JPH038577A - Consumable electrode arc welding equipment - Google Patents

Abstract

PURPOSE:To perform short circuiting transfer welding without generating spatters in a stable arc state by changing the proper time of the change of the current and voltage while monitoring the welding transfer state during welding and determining automatically an accurate proper output condition. CONSTITUTION:Short-circuiting or arc being generated is detected from the wire voltage Vp and the short circuit accompanied by the sufficient droplet 4 transfer or the short circuit without being accompanied by the droplet 4 transfer during welding is discriminated according to its signal. As a result, the next time interval t3 between the short-circuits is measured repeatedly from the short circuit completion accompanied by the sufficient droplet 4 transfer to obtain an average value t0m of the time interval between the short circuits. In addition, short circuiting transfer welding is attained by maintaining a high arc current Ip for a period shorter by a specified value t4 than the average value t0m obtained by that time with a next moment of completion of the short circuit accompanied by the sufficient droplet 4 transfer as the starting point and controlling the output of a welding power source 9 so as to change it to a low arc current Ib after passing that period and maintaining the low current Ib when the short circuit ia completed.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an arc welding device, and in particular to welding for consumable electrode arc welding, in which arc welding is repeated between a consumable electrode and a base metal by generating an arc and short-circuiting. It is related to power supply.

C. Prior Art] FIGS. 4(A), (B), and (C) illustrate the relationship between the form of droplet transfer and the wire voltage waveform and wire current waveform in consumable electrode arc welding. An arc 3 is generated between the consumable electrode (hereinafter referred to as wire) ■ and the base metal 2, and a droplet 4 is formed at the tip of the wire 1, and the droplet 4 forms a molten pool 5 on the base metal 2. Welding progresses by repeating contact (short circuit from an electrical point of view) and transfer to the molten pool 5 side.

Conventionally, consumable electrode arc welding, in which welding is performed while repeatedly generating an arc and shorting, requires a power supply unit 6 that outputs a constant voltage and a power supply unit 6 that outputs a constant voltage, as shown in the equivalent circuit of Fig. In order to suppress this to an appropriate level, a welding power source 8 having a configuration in which a reactor 7 is connected in series with the power source section 6 has been adopted. FIGS. 4B and 4C show examples of wire voltage and wire current waveforms when such a power source is used.

In FIG. 4, when the droplet 4 comes into contact with the molten pool 5 (that is, short-circuits) [(d) in FIG. 4(A)], the wire current gradually increases as shown in FIG. 4(C). Droplet 4 becomes molten pool 5
When the short-circuit transition to the side ends, the connection between the wire and the molten pool becomes constricted (g), and finally the separation, that is, the short circuit is broken (t), and the arc is regenerated. The higher the current at the moment of arc regeneration, the more rapidly evaporation occurs due to expansion of the arc plasma and overheating of the thinned droplets, and the droplets remaining at the tip of the wire and the molten metal forming the molten pool 5 The force that causes the particles to scatter increases, generating a large amount of spatter. Moreover, the uniformity of the bead appearance is often impaired.

Subsequently, the melting of the wire tip by arc 3 progresses (
A) to (B), a large droplet 4 is formed at the tip of the wire and is supported by the arc 3 from below (c), but at this time it momentarily comes into contact with the molten pool 5, and the contact A high current flows through the molten pool, causing overheating, or electromagnetic force causes the droplet 4 and the molten pool 5 to be separated, resulting in instantaneous contact, that is, an instantaneous short circuit.

During this period (c) when instantaneous short circuits occur frequently, the movement of the arc 3 and the droplet 4 becomes extremely unstable, and the droplet 4 often falls outside the molten pool 5 in the form of pieces, resulting in large droplets. Generates spatter. While this instantaneous short circuit is repeated for a while, the arc 3 is unable to support the droplet 4 and comes into contact with the molten pool 5 (d), starting full-scale droplet transfer (e).

In this way, spatter is generated when the droplet 4 is attached to the welding wire 1.
There are many cases at the stage (c) when the droplet 4 grows large at the tip and tries to short-circuit with the molten pool 5, and at the moment when the droplet 4 is cut and separated from the tip of the welding wire 1 and the arc is regenerated (g). As the elucidation of this phenomenon has progressed, recently, in order to prevent the formation of spatter,
The effect is to lower the current during arc regeneration (g) and to lower the arc current just before the short circuit (c) to stabilize the droplet 4 and to reduce the occurrence of instantaneous short circuits as much as possible. This knowledge has become common knowledge. Recently, inverter circuits have been adopted in welding power sources, making it easier to control welding current waveforms, and various current waveform controls have been implemented based on the above knowledge.

FIG. 4(D) shows an example of wire current control according to the invention described in JP-A-60-108179, which has been devised based on such knowledge. To explain this in relation to the period of droplet transfer in (A), it will be explained for a while after the short circuit starts (d), that is, the period Ts during which contact is sufficiently formed.
s (1 to 4 m s), then increase the current and heat the shorted part. Immediately before the end of the short circuit, a constriction occurs in the droplet, and the increase in resistance at that part is detected from changes in wire voltage and wire current, and the output current from the power supply is reduced. When the arc is regenerated (g), the current is momentarily lowered, and when the short-circuit force has expired (g), a constant high arc current period TAP is maintained, and then the arc current is set to a low arc current lAl1, waiting for the occurrence of a short circuit.

Here, an attempt is made to momentarily lower the current after detecting the rise in voltage due to the constriction of the droplet 4, but as described in Japanese Patent Application Laid-Open No. 62-212069,
The current at the moment of separation, which is a problem due to the generation of spatter, cannot be lowered sufficiently, and there are considerable difficulties in detecting constrictions using voltage. For this reason, there is still room for improvement in terms of suppressing the current when the arc is regenerated after the short circuit ends.

Further, the high arc current period TAP and the arc current value rap at that time are controlled to predetermined values corresponding to the wire feeding speed, and after that, the arc current remains low to wait for the occurrence of a short circuit. However, if the current value remains high just before a short circuit occurs, large spatter particles will be generated, and the periodicity of short circuit occurrence cannot be expected to be that great. In addition, there are factors that are significantly related to the droplet transfer phenomenon, such as the wire material, extension, welding posture, etc., which are not considered as variables in the pre-condition setting, so in the end, the influence of these factors can be ignored before short circuiting. The period TAll of low arc current had to be lengthened. The period of low arc current results in the period of high arc current TA
It is on the same level as P.

However, this period is only useful for waiting for a short circuit, during which time the arc becomes quite unstable in terms of duration, so it is preferable that it be as short as possible. Furthermore, if this period is too long, the heat input to the base material will be insufficient, resulting in a phenomenon called poor swirl flow. In reality, due to these considerations, a compromise is ultimately made with shorter conditions that involve some spatter.

In the above example, the high arc current condition is set by focusing only on the wire feed speed, but in reality, the wire feed speed
Since the extension and welding posture change, and the wire material also changes, it is not possible to deal with changes in the optimal transition period due to these changes in conditions. For this reason,
This does not necessarily result in optimal conditions for welding work.

[Problem to be solved by the invention]

Most of the conventional technologies are based on the wire current waveform based on the optimum short-circuit transition state, short-circuit transition period, etc. set in advance for limited conditions, while taking into consideration some welding condition factors such as wire feeding speed. Control was sought to reduce spatter formation. Alternatively, an attempt has been made to detect a sign of a short circuit or a sign of the end of a short circuit and control the wire current to reduce the occurrence of spatter. However, in reality, the problem remained that it was not sufficiently effective because it was quite difficult to detect the signs.

As described above, the conventional technology does not take sufficient measures to stabilize the period in which droplets migrate to short circuits, stabilize the behavior of droplets before short circuits, and suppress the current when the short circuit ends and the arc regenerates. Although the amount of spatter has decreased compared to the past, a considerable amount still occurs, and improvements are still desired.

The purpose of the present invention is not only the wire feeding speed but also the wire speed.
Even if various welding conditions such as extensions and welding postures change, the short-circuit cycle is always stabilized to keep the short-circuit transition state of the wire droplets in the optimum condition, and short-circuit transition welding is performed in a stable arc condition with almost no spatter. The purpose is to provide the means.

[Means to solve the problem]

The purpose of the above is to detect whether a short circuit or arc is occurring from the wire voltage and use the signal to weld in a consumable electrode arc welding device that welds by controlling arc generation and short circuit between the consumable electrode and the base metal. determine whether the short circuit is with sufficient droplet migration or without droplet migration, and repeat the inter-short time interval from the end of a short circuit with sufficient droplet migration to the start of the next short circuit. While measuring to determine the average value of the inter-short circuit time interval, during a period that is a predetermined value shorter than the previously determined average value starting from the moment of termination of the short circuit with the next sufficient droplet transfer. This is achieved by controlling the output of the welding power source to maintain a high arc current and reduce the arc current to a low arc current after that period, and by leaving the current low at the end of the short circuit. In other words, the high arc current period ends, the short circuit enters, and the short circuit ends a little before the time when the next short circuit is estimated to be formed based on the short circuit amount time interval measurement results that take into account the transfer state of the droplet. This is sometimes achieved by keeping the current low.

[Effect]

In consumable electrode arc welding, in which welding is performed by repeatedly shorting the arc light body, various factors affect the droplet transfer phenomenon, including not only the wire feed speed but also the wire material, extension, welding posture, shielding gas composition, etc. . However, in the end, the transition state can be collectively expressed by the period, period, etc. of the short circuit. Looking at the occurrence of short circuits during welding, short circuits accompanied by droplet transfer tend to occur periodically, but irregular instantaneous short circuits disrupt this periodicity. It is known.

FIG. 6 is an example of measuring the short circuit time and the frequency of occurrence of 100 short circuits that occurred during short circuit transition welding.

After the short circuit, the distribution shows two peaks with a valley at about 1 ms. A short circuit that forms a peak in less than 1 ms has traditionally been referred to as an instantaneous short circuit, and an investigation of the phenomenon reveals that in most cases, droplet migration does not occur. Ta. It was found that the short circuit forming a peak at 2-5 ms was a short circuit in which a sufficient amount of droplets had migrated. In this way, it was found that the two types of short circuits can be approximately distinguished based on whether the short circuit time is longer or shorter than l m s. To perform stable welding,
Short circuits accompanied by droplet transfer occur periodically and regularly, and 1
It is desirable that the amount of droplets short-circuited and transferred during the process is small. In other words, for the same wire feeding speed, one criterion for selecting optimal conditions is to exclude instantaneous short circuits, that is, to have a large number of short circuits that transfer a substantially sufficient amount of droplets. In such cases, the arc length can be shortened and welding work becomes easier.

In order to lead to such a state, it is preferable to cause a short-circuit transition to occur as soon as a suitable amount of wire melting progresses.

To do this, it is best to melt the wire with a high arc current, and then suddenly reduce the arc current when the droplet becomes large enough to make contact with the molten pool, that is, to short-circuit it.

Then, due to the action of the contraction of the plasma column and the surface tension that causes the droplet to become spherical, the droplet and the molten metal in the molten pool were attracted to each other and came into contact, forming a droplet at the tip of the wire. This is because the molten metal begins to migrate to the molten pool side.

By the way, an instantaneous short circuit occurs after a droplet large enough for droplet transfer has been formed, so the period from the end of a short circuit accompanied by droplet transfer to the start of the next short circuit, including the instantaneous short circuit, is can be regarded as the period during which wire melting progresses to an appropriate amount for droplet transfer.

On the other hand, the period from the end of the instantaneous short circuit to the start of the next short circuit, including the instantaneous short circuit, is noise that has nothing to do with detecting the period in which wire melting has progressed to an appropriate amount for droplet transfer. . Therefore, a short circuit with a short circuit duration of l m s or more is regarded as a short circuit accompanied by droplet transfer, and the period from the end of the previous short circuit of l m, s or more to the start of the first short circuit is defined as the regular inter-short circuit time. Measured as an interval. Short circuit period is 1m
A short circuit of 1 ms or less is judged as an instantaneous short circuit, and the time interval from a short circuit of 1 ms or less to entering the next short circuit is excluded from the measurement value. Based on the values measured in this way, find the average time interval from the end of the short circuit transition to the start of the next short circuit, and based on that value estimate the time immediately before the next short circuit starts,
Reduce the current to a low level immediately before a short circuit.

If a short circuit occurs before the estimated time, the current is immediately reduced to a low level and the current is maintained at a low level until the short circuit transfer accompanied by droplet transfer is completed.

As described above, even if an instantaneous short circuit of l m 5 or less occurs at first, based on the short circuit signal including it, the time interval between short circuits becomes the shortest short circuit transition state interval, that is, substantially The welding state will be drawn into a state where the number of short circuits accompanied by droplet transfer is the greatest in that welding state.

Thus, in the present invention, welding is performed while determining the appropriate value for the high arc current period based on the period of transition under the conditions from the short circuit signal during welding, rather than using a preset setting. Short circuits that occur irregularly become synchronized and self-oscillate, so that the transition state becomes stable as an extremely periodic phenomenon. In this way, instantaneous short circuits no longer occur, and large spatter, which tends to occur when the droplets become large, no longer occurs.

If you lower the current just before the short circuit, and then continue to keep the current low even when the droplet short circuit transfer progresses and ends, there will be less overheating of the constricted part when the droplet constricts and separates. Even during arc regeneration, only a small plasma column with a low arc current is formed, so the expansion of the plasma column is gradual and no spatter occurs.

In the present invention, the short-circuit transition state during welding is cumulatively corrected so that the short-circuit transition period becomes optimal. Condition settings etc. are essentially unnecessary.

[Embodiments of the invention]

FIG. 1 shows (A) a transfer state of a droplet, (B) a wire voltage waveform, and (C) a wire current waveform according to an embodiment of the present invention.

After the short circuit of 1 m s or more ends (Ta) and a lapse of 0.3 ms (for example, 0.3 ms), the control of the power supply is switched to constant voltage characteristics so that the wire voltage becomes Vp, but as a result, the arc current becomes high and enters a period of high current 1p (for example, 200A), and the melting of the base metal and wire progresses. After 2 hours have elapsed since the short circuit of 1 m s or more is completed (Ta), the power supply is turned off. Control is set to constant current characteristics and switched to base current Ib (for example, 40 A). Then, due to the sudden decrease in arc current, the arc force decreases, the plasma column contracts, the surface tension of the droplet, etc. acts, and the wire feeding speed remains constant and the wire melt amount decreases, so the wire tip gradually approaches the base metal. This also induces short-circuiting of the droplets. The previous short circuit of l m s or more is completed (Ta'
) until the next short circuit occurs (Ts) t3
Measure. This +3 is measured every time there is a short circuit, but this time the set value is t: +1, t:1)% = 0.8 t ym' + 0
．． 2 t ff---L3 obtained from (1)
Set m as the next set value, and set the gauge time t4 to keep the arc current low to 1 ms, tz = txIIj 4
------- From the relationship (2), find the set value for the next time t2 for switching from high current to low current.

As long as the set value of +3□ at the start of the arc generation of the first weld bead is set to a long value, there is no problem with any value, but spatter will occur during the period when this value is inappropriate, so it is necessary to quickly set the value to the appropriate value. In order to converge, we usually start with a setting of 100 ms. The set value of +31 at the time of starting arc generation of the weld bead from the next time onwards is used by storing the value during the previous arc generation.

In addition, if a droplet short circuit happens to occur before reaching +2 calculated according to equation (2), even if it is an instantaneous short circuit of 1 ms or less, the current is immediately switched to the lower base current rb. . In this way, the current rb is always low at the end of the short circuit transition. The lower the current at the end of the short circuit transition, the less spatter occurs, but the practical range is 60 A or less. However, if the current is less than 20 A, the stability of arc regeneration deteriorates. Considering these considerations, the base current rb was selected to be 40A. 4
At around 0 A, no spatter occurs, and there is no practical problem with the stability of arc regeneration.

At arc start, when the short circuit transition period is still not stable, and even during normal welding, there are rare cases where the short circuit does not break easily.
A short circuit phenomenon lasting longer than 1 Q m s may occur.

In such cases, as adopted in the prior art,
When the length exceeds 10 m5, a large current such as 300 A is immediately applied to heat the short circuit and cause a melt to occur, thereby regenerating the arc. It freezes when I poke it.
This prevents "stipping". That is, the short circuit time t.

The current is also measured, and if this exceeds 10 ms, it is assumed that there is a risk of tapping, and the current is switched to a high current to quickly cause a blowout and regenerate the arc. Spatter is generated during this melting, but since this occurs only rarely, it can be said that the amount of spatter generated as a whole has been extremely reduced. Since this type of control is adopted, arc sustainability is maintained at the same level as conventional power supplies, and there is no instability due to arc breakage or tapping.

In this embodiment, after 1+ (for example, 0.5 ms) has elapsed after the short circuit of 1 ms or more ends (T a ), the control of the power supply is switched to constant voltage characteristics so that the wire voltage becomes Vp, As a result, the arc current increases and enters a period of high current 1p (for example, 200A), but this is provided as a time for the surface shape of the molten pool and the molten metal at the tip of the wire to stabilize after the short circuit ends. be. In reality, there is a problem that the rise of the current is suppressed by the actance associated with the output cable of the power supply, so it is actually t.

may not be added as Qms.

In addition, the current Ip at the time of high arc current in this embodiment is set to a value that is determined dependently from the balance with the wire feeding speed, with the power supply 9 (see Fig. 2) having constant voltage output characteristics. It is also possible to use a constant current determined from a functional relationship, such as proportionality, to the wire speed.

The target time L4 for maintaining the arc current at a low level is provided so that the droplet, which has grown large at the tip of the wire, can stably contact the base material. If it is too short, a short circuit will occur during a high current period when the period of short circuit transition is slightly disturbed, increasing the frequency of forming large spatter particles. If t4 is made too large to avoid this, the average value of the arc current will decrease, and the melting of the base metal will be less compared to the amount of wire melting, that is, the amount of deposited metal forming, resulting in poor bead shape. The stability of the arc also deteriorates. The appropriate range is about 0.5 to 3 ms, and in this example, it was set to 1 ms.

FIG. 2 is a diagram illustrating the main configuration of the welding power source according to the present invention. A three-phase AC voltage is input, and a DC main power circuit 10 consisting of an inverter circuit outputs a DC welding current.

Its output characteristics are that the signals v, , v2 from the output current detector n1) or the output terminal voltage detector 12 are output to the control circuit 1.
), and is controlled by switching to constant current characteristics or constant voltage characteristics at any time. In addition, the short circuit detection circuit 14 receives the signal 2 from the output terminal voltage detector 12 and determines whether the droplet 4 is in contact with the molten pool 5 (see FIG. 1), that is, whether a short circuit is occurring. The output V3b is sent to the period measurement/short circuit discrimination circuit 15. The period measurement/short circuit detection circuit 5 measures the short circuit time and determines whether the short circuit has continued for 1 ms or more or 10 ms or more for the determination reference time set by the determination reference time setting circuit 16. The determination is made, and the time from the end of a short circuit of l m,s or more to the start of the next short circuit is measured, and the result is outputted to the control signal circuit 17 v4. The control signal circuit 17 is mainly composed of a microcomputer, performs calculations based on the input signal v4, and sends a period signal 5 for producing a high arc current as a constant voltage output to the output control circuit 1). The output control circuit 1) forms an output control signal v6 for a high arc current period of constant voltage output, a period of constant current low current output, etc., and sends it to the DC main power circuit 10. With this configuration, the control explained in FIG. 1 is performed.

FIG. 3 shows (A) wire voltage and (B) wire current according to another embodiment of the present invention.

The main configuration of the welding power source is almost the same as that shown in FIG. 2, and the only difference is a part of the internal configuration of the period measurement/short-circuit discrimination circuit 15, so illustration thereof is omitted.

If the wire current during short-circuit transition is low, it will take a long time to constrict the droplet, and if the welding wire feeding speed is high, the unmelted part of the wire will enter the molten pool during that time, and the welding wire There is a strong possibility that "stapping" will occur, where the metal will be welded to the base metal and the arc will not re-occur. To prevent this, when the wire speed is high, a high current is allowed to flow even during a short circuit. In this case as well, regardless of the value of the wire current, control is performed focusing on the short-circuit time in that state.

The embodiment shown in FIG. 3 has a function to control the high current arc generation time by measuring the time t from the end of the short circuit (Ta') of 1 ms or more to the start of the short circuit (Ts) as explained in FIG. While holding it, a period of short circuit transition is also detected, and 1
A certain period of time tb from the short circuit start point (T s ) of ms or more
After (1ms) elapsed, high current 1s under constant current control
(150A) before the short-circuit transition of the droplet is completed. The current is then returned to the low current Ib (40A). That is, the short-circuit time t during the short-circuit transition of more than l m s before that.

The average value of the short circuit time obtained by measuring each time is tl,
′ and the newly measured short circuit time as t, calculate the new average value jsm of the short circuit time from the following equation.

tsm=0.8 j5m' + 0.2 ts
-----------(31Actually, 1ms from the start of the short circuit
After a period of time, the high current Is is applied, and t 7 = j S+++ o, 5 rn 3
------(It is determined from 41, but after 7 hours have passed from the start of the short circuit, the current is returned to the low current Ib again.

In this embodiment, a certain period tb from the short circuit start time (Ts)
(1ms), high current Is in constant current control state after lapse of 1ms.
The period of a j & set at (150A) is the period necessary for the droplet to make sufficient contact with the molten pool to the extent that it will not burn out immediately even if a large current is applied, and such a contact state is maintained. Normally, 0.5 ms or more is required for formation.

In addition, at t7 before the end of the short-circuit transition of the droplet, the low current 1b is again applied.
(40A), but LSII j? The value of is determined from the time required for the wire current to be sufficiently low to prevent spatter at the end of the short circuit (g). Although this is related to the dynamic characteristics of the power supply, it is preferable to make it as short as possible, and in this example it is 0.5 m.
It was set as s. According to this embodiment, the current at the end of the short circuit is sufficiently low so that no spatter is formed, and the droplets during short circuit transition are resistively heated by the high current, so that the short circuit transition is smoother, and the transition time is is also shorter, so
This has the effect of making arc regeneration more stable after the short circuit ends.

In the embodiments listed here, the calculation of the average time L3m from the end of the short circuit to the start of the short circuit or the average short circuit time jsm takes economic efficiency into consideration, and because it can be easily calculated with a simple microcomputer, the above (1) is used. Although Equation and Equation (3) are used, the present invention is not limited to these equations, and a weighted average time estimated based on fuzzy theory or knowledge processing theory may also be used. Furthermore, the measurement and control of these times can be implemented using an analog circuit instead of relying on a microcomputer.

〔Effect of the invention〕

All conventional spatter prevention measures have been based on preset cycles and times of current and voltage changes. In contrast, in the present invention, the transition state is monitored during welding and the appropriate time for changes in current and voltage is changed to automatically determine the appropriate output conditions. The transition state has become more periodic and regular, and welding with excellent arc stability and very little spatter can now be applied to a wide range of welding conditions.

In conventional consumable electrode arc welding with short-circuit transition, the generation of spatter was unavoidable to varying degrees, but according to the present invention, short-circuit transition arc welding can be performed with almost no spatter. As a result, short-circuit transitional arc welding can be applied to welding in locations where spatter generation is averse, and also has the effect of eliminating the need for spatter removal operations after welding even in normal applications. Furthermore, since the cycle of short-circuit transition is stabilized, the weld bead surface becomes more beautiful.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of droplet transfer phenomenon, arc voltage waveform and arc current waveform during welding by the welding apparatus of the present invention, and Fig. 2
3 is an explanatory diagram of another embodiment of the present invention; FIG. 4 is an explanatory diagram of the prior art; FIG. 5 is an explanatory diagram of a conventional welding power source; The figure is a characteristic diagram showing short circuit time distribution. ■...wire, 2...base material, 3...arc, 4...
...Global droplet, 5... Molten pool, 9... Welding power source, 10.
...Main power circuit, 1)...Current detector, 12...Voltage detector, 1)...Output control circuit, 14...Short circuit detection circuit, 15...Period measurement/short circuit discrimination circuit , 16...
-Discrimination reference time setting circuit, 17...control signal circuit. D￥￥n Kanano Nori'V and general

Claims (5)

[Claims] (1) A consumable electrode arc welding device that is equipped with an arc power source, a wire feeding device, an arc torch, and a shielding gas supply device, and performs welding by repeatedly generating an arc and short-circuiting between the consumable electrode and the base metal in a shielding gas. Output control means for controlling the output current or output voltage; Short-circuit detection means for detecting whether or not a short-circuit transition between the consumable electrode and the base material is occurring from the wire voltage; Reference time setting means for setting a reference time; a timer for measuring the short circuit time and the short circuit interval; a discriminator for determining a short circuit for a time longer than the reference time; and a discriminator for determining a short circuit for a time shorter than the reference time; and based on the results of repeatedly measuring the short circuit interval, A control signal output means for determining a period for outputting a high arc current between the end of a short circuit for a time longer than a reference time and the start of the next short circuit, and outputting a control signal to the output control means during the period. A consumable electrode arc welding device characterized by: (2) The consumable electrode arc welding apparatus according to claim (1), wherein the reference time is 1 millisecond. (3) In claim (1) or claim (2),
The control signal output means enters a high arc current period of 100 A or more within 1 millisecond after a short circuit lasting longer than a reference time ends, and it is estimated that the next short circuit will occur based on the measurement of the short circuit interval. A consumable electrode arc welding device characterized in that it is configured to output a control signal so as to enter a period in which the current is controlled to a constant current of 60 A or less during a period of 0.5 to 3 milliseconds before the point in time. (4) In claims (1) to (3), when a short circuit occurs during the set high arc current period, a control signal is output so as to immediately enter a period in which the current is controlled to a constant current of 60 A or less. A consumable electrode arc welding device characterized by being configured to: (5) In claims (1) to (4), there is provided a means for measuring the short circuit time for a short circuit longer than the reference time; A means for determining a period for controlling the output to a constant current of 100 A or more during a short circuit and outputting a control signal to the output control circuit; 60 milliseconds before the time point when the current short circuit is estimated to end from the short circuit time measurement results of the short circuit that has entered the control period and the time longer than the reference time.
A consumable electrode arc welding device characterized by comprising means for maintaining an output current at a current below A.