Electronic welding current generator for pulsed-arc welding.

Technical field:

The invention relates to an electronic welding current generator according to the preamble of patent claim 1.

State of the art:

The real arc voltage cannot be directly recorded as actual value for the purpose of regulating the pulse voltage, since electrical tapping is not possible on the arc. In most cases, the actual voltage value is therefore tapped at the output terminals of the welding power source.

The setpoint value Up of the pulse voltage is set taking into account the other relevant parameters until the welding process runs optimally. The effective arc voltage is less than the regulated generator output voltage. The difference results from the voltage drop at the welding cables and the voltage drop at the contact nozzle.

These voltage drops increase over time due to an increase in the resistance of the heating welding cables and the wearing contact nozzle. The voltage available on the arc drops accordingly, the arc becomes shorter and the spatter frequency increases up to the point where the process is completely derailed.

By observing the arc length and correspondingly increasing the setpoint potentiometer of the pulse voltage Up, the tendency to reduce the arc length can be corrected manually and thus had the desired constant. However, this method overwhelms the operating personnel and is prohibited by itself due to cost reasons and a lack of quality assurance.

For this reason, experience is used in practice. After a determined number of welding cycles, the contact nozzle is changed as a matter of precaution. Although this method is practicable, it leads to high nozzle consumption and relatively long downtimes and thus to avoidable high costs.

Since this time-increasing, harmful voltage drop at the contact nozzle and the warming welding cables is only significant in the pulse voltage phase, i.e. occurs during the pulse phase tp, it would be obvious instead of the Up-I_G-Modulation the principle of Ip-I_G-Modulation to apply, since constant currents at the process resistor produce constant voltage drops. However, here the so-called "internal regulation" of the Up-I_G-Modulation is not available, the process stability must be enforced using the pulse time control (tp control). This method undoubtedly increases the nozzle service life, but has been shown to lead to poorer welding qualities.

Technical task:

The invention is based on the object of immediately recognizing and effectively suppressing or regulating a set length in an electronic welding current generator of the type mentioned at the beginning of tendencies to shorten or extend the arc.

Presentation of the invention and its advantages:

The object is achieved by an electronic welding current generator according to the following features of claim 1:<ul id="ul0001" list-style="none"><li>a) starting from the Up-I_G-Modulation is the pulse voltage (Up) during the pulse phases (tp) and during the basic phases (t_G) the basic current (I_G) regulated</li><li>b) during the basic phases (t_G) the basic voltage (U_G) measured, the measured value (E) saved and updated after each pulse period (T)</li><li>c) the updated measured value (E) of the basic voltage (U_G) is obtained in direct proportionality to the real effective length (LL) of the arc</li><li>d) the updated measured value (E) serves to control the welding process by comparing the measured value (E), which represents the length (LL) of the arc, with an adjustable target value (SWLL) for the length of the arc. The result of the updated measured value (E) and the setpoint (SWLL, SWU_u) resulting control deviation (± X_W) is given to a downstream PI controller and amplified to the manipulated variable (± Y), which is based on the process parameters such as setpoints (Up, tp, t_G) is able to act incrementally or decrementingly, the degree of intervention (control factor) being adjustable or rigidly determinable. Further advantageous embodiments of the invention are characterized in the subclaims.</li></ul>

The invention has the salient advantage that the arc length control in pulsed arc welding increases the process stability in both directions, the contact nozzle consumption is considerably reduced and the downtimes are reduced. In a practical test arrangement, the contact nozzle service life could be increased to 5 times the welding cycles when using the invention. The arc length also remains stable during manual welding, such as swinging in a fillet weld; in addition, the frequency of spatter is greatly reduced. Furthermore, the influence of the increase in welding cable resistance when heating the welding cable is now without influence on the welding process.

The essence of the invention lies in the acquisition of a measured value which is directly proportional to the real arc length, the principle of pulse technology with Up-I being advantageous_GModulation is used. The pulse voltage Up is regulated during the pulse phase tp, during the basic phase t_G however, the basic current I_G. During the basic phase t_G time-selectively the generator voltage is measured, stored and updated after each pulse period, a measured value proportional to the arc length is obtained in this way, which is excellently available for a higher-level control that maintains the arc length.

This actual "arc length" value is compared with an adjustable "arc length" setpoint and the resulting control deviation ± X_W amplified by a downstream PI controller to the manipulated variable ± Y. The setpoint follows the relationship U_LL = Uo ⁺ m • basement; mean:<ul id="ul0002" list-style="none"><li>U_LL = Setpoint arc length</li><li>Uo = basic voltage (adjustable e.g. via potentiometer between 13-20V)</li><li>m = inclination factor with adjuster, adjustable (e.g. on circuit board) from 0.02-0.05</li></ul>

U_IG = Control voltage for the base current I_G In this way, a characteristic curve that takes into account all process influences can be set in an advantageous manner, which can be obtained similarly to the rigid VDE characteristic curve.

With the controller output variable ± Y, the process parameters required in pulse technology, such as setpoints Up, tp or t, can be varied_G, incrementing or decrementing intervention, the degree of intervention, the control factor, can be made adjustable or rigid.

The following types of rules are intended to illustrate the principle according to the invention:<ul id="ul0003" list-style="none"><li>Rule type "pulse voltage increase with tendency to arc shortening":</li><li>Effective setpoint Up = setpoint Up + ΔUp, where:</li><li>Up = α • (+ Y); the value -Y is suppressed a = degree of intervention (control factor)</li><li>Rule type "pulse voltage increase and pulse time increase with tendency to arc shortening":</li><li>Effective setpoint Up = set setpoint Up + ΔU_p</li></ul>

Effective setpoint U_tp = setpoint U_tp + DU_tp; where mean:<ul id="ul0004" list-style="none"><li>ΔUp = α • (+ Y); the value-Y is suppressed<maths id="math0001" num=""><img file="EP0520439A2_D0001.tif" /></maths></li><li>The two types of control mentioned are particularly suitable for automatic welding and in particular for increasing the contact nozzle service life and eliminating the influences of the supply cables.</li><li>Rule type "increase in pulse voltage when there is a tendency to shorten the arc and increase in base time when there is a tendency to extend the arc"</li><li>Effective setpoint Up = set setpoint Up + ΔU_p</li><li>Effective setpoint U_tG = setpoint U_tG⁺ Automobile; where mean:<ul id="ul0005" list-style="none"><li>ΔUp = α • (+ Y); the value - Y is suppressed</li><li>AUto = a2 '(- Y); the value + Y is suppressed.</li></ul></li></ul>

This up-t_G-Regulation type is particularly suitable for manual welding.

Depending on the application, one of the three aforementioned control types can be specified or switched on. The prerequisite for this is an optimally set welding process, with an existing control switch in the adjustment position. The setpoint of the voltage "arc length" SWU_LLis set while the process is running until the manipulated variable ± Y = 0. For this purpose, the manipulated variable can be switched to a differential display of any type, for example to a zero point instrument or LED line display or the like, in order to recognize positive and negative deviations from zero. In this adjustment position, the PI controller only acts as a P controller with a defined gain and thus functions as a display amplifier. After the adjustment has been made, you can then switch to the desired control type; afterwards the process parameters should no longer be adjusted. If this is necessary, for example, for another welding task, a one-off readjustment is necessary first. This adjustment can also be made automatically (automatic zero point adjustment).

A control value can be assigned to the control manipulated variable ± Y, which signals to the user when the increased service life has been reached. The limit value detector preferably receives the arithmetic mean value of the manipulated variable as an input variable, thereby suppressing rapid manipulated variable fluctuations.

The limit value detector is preferably blocked in the adjustment position and during the ignition phase of a welding process in order to avoid false messages; it is thus released with a delay. The limit value can be set with an assigned potentiometer for setting the limit value voltage U_Gw can be set.

Brief description of the drawing, showing:<ul id="ul0006" list-style="none"><li>Figure 1 is a block diagram of the electronic welding current generator</li><li>Figure 2 the detection of the basic voltage by sample and hold and</li><li>Figure 3 shows the relationship between the control factor, the effective voltage setpoint Up of the pulse phase t_G at the Up-t_G-Regulation.</li></ul>

Preferred embodiment of the invention:

A period T of the welding voltage 1 is shown in the block diagram in FIG. 1 of the electronic welding current generator. In pulse mode, a higher generator pulse voltage Up is generated during the pulse phases tp and t during the basic phases_G a lower generator base voltage U_G generated, preferably the basic phases t_G take longer than the pulse phases tp. The welding voltage signal is given to an adaptation network 2, if necessary for adaptation of the measured value and / or for standardization and possibly for electrical isolation. The output signal of the network 2 is used for time-selective detection of the straight branch of the basic voltage U_G the welding voltage signal fed to a sample-and-hold circuit 3; at the same time, the output signal from the network 2 is parallel to a module 4 for the selection and determination of the basic phase t_G given up. This module 4 serves to delay the sample signal when the measuring voltage drops, but to switch it off immediately when the measuring voltage increases, so that in fact only the horizontal branch of the basic voltage U_G is detected. This measured value E obtained in this way is directly proportional to the real-acting length LL of the arc. The measured value E is updated by every period T of the welding voltage, so that an updated measured value E proportional to the arc length LL is always available at the output of the sample-and-hold circuit 3; this measured value E thus gives the actual value of the basic voltage U_G again, which is directly proportional to the real-looking arc length LL.

Now this measured value E (actual value U_G) with an adjustable setpoint for the arc length SWLL, which is preferably a voltage value SWU_LL is, the setpoint SWU_LL is obtained from a block 6. The setpoint of the voltage SWU_LL follows the relationship: SWU_LL = Uo + m UG, the inclination factor m being adjustable by means of a potentiometer 18 which is connected to the supply voltage U_IG of the base current is multiplied. The basic voltage U o is set via a potentiometer 19 of the module 6. In this way, the control deviation ± X is obtained at the input of the PI controller 5, which is fed to the PI controller 5.

This equation can be used to set a characteristic curve that takes all process influences into account, similar to the rigid VDE characteristic curve. The control deviation ± X_w is amplified by the PI controller 5 to the manipulated variable ± Y, which represents the control output variable. With the control output variable ± Y, the process parameters required in pulse technology, such as setpoints, pulse voltage Up, pulse phase tp or basic phase t, can now be varied in many ways_G incrementing or decrementing intervention. The degree of intervention, that is the control factor a, can be designed to be adjustable or rigid.

In the block diagram of Figure 1, the manipulated variable ± Y is given in parallel through a diode 7 in the forward direction to a switch 9 for the control type "pulse voltage increase with a tendency to arc shortening", that is, with decreasing arc length, the pulse voltage is increased until the original Arc length is reached again. This type of control optimizes the welding process in particular.

At the same time, the manipulated variable ± Y is given via a diode 7 'which is polarized in the forward direction to a switch 9' for the control mode "pulse voltage increase and pulse time increase with a tendency to arc shortening", that is, with decreasing arc length, the pulse voltage and the pulse width are increased until the original Arc length is reached again.

At the same time, the manipulated variable ± Y is applied to a switch 9 "via a diode 8 switched in the reverse direction in order to generate the control type" pulse voltage increase with a tendency to shorten the arc and increase in the base time with a tendency to extend the arc ", ie the control type serves to increase the pulse voltage when the arc length decreases and to increase the base time when the arc length increases Arc voltage and is mainly used for manual welding.

The switches 9, 9 ', 9 "are used to set or select the control type. A potentiometer 10, 10', 10" is arranged in front of each module 11, 11 ', 11 "in order to control factor a or a1 or to set a2. The modules 11, 11 'and 11 "consist of active adding networks in which the corresponding setpoint value SW, which is preferably a voltage, namely one of the setpoints: SWUp = Up +, is generated by adding the corresponding reference voltage plus the differential of this reference voltage dUp for block 11 SWU_tp = U_tp ⁺ you_tp for block 11 'or SWU_tG = U_tG⁺ you_tG for module 11 ". This respective setpoint can now be used as a control variable in a suitable power unit.

The manipulated variable ± Y is given in the same way, preferably via an RC network 15, for averaging to a limit indicator 12 and, in parallel, a module 16 for adapting the display, the output signal of which is given to a display instrument 17 which is capable of displaying a ± zero deviation. The limit value detector 12 signals to the user when the increased service life has been reached due to the higher-level regulation of the control types mentioned. The arithmetic averaging of the manipulated variable suppresses rapid manipulated variable swings. The limit value detector 12 is blocked by means of a switch 14 in the adjustment position and during the ignition phase of a welding process in order to avoid incorrect messages; it is released with a delay. The limit voltage U_Gw can be specified to the limit indicator 12 via a potentiometer 13.

FIG. 2 shows the detection of the basic voltage by the sample-and-hold circuit 3. The sample and hold signal is only during the basic phase t_G applied to the measuring voltage if the basic voltage U_G is reached, ie only the horizontal branch of the measuring voltage is detected in this way. The basic voltage is stored and is available in the hold phase as a preferably positive actual value signal for the subsequent setpoint / actual value comparison.

The setpoint / actual value comparison takes place at the input of the PI controller 5, whose PI behavior can be predetermined in a changeable manner.

Figure 3 shows the relationship between the control factor and the basic phase t_G as well as the control voltage U_tG. It can be seen that at small t_G-Times the controller intervention is weakened because the so-called "internal control" is more effective, ie the tp phase is relatively large compared to the period duration T. At large t_G-Times, starting at around 15 ms, the controller intervention becomes degressive so that the pulse frequency does not become too low.

List of reference numerals:<tables id="tabl0001" num="0001"><img file="EP0520439A2_D0002.tif" /></tables><tables id="tabl0002" num="0002"><img file="EP0520439A2_D0003.tif" /></tables>