Weld- Wire-Filler Identification Apparatus and Method

The present invention relates to a filler wire identification apparatus according to Claim 1.

The invention also relates to a method for identifying a filler wire.

According to the prior art, in MIG and MAG welding, the material of the filler wire affects the control parameters, the welding current and the welding voltage used in the welding. According to the prior art, the parameters relating to the type and diameter of the filler wire have been entered in the welding device manually, for example on the basis of data in the welding- wire package. The information may be in the package in either verbal or code form.

The manual entry of data is difficult and an inexperienced operator can make mistakes when entering filler wire information. Erroneous information can lead to quality problems in the final weld.

The present invention is intended to eliminate the defects in the prior art and create an entirely new type of apparatus and method for identifying the type of filler wire.

The invention is based on measuring the properties of the filler wire electrically, so that the welding system receives accurate information on the filler wire being used.

More specifically, the apparatus according to the invention is characterized by what is stated in the characterizing portion of Claim 1.

For its part, the method according to the invention is characterized by what is stated in Claim 9.

Considerable advantages are gained with the aid of the invention.

The type of the filler wire can be identified reliably and errors made by the operator can

be eliminated. As a result, the quality of the weld will improve.

In the following, the invention is examined with the aid of examples of applications of an embodiment according to the accompanying drawings.

Figure 1 shows schematically one welding-device assembly, which can be applied to the solution according to the invention.

Figure 2 shows a side view of one wire-identification apparatus according to the invention.

Figure 3 shows schematically one electrical circuit for use with the wire-identification apparatus according to Figure 2.

Figure 4 shows schematically a second electrical circuit for use with the wire- identification apparatus of Figure 2.

Figure 5 shows a schematic circuit scheme of one possible way to implement the exciter- generator in the solution according to Figure 3 or 4.

Figure 6 shows a schematic circuit scheme of one possible way to implement the current generator hi the solution according to Figure 4.

Figure 7 shows a supplemented system scheme of the solution according to Figure 4.

In the description of the invention, the following terminology will be used:

1 identification coil

2 filler wire 3 exciter-generator

4 ballast resistor

5 amplifier

6 detector

7 adjustable comparator

8 reference voltage control

9 interpretation logic

10 logic output

11 reference coil

12 first power supply

13 second power supply

14 multiplier

15 crystal

16 pull-up resistor

17 transistor

18 resistor

19 capacitor

20 current output

21 transistor

22 input

23 output

24 resistor

25 pull-up resistor

26 processor

27 information output

28 multiplier output

30 welding power supply

31 wire-feed device

32 cable

33 gun

34 workpiece

According to Figure 1, the energy required in MIG and MAG welding methods is obtained electrically from an arc. The arc is created between the filler wire 2 and the workpiece 34 by means of a welding power supply 30, which typically also includes a wire-feed device 31. Both the necessary electrical energy and the filler wire 2 are fed to the welding gun 33 through a cable 32.

In MIG/MAG welding, the arc bums between the melting filler wire and the workpiece. During welding, the filler wire 2 melts onto the workpiece.

According to the invention, the filler wire 2 is arranged to travel through a cylindrical identification coil 1 according to Figure 2, before it is fed to the welding nozzle. The identification coil 1 is a normal electro-technical coil, in which one or more loops of an electrical conductor are formed around the same axis. The coil thus typically forms a cylinder, on top of which there are one or more layers of a spiral, continuous conductor. According to the invention, the filler wire 2 is arranged to travel through this cylindrical structure. This can be implemented, for example, inside the wire-feed device 31 of Figure 1, around the cable 32, or in the welding gun 33. The filler wire 2 will then affect the inductance of the coil 1 in such a way that the ferromagnetic substances in the wire 2 increase the inductance of the coil 1 and the non-magnetic substances of the filler wire 2 correspondingly reduce the inductance. In addition to this, the diameter of the filler wire 2 affects the inductance in such a way that the effect is approximately proportional to the cross-sectional area of the filler wire 2, so that the diameter can be identified quite well. The appropriate measurement frequency depends on the material of the wire, in such a way that with ferromagnetic materials a relatively low frequency (10 - 20 kHz) can be used, but with non-magnetic materials, such as aluminium, the appropriate frequency is much higher, being preferably as much as several megahertz.

Figure 3 shows the principle of the construction of a pilot device. The alternating excitation current (AC) obtained from the exciter-generator 3 is taken via a resistor 4 to an identification coil 1, the voltage acting over which depends on the kind of filler wire 2 running through the coil 1. The said voltage is amplified by an amplifier 5 and detected by a detector 6. The detected signal is taken to window comparators 7, the control resistors 8 of which are each set to a threshold value corresponding to the filler wire 2. The information coming from the window comparators 7 is interpreted by interpretation logic 9, from which information corresponding to each filler wire 2 is obtained.

If better discrimination is required, it will then be advantageous to make a bridge

measurement according to Figure 4, in which case the discrimination will improve substantially. The measuring principle as such remains as before, i.e. the measurement is still based on the change in the impedance of the coil 1 caused by the filler wire 2 travelling through the identification coil 1. If the measurement is made at several different frequencies, a significant improvement in discrimination will be possible. Figure 4 shows the construction principle of the bridge measurement. In this case, the bridge is formed by the actual identification measurement coil 1, a reference coil 11 that is of the same kind as the identification coil 1, inside which there is not, however a wire, and two mutually identical current generators 12 and 13, which feed the coils 1 and 11. The current generators 12 and 13 are controlled using the alternating electrical signal obtained from the exciter-generator 3, the frequency of which can vary from 1 kHz to 100 MHz.

The alternating current developed by the current generators 12 and 13 is led to the coils 1 and 11, over which a voltage is induced, which depends not only on the current

(frequency and amplitude), but also on the impedance of the coils 1 and 11. The signals obtained from the coils 1 and 11 are taken to a multiplier 14, which multiplies the difference between the signals of the coils 1 and 11 by the exciter signal. The device is a synchronous detector, well known from radio technology. As is known, the output signal 28 of the said multiplier 14 depends on both the amplitude and the phase of the input signals. Thus, because wire 2 travelling through the identification coil 1 influences the impedance in a manner that depends on both the material and the dimension of the wire 2, it is possible to identify the filler wire 2. For the identification to be sufficiently reliable, the identification must be made at several different frequencies, unless previous information on the wire 2 exists. If limiting conditions (e.g., the material of the wire) are available, the dimension can be identified using only a single frequency.

A crystal oscillator according to Figure 5, for example, the frequency of which can be changed by changing the crystal 15, can be used as the exciter-generator 3. The exciter- generator can thus be implemented with the aid of a crystal 15, a transistor 17 and resistors 16, 18 and capacitors 19 connected to it. The oscillation frequency can be changed electronically, for example with the aid of a semiconductor switch. Alternatively, a crystal oscillator and a frequency limiter connected after it, from which

signals of different frequency can be obtained, can also be used as the exciter-generator. A microprocessor 26 or hardwired logic according to Figure 8 can be used in a known manner for controlling the frequency change and for analysing the output signal 28 of the multiplier 14. The analysed signal is obtained as the output 27.

If the measurement must be made over a large frequency range, a good alternative to an exciter-generator is the use of direct frequency synthesis.

For its part, the current generator 12 and 13 can be implemented, for example, in the well-known manner shown by Figure 6, with the aid of a transistor 21 and resistors 25 and 24 connected to it, in which case the control signal is fed to the input 22 and the output signal is obtained from the output 23.

The multiplier 14 can be, for example, a Gilbert cell, such as are used widely, for example, in general radio receivers. The multiplication calculation can also be made digitally, in such way that samples are taken of the voltages of the coils 1 and 11 at at least the frequency required by the sampling theorem and the samples are converted by an AD converter into digital signals, which are multiplied by software.

In the present application, the term electrical properties of a filler wire typically refers to the identification of the type of filler wire. The term type of filler wire refers in turn usually to the dimension of the wire and its material (raw material).

According to the invention, the measurement of the filler wire can take place, for example, by frequency scanning over the desired frequency range, or alternatively by using predefined point-type measuring frequencies. There can be, for example, three point-type measuring frequencies.

Low frequencies are well suited to the identification of ferromagnetic filler wires, whereas higher frequencies are well suited to the identification of non-magnetic or slightly magnetic materials such as aluminium and certain stainless and acid-resistant steels.

Inductive measurement can be implemented using either a single measuring coil, or alternatively in such a way that each frequency range has its own coil. A combination of the aforementioned solutions is also possible according to the invention.

The invention can also be used for detecting the ending of the wire, in which case the apparatus will give an alarm if the wire ends.