a working winding, one terminal of which is connected to a magnetic key input, the other terminal to an output via a capacitor, characterized in that a bias winding is provided for series connection with power transformer primary winding to the saturating magnetic circuit of the electric arc power source. 9. An excitation voltage pulse source, comprising a capacitor, a spark-gap, a transformer saturating magnetic circuit with working and high-voltage windings, a capacitor and a spark-gap are connected to said high-voltage winding, characterized in that a bias winding is provided for series connection with power transformer primary winding to the saturating magnetic circuit of the electric arc power source.

Description

The invention relates to the field of electrical engineering, and more specifically, to a method and apparatus for powering an AC welding arc and an auxiliary arc of a plasma torch and can be used in apparatuses capable of welding oxidized and rusty metal, steel, stainless steel, aluminum and other metals, welding of which by apparatuses AC was previously considered impossible.

State of the art

The electric arc was accidentally discovered by Petrov, and then by Devi. However, the essence of the phenomena was not understood correctly by them. Even now, calculations of electrical circuits with negative resistance are complicated, although they can be considered mastered for high-level engineers. The arc has an 8-type current-voltage characteristic and is stable only when powered by a current source with inductive or sufficiently large active resistance. Connecting a capacitor of a sufficiently large capacity parallel to the arc (especially at low currents) inevitably breaks the arc. The AC arc, opened by Yablochkov, with an inductive current limiter, the power of which does not dissipate, is difficult to perceive. The moment of reversal of the sign of the current with arc rupture has not been sufficiently studied. The oscillator invented by Vologdin in the thirties remained a purely empirical invention, despite its widespread use. Starting an arc with a spark from the first impulse to a cold part has not been investigated. To date, there have simply been no devices capable of implementing it. Some provisions that were considered generally accepted were simply erroneous. So, the recommendations to use pulses with an energy of 1 J or more to start up, look surprising if in our devices the pulse energy is measured in millijoules. This position can be understood if we take into account that there was no electronic equipment that allowed one-time processes to be recorded in the time range starting in fractions of a nanosecond.

A spark discharge of 1–3 mm at a voltage of 5–15 kV occurs in tenths of a nanosecond in a channel with a width of a few microns and a temperature of 100,000 K and higher (a characteristic violet color). Pressure 500 atm. The energy intensity of such a plasma is 100 times greater than the TNT equivalent. The shock wave expands the channel at a speed of about 10 microns per nanosecond (10,000 m / s). The temperature drops to 8 - 6 thousand K, characteristic of an arc discharge (white color). The heat content of such a plasma is 3 mJ / mm ³ . Plasma conductivity 1500 Sim / m. The thermal conductivity of air (nitrogen) is λ = 0.0255 for T = 273 K and λ = 0.151 for T = 5000 K. The thermal conductivity of the plasma is λ = 4 for higher temperatures, i.e., it exceeds the thermal conductivity of air by more than two orders of magnitude. This numerical data is important for understanding the statics and dynamics of an arc.

In a stationary (static) state, the heat release from the conduction current is exactly equal to the heat sink. The shape of the body of the arc can be found by solving the equations of thermal conductivity and electrical conductivity under obvious boundary conditions. Both of them are sharply dependent on temperature, that is, the equations are nonlinear. As indicated above, the thermal conductivity of the plasma is more than two orders of magnitude higher than the thermal conductivity of air, the electrical conductivity of the plasma is given above, and air is generally an insulator. A similar (but simpler) problem was solved in 1950 under the guidance of A.D. Sakharov (Nature magazine No. 8, 1990, article by Yu.A. Romanov, Father of the Soviet hydrogen bomb).

Therefore, the description of the excitation of the welding arc in the literature does not correspond to the truth.

Known power sources for an electric arc with a saturable transformer and a current-limiting inductor in the primary circuit (8and No. 1839648, 09.21.90.).

The arc is excited in them by high voltage pulses generated from the voltage front when the magnetic circuit exits saturation. Oscillograms show that the arc is not clearly excited, only from the third or fourth pulse from the train of pulses following 50 μs does the excitation occur.

A known method of excitation of an electric arc, in which, before applying a pulse of excitation to the arc gap, short circuit the output of the power source of the electric arc for the duration of the rise of the short circuit current to the level of a stable arc current and a device for exciting an electric arc containing a source of excitation pulses, an electrode holder, an earth terminal, power supply of an electric arc, in parallel with the tires of which a blocking capacitor and a key with a threshold element are connected (KI No. 2011493, 07.27.1991.).

In the known inventions, it is proposed to short-circuit the power source with the equivalent inductive output impedance and keep it closed until the current in the inductance rises to a level sufficient to maintain the arc current, then open and supply an excitation pulse. Practice shows that such exciters provide excitation of the welding arc from the first pulse remotely, at a distance of 0.5 - 1 mm from the part.

Oscillographic observations show that in the first nanoseconds after the passage of the spark, the voltage in the arc gap decreases from 7 - 5 kV (this is the spark voltage) to a voltage of about 500 - 300 V, and then decreases to 150 - 70 V in tenths of a microsecond. Further it should decrease to a voltage of 20 V, characteristic of an arc, in a time of a few microseconds. However, it persists 150 - 70 V for a time in tenths of a millisecond. If the voltage of the power source at the time of excitation is 35-50 V, then the voltage on the arc reduces the current of the inductive source and the excitation may not take place. The reasons for this behavior of the voltage across the arc are unknown. It has been observed that this occurs in the case of a contaminated and oxidized part. Although the spark arises at a distance of 2-3 mm from the part, the arc does not light up until it touches, since the main voltage drop occurs in a narrow punched oxide channel with an energy intensity of about 10 J / mm ³ .

Disclosure of invention

The basis of the invention is the task of developing a method and device capable of ensuring the normal process of arc excitation and cooking of contaminated and oxidized metal, steel, aluminum, etc.

This problem is solved in that in the method of excitation of an electric arc, in which, before applying a pulse of excitation voltage to the arc gap, the output of the electric arc power source is shorted for the duration of the short-circuit current rise to the level of the stable arc current, according to the invention, after the short-circuit current reaches the level of stable the arc current to the output of the power source of the electric arc serves a boost voltage of the same polarity, exceeding the voltage of the power source, and then on the arc the interval is supplied with a time delay, the pulse of the excitation voltage of the opposite polarity with respect to the power supply of the electric arc, exceeding the nominal voltage and having a shortened front duration, while the delay time of the pulse of the excitation voltage is selected in the range from 3 to 300 ns, and the time ratio the delay of the excitation voltage pulse to the duration of its front is 1 - 30, in addition, the boost voltage is formed by transforming the energy of the power source from short-circuiting minutes of its circuit, and the excitation voltage pulse is formed by transforming from the power supply circuit shorting its energy.

The problem is also solved by the fact that in the device for exciting an electric arc, containing a source of pulses of the excitation voltage, the first and second terminals of the arc gap, a power source of the electric arc, in parallel with which a blocking capacitor and a key with a threshold element are connected, according to the invention, an internal boosting source is introduced voltage, blocking choke and the element is a long line, while the source of internal boost voltage is connected according to the power source of an electric arc, the source of excitation voltage pulses is connected to the electric arc power source with one of its outputs — through a blocking inductor and the other — directly, and through a long line element, the conclusions of the excitation voltage pulse source are connected to the first and second terminals of the arc gap, while the source of the internal voltage can be made in the form of series-connected key element, resistance and inductance elements and a capacitive voltage source and I.

In addition, to enhance the effect, a bias winding is introduced into the key for the electric arc excitation device, which contains a capacitor, a saturable magnetic circuit with a working winding, one terminal of which is connected to the input of the magnetic key, and the other terminal through a capacitor with an output, according to the invention connections to the primary winding of a power transformer with a saturable magnetic circuit of an electric arc power source.

In addition, in one embodiment, a key for an electric arc excitation device comprising a capacitor, a spark gap, a saturable magnetic core of a transformer with working and high voltage windings, the last of which is connected to a capacitor and a spark gap, according to the invention, a bias winding is introduced for series connection with the primary winding power transformer with saturable magnetic circuit of an electric arc power source.

In addition, it is preferable in the method for preparing the key to operate the electric arc excitation device made in the form of a cascade chain of η magnetic keys, including setting the magnetic circuit of the 1st key in the initial magnetization state, according to the invention, an offset current from ( 12) -th magnetic key and noise immunity current from the (1-3) -th magnetic key.

The essence of the invention lies in the fact that due to the introduction of the above elements and the connections between them, conditions are created for arc excitation, namely: during the passage of a negative pulse along a long line and an arc gap, a current with a polarity coinciding with the polarity of the future arc current accumulates in the blocking inductor and the source of internal boost voltage provides a significant arc current necessary for the removal (evaporation) of contaminated or oxidized surfaces.

The essence of the invention for a method of preparing a magnetic key for operation is that a current pulse is supplied to the interference protection winding of the ί-th magnetic key before the interference from the (1-3) -th magnetic key arrives and ends simultaneously with the end of this interference.

Brief Description of the Drawings

The invention is further explained in the description of examples of its implementation and the proposed drawings.

In FIG. 1 shows a functional diagram of the proposed device, in FIG. 2 is an equivalent circuit at the time the spark passes, in FIG. 3 - the shape of the body of the arc at the end of the spark, in FIG. 4 is a timing chart for arc currents and voltages; FIG. 5 is a circuit diagram of a first embodiment of the device of FIG. 6 - a device for exciting the auxiliary arc of the plasma torch, in FIG. 7 - waveforms of operation of a variant of the device shown in FIG. 5 and stabilization timing diagrams, in FIG. 8, 9 and 10 are circuit diagrams of the second, third and fourth variants of the proposed device, FIG. 11 - embodiments of magnetic keys, in FIG. 12 is a fifth embodiment of the device of FIG. 13 is a circuit diagram of an electronic key, in FIG. 14, 15 and 16 - the sixth, seventh and eighth variants of the proposed device, in FIG. 17 is a circuit diagram of an attachment for reducing a welding current (ballast inductor).

The best embodiment of the invention

The device (Fig. 1) contains a source 1 of excitation voltage pulses, the first and second terminals 2 and 3 of the arc gap, an electric arc power supply 4, parallel to the buses 5, 6 of which a blocking capacitor 7 and a key 8 with a threshold element are connected. In addition, the device comprises an internal boost voltage source 9, a blocking inductor 10, and a long line element 11.

The findings of the internal boost voltage source 9 are connected according to the buses 5 and 6 of the electric arc power supply 4, the first output of the excitation voltage pulse source 1 is connected through a blocking inductor 10 counter to the first bus 5 of the electric arc power source 4, the second terminal is connected to the second source bus 6 4 of the electric arc supply directly, and the first and second conclusions of the source 1 of the excitation voltage pulses are connected through element 11 a long line to the electrode holder - the first terminal 2 and the second terminal e 3 (ground) respectively.

Element 11 a long line is characterized by a wave impedance of 150-300 Ohms and a delay of 4 ns per 1 m of length. Element 11 is made in the form of welding cables.

The key 8 with a threshold element for an electric arc excitation device can be made electronic (Fig. 13) or magnetic (Fig. 5) and contain a key element 12 and elements 13, 14, 15 of inductance, resistance, and capacitance, respectively.

The source of internal boost voltage may contain a key element 16 (or switched by other keys), elements 17 and 18 of the resistance and inductance, respectively, and a capacitive voltage source 19.

The source 1 of the excitation voltage pulses may contain a key element 20 (or switched by other keys), an inductance element 21 and a capacitive voltage source 22 of reverse polarity.

The body 23 of the arc (Fig. 3) with a gap of several microns touches in a circle around the cold electrode (terminal 2), where there is heat sink and a voltage drop of 6 - 10 V, in the middle part the body 23 of the arc is expanded and the voltage drop is minimal. The entire voltage drop occurs in a narrow region of oxide 24, which has evaporated from the spark discharge along the discharge channel.

Schematic diagram of the first embodiment of the proposed device (Fig. 5) contains an electric arc power source 4, including AC bus 25, 26, a switch 27, a transformer with a saturable magnetic circuit 28, a primary winding 29, a secondary winding 30, and an excitation winding 31 with an active resistance 32.

The key 8 is made magnetic on a saturable magnetic circuit with a working winding 33, a resistor 34, and with a bias winding 35. In addition, the key 8 contains a capacitor 15, a transformer 36 with windings 37 and 38, working and bias, respectively.

The function of the threshold element of the key 8 performs the volt-second area, responsive to the voltage integral.

The auxiliary winding 39 of the transformer 36 with a capacitive source 19 and a resistance element 17 form a source 9 of internal boost voltage.

The source 1 of the excitation voltage pulses may contain a high voltage winding 40 of the transformer 36, a capacitive voltage source 22 of reverse polarity with respect to the voltage on the buses 5, 6, a key element 20 in the form of a spark gap and an inductance element 21, the role of which is the inductance of the wiring wire.

An embodiment of a blocking inductor 10 in the form of a secondary winding of a transformer having a primary winding 42 is possible.

A device for excitation of the auxiliary arc of the plasma torch (Fig. 6) contains a power source 4 of the electric arc (auxiliary arc) of the plasma torch, including busbars 25, 26 of an alternating voltage network, a transformer with a saturable magnetic circuit 28, a primary winding 29 and a secondary winding 30 and a power winding 43 of the power supply the main arc of the plasma torch, as well as the throttle 44.

In addition, the device contains a source 1 of excitation voltage pulses, a key 8 with a threshold element, an internal boost voltage source 9, a blocking capacitor 7, a blocking inductor 10, a long line element 11, the first, second terminals 2 and 3 of the electric arc (auxiliary arc of the plasma torch) and ground terminal 45 of the main arc.

Another embodiment of the device shown in FIG. 8 further comprises a shortening pulse width of a magnetic key including a saturable magnetic circuit 46 with windings 47 and 48, respectively, of the bias of both the working and the capacitor 49.

In the embodiment of the device of FIG. 9, the key element 12 must be made non-magnetic, mainly transistor, able to hold the voltage of both polarities and interrupt the direct current. In this case, it is possible to use an inductive drive 50 connected through a serial circuit from a resistor 51 and a capacitor 49 to the output of the winding 37.

In an embodiment of the device of FIG. 10, used a magnetic key with a magnetic circuit 52, a working winding 53, a bias winding 54 and a capacitor 55, a magnetic key with a magnetic circuit 56, a working winding 57, a boosting coil 58, a bias winding 59 and a capacitor 60, a magnetic key with a magnetic circuit 61, a working winding 62 a raising winding 63, a bias winding 64 and a capacitor 65, and a magnetic key with a magnetic circuit 66, a working winding 67, a raising winding 68, a bias winding 69 and a capacitor 22.

In addition, a charging magnetic key with a working winding 70 and a bias winding 71 was used, as well as a bias circuit with a resistor 34, a filter choke 72 and bias windings 73 - 77 with their belonging to the corresponding aforementioned magnetic keys.

At the bottom of FIG. 10 shows a variant of the bias circuit without including it in the circuit, where the element 78 does not represent the resistor 34, but the resistance of the closed conductor passing through the filter inductor 72, the aforementioned bias windings 73 - 77 and the coil 79 on the magnetic circuit 46.

In FIG. 11 shows various options for connecting magnetic keys and diagrams of their operation, explaining their operation.

In FIG. 12 shows an embodiment of the inclusion of the proposed device.

The electronic switch (Fig. 13) contains a rectifier bridge on diodes 80 - 83, the input diagonal of which is connected to the power supply buses 5, 6, to which the varistor 84 is connected. A serial circuit from the transistor transition KE is included in the output diagonal of the aforementioned bridge, resistor 86, emitter winding 87 of base transformer 88, and emitter winding 89 of accelerating transformer 90.

The base winding 91 of the base transformer 88, the resistor 92 and the series circuit of the diodes 93, 94 and the base winding 95 of the accelerating transformer 90 are connected to the emitter-base circuit of the transistor 85. The reverse diode 96 is connected to the base of the transistor 85, as well as a threshold element on the thyristor 97. capacitor 98 and resistor circuit 99 - 102. In addition, a protective capacitor 103 is used in the circuit.

Presented in FIG. 14 variant of the proposed device requires the use of an electronic key.

Presented in FIG. 15 option of the proposed device also requires the use of an electronic key with an inductive storage 104, separated from the transformer with a magnetic core 36 by a series circuit on the resistor 105 and the capacitor 106.

Presented in FIG. 16 variant of the proposed device requires the use of an electronic key, inductive storage 104, shortening magnetic keys, which avoids the use of a spark gap.

In FIG. 17 shows a variant of the proposed device using a prefix for welding with low current.

The prefix contains a inductor 107 with a winding 108, in parallel with which a protective capacitor 109 and a series circuit of a resistor 110 and a capacitor 111, and terminals 112-115 are connected.

For a correct understanding of the essence of the proposed method, we give our understanding of the process of arc excitation. The passage of the spark creates a channel width of a few microns, stable in the longitudinal direction. Entering the current ionizes evenly and extends its channel, the conductivity is 1500 Sim / m, and energy consumption of 3 mJ / mm ^3. Then the behavior of the channel is determined by the heat conduit to the cold electrode and the part (since 1500 K are cold to 8000 K). The channel swells in the middle, the voltage drop and heat dissipation decrease until the heat sink is exactly equal to the heat dissipation. In the literature, this is interpreted as contraction (compression) of the arc at the electrodes. Calculations show that an electrode forms a spot with a diameter proportional to the current - (the flat part of the plasma in Fig. 3 above). The voltage drop on the spot is 8 - 10 V in the current range, and this is a drop on the plasma, not on the contact. The approximate spot diameter is 0.1 - 0.2 mm for a current of 1 A. Note that the entire plasma volume in FIG. 3 less than 1 mm ³ , that is, there is only 5 mJ of energy. A cube of 1x1x1 mm at δ ₀ = 1500 Sim / m has a conductivity of 1.5 Sim, or a drop of 5 V occurs at a current of 7.5 A, in the remaining 2 x 8 V near the contacts.

Real falls are much larger on contaminated and oxidized parts. Presumably, this is due to the huge heat capacity of oxide evaporation, equal to 10 J / mm ³ , that is, 3000 times the heat capacity of the plasma. Then shown in FIG. The 3 hole in the oxide, pierced by a spark discharge, 0.1 x 0.1 x 0.1 mm, has a conductivity of 0.15 Sim, that is, 50 V at a current of 7.5 A and 200 V at a current of 30 A, i.e. it determines the voltage drop across the arc.

Arc excitation by the proposed method (see Fig. 1) begins with closing the key 8 and holding it for a time when the current in the inductance of the power supply 4 rises to the rated arc holding current. Thus, shorting the output of the source 4 of the power supply of the electric arc for the duration of the rise of the short circuit current to the level of a stable arc current.

Then, after opening the output short circuit (terminals 5, 6) of the electric arc power supply 4, a boost voltage of the same polarity, but exceeding its value, is supplied to it.

After that, an impulse of reverse polarity excitation voltage with respect to the electric arc power supply 4, which exceeds the aforementioned boost voltage and has a shortened front duration, is supplied with a time delay to the arc gap (to the holder — electrode terminal 2 and ground terminal 3).

The delay time of the above-mentioned pulse of the excitation voltage should be in the range from 3 to 300 ns, which is determined experimentally.

Moreover, the ratio of the delay time of the above excitation pulse to the duration of its front should be in the range of 1-30.

While the boost voltage is formed by transforming the energy of the power source 4 from the circuit shorting it. The excitation voltage pulse is also formed by transforming the energy of the power source 4 from the circuit of its short circuit.

The device operates as follows.

The power source 4 (Fig. 1) at the time of arc excitation is an EMF source with equivalent output inductance (for example, the secondary winding of a welding transformer), and its output terminals 5 and 6 must be bridged by a blocking capacitor 7, which protects the transformer isolation from high-voltage trigger pulse.

Upon reaching the instantaneous voltage value of the power supply 4 of the threshold level, the key 8 closes, in the inductance of the power supply 4, an increase in current begins. By the time when the current reaches a level sufficient to maintain the arc, the key 16 of the source 9 closes with the voltage source 19 of the same polarity as the source 4, and the exciting pulse has the voltage of the source 22 of the opposite polarity to the voltage of the source 4, and the pulse choke 10 is fundamentally not saturated.

An excitation pulse of reverse polarity (let it be negative for definiteness) is sent to the output of the device and, through element 11, a long line extends to the electrode holder - terminal 2. A negative voltage, for example, 7.5 kV, is applied to the right end of the inductance of the blocking inductor 10, let its inductance equal to 5 μH. Then the current rises in it from left to right, 1.5 A each ns. Having reached the end of element 11 a long line, the pulse is reflected, doubled in voltage. Let the arc gap break through with a spark discharge. The spark current is negative. The sparking closure is reflected and a long line returns to the beginning of element 11, and after several damped circulations in the circuit of the inductor 10, element 11 and the arc gap, a current is established that coincides with the future arc current. It can be seen that the current is 20-30 A. An arc body is formed, as shown in FIG. 3. The process can then be illustrated by the equivalent circuit in FIG. 2. It shows voltages and current directions. This scheme is described by a system of differential equations:

ΐ ^d ΐ ^- Catfish ⁺ (| ohm

1pom (Npom ^- Su;) ^/ P _And ohm and d = IE / 2 + 1E / U

EU / A1 = a · 1d ² / U

E1 / d! = (e - and _uz ) / b ^d ^ com / E1 = (uz - ' ^and com ^{) /} ^ com ^d 1E ^{/ d} 1 = ^(and pom ^- ΐ-pom ^ ^ pom ^{- id} ) ^/ bdi ^di com ^{/ d} 1k ^{/ C} com ^di IOM ^{/ d} ^ 1pom ^{/ C} pom where id - steady arc voltage at medium currents, for example, id = 20;

and - coefficient, with an oxide thickness of 0.1 mm, its energy intensity of 10 J / mm ³ and plasma conductivity of 1500 Sim / m, a = 15;

Υ is the plasma conductivity in the channel in the oxide, Sim;

The designations of the elements, currents and voltages are shown in FIG. 2. For example, we take: b = 600 · 10 ^-6 H, bcom = 70 · 10 ^-6 H, bdi = 5-10 ^-6 H, Com = 140 · 10 ^-6 F, Spom = 2.0 · 10 ^{- 6} Ф, Кпом = 10 Ohm, е = 50 V.

The initial conditions are 1E (0) = 30 A, Υ (0) = 0.10 Sim, 1 (0) = 7 A, 1 _K ohm (0) = 7 A, and _K ohm (0) = 60 V, Lp ( 0) = 120 V.

In FIG. 4. The solution of the equations for the indicated parameters is given. You can see that at the beginning the arc voltage is 300 V and the current is supported by the inductance current b _di . It coincides with the future arc current precisely because the polarity of the excitation pulse is inverse to the voltage of the power source. If it were the other way around, then the arc current would have to go through zero, which would slow down the establishment of the necessary conductivity and could cause the arc to break, that is, the start could not take place. It is important that b _di is not saturated during excitation (during arc operation, the current is hundreds of amperes and saturation is inevitable), its magnetic circuit must be made with an air gap. Calculations of the variants according to the given system of equations allow us to find the optimal value of b _di . As indicated above, the accumulation of current depends on the time of maintaining a high voltage, until the pulse passes through the line and returns reflected.

Then the arc current from the region of tens of amperes passes into units and is supported by the current of the assisting circuit. During this time, a current ί is intercepted, at first completely passing through the inductance of the switching key b _di . The solution of the system of equations for various parameters of the elements of the equivalent circuit and the initial conditions shows the direction of efforts to create a device according to the proposed method.

In FIG. 5. shows a circuit diagram of the first variant of the proposed device, designed to work with a transformer with a saturable magnetic circuit 28. The inductor in the primary circuit and the transformer with the primary winding (it is the winding of the inductor) are combined. The magnetic circuit of the inductor and transformer, that is, combined, is not saturated. The magnetic circuit on which the secondary winding of the transformer is located has a smaller cross section and is saturated at idle. The magnetic circuit 28 has windings 29, 30, 31, respectively, primary, secondary and excitations wound from the side of the secondary winding 30. As was shown in the example of excitation calculation, the capacitance of the switching capacitor was 140 microfarads, which is not constructive. When using the field winding, the capacitance is transformed, in this case it decreases to 2-5 microfarads. But an additional inductance appears in the circuit of the switching key due to incomplete coupling of the windings 31 and 30. In order to reduce this inductance, the field winding 31 is switched on in series with the secondary winding 30, that is, an autotransformer inclusion takes place.

As a key element 12 (in Fig. 1), this device uses a magnetic key having bias windings 35 and 33 and a working one, respectively. The primary winding 37 of a high voltage transformer 36 having a bias winding 38 and a secondary winding 40 is sequentially connected to this circuit. A capacitive voltage source 22 is included in parallel to this winding, and a spark gap is used as a key excitation element 20. The assisting resistor of the resistance element 17 and the assisting capacitor of the capacitive source 19 are included on the auxiliary winding 39 of the high voltage transformer 36.

For the operation of the pathogen with magnetic keys, it is important to synchronize them for accurate inclusion in accordance with the method. This synchronization is achieved by bias windings 35, 38, connected in series in the primary winding circuit 29. Due to saturation of the transformer at idle, significant currents pass through its primary winding 29 (up to half the amplitude of the operating currents), which set the magnetic switch and transformer 36 to state induction of readiness for work.

In FIG. 7 shows experimental oscillograms of the pathogen idling. Synchronization is performed at the beginning of the sinusoidal mains voltage. To orient on the waveform, a virtual voltage is applied. The current of the magnetic switch is measured by a current transformer connected in series with a resistor 34 (Fig. 5). During the saturation of the power transformer, the bias current calms the oscillations in the magnetic key circuit. Voltage and ₀ (1) are close to zero when the transformer begins to exit saturation. The voltage and (!) On the primary winding of the transformer is close to zero by this time. The magnetic key current is also zero. The voltage and ₀ (1) is applied to the working winding 33 of the magnetic key, and the induction moves in its magnetic circuit from -V to + V ,. When saturated, the magnetic key is short-circuited according to the method, through the field winding this is transmitted to the secondary winding. The current ί (!) Charges the switching capacitor 15, which causes the voltage ₀ (1) to increase, and through the winding 40 it charges the capacitor of the capacitive source 22. Upon reaching the voltage of the triggering level of the spark gap, a high-voltage pulse is supplied to the device output, with a negative polarity, according to the method. At the same time, the discharge causes a voltage jump on the winding 37, producing a positive helping voltage, according to the method. Arc excitation described previously. In FIG. 5 shows an application of a pulse transformer with a primary winding 42 instead of a choke.

Arc excitation occurs from idle. If the arc was burning, then the current passes through zero, and it is necessary to resume the arc again, then there is a stabilization mode, which, generally speaking, differs from the excitation. The difference is that when the arc is stabilized, the voltage and ₀ (ΐ) follows not from the zero level, but from the negative voltage of the arc burning.

The calculated stabilization timing diagrams are shown on the right in FIG. 7. It can be seen that the voltage u (ΐ) lags behind the voltage and ₂ (ΐ) on the discharger (or the voltage on the discharger is ahead of it), the discharge occurs at the moment when the voltage and ₀ (ΐ) did not reach the desired level 40–60 In, and stabilization of the arc does not occur. The device switches back to mode XX, and the excitation will be only in the next half-cycle. When welding, this is expressed in notches - dips on the oscillograms of the current of the welding arc, metal spatter.

The diagrams show the stabilization process in the presence of a resistor 34 in the switching circuit (resistance is given to the secondary circuit). In this case, the rate of increase in voltage across the arrester slows down and the discharge occurs at the right time. There is an optimal value for this resistance. In addition, with zero or too low resistance, transients do not have time to end by the time the transformer leaves saturation. In FIG. 7 on the right shows the calculated voltage diagram and ₂ (ΐ) with zero resistance, and when r = 3 and current 1 Ohm _com (ΐ) key (shown to the secondary circuit) when r = 0 and r = 3 ohms.

This variant of the exciter is quite reliable in operation, although the transformer is obtained with the number of turns in the secondary winding 40 of the order of 12000 - 4000. In order to reduce the dimensions of the transformer and slow down the operation of the arrester, a shortening magnetic pulse with a bias winding 47 and a working winding 48 and a shortening capacitor 49 is introduced shown in FIG. 8. The directions of the windings are shown by common signs. It must be borne in mind that the shortening key in this switch changes the polarity of the pulse, therefore, the assist winding 37 and the primary winding 40 have a reverse switch polarity. The capacity of the shortening capacitor 49 can be selected so that the turns of the primary winding 40 and the winding 37 are equal, and since in this case the voltages are in the same phase, this can be one winding, which is done in FIG. 8.

In this variant of the pathogen, despite the introduction of another magnetic key, more acceptable design parameters are obtained. The turns of the transformer are reduced to 3000 - 1500, the volume of steel - four times. The delayed response of the shortening magnetic key facilitates stabilization, but the resistor 34 is useful. Note that in these two variants of the pathogen, after the current of the switching key is intercepted and becomes zero, the magnetic key automatically opens, which favors the start. After a while, its reverse operation follows, but its current is directed into the arc, that is, this does not interfere with the launch. We also note that the necessary bias currents are obtained in the bias windings when using a saturable transformer. Magnetic keys do not respond to voltage amplitudes, like conventional threshold devices, but to the voltage integral at the front of the transformer core coming out of saturation.

The energy for the excitation pulse in these pathogens is obtained from part of the energy of the switching capacitor. The magnetic key, in principle, cannot switch the constant component of the voltage, in these circuits the magnetic key circuits are separated by capacitors.

The embodiment of the device of FIG. 90 uses an inductive energy storage device. In this case, the key element 12 must have the ability to hold a constant component of the voltage, that is, it must be an electronic key. Such a pathogen is able to work both with a saturable transformer and with a conventional welding transformer. Therefore, all necessary offsets must be developed within the circuit. The key element 12 with a threshold device is triggered when the voltage of the source 4 of a predetermined level is reached, its inductance is closed through the inductive storage 50.

After reaching the current of the key element 12 of a given level, it opens (it will be described in detail later), the inductance current of the drive 50 through the isolation capacitor 49 and the current-limiting resistor 51 is supplied to the primary winding 37 of the transformer. The presence of the capacitance of the source 22 limits the increase in voltage applied to the key element (for definiteness, it is selected 400-600 V, according to the quality of the transistors). The arrester is triggered when a predetermined voltage level is reached, the excitation pulse is supplied to the blocking inductor 10 with negative polarity, according to the method.

Inductance current cannot stop abruptly. But he can jump abruptly from one circuit to another. In this case, 1 _com stops in tenths of a microsecond, and the inductance current of the drive 50 jumps into the primary winding 37 of the transformer. But there remains a current 6 of the inductance of the source 4. It jumps from the key element 12 to the circuit of the source 9. It generates a boost voltage. You can see that in this circuit there is no inductance of the switching key in the circuit of the increasing arc current. This makes starting easier.

To excite sparks and arcs, pulses with nanosecond fronts are used. So that high voltage could not penetrate the welding transformer, it is blocked by the inductance of the blocking inductor 10 and the capacity of the blocking capacitor 7. These pulses are generated by a spark gap. Experimentally observed cut-through (skipping) pulses in the apparatus used several arresters connected in parallel.

In principle, the necessary shortening of the pulses can also be obtained using magnetic keys (Km), but the Km theory is still insufficient for their practical use in pathogens. Therefore, we present the results obtained. In FIG. 11a, two types of Km magnetic keys are shown — P-type or Km type A, and T-type Km or Km type B in common terminology. So, as they are shown in the figure, they can be switched on in cascade (according to the custom accepted in electronics, we assume that the energy propagates from left to right).

In FIG. 11g, h real hysteresis cycles of magnetic materials used in km are given. Recall that at zero voltage H (zero sum of currents in all turns), induction B (0) can be at any point inside the hysteresis loop. This is fundamentally different from the state of an ordinary electric circuit, where before the start of the process, if the pause lasted long enough, the voltage across the capacitors is zero, the currents in the inductors are zero, and the power dissipated by the resistors is zero. This follows from the law of conservation of energy, since non-zero power means a continuous flow of energy. But non-zero induction in the magnetic circuit can be at zero energy, and therefore possible.

In addition, phenomena similar to those studied in the new branch of mathematics, known as the theory of catastrophes, are observed in Km chains. Moreover, unequal consequences, that is, causeless phenomena, follow from equal initial conditions (causes). An example is an object called an attractor, or fluid motion in a turbulent mode. That is why the weather, in principle, cannot be predicted for a long period, and the problem is not the imperfection of computers. In Km chains, this is expressed in nonperiodic transients in each half-period of the supply voltage of the network, in an abrupt change in the modes of oscillations, in changes in the time of appearance of the excitation pulses and even its sign. This makes the work of the pathogen unstable, and therefore, close attention should be paid to displacement issues.

The operation of Km can be understood by the example of a Km chain of type A in FIG. 11b. First, let all the magnetic cores be shifted to position -B, by an external source. A voltage step is applied to the first capacitor on the left. The first left Km moves from -V, to + B ,. When its volt-second area is exhausted, the magnetic permeability of the magnetic circuit drops (in ferrite from 1.2 to 2.0 of the vacuum permeability, this is more than unity, since the cross-sectional area of the winding coil is larger than the magnetic circuit, in permalloy and amorphous steel - similarly, in electrical steel permeability from 20 to 3 remains, depending on the depth of saturation and the quality of steel annealing). The inductance of the saturated first CM turns out to be connected in series with the first and second capacitors, and at this moment the voltage u on the first capacitor is maximum, and ₂ on the second capacitor is zero. A current passes in the circuit in the form of half a sinusoid, the voltage u drops to zero, the voltage ₂ increases to maximum, the current strives to become negative, but this means that the reverse voltage is applied to the first Km, which it can hold to the best of its volt-second area. In FIG. 11c it can be seen that the voltage u appears on the second Km, then the same process occurs, each Km shortens the pulse duration, leaving the voltage amplitude, and, therefore, increasing the current.

In FIG. Figure 11g shows a Km chain of type B. Upon receipt of voltage and ₁ through the capacitor of the previous Km, current is transformed into capacitor C1, and the current through Km 2 saturates it and maintains the charging current C1 in a saturated state all the time. At this time, Km 1 works as a transformer, induction moves from -V to + V. When saturation is reached, Km 1 turns into a small inductance, charged C1 is recharged by reverse current to a reverse voltage, that is, there is a double voltage amplitude drop. In this case, Km 2 goes out of saturation, the capacitors C1 and C ₂ turn on in series, on C ₂ it turns out half of the double voltage drop, that is, a single drop of the opposite sign. (For a better view of the operation of Type B km, it is useful to imagine the windings turned on according to, as shown in Fig. 11d, and the turns equal, then these windings can be connected and the transformer turns into a choke, all capacitors are equal and connected in a series circuit). You can see that Km type B allows you to change the polarity and make the transformation, that is, go from relatively low voltages to high, which will be used in the future.

Note that the voltage and ₁ per Km 1 represents a series of cosine waves with decreasing half-periods from the operation of each subsequent Km with a change in the sign of the voltage (it can be shown that the difference between the transformation coefficients in the Km circuit from unity does not change the voltage and). Note that the voltage at the output of Km type A does not change the sign and there is no dribbling there.

This is important for arc generation, as will be shown later. It is believed that at the end of the Km chain there is a load in the form of a resistor or spark gap, where the pulse energy is absorbed. In this case, in the Km type A chain, all capacitors are discharged, and all the magnetic cores are in the state + B ₈ . In a Km type B chain, the last dribbling pulse has zero voltage. This is followed by a new half-cycle of the mains voltage with the bias setting.

However, it may be that at the end of the Km chain there is no load (there was no spark) or there is a short circuit or XX. In this case, there is a reflection of the impulse to the left according to the scheme, and there are different options. For example, in a Km type A chain (Fig. 11b), the last Km 4 arranges dribbling between the last capacitor C ₄ , Km 3 has zero voltage on the left, alternating zero on the right and full positive amplitude, until this positive voltage across the area exceeds the volt-second area Km 3, which is triggered (and all comments regarding the attractor have an effect), the impulse propagates to the left.

The process in the Km chain of type A is simpler if the output is short circuit. In this case, the positive voltage at the last capacitor is inverted, that is, it becomes negative, and the whole chain Km of type A, in the state + B ,. transfers this impulse to the left, and the times correspond not to the exposure time Km, but to the time of their switching. This is expressed in the fact that when oscillating the current, for example, Km 1, after the current pulse, a second current pulse of a slightly lower amplitude immediately appears (due to losses).

If a process takes place in a type B chain, and a short circuit occurs at the output, then a reverse polarity pulse propagates from right to left, i.e., a double current of Km takes place. If the output is XX, then the problem first arises as to how the capacitor charges at XX. If measures are taken for this, then, when the last Km is triggered, the process seems to end and the impulse does not propagate back at all.

The current biases for the correct operation of Km, taking into account that the characteristics of magnetic materials can be such as shown in FIG. 11, i.e., INGs (with a rectangular hysteresis loop) or closer to linear ones (ferrites) are shown in FIG. 11. It is convenient to begin the examination of the work of Km from the last stage in time, then moving forward in time to earlier stages. Moreover, the duration of each previous stage increases in proportion to the shortening of each Km.

Stage 1 - switching. Saturated Km is the inductance connected between two equal capacitances, one of which is charged, the other at zero voltage, that is, the capacitance of the circuit is equal to half the capacitance of the capacitor (in Km type B, the indicated capacitance is meant).

Stage 2 - integration or exposure. For Km type B it can be called a charge or transformation. If Km type B has equal windings according to the included conditions, or if it is Km type A, then they behave like a choke whose current is the magnetization current in accordance with its magnetization curve. If Km is of type B with separated windings, then transformed currents flow in them in opposite directions. At this stage, the induction goes from -V to + B, then - stage 1.

Stage 3 takes place only for Km type B. It can be called a preset or an offset. At this stage, the capacitor of the previous Km is charged through the current Km switching current of the previous Km. This is favorable for this Km, since this current shifts it to the state of readiness for stage 2, that is, -B. Km type A does not have this stage, therefore, measures must be taken for the desired displacement.

Stage 4 may be called the noise immunity stage. Let there be a chain Km of type B with unit transformation coefficients. Then the capacitors are connected in series and they have dribbling with decreasing periods. If the current of this km at stage 3 is favorable for it, then the current at the previous stage gives a reverse bias (although the amplitudes of the previous currents decrease in proportion to the increase in their periods). Care must be taken to prevent interference.

Although at stage 3 this Km will be biased in the right direction, we are not dealing with a passive KS chain, but with nonlinear inductances connected by capacitors. The displacement of the magnetic circuit in this direction is proportional to the voltsecond area of the voltage applied to the winding. To obtain a finite area at a finite displacement time, a voltage other than zero is required. But this voltage is on the capacitors. This is energy, moreover, lost. This in itself is tolerable, since the energy depends on the square of the voltage and can be small, but these small voltages for a long time move other Km in different unforeseen directions, the system of coupled Km turns into a typical attractor with unpredictable behavior. Therefore, bias currents should be organized with the lowest possible slew rate. We give a numerical example for a typical Km chain. Suppose that the last three Km are ferrite, they are shortened five times, the previous ones are steel, with a shortening of 10 and 20. The current of the last Km is 200 A at 7000 V for 23 ns, the voltage is 35 000 A / m. (It is useful to compare the Km parameters with the known parameters of transistor or thyristor switches). In the previous stage, the tension was 7000 A / m, in stage 3 - 1400 A / m, in stage 4 - 140 A / m. Turning to the characteristics of magnetic materials in FIG. 11, one can see the scale of the troubles in step 4 and the magnitude of the displacements necessary to avoid them.

In the embodiment of the excitation device with magnetic keys, the following bias order is organized: a bias current is started at a given Km from the pre-Km, which prepares it for stage 2 (integration), and an interference current with an appropriate polarity is started from a pre-Km, preparing it for stage 3 (bias ) and protecting it in step 4 (noise immunity). This is shown in FIG. 11b for a Km chain of type A, in FIG. 11g for the Km chain of type B and in FIG. 11e for the Km chain of types B, B, A, B. In the event that the preceding Km is of type B, it itself gives the bias current, so there is no need to start the bias specifically, although it does not hurt. In the numerical calculation of bias currents, the right decision can be made.

The last Km in the chain, which has an output to the output terminals of the welding machine and to the welding cables, has a special mode of operation. If it is Km type B, then the charge current at its stage 2 and the bias current at its stage 3 must pass through it and the load. But there are welding cables with a small capacitance and a pulse inductor with a relatively large inductance at the output. It is necessary to provide paths for the passage of these currents. For this, it is necessary to use an additional Km _zar (charging key) with an appropriate offset. This will be described in more detail on the example of a specific implementation of the pathogen.

According to the proposed method, a helping voltage should be supplied to the chain K. _pom C _pom using the help key. For this purpose, in the Km exciter, it is proposed to use a voltage on one of the Km type B chains, whose integration time (stage 2) is longer and the switching time (stage 1) is less than the time constant of the assisting chain. The polarity of the helping pulse must be appropriate for the method, and its amplitude can be calculated by known methods. On the selected Km type B, an additional winding of the helping voltage can be made. As indicated above, on km Type B after the end of stage 1 (switching), alternating dribbling occurs (Fig. 11e), ending with zero voltage in the event that the energy of the Km chain completely goes into the load. Then, the assist voltage remains on the help capacitor, to which it was charged in stage 2 of its Km. It will contribute to arousal.

In FIG. 11e shows an embodiment of obtaining a helping voltage with Km type A with times of steps 1 and 2, as indicated above. A definite advantage in this case is the absence of dribbling on Km type A.

As can be seen from the description of the operation of Km type A or Km type B with equal windings turned on according to (inductor), a significant current flows only at stage 1. This allows you to charge C and store the charge on it until all processes in Km are complete until the start of a new cycle . In FIG. 11i shows the connection of the help capacitor with one grounded end (which is necessary for the help circuit to work correctly), included in the serial circuit with Km type A and its load - Km type B. At Km type A at stage 2 there is no charge current, therefore, it will pass only the charge current of the previous Km type B, which is much smaller, since it is more stretched in time. It passes through the tanks and B and goes into the circuit of the welding current source.

In FIG. 11k shows an embodiment of including a helping circuit in a Km circuit of type B throttle type. Often this option is convenient because of the desired polarity of the pulses per km. The response time of Km4 in this particular case is about 0.8 μs, that is, the assist current is held for a sufficient time after start-up, and zeroing occurs only after hundreds of microseconds in the next displacement cycle.

As described above, the currents in these stages of operation Km are alternating and, therefore, pass through a zero value. But at a zero value of the currents, the shape of the hysteresis loop becomes significant, since ferrite slides to - Br.

In order to prevent the appearance of an attractor, it is necessary to have an offset of at least a certain value (Fig. 11 g, h).

In FIG. 10 shows a diagram of the pathogen according to the proposed method, made entirely on Km, without a spark gap. The causative agent is designed to operate with a saturable power transformer having a significant XX current and a rather steep front of saturation exit. The first three Km are made of steel and have a synchronizing bias current, as well as the pathogen in FIG. 8, but instead of a high-voltage excitation pulse transformer, Km 3 of type A is included.

Its parameters are selected such that it is saturated at a level corresponding to the level of operation of the arrester in FIG. 7 (see waveforms).

The current ΐ ₁ in FIG. 10 is the current ί (ΐ) in FIG. 7. You can see that the current reaches a maximum, then decreases, and the excitation occurs at a current of 2/3 - 3/4 of the maximum (this is selected during debugging by changing the number of turns of Km 2). The time of this current is 700 μs for this example, the response time of the Km chain to the moment of arc excitation is 50 - 30 μs, that is, during this time this current can be considered unchanged. In series with this current circuit, a chain of two parallel branches is connected, in one of which is the unsaturated inductance of the 72 b choke with successive bias windings at all Km, starting from Km 3, and also Km _charge , and in the other resistor 78 of the Kf filter, so the time constant Lp / Kp is commensurate with the duration of the current 1 ₁ . Then the current in the inductance of the inductor 72 at the time of excitation will be close to the current ΐ ₁ , and then it will remain for the duration of the remaining Km and arc excitation. The KS circuit of the resistor 17 and the capacitor 19 is connected to the input of Km 3, as described previously. The polarity of the pulses at the inputs of Km are also given.

When Km is triggered, impulses appear on the displacement turns, each time increasing the displacement current of subsequent Km. It can be seen that at Km 6 the voltage at the bias turns is half of the total voltage Km, that is, 4 kV or more. This is inconvenient in terms of insulation of the winding.

Below in FIG. 10 shows a variant of bias from voltage at Km 2. This voltage u (щ) (Fig. 7) becomes zero and even changes sign at the moment Km 2 is triggered, that is, it is as if unsuitable for creating a bias current. But due to the inductance of the inductor 72, the time constant is sufficient to preserve the accumulated current. In this embodiment, one turn of displacement permeating Km 2 and beyond is sufficient; the problem of high voltages is removed. Resistance element 78 is the resistance of the wire itself, selected accordingly.

In FIG. 12 presents a fifth embodiment of the proposed device, designed to operate a transformer with saturation. The common bias circuit from Km 2 to Km 3, 4, 5, 6, and Km is made in the form of a single turn, as in the previous pathogen. Offsets from currents ΐ ₁ , ί ₂ , ί ₃ , ί ₄ compensate for the influence of these currents at the noise immunity stages of each Km. The auxiliary voltage is generated on the capacitor 19, connected in series in the Km 4 circuit of type B. The rise time of the auxiliary voltage is about one microsecond, during which time the auxiliary circuit will not discharge. Also, the outflow of the assist current to the inductance of the switching circuit does not occur (see Fig. 2). At the same time, the charge current of the capacity of the third key, Km 3, when Km 2 is activated, passing through Km 4 and the resistor 17, turns out to be small, that is, the time constant of the assist circuit should be between the response time of Km 4 and the accumulation time of Km 3, i.e. a wider interval than in previous schemes, which facilitates the implementation of the help chain.

In this example, the pathogen is convenient to consider the work of the charging magnetic key Km _zar . It is made on ferrite of much smaller dimensions than the blocking inductor 10. This is due to the fact that the influence of magnetic viscosity is already significantly affected at nanosecond times, and the smaller the volume of the magnetic circuit, the less loss. It can be seen that for Km, the current is ί ₄ and the bias current is ί ₅ . Although Km _{zar is} biased by the total current of the coil in the right direction, numerical calculation shows that the interference current requires at least 2-3 turns from current от ₄ to compensate for it. It is held in state -B at the stage of noise sensitivity, the total bias current and current ί ₄ , then continues to be held in this state by the more significant current of the bias stage from current ί ₅ , the capacitor 22 is charged, and then during the switching step for Km 6 a negative pulse fed to the inductor 10 and the output of the line.

As follows from the description of the operation of the magnetic keys Km, the threshold for their inclusion is determined not by the voltage level, but by its integral, and by its nature Km cannot hold a constant voltage. Therefore, in the Km circuit there must necessarily be a series capacitance that blocks a constant voltage. All exciters with Km proposed above are designed to work with saturable transformers. In addition, since transformers for different welding currents have different inductances, different process times are obtained, and therefore for each type of trans23 transformers for different welding current there must be an exciter with different parameters Km.

To work with transformers of various types, an electronic switch is more convenient, it can open and block a constant voltage.

In FIG. 13 is a diagram of a key element 12 satisfying the requirements noted in the description of the device of FIG. nine.

The operation of the key element 12 is as follows. When voltage of any polarity arrives at busbars 5-6 in the output diagonal of the bridge, the polarity is positive. First, transistor 85 and thyristor 97 are closed. As the voltage increases from the power supply 4, the voltage at the reference resistor 102 increases and reaches the threshold of the thyristor 97 (0.5 - 0.6 V). The current of this chain is of the order of a milliampere, it is not enough to open the transistor 85, locked by a resistor 92. But because of the capacitor 98, the current of the thyristor 97 will be much larger and limited by the resistor 100. This current opens the transistor 85, a regenerative process occurs in the base transformer 88, and then the current from the power circuit will be transformed into the base circuit of transistor 85, keeping it open. If the source has an inductive resistance (welding transformer), then the current in it increases gradually over a period of hundreds of microseconds. The voltage at the base of the transistor 85 relative to the assembly point is fixed by a chain of limiting diodes 93-94. As the current increases, the voltage drop across the current-limiting resistor 86 increases, leading the transistor 85 to the gate. Transistor 85 goes out of saturation, a regenerative process occurs. An accelerating transformer 90 with very small turns and with a minimum leakage inductance serves to accelerate the locking. The current in the inductance cannot stop abruptly, it raises the voltage at the external capacitor outside the circuit of the key element 12 (see Fig. 9), an oscillatory process occurs in the external circuit. The protective capacitor 103 partially limits the overvoltage on the transistor 85, as well as the varistor 84, but its role is not only in this. It does not allow the voltage and current of the thyristor 97 to drop, so that the current of the thyristor 97 is supported, although quite small, since the entire chain of resistors limits it, and the capacitor 98 is discharged.

During the direct current of the key element 12, the inductive current accumulates on the base transformer 88 with the cut-off magnetic circuit 85, and since the emitter circuit of the transistor 85 is closed, this current of negative polarity (that is, closing the p-p-p type transistor) goes through the reverse diode 96. For example, if the emitter current was 5 A, the base current was 2.5 A, then the inductive current can be selected 1 A, and 1.5 A remains in the base, which is quite enough. With this closing current, the thyristor current 97 cannot open the transistor 85 a second time. Thyristor 97 remains open milliampere current until the end of the half-cycle, until the current drops below the holding current of thyristor 97 with a chain of resistors 99-102 (about 0.5 mA can be made). Then the thyristor 97 closes and remains closed the rest of the half-cycle, and the next opening will be in a new half-cycle with a new increase in the voltage of the source 4 to the threshold. Note that the closing current of the transistor 85 lasts about twice its opening time, since the reverse voltage on the reverse diode 96 is approximately half the direct voltage on the two diodes 93, 94.

Thus, the requirements for the key element 12 are fulfilled: tripping at the voltage level set using the resistor 102, releasing when the maximum current is set, controlled by the resistor 86, blocking the second tripping in half a period. If for some reason the current of the transistor 85 does not increase rapidly enough (with a large inductance of the external circuit), then the base transformer 88 exhausts its volt-second area, saturates and opens the key element 12. This one-sided saturation is conventionally shown in the diagram in the conventional notation. Usually, this transformer is designed for the entire range of the exciter from transformers with different operating currents and, therefore, different inductances. For example, the pathogen can operate in the range of welding currents of 300 - 30 A without restructuring.

Please note that the transistor switch closure time is several hundred microseconds, depending on the inductance of the welding transformer. At this time, the voltage at the collector is close to zero, the voltage at the resistor circuit and at the capacitor 98 is zero, the thyristor 97 inevitably closes. This is followed by a voltage surge of up to several hundred volts, the thyristor 97 opens, but the transistor 85 is closed at this time by the ampere current and the opening of the thyristor 97 does not open it. When the ampere closure currents end, the high-voltage transients have long ended and thyristor 97 remains open in the mode of milliampere currents, which are not able to cause the transistor 85 to trip again.

In FIG. 14 shows a driver circuit in which the bias is provided by the use of a high-voltage pulse transformer with a split magnetic circuit, which itself is an inductive storage. This halves the difference in induction, that is, it requires a double number of turns, but it allows the use of a pathogen without bias, that is, it can be used for any transformer that is not necessarily saturated. The pathogen works similarly to that described.

In FIG. 15 shows a driver circuit in which bias is provided from the current ΐ _{1 of the} inductive storage 104. The current from the bias winding is transformed into the primary winding and charges the isolation capacitor 106 so that saturation is ensured for any initial induction value. For this, usually 1/10 of the number of turns of the displacement winding from the turns of the winding 37 is sufficient.

In FIG. 16 shows a driver circuit with an electronic key element 12 with magnetic keys without using a spark gap. The electronic key element 12 gives such a shortening of the front length that Km 2 is not required, the Km chain begins immediately with Km

3. The offset of Km 3 is provided by current ΐ ₁ , as described above for the bias of the transformer. In addition, current ΐ _{1 is} used for protection at the noise immunity stage of Km 5, as described previously. Km 6 has an offset from current ί ₃ , and Km _charge - from a current ί ₄ , as described previously. There is a common bias circuit from the current 1 _{1 of the} filter choke 72 and the resistance of the filter resistor 34, as described previously for magnetic keys.

Pay attention to Km 4 with separate windings. They are used to change the polarity (sign), but since they can be galvanically disconnected, part of the circuit can be placed relative to the zero bus, and part relative to the top, as it is convenient. An excitation pulse with Km 6 is supplied relative to the upper tire. This slightly changes the ratios at startup, but insignificantly, since the voltage on the blocking capacitor 7 changes by the charge current of the inductance of the source 4. Recall that the charge time is the transit time of the negative excitation pulse through element 11, a long line of welding cables and return, that is, tens of nanoseconds . When a spark and an arc are excited, the decay time of the inductance current from tens of amperes is tenths of a microsecond, and during this time, the capacitor 7 is charged to hundreds of volts. Therefore, it can be assumed that the connection of the excitation pulse circuit with respect to the ground bus or the upper bus is the same and it can be chosen for reasons of structural convenience.

Apparatuses with a pathogen according to the proposed method have such an easy arc excitation and such stabilization that the problem of choosing a welding current appears in a completely different light. The welder does not feel a change in the welding current, for example, one and a half times. Moreover, it is possible to weld metal 0.8 mm at an arc current of 200 A if the welder interrupts this current at high speed. Nevertheless, for welding particularly thin parts it is advisable to have currents of the order of 50 - 20 A. In principle, such adjustment is possible using a movable magnetic shunt. In this case, the inductance changes inversely with the current. This creates problems for magnetic keys (for Km 1 and Km 2). Electronic key exciters operate over a wider range of inductances. However, the additional ballast inductance agent shown in FIG. 17. A ballast inductor 107 with a winding 108 on steel with a conventional air gap is shunted by an excitation pulse transmission capacitor 109 and a supporting chain K. From resistor 110 and capacitor 111, it can be in the range from fractions of nanofarads to units of microfarads, supporting chain, for example, 10.0 uF and 20 ohms. It maintains the arc current, while the inductor 107 picks up a current of several amperes, sufficient for arc burning. Then the current reaches 50 - 20 A, for which ballast is intended. The inductor 107 may be a single current, or have taps, or have a smooth adjustment by the movement of the coils or by changing the magnetic gap. To reduce the size and weight of the throttle, it is proposed to saturate it. For example, if the welding transformer itself is designed for a current of 200 A, and the volt-second area of the inductor is taken 0.65 of the applied voltage, then the current after saturation (effective value) is only 0.2 of the source current, i.e. 40 A for the given example. Against the background of a smooth current of 20 A, a current pulse appears with an effective value of 40 A. Such pulses even contribute to improving the quality of welding, as is known, and a choke with saturation is 35% lighter. Practice shows the possibility of having a stable arc at currents in fractions of an ampere.

As can be seen from the description of pathogens above, the ballast choke does not affect their operation with both electronic and magnetic keys.

Industrial applicability

The stated advantages of the proposed method and devices for its implementation provide them with the possibility of widespread use in apparatuses capable of cooking oxidized and rusty metal, steel, stainless steel and other metals.

Claims (9)

CLAIM 1. The method of excitation of an electric arc, in which before applying an excitation voltage pulse to the arc gap, the output of the electric arc power source is shorted for the duration of the short-circuit current rise to the level of the stable arc current, characterized in that after reaching the short-circuit current level of the stable arc current to the output of the power source of the electric arc is supplied with a boost voltage of the same polarity, exceeding the voltage of the power source, after which it is supplied to the arc gap erzhkoy time pulse of reverse polarity excitation voltage relative to the source of electric arc power in excess of nominal boost voltage and having a shorter rise time. 2. The method according to claim 1, characterized in that the delay time of the pulse of the excitation voltage is selected in the range from 3 to 300 ns. 3. The method according to claim 1, characterized in that the ratio of the delay time of the pulse of the excitation voltage to the duration of its front is 1-30. 4. The method according to claim 1, characterized in that the boost voltage is formed by transforming the energy of the power source from a circuit shorting it. 5. The method according to claim 1, characterized in that the excitation voltage pulse is formed by transforming the energy of the power source from the circuit shorting it. 6. A device for exciting an electric arc, containing a source of pulses of the excitation voltage, the first and second terminals of the arc gap, a power source for the electric arc, in parallel with which a blocking capacitor and a key with a threshold element are connected, characterized in that a source of internal boost voltage is introduced into it, blocking choke and element long line, while the source of the internal boost FIG. 1 voltage is connected in accordance with the power supply of the electric arc, the source of pulses of the excitation voltage is connected to the power source of the electric arc counter-one of its output - through a blocking inductor and the other directly, and through the element a long line the conclusions of the source of pulses of the excitation voltage are connected to the first and second terminals of the arc gap. 7. The device according to claim 6, characterized in that the source of internal boost voltage is made in the form of series-connected key element, resistance and inductance elements and capacitive voltage source. 8. A key with a threshold element, comprising a capacitor, a saturable magnetic circuit with a working winding, one output of which is connected to the input of the magnetic key, and the other output through a capacitor with an output, characterized in that a bias winding is introduced for series connection with the primary winding of the power transformer with saturable magnetic circuit of an electric arc power source. 9. A source of excitation voltage pulses, containing a capacitor, a spark gap, a saturating magnetic core of a transformer with working and high voltage windings, the last of which is connected to a capacitor and a spark gap, characterized in that a bias winding is introduced for series connection with the primary winding of the power transformer with a saturable magnetic core of the power source electric arc.