JPS58196169A - Method and apparatus for pulse arc welding - Google Patents

Abstract

PURPOSE:To enlarge the range of current that can perform welding with a consumable electrode of one kind, by decreasing the product of n-th power of pulse current value and pulse duration when the average value of welding current is set to a large value in pulse arc welding. CONSTITUTION:When base current value Ib, pulse current value Ip1, pulse duration Tp and pulse frequency f1=1/T are set as welding condition suitable for the first object to be welded, the average value Ia1 of welding current is shown by an equation I. Next, to obtain an average value Ia2 of welding current suitable for another object to be welded, pulse duration Tp is fixed, and the average value of welding current is enlarged by increasing pulse frequency from f1 to f2=1/T2. That is, by decreasing pulse currnt value Tp1-Tp2 shown by alternate long and two short dashes line simultaneously, the product of n-th power of pulse current value Ip and pulse duration Tp is decreased. The average value Ia2 of welding current at this time is given by an equation II.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a pulse arc welding method and apparatus in which a consumable electrode (hereinafter referred to as a wire) is fed at a preset substantially constant speed and brought into contact.

In pulsed arc welding, important parameters that affect mIgi transition are the instantaneous value during the pulse current application period (hereinafter referred to as the pulse current value) ip, and the pulse current application period (hereinafter referred to as the pulse current value), as shown in Figure 1. IP (referred to as pulse duration) and the reciprocal of the repetition period T of the pulse current (referred to as
Hereinafter, there is a pulse frequency) f. In order to ensure smooth droplet transfer, conventionally, as shown in Figure 2 (horizontal axis TP, vertical axis ip),
), (IP)·Tp is controlled to be a constant value.

Here, n is a value determined by the material and experiments, and is greater than 1 and smaller than 3. For example, n = 2.4 in the case of aluminum and stainless steel, and n = 2.4 in the case of mild steel. is Kl
= 2.8. That is, even in the current pulse arc bath welding method in which the welding current is controlled by a transistor, (ip)·Tp is kept constant over the entire welding current range. That is, as shown in FIG.
With ip and Tp set at constant values, the average value of welding current Ia - (IP, TP + Ib - (1" - 'rp)) /
r ·-ill is controlled. However, 1b is the instantaneous value during the base current energization period (hereinafter referred to as base current)
It is.

However, such (Ip)”−Tp = (constant value)
In a control method such that the average value of welding current is 1
The pulse current value 1p and pulse duration Tp are set so that optimal droplet transfer is performed when a is a small current, for example 100 (A).
and set these hα to a constant value lζ,
If the pulse frequency or base current value is increased and the average value Il of the welding current is set to the current value, for example, a o o (A), the arc force becomes too strong, causing the molten metal in the molten pool to fluctuate violently and melt. A puckering phenomenon occurs that makes the pond unstable, making undercuts and pores more likely to occur. Furthermore, when the arc length is shortened due to camera shake, there remains the problem that short circuits occur and spatter is likely to occur.

Therefore, in order to investigate this problem, we varied the average value 11 of the welding current and observed whether a "C smooth/j spray transition was performed.The average value 1a of the first welding current was set as a parameter (100, 200 300 (A)), the relationship between the pulse duration Tp and the pulse current value ip that can maintain the spray transition is as shown in Figure 3. In the figure, the horizontal axis shows the pulse duration Tp (ms). , the vertical axis represents the pulse current value 1p (A), and the average value of the welding current 1s-
It is a diagram showing the relationship between the pulse current value ip that allows spray transfer and the pulse duration 1゛P for each average value of the welding current, using too, zoo, and a o o (A) as parameters, Above each curve, at the average value 11 indicated by the parameter Ia of each curve,
This is within the range where spray transfer is possible. As is clear from the figure, at welding current I@= 100 [A], (Ip)・
Tp = s, (ip) −T for In-200(A)
For p = b and 1a = 300 (A) (ip) ・
'rp=c (however, -9b and C are each constant value), and by the average value 1a, (ip)・T
The value of p, i.e. λ.

As b and C change, the curve of (Ip)·'rP appears to be shifted in parallel. The average value 1a is 100°20
0 and 300 (A), the constant a.

The values of b and C are small. This means that Ia
= 300 (A) (7), then Im-100 (A)
or 200 (Ip)・Tp smaller than that of (A)
However, it shows that spray migration is possible. Furthermore, to explain in detail, the pulse duration TP is l, Q (ms)
When the average [1 = 100, 200 and 300 (
The curve A) intersects the straight line A indicating Tp = l, Q (m s) at points At, Am, and As. These intersection points have the same pulse duration Tp of l and Q (ms), but the pulse current value I that allows spray transition
p indicates that they are 450.425 and 400 [A], respectively. This means that the pulse duration is 1.
o (ms), the pulse current values that allow spray transition have an average value of 100, 200, and 300 (A).
This shows that the current values that can be transferred to spray are decreasing to 450, 425, and 400 (A), respectively. Next, in contrast to the case described above, in FIG.
Keeping it constant as (A), average value 1a = 100°20
0 and 300 (A) curves are 1p = 400 (A)
It intersects with the straight line B'' at points Bs, Bs, and Bs, respectively. These intersection points are the pulses il[[4d[
IP=400 (A), both are the same, but the pulse duration IP that allows spray transition is 1.5 in each case.
1.25 and l, Q (ms).

This means that when the pulse current value ip is 400 (A), the pulse duration that allows spray transition is an average value of 1
00, 200, and 300 (A), whereas the pulse durations that can be sprayed decreased by 1.5, 1.25, and 1.0 (ml), respectively. It shows.

Hereinafter, from the study results of E, the present invention has an average value of welding current of 1
The higher a or the wire feeding speed Vf is set, the product (Ip) of the n-th power of the pulse current value 1p and the pulse duration time Tp.
The present invention provides a pulse arc welding method for welding with reduced IP and an apparatus for carrying out the method.

First, the invention of sl will be explained with reference to FIGS. 4 to 7.

The wi4 diagram is! FIG. 3 is a waveform chart of welding current for explaining a first embodiment of the welding method of the invention of J1. In the same area, the welding conditions suitable for the first workpiece are set: base current value 1b, pulse current value 1pt, pulse duration.
When P': Ji and pulse frequency f I = l 7'TI are set, the average value of welding current is Ia + = [Ipt
−Tp+Ib(Tt−Tp)]/T+ −121. Next, in order to obtain the average value lag of the welding current suitable for other workpieces, as shown in the same figure (b), the base current value 1b and the pulse duration ■゛P are kept constant, and the pulse The average value of the welding current is increased by intentionally increasing the frequency from El to b=1/I''.In this case, in the present invention, the pulse current value shown by the two-dot chain line in FIG. By simultaneously decreasing 1p+ -1pt, the product (I
p) Decrease par-rP.

The average value of the welding current at this time is lag - [1pz - 'l'p + 1b (Tx - 'I''P
) ]/T2 −+31.

Next, FIG. 5 is a waveform diagram of welding current for explaining a second embodiment of the welding method of the invention of Part 1. In the same area, when the welding conditions suitable for the initial workpiece are set, that is, the base current value 1b, pulse 11 ffi 4di, pulse duration Tp1 and pulse frequency fl = 1/T+, the average welding current The value is Ias-(ip-
Tpl+1l)(Tl-"PI))/Tt...i4
It is 1. Next, in order to obtain an average value 111 of welding current suitable for other workpieces, the base current value 1b and the pulse 4 current value IP are kept constant as shown in FIG. From f! to increase the average value of the welding current. In this case, in the present invention, the pulse duration Tpt − indicated by the two-dot chain line in FIG.
By simultaneously decreasing Tpm, the pulse current value is reduced to 1p.
The product (ip)·TP of the pulse duration time TP and the first power of the pulse duration TP is decreased. The average value of the welding current at this time is 1az=(Ip'l'p!+Ib(Tx-'rp*))
/T's =151.

Next, FIG. 6 is a waveform diagram of welding current for explaining a third embodiment of the welding method of the first invention. Same figure (
) is the average value of the welding current when the welding conditions suitable for the first welded proboscis are set, that is, the base current value ib+, the pulse current value 'P', the pulse duration Tp1 and the pulse frequency i=1/T. is 1 set = (Ipl-Tp
t+Ibt(Tt-Tpt))/Tt-(6). Next, the average value l of welding current adapted to other workpieces is
In order to obtain ag, as shown in the figure (b), the pulse frequency f is kept constant and the base current value is increased from Ibl to lb to increase the average value of the welding current. In this case, in the present invention, by simultaneously decreasing the pulse current value 1p+ -Ipm, g and the pulse duration time Tp+ -'rpg shown by the two-dot chain line in FIG. The product (ip)·Tp with the pulse duration time TP is decreased.

The average value of the welding current at this time is 1az=(Ipt-"l'px+Ibm(T-Tpg)
)/T ・−・(71)

In Figs. 4 to 6 above, the average value of welding current I
Although the case where a is changed by changing either the pulse frequency f or the base current value Ib has been described, both of these may be changed at the same time. In addition, in Fig. ts4 and Fig. 5, when the welding current I@ is increased, the pulse current value ip or the pulse duration time Tp
We explained the case where one of the two is changed, but
As in the example shown in FIG. 6, both of these can be decreased at the same time, or one of them can be increased and the other can be decreased at a higher rate so that both M(Ip)n·Tp can be decreased. It may be changed. Furthermore, as an operation for increasing the average value 1a of the welding current and decreasing (Ip)・1゛P, both may be set individually, but (lp)・′1゛P is controlled. By mechanically or electrically interlocking the signal with a signal corresponding to the average value Iλ of the welding current, operation is simple and setting errors can be prevented. Furthermore, the rate of increase in the average value 1a of the welding current and the rate of decrease in (Ip)·'rP do not necessarily have to be changed proportionally, as shown in FIG. 7 or FIG.
It may be changed in a 18k shape or in a step shape. Also,
When No. 4M, which controls (Ip) and Tp, is moved with the average value i of the welding current as described above, the average value 1a of the welding current is detected to control (lp)' and 'rP. Although the signal may be changed, a signal corresponding to the pulse frequency 1 that sets the average value of the welding current, a signal corresponding to the base current value 1b, or a signal that is a combination of these may be used.

Next, the invention of the second welding method will be explained.

The wire feeding speed Vf and the average value 1a of welding current are
As shown in the figure, there is a proportional relationship. The second invention utilizes this relationship to generate a signal l corresponding to the output signal s1 or 811 of the wire feeding speed setting circuit 10 shown in FIG. 12, which will be described later.
This is compatible with Pulse II! What is necessary is to change the flow value 1p or the pulse duration time Tp. Therefore, the pulse current value t
p or pulse qs+ M time 1゛P is the 10th
As shown in the figure and FIG. 11, ip = Kl (1-cap・St)...-
(8)'rp = K2(1-dx-Ss)
-:9). Here 1. ．． 18
) In the formula, 1 and αl are appropriately values corresponding to the intersection of the vertical axis IP and the straight line connecting points At, At, and A1 on FIG. 17, and the slope of this straight line. Similarly, te, (shi)
) of the ceremony! Appropriate values for and α- correspond to the intersection of the vertical axis 'I'p with the curve connecting Bt, Bm, and Bl on Figure 118, and the slope of this straight line. In this case as well, the signal S! corresponds to the wire feeding speed! or Sl
Both the pulse current value 1p and the pulse duration Tp may be decreased at the same time as
It may also be possible to reduce the Also, a signal S! corresponding to the wire feeding speed Vf! Alternatively, Ss and the signal corresponding to the pulse current value ip or the pulse duration time 1゛P do not necessarily have to be changed proportionally, but rather in a curved manner or in a step-like -ζ change as shown in Fig. 10 or IJ11. You may let them.

Next, the decoration for carrying out the first welding method will be explained. FIG. 12 is a schematic configuration diagram of an apparatus for carrying out the invention of the first welding method, in which C11 is a base current supply power source, 2 is a pulse current supply source, and 3 is a consumable electrode (wire ), 4 is an arc, 5 is a wI welding object, 6 is a base current value regulator and a contact tip that supplies a pulse current 1ρ to the wire, 7 is a wire feed control circuit that controls the wire feed motor 8, 9 10 is a wire feeding speed setting circuit, 11 is a base current value setting circuit, 12 is a pulse current value setting circuit, and 13 is a pulse duration setting circuit. ,
14 is a pulse frequency setting circuit.

Circuits 12 to 14 constitute a pulse current control circuit.

Reference numeral 15 denotes a comparison circuit that supplies a signal representing the difference between the output signal Ef of the pulse current setting circuit and the arc electric ratio Ea to the pulse current supply power supply 1. 22 is the output signal sb of the base jet value setting circuit 11 or the output signal S of the pulse frequency setting circuit 14.
A function conversion is performed using f as an input, and a pulse current value correction signal Sp is obtained.
This is a pulse current value correction circuit that outputs. Reference numeral 23 denotes a pulse duration correction circuit which inputs the output signal sb of the base current value setting circuit 11 or the output signal St of the pulse frequency setting circuit 14, performs function conversion, and outputs a pulse duration correction signal S[.

Here, we will explain the operation of the fPi and the heel shown in FIG. 12. When the signal set by the base current value setting circuit 11 is supplied to the base flow supply power supply IK, the power supply j11 supplies the base current 1b to the contact tip 6, wire 3, arc 4, and workpiece 5. When each signal set by the pulse current value setting circuit 12, pulse duration setting circuit 13, and pulse frequency setting circuit 14 is supplied to the pulse current supply power supply 2,
The power supply 2 supplies a pulse current similar to the base current described above. In this case, since the pulse frequency is controlled by the signal of the difference between the output signal Kf of the pulse frequency setting circuit 14 and the arc voltage E&, even if both the pace electric @ and the pulse current have constant current characteristics, The arc length is maintained at a substantially constant value.

That is, when the arc length becomes large and the arc voltage Ea becomes large, the difference hf - Ea between the output signal Ef of the pulse frequency setting circuit 14 and the arc voltage Ea (c) becomes small, and the voltage from the pulse current supply power source 2 becomes small. By controlling the frequency of the output pulse current to be small, the melt transfer speed is reduced, and the arc length is controlled to be shortened, and recovery is performed.

When the inverse l arc length becomes shorter and the arc voltage Em becomes smaller, the pulse frequency C becomes larger, the droplet transfer speed increases, and the arc length is controlled to become longer and the return is made. The operation described above is the same as the conventional device, but ll'l! 55
The circuits connected to the double solid line, the dotted line, and the one-dot chain line shown in the figure are pulse current correction circuits newly provided in the device of the present invention.

In the first embodiment of the device of the present invention, as shown by double solid lines, the output signal Sf of the pulse frequency setting circuit 14 is functionally converted in the pulse current value correction circuit 22, and A pulse current value correction signal Sp is supplied. The simplest function transformation is '(gradient=5P=Ks(1-t0t+-8f)) However,
Constants are used for Ks and σ1. It is configured such that as the signal Sf becomes larger, the pulse current value correction signal Sp becomes smaller, and as the output signal of the pulse current value setting circuit 12 becomes smaller, the pulse current value tp decreases.

In the second embodiment of the device of the present invention, the binary dotted line No. 8f is input, the pulse duration correction circuit 23 performs function conversion, and the pulse duration correction signal S is supplied to the pulse duration setting circuit 13. do. As a simple example of function transformation, g(→= at = -class 4(1-αt-8f) However,
Constants are used for k4 and b. It is configured such that as the wire feeding speed setting signal St becomes larger, the output signal of the pulse duration correction circuit 23 becomes smaller and the pulse duration IP decreases.

A third embodiment of the device of the present invention has 2! As shown by the dashed line, the pulse current value correction circuit 22 receives the output signal sb of the base current value setting circuit 11 as an input.
The pulse current value correction signal Sp is supplied to the pulse current value setting circuit 12, and the circuit 12 performs a function conversion on the base 'I
@As the current value Ib increases, the pulse current value ip is decreased. Further, in the fourth embodiment of the present invention, the output signal sb of the base current value setting circuit 1411 is inputted, and as described above, the pulse duration correction circuit 23 performs a function conversion, and the pulse duration setting circuit 1411 performs function conversion. The circuit 13 supplies the pulse duration correction signal S to the circuit 13, and decreases the pulse duration TP as the base current value 1b increases.

In the device of the present invention, in addition to applying the first to third embodiments described above alone, two or more of the circuits described above may be applied in combination. In this case, in a single circuit, even if either one of the output signals Sp or S of the correction circuit increases with an increase in the base current value setting signal sb or the pulse frequency setting signal St, a combination of two or more In the whole circuit,
As the signal Sf or sb increases, (Ip)・'l
The signal S4 indicated by the two-dot chain line in FIG. , with a pulse frequency f
is controlled by the wire feed rate.

FIG. 13 is a schematic configuration diagram of an apparatus for carrying out the invention of the second welding method, and the configuration of each part is the same as that in FIG. The input signals to the current value correction circuit 22 and the pulse duration correction circuit 23 are similar to those shown in FIG. The first embodiment of the fourth invention is as shown by the double solid line in FIG.
With the output signal S2 of the wire feeding speed setting circuit 10 as input, the pulse current value correction circuit 't? ! I22
The pulse current value is converted into a function at , and the pulse current value is supplied to a circuit 12, which decreases the pulse current value ip as the wire feeding speed Vf increases. Further, in the second embodiment of the fourth invention, as shown by the double dotted line in FIG. The function is converted at , and supplied to the pulse duration setting circuit 13 .
As the wire feeding speed Vf increases, the pulse duration 1
'P is decreased.

In the device of the present invention, in addition to applying the first and second embodiments described above alone, the two circuits described above may be combined. In this case, in one circuit, as the pulse frequency 1 increases, either the output signal Sp or St of the correction circuit increases, and in the entire 62 combined circuits, as the signal St increases, (Ip)・TP only needs to decrease. In addition, in FIG. 13, the signal sb indicated by the dashed line is
This is a signal for unified control that determines the base 11112 value 1b in correspondence with the output signal sb of the wire feeding speed setting circuit 10. Further, a signal S4 indicated by a two-dot chain line is a signal for unified control that interrupts the pulse frequency f in accordance with the output signal S4 of the wire feeding speed setting circuit 10.

Furthermore, the base current value setting circuit 11 in the embodiment of FIGS. 12 and 13. Pulse current value setting circuit 12. Pulse duration setting circuit 13 and pulse frequency setting circuit 1
4 is a circuit with adjustable bending where all the circuits can set the conditions arbitrarily.No. In the device of the present invention, these non-adjustable circuits are also included in the setting circuit.

As described above, according to the present invention, the range of current that can be welded with one type of consumable electric current is significantly expanded, and stable droplet transfer is always performed in synchronization with the pulse over a wide range of current values. This has the effect of shortening the arc length and reducing the occurrence of undercuts or spatter. Furthermore, by centrally adjusting the average value 11 of the welding current or the wire feeding speed Vf and the pulse current value ip or the pulse duration Tp, the operation is simplified and setting errors are prevented.
Reliability and reproducibility can be improved.

[Brief explanation of the drawing]

Figures 1 and 2) are diagrams showing pulsed arc welding current waveforms when the average value of welding current is increased from 11 to lx using a conventional welding method, and Figure 2 is a diagram showing the horizontal axis 'I'p.・The vertical axis is 1p, and the pulse current value I
Figure 3 shows the characteristics of the conventional pulsed arc welding method, which is controlled so that the product of p to the nth power and pulse duration Tp (Ip)・1゛P is constant. The average value I3 of the parameter (100
, 20α and 300 [A)) A diagram showing the relationship between the pulse duration '1' p that can maintain spray transition and the /<7less current value 1p, Figures 4^ and ＜), Figure 5^ and伜) and 6th
Figures (A) and (C) show the average value of welding current l by @i to the third embodiment of the welding method of the present invention, respectively.
From a+! FIG. 7 is a diagram showing the waveform of the pulsed arc welding current when increased to 2, and shows the difference between the average value Il of the welding current and the pulsed current value ip obtained by the first embodiment of the welding method of the present invention. FIG. 8 is a diagram showing the relationship between the average value 1a of the welding current obtained by the $2 embodiment of the welding method of the present invention and the pulse duration Tp, and FIG. Average value 1a of substitute current and wire feeding speed Vf
Figure 10 is a diagram showing the relationship between the signal -82 corresponding to 1 wire feed speed of the device of the present invention and the pulse current value 1p, Figure @11 is the diagram showing the relationship between the pulse current value 1p and A diagram showing the relationship between the signal SS corresponding to the wire feeding speed of the device and the pulse duration Tp, FIG. 12 is a configuration diagram of a pulse arc welding device that implements the welding method of the first invention, and FIG. FIG. 2 is a configuration diagram of a pulse arc welding apparatus that implements the welding method of the second invention. DESCRIPTION OF SYMBOLS 1... Power supply for base current supply, 2... Power supply for pulse current supply, 10-... Wire feeding speed setting circuit, 11.
... Base current value setting circuit, 12... Pulse current value setting circuit, 13... Pulse duration setting circuit, 14...
- Pulse frequency setting circuit, 22... Pulse current value correction circuit, 23... Pulse duration correction circuit. Agent Patent Attorney Hiroshi Nakai 4th FuJni(A) Maru', 4I(B)

Claims (1)

[Claims] 1. In a pulsed arc welding method in which welding is performed by presetting a base value and a wire feeding speed, the larger the average value of welding current 1m is set, the higher the pulse current value 1p.
A pulsed arc welding method in which welding is performed using a pulsed current in which Tp is reduced to the nth power and the pulse duration 'rP. 2. The pulsed arc welding method according to claim 1, in which the welding method is performed using a pulsed current determined by interlocking the welding current (Ip)/TP in ms with the average value la of the welding current. 3.1 The pulsed arc welding method according to claim 1 or 2, in which the average value 1a of the welding current is determined by setting the pulse frequency f. 4. The pulsed arc welding method according to claim 1 or 2, wherein the average value 1a of the welding current is determined by setting a base current value ib. 5. The above III (I
The pulsed arc welding method according to claim 1 or 2, wherein the pulsed arc welding method uses a pulsed mill flow with reduced p).TP. 6. The above product (I
The pulsed arc welding method according to claim 1 or 2, wherein the pulsed arc welding method uses a pulsed current with reduced p)·Tp. 7. In the pulsed arc welding method in which the pace current value and wire feed speed are set in advance for welding, the larger the wire feed speed Vt is set, the greater the difference between the pulse mill current value 1p to the nth power and the pulse duration Tp. A pulsed arc welding method in which welding is performed using a pulse*m with a reduced product (ip)/Tp. & The pulse arc welding method according to claim 7, in which welding is performed using a pulse current value determined by interlocking the product (Ip)·Tp with the wire feeding speed Vf. - 9. The pulsed arc welding method according to claim 7 or 8, in which the pulse frequency f is determined by setting the wire feeding speed V (). 10. The wire The above product (ip
)・The pulsed arc welding method according to claim 7 or 8, in which the welding is performed using a pulsed current with reduced Tp. 11. By reducing the pulse duration Tp by a signal corresponding to the wire feed speed Vt, the product (
The pulsed arc welding method according to claim 7 or 8, wherein the pulsed arc welding method uses a pulsed current with reduced Ip) and Tp. 11 In a pulsed arc welding device equipped with a base current value setting circuit, a pulse current value setting circuit, a pulse duration setting circuit, a pulse frequency setting circuit, and a wire feeding speed setting circuit, the pulse frequency setting circuit or the base A pulse current value correction circuit that converts a signal corresponding to the output signal of the current value setting circuit and outputs it to the pulse current value setting circuit, or converts a signal corresponding to the pulse duration time to a function and supplies it to the pulse duration setting redundancy circuit. The pulse duration correction circuit is provided with at least one of a pulse duration correction circuit that calculates the sum of the pulse current value ip determined by the output signal of the correction circuit to the nth power and the pulse duration time Tp by the pulse frequency f or the base. Pulsed arc welding in which the current value I@ is corrected to decrease as the current value I@ is increased [11°13, The pulse frequency setting circuit inputs a signal corresponding to the output signal of the wire feed speed setting circuit and sets the pulse frequency. A pulse arc welding device according to claim 12, which outputs a signal corresponding to f. 14. In a pulse arc welding device equipped with a base current value setting circuit, a pulse current value setting circuit, a pulse duration setting circuit, a pulse frequency setting circuit, and a wire feeding speed setting circuit, the wire feeding f1 degree The signal corresponding to the output signal of the turf circuit is converted into a function and outputted to the pulse current value setting circuit.The signal corresponding to the output signal of the wire feeding speed setting circuit is converted into a function and the pulse duration value is output to the pulse current value setting circuit. At least one of the pulse duration correction circuits output to the setting circuit is provided, and the pulse current (the product of the n-th power of mlp and the pulse duration Tp (Ip)·Tp) is determined by the output signal base ζ of the correction circuit. , a pulsed arc welding device that performs a correction to increase and decrease the wire feeding speed Vf. 15. The pulse frequency setting circuit inputs the output No. 13 of the wire feeding speed setting circuit and changes it to the C pulse frequency f. A pulse arc welding device according to claim 14, which outputs a corresponding signal.