JPH0118831B2 - - Google Patents

Description

Detailed Description of the Invention "Field of Industrial Application" The present invention is a consumable welding process in which V-shaped, L-shaped or similar grooves are welded by weaving the welding torch left and right along the welding line. The present invention relates to an electrode arc welding method, and particularly to a welding method using automatic welding line tracing control that causes a welding torch to accurately follow the welding line.

"Prior Art" Arc tracing is possible by detecting welding current or welding voltage, but here we will explain it by taking welding current as an example.

In short, the problem of weld line tracing as described above is solved by the fact that the weld line detector has a wide range of applicability in response to the structural constraints of the welded structure and changes in the shape of the weld groove, and also in response to changes in various conditions during welding. The problem also comes down to whether or not it is possible to accurately detect the weld line to be welded, and methods that utilize welding phenomena are often used as a method for tracing such weld lines. This method detects the weld line in the groove by using the correlation between the protruding length of the welding wire and the welding current, and has the advantage of not requiring a dedicated detector.

Japanese Patent Publication No. 52-10773 and Japanese Unexamined Patent Application Publication No. 1983-1977 are a method for automatically tracing welding lines using this type of welding phenomenon.
A method described in the specification of Publication No. 78654 is known. This method detects and compares welding current values at both left and right end positions of periodic weaving, and controls movement of the center of the weaving so that both welding current values become equal.

Another known method is the consumable electrode type arc welding method described in Japanese Patent Application Laid-Open No. 58-53375. In short, this method compares the maximum and minimum welding current values in each half period of periodic weaving, calculates the difference, and calculates the maximum and minimum current values for each adjacent half period. By comparing the difference and controlling the movement of the center of the weaving in the direction where the difference between the left and right sides becomes smaller, the weaving is made to move in a manner that accurately follows the welding line. and Japanese Patent Application Publication No. 1973-
Compared to the method described in Publication No. 78654, this method has the advantage of superior detection resolution for deviations from the weld line and the ability to perform precise weld line tracing control.

Next, an example of the welding line tracing method which is the basis of the present invention will be explained with reference to the welding method described in the above-mentioned Japanese Patent Application Laid-Open No. 58-53375.

Now, for example, using a welding current with constant voltage characteristics, weave the welding torch 3 in the horizontal direction with respect to the welding line S _w for the horizontal plate 1 and vertical plate 2, which are the base materials, as shown in Fig. 5. Horizontal fillet welding is performed by progressing welding in the welding line S _w direction (perpendicular to the plane of the paper). When the welding wire 4 is fed at a constant speed, the protruding length l of the wire 4 changes along the groove shape.

FIG. 5 shows a case where the center position _S2 of the weaving width of the welding torch 3 coincides with the welding line _Sw , and the welding line is correctly followed. FIG. 6 shows the forward stroke of weaving in which the torch 3 reaches the right end position S ₃ from the left end position S _{1 of the weaving via the weaving center S 2 and} vice versa in the case of FIG. ₅ .
3 is a graph showing changes in welding current during the backward stroke of weaving from 1 to ₁ to the left end S1. The details are leg length
The welding current when performing 7.5 mm horizontal fillet welding was measured and recorded using an optical electromagnetic oscilloscope.
This is a waveform diagram showing the results, and the welding conditions are shield gas composition Ar + 20% CO ₂ , wire diameter 2 mm, average welding current 305 A, welding voltage 29 V, welding speed 35
cm/min, weaving cycle approximately 150 times/min, and weaving width L=6 mm. In addition, when measuring the welding current, a low-pass filter with a cutoff frequency fc of 10 Hz is used to remove high frequency components caused by the commercial frequency of the welding power source and components caused by irregular short circuits of the arc. .

The left and right end positions of the weaving are detected by a proximity switch attached to the weaving device.

As is clear from Fig. 6, when the center _S2 of the weaving width of the welding torch 3 coincides with the welding line _Sw and the welding line is correctly followed, the wire protrusion length l and the welding current are From the correlation characteristics, the weaving left end current IL ₁ and the weaving right end current value IR ₁ are equal to each other, and the minimum welding current IL ₂ in the forward stroke of weaving and the minimum welding current value IR ₂ in the backward stroke are equal to each other.

In addition, the left end current value IL ₁ is from the time when the weaving left end position S ₁ is reached, and the right end current value IR ₁ is from the time when the weaving right end position S ₃ is reached by the phase delay time τ of the welding current signal due to the low-pass filter. It can be seen that an appropriate response can be obtained if the detection is performed at each delayed time point.

Furthermore, the minimum welding current value IL ₂ is the left end current value IL ₁
After detecting the welding current, the welding current in the forward process leading to the weaving right end position S ₃ was measured, and the minimum value between them was detected. Similarly, the welding current minimum value IR ₂ was detected as the right end current value IR ₁ . Thereafter, the welding current during the backward stroke to the weaving left end position _S1 may be measured and the minimum value during that period may be detected.

Next, Fig. 7 shows the weaving state when the center position _S2 of the weaving width is shifted toward the vertical plate 2 side with respect to the welding line _Sw , and Fig. 8 shows the forward and reverse weaving strokes in this case. This is a specific example showing a change in welding current during a process.

Figure 8 shows the results of measuring and recording the welding current using an optical electromagnetic oscilloscope when horizontal fillet welding was performed under the welding conditions explained in Figure 6, and the deviation from the weld line S _w was D = 2 mm. FIG.

From FIG. 8, when it shifts toward the vertical plate 2 side, a deviation occurs between the right end current value IR ₁ and the left end current value IL ₁ .
It can be seen that IR ₁ > IL ₁ .

By the way, what should be noted in Figure 8 is that
Minimum value of welding current in the forward stroke of weaving
IL ₂ is not equal to the minimum value IR ₂ of the welding current in the backward stroke, and IR ₂ <IL ₂ . That is, even though IR ₂ should be equal to IL ₂ by simply applying the correlation between the wire protrusion length and the welding current, they are not equal. The reason for this is not necessarily clear, but if we estimate from the arc observation results during the experiment, if the center _S2 of the weaving width deviates from the welding line _Sw as described above, the Since the behavior of the molten pool at the weld line is different from that in the case where there is no deviation, the behavior of the arc with respect to the molten pool is different, and it is presumed that a change in the welding current as shown in FIG. 8 occurs.

Furthermore, in the experiment, the minimum value in the above outgoing process was
In contrast to IL ₂ , the minimum value IR ₂ in the backward stroke is always small, and the difference between the two minimum values (IL ₂ - IR ₂ ) is
It was confirmed that the value was proportional to the magnitude of the deviation between the center position _S2 of the weaving width and the welding line _Sw .

Conversely, if the center position _S2 of the weaving width shifts toward the horizontal plate 1 side, the result will be opposite to that shown in Figure 8.
The magnitude relationship between IL ₁ >IR ₁ and the minimum value of the welding current in the forward stroke and backward stroke is IR ₂ >IL ₂ .

To summarize the above relationship, let IL ₁ be the current value at the left end of the weaving in the forward stroke of weaving from the left end position S ₁ of the weaving to the right end position S ₃ via the weaving center S ₂ , and the minimum value of the welding current.
Assuming that IL ₂ is the current value at the right end of weaving in the backward stroke from right end position S ₃ to left end position _{S 1} and IR ₂ is the minimum welding current value, (IL ₁ − _{IL 2} ) = (IR ₁ − IR ₂ ), it may be determined that the weaving center _S2 and the welding line _Sw coincide.

If (IL ₁ −IL ₂ )>(IR ₁ −IR ₂ ), it is determined that the weaving center S ₂ is shifted to the left with respect to the welding line S _w .

If (IL ₁ −IL ₂ )<(IR ₁ −IR ₂ ), it is determined that the weaving center S ₂ is shifted to the right with respect to the welding line S _w .

This provides a criterion for determining the direction of deviation.

In addition, the difference current values (IL ₁ - _{IL 2} ), (IR ₁ - IR ₂ ) and the deviation of the two difference currents | (IL ₁ - IL ₂ ) - (IR ₁ - IR ₂ )| are the welding center S ₂ of the weaving center. Proportional to the deflection value relative to the line.

Therefore, the current value in each weaving half cycle
Detect IL ₁ , IL ₂ and IR ₁ , IR ₂ , and then use these detected values to calculate the difference current value (IL ₁ −IL ₂ ), respectively.
(IR ₁ − IR ₂ ) is calculated and the difference current values are compared with each other, and based on the comparison result, the difference current values become equal to each other or in the direction of a constant value based on the deviation direction judgment criteria described above. By controlling the movement of the weaving width center position according to the deviation between the two difference current values, automatic tracing of the weld line becomes possible.

"Problems with the Prior Art" By the way, in the welding line tracing method as described above, four welding current values IR ₁ , IR ₂ , IL 1 , IL ₂ are obtained by performing one cycle of weaving motion in the right and left _directions. The weaving center in the next weaving cycle is corrected based on this collected data, and the correction is made every time four welding current values are sampled, that is, every weaving cycle, so the center of weaving is rough. It was only possible to make corrections, and it was not possible to improve the tracking accuracy of copying. Furthermore, with a control method in which the weaving center can only be moved by a certain distance in one correction, the tracking range is narrow, and there is a problem that it is not possible to follow a weld line that bends sharply in a small arc. It had a point.

``Object of the Invention'' Therefore, the object of the present invention is to double the number of corrections to the weld line compared to the conventional method, and to enlarge the tracing range even when the amount of correction per time is constant. The purpose is to improve the accuracy and responsiveness of tracing and following by adding correction.

"Structure of the Invention" In order to achieve the above object, the main means adopted by the present invention is to move a consumable electrode type welding torch in the direction of the welding line while weaving, and detect it in the left and right half cycles of the weaving. This is a consumable electrode arc welding method in which welding is carried out by repeatedly correcting the center position of the weaving by a predetermined correction amount in the direction of eliminating the difference between the welding currents or welding voltages, and the correction is made half a cycle before. If so, the predetermined correction amount is converted into a corresponding welding current value or welding voltage value, and this converted welding current value or welding voltage value is applied to the weaving half cycle half a cycle before (more specifically, the weaving half cycle before a half cycle). If the correction direction and weaving direction match, subtract from the welding current or welding voltage detected in the weaving half cycle immediately before the correction, and add it if the weaving direction is opposite.
If the correction was made half a cycle ago, the subtracted or added welding current or welding voltage, and if the correction was not made half a cycle ago, the welding current or welding voltage detected in the previous weaving half cycle. , the next correction direction is obtained based on the difference between the welding current or welding voltage detected in the previous half cycle of weaving, and its scope of application is disclosed in the above-mentioned JP-A-58-
Not only the arc welding method described in Publication No. 53375, but also Japanese Patent Publication No. 52-10773 and Japanese Patent Application Laid-open No. 52-78654.
The present invention can be applied to the welding line automatic tracing method described in the above publication and all other consumable electrode type arc welding methods.

"Embodiments" Next, embodiments embodying the present invention will be described with reference to the accompanying drawings of FIGS. 1 to 4 to provide an understanding of the present invention. Here, FIG. 1 is a diagram showing changes in welding current when a welding method according to an embodiment of the present invention is implemented, FIG. 2 is a flowchart showing the control procedure of the welding method, and FIG. 3 is a diagram showing the same. FIG. 4 is a side view of the entire robotic device for carrying out the welding method according to the embodiment, and is a block diagram showing the flow of signals in the robotic device. In this embodiment, the same reference numerals are used for the same elements as those in the conventional example.

Fig. 3 shows the V of the base material M by the welding robot R.
A state in which arc welding is being performed on the shaped groove portion 5 with the welding torch 3 is shown, and the current flowing through the welding wire 4 is detected by the shunt 6 and transmitted to the robot control panel 7. The robot control panel 7 is a welding robot R.
A control section for guiding the welding torch 3 along a predetermined welding line by driving the welding robot R, a robot body control section 8 for specifying the attitude and position of the welding robot R, and furthermore a weaving waveform, and a shunt 6. A tracing control section 9 samples the welding current input from the controller and sends a correction command signal to the robot main body control section 8 in order to correct and move the welding torch 3 in the direction of an appropriate welding line _S. (4th
figure).

The welding robot R is driven by a servo control unit 10 (FIG. 4) of the robot main body control unit 8, driven by a signal from the servo control unit 10, is attached to a foundation 11, and is rotated around a vertical axis by an arrow θ. ₁ , a vertical arc 13 mounted on the swivel base 12 that performs a driving motion in the direction of the arrow θ ₂ in the vertical plane, and a swivel base 13 mounted on the tip of the vertical arc 13 Horizontal arc 14 driven to swing in the direction of arrow θ ₃ in the same driving vertical plane
and a wrist portion 15 that is attached to the tip of the horizontal arc 14 and holds the welding torch ₃ . The welding robot R is able to perform a pivoting motion around the axis of the wrist portion 15 as indicated by θ ₅ , and as a result, the welding robot R is constructed with five degrees of freedom.

Next, with reference to FIGS. 1, 2, and 4, the control procedure for the arc welding method using the microcomputer will be explained in detail. Figure 1 is a diagram showing changes in welding current in a conventional welding method;
This corresponds to FIG. 8, and FIG. 2a shows the control procedure in the robot main body control section 8, and FIG.
2 is a flowchart showing the processing procedure in the copying control section 9, and FIG. 2c is a flowchart showing the correction subroutine of I shown in FIG.

First, for control, the robot has already been taught the welding line using the PTP method, for example.
It is assumed that the teaching data is stored in a storage device (not shown) within the robot main body control section 8. Furthermore, whether or not to correct the welding line, which is the main theme of the present invention, is determined by the operator's intention regarding the groove shape. By operating it, it is possible to select whether or not to perform correction processing.

First, at the start of welding work, the processing procedure on the robot main body control section 8 side shown in Fig. 2a and the processing procedure on the copying control section 9 side shown in Fig. 2b are synchronized and executed in parallel. will be processed. In the program shown in FIG. 2A, the microprocessor first retrieves teaching data from a storage device (not shown), performs interpolation processing on this data, and calculates the position of the weld line based on the teaching data. This calculation is performed in the position calculation section 16 of the robot main body control section 8. In addition, the microprocessor has a weaving pattern (e.g. weaving width, weaving period, etc.)
is specified, and based on this weaving pattern, the weaving waveform calculation section 17 performs weaving waveform calculation, that is, calculation of the weaving locus. Here, in step a, it is determined whether a correction command signal has been transmitted from the copying control section 9 to the copying input section 18 of the robot main body control section 8.

On the other hand, on the side of the scanning control section 9, as shown in FIG. 2b, in step b, the welding torch is in a state of movement of the torch 3 from the left end position _S1 to the right end position _S3 shown in FIG. 5 (rightward process). Determine whether or not.
At this time, assuming that the torch is moving from the left end position _S1 to the right end position, the welding current at that time is during the first half period of the curve P as shown in Figure 1.
It is represented by the part C _1l . As shown in FIG. 4, a weaving left end signal and a weaving right end signal are output from the weaving waveform calculation section 17 of the robot main body control section 8 to the copying calculation section 19 of the copying control section 9. The start and end points of each half cycle and left half cycle, that is, the position of the left end S ₁ and the position of the right end S ₃ are informed.

Here, to explain Fig. 1 based on the flowchart of Fig. 2b, in the first left half circumference _C1l after welding is started, the welding torch 3 moves from the left end (L) to the right end (R). Since it is an interval, processing proceeds in the direction of YES in step b, and the left half period C _1l
The maximum welding current value IL ₁ and minimum value IL ₂ of the welding current P at are sampled, and in the next step c, it is determined whether welding processing for a half period of the weaving operation has been completed. The criterion for this judgment is whether or not the above-mentioned weaving right end signal has been input from the weaving waveform calculating section 17 to the copying calculating section 19. If the judgment is NO at this time, the process returns to step b. If the answer is yes,
This indicates that the welding torch 3 has arrived at the right end position _S3 , and the process proceeds in the yes direction to step d.
Let's move on to the judgment. At step d, step c
It is determined whether or not the half cycle determined in step B is the first half cycle at the start of welding, that is, the left half cycle _c1l in FIG. 1, and if YES, the process returns to step b. At this time, since the welding process has already moved from the right end position S ₃ to the left end position S ₁ (right end half cycle C _1r in FIG. 1), a negative determination is made in step b. Note that since this is not the first half cycle at the start of welding, it is always determined as NO. Therefore, the maximum value IR ₁ and minimum value IR ₂ of the welding current P in the leftward stroke are sampled. Then, in step c, half a period, in this case, the right half period.
c After waiting for the completion of _1r , the process goes through step d and then goes to step e. The first one weaving cycle has now passed, and it becomes possible to calculate the differential current value described in the above-mentioned Japanese Patent Application Laid-Open No. 58-53375. Here, in step e, it is determined whether the mode changeover switch described above has been switched to the correction mode,
If the mode is switched to the correction mode, the first period part, that is, the first left half period c _{1l is} selected in step f.
Then, it is determined whether weaving in the right half cycle c _1r is currently completed. At present, the answer is yes, so calculate the difference ΔIL in current values in the left half cycle by IL ₁ - IL ₂ , and calculate the difference ΔIR in welding current values in the right half cycle by IR ₁ - _{IR 2} . Calculate and further welding line S _w of weaving center S ₂
Difference current value ΔO = ΔIL− proportional to the deflection value for
Calculate ΔIR. Further, in step g, it is determined whether the difference current value ΔO is a negative value. If the difference current value ΔO is 0 or a positive value, the R flag is set to 1, and if it is a negative value, the R flag is set to 0. After that, in step h, the absolute value of the difference current value ΔO is within the allowable range (dead zone).
It is determined whether or not it is within the dead zone, and if it is far from the dead zone, a correction command signal is sent to the input section 18. If the absolute value of the difference current value ΔO is within the allowable range, the program is returned to the beginning without modifying the welding trajectory.

On the other hand, the robot main body control section 8 shown in FIG.
If it is determined that there is a modification command after receiving the modification command signal in step a of the program on the side, the modification command position is given to the position calculation unit 16, and in response to this, the position calculation unit 16 changes the fixed shift amount to the original position. Add to the target position data. Thereafter, the microprocessor superimposes the data corresponding to the weaving waveform sent from the weaving waveform calculating section 17 and the calculated position data corresponding to the corrected welding line locus, and according to the signal from the corrected timing generating section 20. The corrected target position signal is sent to the servo control unit 10, and the welding robot R is servo-controlled.
The first correction (correction 1) is performed at (L ₁ ) in the figure.

After this modification, the determination as to whether or not the modification is made at the end of the first cycle in step f shown in FIG. '', the process moves on to calculation of the difference current value ΔO.

For example, in Fig. 1, if a correction is to be made after the fifth half period c _3l has passed (correction 4), if the data is corrected (correction 3) to shift to the right in half period C _3l . If the shift was performed, the data used in modification 4 is the above half cycle C _3l
IL ₁ and IL ₂ in , and the data IR ₁ and IR ₂ in the half cycle C _2r half a cycle before this, but IR ₁
and IR ₂ are affected by modification 3, so if the data is corrected by the constant shift due to modification 3, the calculation of the difference current value ΔO will be more accurate and responsiveness will be improved.

Correction of such data is performed in step f.
Since this is after the end of the first cycle in step f, when it is judged as no in step f,
The subroutine "correction of I" shown in FIG. 2c is executed.

That is, to correct the welding current I, as shown in FIG. 2c, it is first determined whether the correction was made half a cycle ago (step i). If we are currently at the stage of making modification 4 in Figure 1, the modification (in this case modification 3) has been made half a cycle before, so the process proceeds in the direction of YES at step i, and the left end is changed at step j. It is determined whether the correction was made in the welding section from the section to the right end section (rightward process). In the example shown in FIG. 1, the answer is YES, so the process proceeds to step k. In step k, it is determined whether the R flag is set to 1 or not, and the difference ΔIR between the maximum welding current value IR ₁ and the minimum value IR ₂ obtained in the welding process of half cycle c _2r after modification 2 is set as a predetermined value. Correction amount K
After correcting this to a new current value difference ΔIR, ΔIL=IL ₁ −IL ₂ is calculated. R flag is 0
In this case, correction to the left is required, so
After replacing ΔIR with ΔIR=ΔIR−K, ΔIL=IL ₁ −
Perform IL ₂ calculation.

If it is determined at step j that the stroke half a cycle ago was weaving from the right end to the left end, the process jumps to step m and determines whether the R flag is 1. Since a correction in the direction is required, the difference in current value ΔIL obtained in the welding process of the first half cycle is replaced with ΔIL - K, and then ΔIR = IR ₁ - IR ₂ is calculated, and the R flag is 0. In this case, after correcting ΔIL to ΔIL+K, ΔIR=IR ₁
- Perform the operation of IR ₂ . After completing the above corrections,
The process returns to step n in which a difference current value ΔO is calculated from the current value differences ΔIL and ΔIR obtained in the previous steps, and the same process as described above is performed to reach the stage of sending a correction command.

The correction amount K described above is a conversion value obtained by converting a predetermined correction amount corrected in each correction shown in FIG. 1 into a welding current or a welding voltage. In addition, the conversion method is based on the change in welding current or Find the value of welding voltage. Then, the determined value is stored in a storage device, and when welding is performed under the same conditions, it is called up and used in step k or m in FIG. 2c.

In the above embodiment, when the correction was made half a cycle ago, the welding current or welding voltage detected in the previous weaving half cycle was corrected. This is substantially equivalent even if the welding current or welding voltage detected in half a cycle is corrected. Furthermore, it is substantially equivalent to calculate the difference first and then correct the difference. Therefore, it goes without saying that these cases are also included within the scope of the present invention.

The present invention can be applied, for example, to step b in FIG. 2b.
As clearly shown in the procedure of step c, the maximum and minimum values of the welding current during each half cycle of the weaving operation are sampled, and the weaving process is continued using the existing data of the half cycle. The position of the weld line is corrected every half cycle of operation. Therefore, when the welding line is corrected by a fixed shift amount, the correction amount of δ ₂ will be obtained after four cycles as shown in FIG. On the other hand, in the conventional method of making corrections every cycle, the amount of correction after 4 cycles is δ _{1, and as the number of cycles increases, in the case of the present invention, compared to the conventional method, the amount of correction becomes δ 1} . You will get double the amount of correction.

"Effects of the Invention" As described above, the present invention moves a consumable electrode type welding torch in the welding line direction while weaving, and compares the welding current or welding voltage detected during the left and right half cycles of the weaving. This is a consumable electrode type arc welding method in which welding is carried out by repeatedly correcting the center position of the weaving by a predetermined correction amount in the direction where the difference disappears, and if the correction has been made half a cycle before, the predetermined correction amount is Convert this to the corresponding welding current value or welding voltage value,
This converted welding current value or welding voltage value is subtracted from the welding current or welding voltage detected in the previous weaving half cycle if the correction direction and the weaving direction match, and if the weaving direction is opposite, it is added,
If the correction was made half a cycle ago, the subtracted or added welding current or welding voltage, and if the correction was not made half a cycle ago, the welding current or welding voltage detected in the previous weaving half cycle. Since this is a consumable electrode type arc welding method characterized by obtaining the next correction direction based on the difference between the welding current or welding voltage detected in the previous weaving half cycle, the conventional weaving control for each weaving cycle is not possible. The tracking width of the arc sensing is doubled compared to the previous model, dramatically improving the tracking accuracy and demonstrating high precision tracking even in arcuate weld line parts with small curvature.This reduces bead meandering and improves welding. The appearance of the section is improved. In addition, since the degree of correction for misalignment of the weld line is relaxed, the meandering of the weld bead is further reduced, and the appearance of the welded part is extremely good. Also, if a correction was made half a cycle ago, it will be corrected, so the accuracy
Responsiveness has also been improved.

[Brief explanation of drawings]

FIG. 1 is a diagram showing changes in welding current when an arc welding method according to an embodiment of the present invention is carried out;
Fig. 2 is a flowchart showing the processing procedure of the same embodiment, Fig. 3 is a side view of the entire robotic device that can be used directly to carry out the arc welding method according to the same embodiment, and Fig. 4 is a side view of the entire robotic device that can be used directly to implement the arc welding method according to the same embodiment. Fig. 5 is a side view of the welding part showing the welding operation state when the weaving center and weld line are aligned, and Fig. 6 shows the welding current in the state shown in Fig. 5. A line diagram showing the changes, Figure 7 is a diagram equivalent to Figure 5 showing a state in which the weaving center is displaced with respect to the weld line, and Figure 8 shows the change in welding current in the state of Figure 7 and conventional correction processing. A diagram showing the procedure, and FIG. 9 is a conceptual diagram for explaining the effects of the present invention. (Explanation of symbols), 3... Welding torch, 4... Welding wire, 7... Robot control panel, 8... Robot main body control section, 9... Copying control section, 10... Servo control section, 16... Position calculation section, 17... Weaving waveform calculation section, 18... Tracing input section, 19...
...patching calculation section, 20...correction timing generation section,
I, IL, IR...Welding current, ΔO...Difference current value, P
...Welding current curve, S _w ...Welding line, R ... Welding robot, C _l , C _r ... Half cycle.

Claims (1)

[Claims] 1 (a) A consumable electrode type welding torch is moved in the welding line direction while weaving, and the direction in which the difference between welding currents or welding voltages detected during the left and right half cycles of weaving disappears. A consumable electrode arc welding method in which welding is carried out by repeatedly correcting the center position of the weaving by a predetermined correction amount, and (b) if the correction has been made half a cycle before, the predetermined correction amount is Convert the converted welding current value or welding voltage value to the welding current value or welding voltage value corresponding to If they match, subtract; if in the opposite direction, add; A consumable electrode characterized in that the next correction direction is obtained based on the difference between the welding current or welding voltage detected in the previous weaving half cycle and the welding current or welding voltage detected in the immediately previous weaving half cycle. Formula arc welding method.