JPS6030580A - Method for correcting weaving amplitude of welding torch in weaving welding - Google Patents

Abstract

PURPOSE:To make an actual amplitude value always equal to a command value by calculating the actual amplitude value from the signal indicating the position of a torch with time during weaving and correcting the command value in accordance with the correction factor determined from the actual amplitude value and the command value for amplitude. CONSTITUTION:A command signal X1(t) indicating the position of a torch with time is formed in accordance with respective commands F, A, S for weaving frequency, amplitude and offset generated from a signal generating circuit 11 and a driving motor 17 is controlled to take the signal X2(t) indicating the actual position of the torch with time from a counter 20 into the sampling circuit 22 of an arithmetic circuit 21. The actual amplitude value (a) of the torch is thus calculated and a correction factor K(=A/a) is calculated from the value (a) and a preset amplitude command value A, by which the corrected amplitude command value A<2>/a is obtd. The command signal is corrected by said value. The dependency of the actual amplitude value of the torch on the frequency and the variance in the characteristic by each torch are thus eliminated.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a welding torch position control system for a welding robot, and more particularly to a welding amplitude correction system for a welding torch in weaving welding.

The 5-pin welding method in which welding is performed while swinging the welding torch almost perpendicular to the welding direction is called "weaving welding," but when this 4-pin welding is performed using a welding robot. The welding arm has a configuration as shown in FIG. 1, for example. That is, a sub-arm a that can be rotated around a vertical axis 2 at the tip of the main arm 1, a swing arm 5 that swings around a horizontal axis 4 at the top of the sub-arm 3, and a swing arm 5. It consists of a welding torch 6 fixed to the tip of the welding torch. The swing arm 5 is connected to and driven by a motor (not shown) via a conduit cable (not shown).

FIG. 2 shows the state in which the torch 6 is oscillated, and is oscillated on both sides of the center line 7 with the same amplitude A.

An example of a control circuit for the torch 6 in such weaving welding is shown in a block diagram in FIG. That is, 11
1 is a torch position command signal generation circuit, and a frequency setting circuit 12 generates a signal to set the weaving frequency command value F.
It consists of an amplitude setting circuit 3 that generates a signal that generates the ping amplitude command value A, and an offset setting circuit 14 that generates a signal that generates the offset command value B, and these three command values F, Based on A and S, a command signal χ,(t) is generated which determines the temporal position of the torch 60. This command signal χ, (1) is generally expressed by the following equation.

χ, (t) = A blood ωt (1) The command signal χ, (t) generated by the above equation (1) is D/A
The signal is applied to a torch driving motor 17 via a converter 15 and an amplifier 16 to control the motor 17. The rotation of the motor 17 is detected by the tacho generator 18 and fed back to the input side of the amplifier 16, and also detected by the encoder 19 and digitized, and sent to the input side of the D/A converter 15 via the up/down counter 20. Give feedback.

This stabilizes the motor's 170 rotations. Output χ of counter 20! (t) is a signal indicating the 60-time position of the torch. That is, χ, (t) = a blood (ω is one θ)... (2
) where a is the actual amplitude value and θ is the phase delay.

Figure 4 shows the signal χ shown in equations (1) and (2) above.
1(t) and χ,(t) are waveform diagrams.

By the way, the conventional torch position control method described using the control circuit shown in FIG. 3 has the following drawbacks. That is, the actual amplitude value a of the torch 6 during welding has frequency dependence with respect to the same wave number of weaving, and the frequency dependence differs from torch to torch.

FIG. 5 shows the frequency dependence of the actual amplitude value a, and FIG. 6 is a graph showing the amplitude command value versus the actual amplitude value at the frequency f (Hz) of FIG.

For this reason, it was necessary to measure the frequency characteristics of each torch and adjust the amplitude command value Afr for each torch, and when the torch was replaced, readjustment was required.

Furthermore, since the load fluctuates due to aging of the conduit cable that drives one moke, there is also a drawback that the actual amplitude value a is obtained as being different from the amplitude command value A.

Therefore, the present invention provides the frequency dependence of the above-mentioned actual amplitude value a,
To provide a new amplitude correction method that automatically corrects the amplitude so that the actual amplitude value a is always equal to the amplitude command value AK even if there are variations in characteristics among torches, load fluctuations, etc. The purpose is to

The present invention samples a signal representing the actual temporal position of the torch during weaving welding at a predetermined sampling period, obtains a value representing the position of the torch in this sampling period, and uses this value to obtain a value representing the actual temporal position of the torch. The above purpose is achieved by calculating the value a'4, calculating a correction coefficient from the calculated actual amplitude value a and the preset amplitude command value A, and correcting the amplitude command value A by this correction coefficient. achieved.

The amplitude correction method according to the present invention will be explained in detail below.

FIG. 7 is a block diagram showing a control circuit to which the amplitude correction method according to the present invention is applied, and corresponding parts to those in FIG. , the torch position command signal generation circuit ii
However, it differs from the control circuit shown in FIG. 3 in that it includes an arithmetic circuit 21. This arithmetic circuit 21 includes a sampling circuit 22,
Actual amplitude calculation circuit 23, division circuit 24, and multiplication circuit 25
It becomes more. As in the case of FIG. 3, the signal χ! representing the actual temporal position of the torch is output from the counter 20 as shown in equation (2) above. (t) is fed back to the input side of the D/A converter 15, and at the same time is taken into the sampling circuit 22 and sampled.

The sampling circuit 22 receives the sampling clock signal 0.
is applied, and its sampling period T is set from the sampling same wave number ω (angular frequency) to π/2ω. As is clear from FIG. 4 mentioned above, if the starting point of one sample synchronization Tπ is, for example, t=6, the ending point is t= '2az
Therefore, in this case, the positions of the torch at the beginning and end of the sampling period T are as follows from equation (2): χt (0) = -athθ (3)
The values of equations (3) and (4) are input to the actual amplitude calculation circuit 23, and the following calculations are performed.

The actual amplitude value a calculated in this way and the amplitude setting circuit 1
The amplitude command value A preset in step 3 is applied to the division circuit 24, and the efficiency β (-1) and the correction coefficient K (=-) are calculated. The correction factor 11x, the amplitude command value A, and the width command value output cultin of the multiplier circuit 25IC are shown in FIG.

In this case, the correction coefficient does not need to be corrected every time the command value output routine is executed.

The above explanation has clarified an example of the amplitude correction method of the present invention, but according to the present invention, it is necessary to measure the frequency characteristics of each torch in advance and adjust the amplitude command value even if the weaving frequency is different. Furthermore, there is no need to re-adjust the aI4 when replacing the torch (this makes maintenance of the welding robot easier.Furthermore, according to the method of the present invention, the #i radiation command value m and the actual amplitude value a) Even if there is a phase difference θ between
It uses a calculation method that does not affect the correction accuracy, so it can perform amplitude correction with high accuracy.
can. Furthermore, the influence of load fluctuations due to age-related changes in conduits and cables can be removed, and the actual amplitude value a can be made equal to the normal pressure amplitude command value A.

In the above-mentioned embodiments, the weaving waveform is a sine wave, but the method of the present invention is not limited to a sine wave. For example, the weaving waveform is not limited to a sine wave. It can also be applied to

In this case, the arithmetic expression for calculating the actual amplitude value a in the circuit 23 is K as shown below.

The processing flow in this case is the same as in the case of a sine wave.

[Brief explanation of drawings]

Figure 1 is a perspective view of a welding arm for weaving welding, Figure 2 is an explanatory diagram showing the swinging state of the torch, Figure 3 is a block diagram of a conventional control circuit, and Figure 4 shows that the weaving waveform is sinusoidal. Waveform diagram of the command signal and the signal indicating the temporal position of the torch in the case of a wave, Figure 5 is weaving #l1
lt11. 6 is a graph showing the relationship between the amplitude command value and the actual amplitude value, FIG. 7 is a block diagram of a control circuit to which the method of the present invention is applied, and FIG. 8 is a graph showing the relationship between the amplitude command value and the actual amplitude value. Correction coefficient determination chart for the method, No. 9
The figure shows the same amplitude command value output routine, and Figure 1.theta. is a waveform diagram of the command signal and the signal for determining the temporal position of the torch when the weaving waveform is a triangular wave. DESCRIPTION OF SYMBOLS 11... Command signal generation circuit, 12... Frequency setting circuit, 13... Amplitude setting circuit, 14... Offset setting circuit, 15... D/h converter, 16... Amplifier,
17-...Motor, 18...Tacho generator, Engineering 9
... Encoder, 20... Up/down counter, 21... Arithmetic circuit, 22... Sampling circuit,
23... Meiji width value calculation circuit, 24... Division circuit, 2
5...Multiplication circuit Fig. 1 Fig. 2 Fig. 3 Fig. 2 Fig. 5 Fig. 6

Claims (1)

[Claims] A command signal for changing the temporal position of a welding torch is formed based on a weaving station wave number command value and a weaving amplitude command value set in advance, and the torch is oscillated by this command signal. When performing weaving welding, a signal that detects the temporal position of the torch detected during weaving welding is sampled at a sampling period selected based on the weaving frequency, and the position of the torch in this sampling period is determined. Obtain the hail value, calculate the actual amplitude value of the torch from the value obtained by this sampling, calculate a correction coefficient from this actual amplitude value, the preset amplitude command value, and the trap, and use this correction coefficient. A weaving amplitude correction method for a welding torch in weaving welding, the method comprising correcting the amplitude command value.