JP2013099773A - In-line test method, in-line test device and plasma-mig welding method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an in-line test method and an in-line test device with which the in-line inspection can be performed whether or not a through hole capable of realizing the desired strength is formed.SOLUTION: In the in-line inspection method, penetrated holes are subjected to the in-line inspection when executing the penetration-welding of a plurality of superposed workpieces W. The in-line inspection device 10 includes a storage means 12 for storing the calibration curve data indicating the relationship between the lower hole diameter of the through hole formed in the workpiece and the arc voltage of the plasma arc welding for each thickness and each welding condition of the workpieces, and a processing means 13. The processing means 13 performs a penetration detection step S1 of monitoring the arc voltage during the plasma arc welding, and detecting on the basis of the arc voltage whether the plurality of superposed workpieces are penetrated, and a penetrated hole diameter detection step S2 of detecting the prepared hole diameter of the penetrated hole by monitoring the arc voltage, and collating the arc voltage measured after the elapse of the prescribed time from the workpiece penetration with the calibration curve data.

Description

  The present invention relates to an in-line inspection method and an in-line inspection apparatus for inspecting welding quality, and more particularly to an in-line inspection method, an in-line inspection apparatus, and a plasma-MIG welding method for inspecting through holes in through welding.

  Conventionally, a technique in which two upper and lower steel plates are overlapped and welded through is known (see, for example, Patent Document 1). The procedure of through welding described in Patent Document 1 is as follows. That is, first, a plasma arc is ignited at a position where welding is performed, and then a hole is penetrated using a plasma arc having a relatively large power. As soon as the hole penetrates, the power of the plasma arc is reduced to a value suitable for welding, and the upper and lower steel plates are appropriately melted and welded. At this time, the filler is fed into the plasma arc. Then, the filler melted by the arc heat is supplied to the welding spot, the hole is finally backfilled, and the gap between the upper and lower steel plates is also filled to establish welding between the two steel plates. At this stage, the filler is pulled out of the plasma arc, the feeding is stopped, and then the plasma arc is extinguished. This completes one welding.

Japanese Patent No. 3379965

  However, in through welding, the diameter of the through hole changes due to a change in welding posture, a gap between workpieces, and the like. Moreover, since the hole is immediately filled after the through hole is formed, it is difficult to directly observe the diameter of the through hole. Moreover, if the presence or absence of a through hole is indirectly observed with an existing voltage monitor, the penetration itself can be detected. However, in the conventional method, a through hole capable of realizing a desired strength is formed. Or could not be inlined. The in-line inspection is not a sampling inspection (a destructive inspection such as a tensile test or cross-sectional observation), but a total inspection without breaking.

  Accordingly, the present invention provides an in-line inspection method, an in-line inspection apparatus, and a plasma-MIG welding method that can inspect whether a through-hole capable of realizing the desired strength is formed and solve the above-described problems. The task is to do.

  In order to solve the above problems, the inventors of the present application have made various studies on the relationship between the size and strength of through holes formed in through welding. As a result, there is a correlation between the size of the diameter of the nugget filled in the through hole and the joint strength, and there is a similar correlation between the size of the diameter of the lower hole of the through hole and the joint strength. And found.

  Therefore, an in-line inspection method according to a first aspect of the present invention is an in-line inspection method in an in-line inspection apparatus for in-line inspecting a through hole when a plurality of stacked workpieces are welded through the in-line inspection apparatus. Storage means for storing calibration curve data indicating the relationship between the diameter of the through hole formed in the workpiece and the arc voltage of plasma arc welding, which is determined in advance for each workpiece thickness and welding condition, and processing Means for monitoring the arc voltage during plasma arc welding, and detecting the penetration of the plurality of superposed workpieces based on the arc voltage, and monitoring the arc voltage. The arc voltage measured after a predetermined time has passed since the penetration of the plurality of workpieces is compared with the calibration curve data. And executes the through-hole diameter detection step of detecting a lower hole diameter of the through hole.

  The in-line inspection method according to the second aspect of the present invention is an in-line inspection method in an in-line inspection apparatus for in-line inspection of a through hole when a plurality of stacked workpieces are welded through. Storage means for storing calibration curve data indicating the relationship between the diameter of the through hole formed in the workpiece and the arc voltage of plasma arc welding, which is determined in advance for each workpiece thickness and welding condition, and processing Means for monitoring the arc voltage during plasma arc welding, and detecting the penetration of the plurality of superposed workpieces based on the arc voltage, and monitoring the arc voltage. Then, the cumulative moving average data of the arc voltage is detected, and the arc voltage cumulative transition measured after a predetermined time has elapsed since the penetration of the plurality of workpieces. By matching the average data with the calibration curve data, and executes the through-hole diameter detection step of detecting a lower hole diameter of the formed through hole.

  In the in-line inspection method according to the present invention, the calibration curve data is obtained for each plate thickness and welding condition of the plurality of overlapped workpieces, and through holes having a desired size are formed in the plurality of overlapped workpieces. It is pilot hole diameter-voltage data indicating a correspondence relationship between a pilot hole diameter indicating the inner diameter of the hole on the back surface side when it is formed and a stable voltage indicating an arc voltage when the arc ignited in the through hole is stabilized Is preferred.

  According to such a procedure, in the in-line inspection method, by monitoring the arc voltage when forming a through-hole by plasma arc welding, and collating the measured arc voltage with a calibration curve data after a predetermined time has elapsed from the penetration, The diameter of the through hole is detected. Therefore, the in-line inspection method of the present invention can guarantee the hole diameter of the lower plate that controls the strength.

  In the in-line inspection method according to the present invention, the storage means of the in-line inspection apparatus is already formed on the plurality of stacked workpieces as calibration curve data obtained in advance for each plate thickness and welding condition of the workpiece. When the arc is re-ignited by plasma arc welding into a through hole of a desired size, the arc voltage at the time of voltage rise is stabilized before the arc voltage within the predetermined dielectric breakdown time and the diameter of the lower hole Re-ignition data indicating the correspondence is further stored, and after the processing means detects the lower hole diameter of the through-hole in the through-hole diameter detection step, the arc is once cut and immediately after the arc of plasma arc welding is re-pointed. A re-ignition step for arcing, and the arc voltage after the arc is re-ignited is compared with the re-ignition data which is the calibration curve data, thereby forming a through-hole. It is preferred to perform the re-detection step of detecting the pilot hole diameter.

  According to such a procedure, the in-line inspection method ensures that the desired through-hole is obtained by comparing the rising voltage of the arc voltage measured when the arc is re-ignited with respect to the formed through-hole with the calibration curve data. Can be determined. According to this, since the rising voltage at the time of arc ignition is larger than the stable voltage when the arc is stabilized, the diameter of the prepared hole can be detected with high accuracy.

  In order to solve the above problems, an in-line inspection apparatus according to the present invention is an in-line inspection apparatus that inspects through holes in-line when through welding a plurality of superimposed workpieces, Storage means for storing calibration curve data indicating the relationship between the diameter of the lower hole of the through-hole formed in the workpiece and the arc voltage of plasma arc welding, which is obtained in advance for each welding condition, and the arc voltage during plasma arc welding. Voltage monitoring means for monitoring and measuring the arc voltage at a predetermined frequency; penetration detection means for detecting that the plurality of superimposed workpieces have penetrated based on the measured arc voltage; and measured arc voltage On the basis of the arc voltage measured after a lapse of a predetermined time from the penetration of the plurality of workpieces, by comparing the calibration curve data with the formed penetration A through hole diameter detecting means for detecting the pilot hole diameter, characterized in that it comprises a.

  According to this configuration, the in-line inspection apparatus monitors the arc voltage when the through hole is formed by plasma arc welding by the voltage monitoring unit. Then, the in-line inspection device detects penetration by the penetration detection means, and collates the arc voltage measured after a predetermined time from penetration with the calibration curve data by the penetration hole diameter detection means, thereby determining the lower hole diameter of the penetration hole. To detect. Therefore, the in-line inspection apparatus of the present invention can guarantee the hole diameter of the lower plate that controls the strength.

  Further, the plasma-MIG welding method according to the present invention is a plasma-MIG welding method in which plasma arc welding and MIG welding are combined, and a step of forming a through-hole by a plasma arc in a plurality of stacked workpieces; The method includes a step of inspecting the through hole inline by the inline inspection method, and a step of welding the plurality of stacked workpieces by filling the through hole by MIG welding.

  According to such a procedure, the plasma-MIG welding method includes a through-hole in a series of through-welding steps in which a through hole is formed by a plasma arc on a plurality of stacked workpieces, and the through hole is filled and welded by MIG welding. The arc voltage value measured in the step of forming is collated with calibration curve data obtained in advance for each workpiece thickness and welding condition. Thereby, the plasma-MIG welding method can confirm whether or not a good through hole capable of realizing a desired strength is formed at the same time as through welding. Therefore, the plasma-MIG welding method of the present invention can guarantee the hole diameter of the lower plate that controls the strength.

  According to the present invention, since the nugget diameter that governs the strength and the hole diameter of the lower plate can be in-line inspected, the quality assurance accuracy is improved.

BRIEF DESCRIPTION OF THE DRAWINGS It is workpiece | work sectional drawing which shows the outline | summary of the in-line inspection method which concerns on 1st Embodiment of this invention typically, (a) is a torch height detection step, (b) is an arc length detection step, (c) is bottom. A plate penetration detection step, (d) shows a hole diameter enlargement step, and (e) shows a hole diameter securing step. It is a graph which shows typically an example of the time change of the voltage monitored by the in-line inspection method concerning a 1st embodiment of the present invention. It is explanatory drawing of the in-line test | inspection method which concerns on 1st Embodiment of this invention, Comprising: (a) is a top view of the test piece before welding, (b) is a test piece in which the hole of 3φ was drilled, (c) Is a test piece with a 6φ hole, (d) shows the shape of the hole when the penetration is actually detected, and (e) shows the shape of the hole when the lower hole diameter is actually detected. It is a lineblock diagram of the welding system containing the in-line inspection device concerning a 1st embodiment of the present invention. It is a block diagram showing typically the composition of the in-line inspection device concerning a 1st embodiment of the present invention. It is a graph which shows an example of the arc length-voltage data acquired for every pilot hole diameter which the in-line inspection device concerning a 1st embodiment of the present invention memorized as calibration curve data. It is a graph which shows an example of the pilot hole diameter-voltage data which the in-line test | inspection apparatus which concerns on 1st Embodiment of this invention memorize | stored as calibration curve data. It is a graph which shows an example of the re-ignition data which the in-line inspection apparatus concerning a 1st embodiment of the present invention memorized as calibration curve data. It is a flowchart which shows the flow of the process by the in-line inspection apparatus which concerns on 1st Embodiment of this invention. It is a graph which shows an example of the measurement data of the voltage measured until the arc became stable in the in-line inspection method concerning a 1st embodiment of the present invention. It is a graph which shows an example of the measurement data of the voltage measured before the arc stabilized in the in-line inspection method concerning a 1st embodiment of the present invention. It is a block diagram which shows typically the structure of the in-line test | inspection apparatus which concerns on 2nd Embodiment of this invention. It is a graph which shows an example of the voltage value measured before and after the penetration of three works by the in-line inspection method concerning a 2nd embodiment of the present invention, and (a) is time series data of the voltage measured with a predetermined sampling period. , (B) shows time-series data of time differentiation of (a). FIG. 14A is a graph showing analysis data obtained from the actually measured voltage values shown in FIG. 13A, FIG. 14A is a moving average of the voltage values of FIG. 13A, and FIG. 14B is FIG. ) Shows the measured voltage value and the cumulative moving average in the time range of the two-dot chain line. It is a graph which shows an example of the voltage value measured before and after the penetration of three works with a gap in the in-line inspection method concerning a 2nd embodiment of the present invention, and (a) is measured with a predetermined sampling period. Voltage time-series data, (b) shows time-differential time-series data of (a). FIG. 16A is a graph showing analysis data obtained from the actually measured voltage values shown in FIG. 15A, FIG. 16A is a moving average of the voltage values of FIG. 15A, and FIG. 16B is FIG. ) Shows the measured voltage value and the cumulative moving average in the time range of the two-dot chain line.

DESCRIPTION OF EMBODIMENTS Embodiments (referred to as embodiments) for carrying out the present invention will be described in detail with reference to the drawings.
<First Embodiment>
Below, for convenience of explanation, 1. 1. Outline of inline inspection method 2. Configuration of welding system 3. Configuration of inline inspection device 4. Flow of processing by inline inspection device Each chapter of a specific example of calibration curve data will be described.

[1. Overview of inline inspection method]
As shown in FIG. 1, the in-line inspection method is a method for in-line inspection of through holes when a plurality of stacked workpieces W are welded through. Here, the use, shape, material, and size of the workpiece W are not particularly limited, but will be described as being, for example, a galvanized steel sheet. FIG. 1 shows a cross section of a workpiece when two workpieces W are overlapped as an example. Here, the description will be made assuming that there is no gap between the upper plate and the lower plate. In this in-line inspection method, a through hole is formed by plasma arc welding using a plasma torch 8. The in-line inspection method also requires that calibration curve data indicating the relationship between the diameter of the through hole formed in the workpiece and the arc voltage of plasma arc welding is obtained in advance for each workpiece thickness and welding condition. It is assumed.

  The in-line inspection method is roughly divided into a penetration detection step S1 shown in FIGS. 1A to 1C and a through hole diameter detection step S2 shown in FIGS. 1C to 1E. FIG. 2 shows an example of a correspondence relationship between the penetration detection step S1 and the through hole diameter detection step S2 and the time variation of the measured arc voltage. The horizontal axis of the graph shown in FIG. 2 indicates time, and the vertical axis indicates arc voltage. Here, it demonstrates with reference to FIG. 1 and FIG.

  The penetration detection step S1 is a step of monitoring the arc voltage during plasma arc welding and detecting that a plurality of workpieces W overlapped based on the arc voltage have penetrated. This penetration detection step S1 can be formally divided into, for example, a torch height detection step, an arc length detection step, and a lower plate penetration detection step.

  As shown in FIG. 1A, the torch height detection step is equal to the length from the plasma torch 8 to the surface of the upper plate (work W) based on the measured initial value of the arc voltage and calibration curve data. This is a step of detecting the arc length (hereinafter referred to as torch height). In the example shown in FIG. 2, the arc voltage once suddenly rises due to dielectric breakdown during the initial arc ignition, but thereafter the voltage stabilizes, and this is the initial voltage. The torch height is detected with this stable voltage. Since the torch height changes under the influence of the plate thickness variation or changes the welding conditions, the calibration curve data of the multiple workpieces overlapped as described above. It is required for each plate thickness and welding condition. The welding conditions include, for example, torch height, welding current, gas flow rate, and the like.

  In the arc length detection step, as shown in FIG. 1B, the length of the drilled hole is determined based on the arc voltage and calibration curve data monitored in real time when the workpiece W is drilled. This is a step of detecting the arc length. In the example shown in FIG. 2, the measured arc voltage rises from the initial voltage. Thus, the arc length is detected when the voltage increases from the initial voltage. Detecting the arc length has the same meaning as detecting the corresponding arc voltage.

  As shown in FIG. 1C, the lower plate penetration detection step is a step of detecting penetration to the lower plate based on the measured arc voltage and calibration curve data. The state in which the drilled hole has reached the lower plate is a state in which a hole can be seen on the back side of the workpiece but is not a hole having a desired diameter. At this time, the arc length becomes equal to the length from the plasma torch 8 to the work back surface. In addition, it can be detected by calculating the increment and increase rate of a voltage whether it penetrated. In the graph of FIG. 2, the increase rate of the measured arc voltage becomes moderate at a certain point in time. In this way, it is detected that the lower plate has been penetrated at the inflection point where the voltage increase rate changes from increasing to decreasing.

  The through-hole diameter detection step S2 detects the under-hole diameter of the formed through-hole by monitoring the arc voltage and comparing the measured arc voltage with calibration curve data after a predetermined time has passed since the penetration of the plurality of workpieces W. It is a process to do. This through hole diameter detecting step S2 can be formally divided like a hole diameter expanding step and a hole diameter securing step.

As shown in FIG. 1D, the hole diameter expanding step is a step of expanding the hole diameter of the through hole after penetrating to the lower plate. Here, since the arc from the plasma torch 8 cannot conduct at the hole portion on the back surface of the workpiece, the arc bypasses the hole. Thus, as the time elapses, the arc spreads around the hole, so that the arc length gradually increases and the diameter of the lower hole also increases.
In the example shown in FIG. 2, the hole diameter is enlarged in the process in which the measured arc voltage gradually rises from the inflection point.

  The hole diameter securing step is a step of detecting that the lower hole diameter has reached a desired diameter based on the measured arc voltage and calibration curve data, as shown in FIG. In the example shown in FIG. 2, the plasma arc is extinguished when the measured arc voltage becomes a threshold voltage when a desired pilot hole diameter is formed. When this threshold voltage is reached, it is determined that a desired hole diameter has been secured.

  Various types of data can be used as calibration curve data used in the in-line inspection method. For example, it may be data of voltage change over time as shown in FIG. In this case, the data need not be the entire broken line as shown in FIG.

  Further, as the calibration curve data, for example, as shown in FIGS. 6 to 8, data indicating the relationship between the pilot hole diameter, the arc length, and the arc voltage may be used. Plan views before welding of the test pieces used to construct the calibration curve data shown in FIGS. 6 to 8 are shown in FIGS. 3 (a) to 3 (c). The test piece shown in FIG. 3A has no holes (hereinafter referred to as “no holes”). The test piece shown in FIG. 3B has a hole with a diameter of 3 mm (hereinafter referred to as 3φ). The test piece shown in FIG. 3C has a hole with a diameter of 6 mm (hereinafter referred to as 6φ). A description of the calibration curve data shown in FIGS. 6 to 8 will be given later.

  FIG. 3D shows an example of the shape of the hole viewed from the surface side of the workpiece when the penetration of a two-layer workpiece without holes is actually detected, and FIG. An example of the shape of the hole seen from the workpiece surface side when actually detected is shown.

  In addition, when the inventors of the present application actually measured the relationship between the size of the formed through-hole and the joint strength when three workpieces were stacked and welded under predetermined welding conditions, the first workpiece from the bottom There is a correlation that the joint strength increases as the size of the hole diameter (hereinafter referred to as nugget diameter) at the height of the interface with the second workpiece increases. At this time, there was a similar correlation that the larger the hole diameter (lower hole diameter) on the back surface of the first workpiece from the bottom, the greater the joint strength. That is, it was confirmed that the nugget diameter and the lower hole diameter are the factors governing the strength. Therefore, it can be concluded that the desired strength can be guaranteed by setting the pilot hole diameter to a desired size.

[2. Configuration of welding system]
Here, the configuration of a welding system including the in-line inspection apparatus according to the first embodiment of the present invention will be described with reference to FIG. The welding system 1 is a robot arc welding system for through-welding a plurality of superimposed workpieces W. As shown in FIG. 4, the welding system 1 mainly includes a welding torch 2, a robot 3, a robot control device 4, A welding power source 5, a filler supply device 6, a welding control device 7, and an in-line inspection device 10 are provided. Although not shown, the welding system 1 includes an operating gas cylinder, a shield gas cylinder, a gas flow rate regulator, a remote controller, and the like. FIG. 4 shows a cross section of a workpiece when three workpieces W are overlapped as an example. Here, the description will be made assuming that there is no gap between the upper plate and the lower plate.

  The robot control device 4, the welding control device 7, and the inline inspection device 10 are, for example, a central processing unit (CPU), a read only memory (ROM), a random access memory (RAM), and a hard disk drive (HDD). And an input / output interface.

The welding torch 2 includes a plasma torch 8 and a MIG torch 9.
The plasma torch 8 is a preceding torch used in advance in a stage before filling with filler in penetration welding (hereinafter referred to as penetration welding phase P1), and penetrates a plurality of stacked workpieces W. It is a torch for forming a through hole. The plasma torch 8 is formed with electrodes and nozzles for performing plasma arc welding, and is supplied with an operating gas such as argon and a shielding gas. The plasma torch 8 generates a pilot arc between the tungsten electrode and the water-cooled constraining nozzle, converts the working gas into plasma by the heat of the pilot arc, and generates a plasma arc between the workpiece W and the plasma torch 8. As the shielding gas, a commonly used MAG gas (Ar + CO ₂ mixed gas) or the like is supplied.

The MIG torch 9 is a subsequent torch used after the through-welding phase P1 and filling the filler (hereinafter referred to as through-welding phase P2), and is filled with the through-holes. It is a torch for fusing a plurality of workpieces W. The MIG torch 9 is a torch for performing MIG welding (metal inert gas welding) for filling a through hole with a filler. A filler 61 as a consumable electrode is sent from the filler supply device 6 to the center of the MIG torch 9, and shield gas (Ar + CO ₂ ) is supplied around the filler 61.

The robot 3 is, for example, a multi-axis multi-joint welding robot, and the welding torch 2 is attached to the arm 3a on the distal end side. The robot 3 can move the welding torch 2 by moving each joint with a motor.
The robot control device 4 is connected to the robot 3 and controls the operation and posture of the robot 3 based on an input command such as a welding path or a command stored in advance.

  The welding power source 5 supplies electric power for arc welding to the welding torch 2. Here, as shown in FIG. 4, the welding power source 5 mainly includes a plasma power source 51, a MIG power source 52, a gas supply device 53, and a voltage detector 54. Although not shown, the welding power source 5 includes a current detector, a control circuit necessary for MIG welding, and the like.

  The plasma power source 51 converts a commercial three-phase AC power source into a DC power source using a transformer (not shown), and supplies power to the plasma torch 8 in a phase P1 of penetration welding (during plasma arc welding). The negative electrode of the plasma power source 51 is electrically connected to the tungsten electrode of the plasma torch 8, and the anode of the plasma power source 51 is electrically connected to the workpiece W. The output characteristic of the plasma power source 51 is generally a constant current characteristic, whereby the arc current after arc stabilization is maintained at a constant value. By this constant current control, the arc length can be estimated from the measured arc voltage.

  The MIG power source 52 converts a commercial three-phase AC power source into a DC power source using a transformer (not shown), and supplies power to the MIG torch 9 in the phase P2 of penetration welding (during MIG welding). The anode of the MIG power source 52 is electrically connected to the filler 61 (consumable electrode) of the MIG torch 9, and the negative electrode of the MIG power source 52 is electrically connected to the workpiece W. The output characteristic of the MIG power supply 52 is a constant voltage characteristic, whereby the arc length after arc stabilization is maintained at a constant value.

  The gas supply device 53 supplies a shielding gas for welding to the welding torch 2 (plasma torch 8 and MIG torch 9) from a gas cylinder (not shown). Further, the gas supply device 53 supplies a working gas for forming plasma to the welding torch 2 from a gas cylinder (not shown).

  The voltage detector 54 detects an arc voltage which is a welding voltage by the plasma welding power source 51. The voltage detector 54 outputs the arc voltage detected in the phase P1 of penetration welding (during plasma arc welding) to the in-line inspection apparatus 10. The voltage detector 54 outputs the arc voltage detected in the phase P2 of penetration welding (during MIG welding) to the welding control device 7.

  The filler supply device 6 is connected to the MIG power source 52. The filler supply device 6 sends out a wire-like filler sent from a filler container (not shown) through a delivery path to the MIG torch 9 in the phase P2 of penetration welding (during MIG welding).

The welding control device 7 drives the welding power source 5 and the in-line inspection device 10 in the through-welding phase P1 (during plasma arc welding), thereby penetrating a plurality of workpieces W stacked by plasma arc welding. Form.
The welding control device 7 drives the welding power source 5 in the through welding phase P2 (during MIG welding), thereby filling the through holes with MIG welding. That is, the welding control device 7 drives the MIG power source 52, the gas supply device 53, and the MIG torch 9.

[3. Configuration of inline inspection equipment]
Here, the configuration of the in-line inspection apparatus according to the first embodiment of the present invention will be described with reference to FIG. 5 (refer to FIG. 4 as appropriate).
The in-line inspection apparatus 10 inspects through holes in-line when through welding a plurality of superimposed workpieces W, and includes a voltage monitoring unit 11, a storage unit 12, and a processing unit 13.
The voltage monitoring means 11 monitors the arc voltage during plasma arc welding and measures the arc voltage at a predetermined frequency.

  The storage unit 12 stores, for example, pilot hole diameter-voltage data 21, re-ignition data 22, and arc length-voltage data 23 as calibration curve data obtained in advance for each plate thickness and welding condition of the workpiece W. To do. The storage means 12 is composed of, for example, a general hard disk or memory.

(Lower hole diameter-voltage data 21)
The lower hole diameter-voltage data 21 is calibration curve data indicating the relationship between the lower hole diameter of the through hole formed in the workpiece W and the arc voltage of plasma arc welding, which is obtained in advance for each thickness and welding condition of the workpiece W. It is. In the present embodiment, the lower hole diameter-voltage data 21 is obtained for each plate thickness and welding condition of a plurality of superimposed workpieces W, and the back surface when through holes having a desired size are formed in the plurality of superimposed workpieces. The correspondence relationship between the lower hole diameter indicating the inner diameter of the side hole and the stable voltage indicating the arc voltage when the arc ignited in the through hole is stabilized is shown. An example of the lower hole diameter-voltage data 21 is shown in FIG. The description of the lower hole diameter-voltage data 21 will be described later.

(Re-ignition data 22)
The re-ignition data 22 is the time before the arc at the time of voltage rise when the arc is re-ignited by plasma arc welding to a through-hole of a desired size already formed in the plurality of stacked workpieces W. The correspondence relationship between the arc voltage within a predetermined dielectric breakdown time (for example, 5 milliseconds) and the pilot hole diameter is shown. An example of the re-ignition data 22 is shown in FIG. The re-ignition data 22 will be described later.

(Arc length-voltage data 23)
The arc length-voltage data 23 is obtained from a plurality of pilot hole diameter-voltage data 21 obtained for each thickness and welding condition of a plurality of workpieces W that are superimposed. An example of the arc length-voltage data 23 is shown in FIG. The arc length-voltage data 23 will be described later.

  As shown in FIG. 5, the processing unit 13 includes a through detection unit 31, a through hole diameter detection unit 32, and a through hole diameter guarantee unit 33. Here, the processing means 13 includes the penetration detection means 31 as a configuration for performing the penetration detection step S1 shown in FIGS. 1A to 1C as an in-line inspection method, and the through hole diameter detection means 32. In addition, the through hole diameter guaranteeing means 33 is provided as a configuration for performing the through hole diameter detecting step S2 shown in FIGS. 1C to 1E as an in-line inspection method. It should be noted that although the through hole diameter detecting means 32 and the through hole diameter guaranteeing means 33 are formally divided, it is needless to say that they may be combined.

  The penetration detection unit 31 detects that a plurality of workpieces W that are overlapped have penetrated based on the measured arc voltage. A method for detecting that the workpiece W has penetrated is not particularly limited, and for example, the workpiece W can be detected by using an increment or an increase rate of the measured arc voltage. In the present embodiment, as an example, it has a configuration for detecting penetration by an increase in the measured arc voltage and a configuration for detecting penetration by an increase rate, and either one of them by a predetermined switching instruction operation or presetting. The configuration is configured so that the process can be executed. Therefore, as shown in FIG. 5, the penetration detection unit 31 includes a torch height calculation unit 311, an increment calculation unit 312, an increment determination unit 313, an increase rate calculation unit 314, and an increase rate determination unit 315. I decided to prepare.

  The torch height calculation means 311 calculates the arc length converted from the initial voltage to the arc length-voltage data 23 based on the arc length-voltage data 23 and the initial voltage for the measured arc voltage. Is calculated as follows. This torch height calculating means 311 calculates the torch height from a calibration curve (indicated by a triangle in the graph of FIG. 6) measured using a test piece without holes. For example, in the graph of FIG. 6, when the voltage value indicated by a triangle (about 30.3 V) is measured with a real-time monitor when the arc length is 5 mm, the torch height calculation means 311 has a torch height of 5 mm. Is detected.

The increment calculation means 312 calculates an increment from the initial voltage for the arc voltage measured after the initial voltage measurement, and the arc length in the arc length-voltage data 23 is calculated from the torch height to the total thickness of the workpiece. The voltage increment until it increases by the length is calculated.
The increment discriminating means 313 discriminates whether or not the increment from the initial voltage coincides with the increment of the voltage in the arc length-voltage data 23. If the increment of the voltage coincides, the plurality of superimposed workpieces W have penetrated. Is determined. The increment calculating means 312 and the increment determining means 313 are configured to detect penetration through voltage increments.

The increase rate calculating means 314 calculates an increase rate per unit time for the arc voltage measured after the initial voltage measurement.
The increase rate discriminating means 315 determines whether or not an inflection point at which the increase rate exceeds a predetermined threshold value, and determines that a plurality of superimposed workpieces W have penetrated when the inflection point is reached ( Lower plate penetration detection step: see FIG. These increase rate calculating means 314 and increase rate determining means 315 are configured to detect penetration based on the voltage increase rate.

  The through-hole diameter detecting means 32 collates the measured arc voltage with a lower hole diameter-voltage data (calibration curve data) 21 based on the measured arc voltage after a predetermined time has elapsed since the penetration of the plurality of workpieces W. The diameter of the prepared through hole is detected.

  The through hole diameter detection means 32 collates the measured arc voltage with the lower hole diameter-voltage data 21 (graph shown in FIG. 7) as follows. The horizontal axis of the graph shown in FIG. 7 indicates the pilot hole diameter of the test piece, and the vertical axis indicates the arc voltage at 0.2 seconds after arc firing. Here, the data of the lower hole diameter = 0 mm is a measurement result by the test piece without holes shown in FIG. In the graph, the arc voltage under the condition of a torch height of 5 mm is indicated by a triangle. Similarly, in the graph, the arc voltage under the condition of a torch height of 10 mm is indicated by a circle, and the arc voltage under the condition of a torch height of 13 mm is indicated by a square.

  When the detected torch height is 5 mm, for example, the through-hole diameter detecting means 32 has a calibration curve (torch height 5 mm) connecting three arc voltage values indicated by triangles in the graph of FIG. 7 when the torch height is 5 mm. Calibration curve). At this time, when a voltage value of about 31.4 V is detected as the voltage value monitored in real time, it can be detected that the diameter of the prepared hole has become 3 mm. Similarly, when a voltage value of about 34.1 V is detected, it can be detected that the diameter of the prepared hole is 6 mm. This calibration curve is an example, and the present invention is not limited to this.

  The through hole diameter guarantee means 33 includes an arc firing control means 331, a rising voltage collating means 332, and a hole diameter confirmation means 333.

  When the through hole diameter detecting means 32 detects the diameter of the through hole in the through hole, the arc ignition control means 331 once cuts the arc and immediately re-ignites the plasma arc welding arc. In the present embodiment, the arc ignition control means 331 issues an arc-off / arc-on command to the welding power source 5 via the welding control device 7. The arc firing control means 331 may issue a command directly to the welding power source 5.

The rising voltage collating means 332 detects the lower hole diameter of the formed through hole by collating the arc voltage measured after re-igniting the arc with the calibration curve data. That is, the above-described through-hole diameter detecting means 32 collates with the calibration curve data to perform processing for detecting the lower hole diameter of the formed through-hole, and after re-igniting the arc, this rising voltage collating means 332 However, the diameter of the prepared through hole is re-detected by collating with the calibration curve data.
In the present embodiment, the rising voltage collating means 332 collates whether or not the arc voltage measured at the time of arc re-ignition coincides with the rising voltage (arc voltage) in the re-ignition data 22. The rising voltage collating means 332 collates the measured arc voltage with the re-ignition data 22 (graph shown in FIG. 8) as follows. The horizontal axis of the graph shown in FIG. 8 indicates the hole diameter of the test piece, and the vertical axis indicates the arc voltage at 5 milliseconds. The way of viewing the graph of FIG. 8 is the same as that of the graph of FIG. Further, since the rising voltage collating means 332 and the through hole diameter detecting means 32 have the same detection method using a calibration curve, detailed description thereof will be omitted.

  The hole diameter confirming means 333 determines that a through hole having a desired size is reliably formed when the measured arc voltage matches the rising voltage in the re-ignition data 22 by collation. Here, “match” means that the rising voltage of the arc voltage measured when the arc is re-ignited into the through hole that should have the desired hole diameter is the rising voltage of the desired hole diameter in the re-ignition data 22. It means that it becomes more than the value of.

[4. Flow of processing by inline inspection equipment]
The flow of processing by the inline inspection apparatus 10 will be described with reference to FIG. 9 (refer to FIGS. 4 and 5 as appropriate). FIG. 9 is a flowchart showing an outline of the flow of processing by the inline inspection apparatus according to the first embodiment of the present invention. Here, the welding power source 5 detects the arc voltage by the voltage detector 54 and outputs it to the in-line inspection apparatus 10. Thereby, the voltage monitoring means 11 of the in-line inspection apparatus 10 monitors the arc voltage.

First, the penetration detection unit 31 of the in-line inspection apparatus 10 detects the arc length of the initial voltage as the torch height H ₀ by the torch height calculation unit 311 (step S11). Then, the penetration detection means 31 continues to detect the arc voltage corresponding to the arc length (step S12), and when the penetration cannot be detected by the voltage increment or the voltage increase rate (step S13: No), the process returns to step S12 to return to the arc. Continue to detect the arc voltage corresponding to the length. On the other hand, when penetration is detected (step S13: Yes), the penetration detection unit 31 passes the process to the through hole diameter detection unit 32.

  The through-hole diameter detecting means 32 continues to detect the arc voltage corresponding to the arc length (step S14), and when the lower hole diameter cannot be detected (step S15: No), if the predetermined time limit is not reached (step S16: No) ), Returning to step S14, the detection of the arc voltage corresponding to the arc length is continued. On the other hand, when the lower hole diameter is detected by the arc voltage measured when a predetermined time has passed since the penetration of the workpiece (step S15: Yes), the through hole diameter detection means 32 passes the process to the through hole diameter guarantee means 33, and the through hole diameter The guarantee means 33 performs arc-off control by the arc firing control means 331 (step S17). As a result, the welding control device 7 issues an arc-off command to the welding power source 5 and the arc of the plasma torch 8 is cut off. Immediately thereafter, the through-hole diameter guaranteeing means 33 performs control to re-ignite the arc by the arc firing control means 331 (step S18). As a result, the welding control device 7 issues an arc re-ignition command to the welding power source 5, and the arc of the plasma torch 8 is re-ignited.

  Then, the through-hole diameter guaranteeing means 33 determines whether or not the rising voltage of the arc voltage measured by the rising voltage collating means 332 matches the rising voltage of the re-ignition data (step S19: re-detection step). . When the rising voltages of the two match (step S19: Yes), the through hole diameter guaranteeing unit 33 sets the pass / fail judgment of the through hole to “good” by the hole diameter checking unit 333 (step S20), and ends the process. On the other hand, if the rising voltages of the two do not match (step S19: No), the through hole diameter guaranteeing means 33 sets the through hole pass / fail judgment to “No” by the hole diameter confirmation means 333 (step S21), and ends the process.

  If the lower hole diameter cannot be detected (step S15: No) and the time limit is reached (step S16: Yes), the through hole diameter detecting means 32 passes the processing to the through hole diameter guaranteeing means 33, and the through hole diameter guaranteeing means 33 is reached. However, the hole diameter confirmation means 333 sets the pass / fail judgment of the through hole to “No” (step S22), and the process is terminated.

  Based on the pass / fail judgment result of the through hole, for example, “good” can be classified as a work that can shift to the phase P2 of through welding, and “no” can be classified as a work that cannot shift to the phase P2. If it is determined as “No”, the process proceeds to a process such as repair.

[5. Specific example of calibration curve data]
(Lower hole diameter-voltage data 21)
The lower hole diameter-voltage data 21 (for example, the graph shown in FIG. 7) can be created by using the calibration curve data shown in FIG. Here, the calibration curve data shown in FIG. 10 will be described. The horizontal axis of the graph shown in FIG. 10 indicates time, and the vertical axis indicates arc voltage. The voltage line 101 shown in FIG. 10 is formed in a hole having a diameter of 3 mm already drilled under a predetermined welding condition with a torch height of 5 mm using the 3φ test piece shown in FIG. The result of having measured the time change of the arc voltage over about 1 second when an arc is ignited by plasma arc welding is shown.

  Moreover, the voltage line 102 shown in FIG. 10 shows the measurement result of the time change of the arc voltage similarly obtained when the torch height is changed to 10 mm while using the 3φ test piece shown in FIG. . Furthermore, the voltage line 103 shown in FIG. 10 shows the measurement result of the time change of the arc voltage similarly obtained when the torch height is changed to 13 mm while using the 3φ test piece shown in FIG. . These voltage lines 101, 102, and 103 all show a voltage (hereinafter referred to as a stable voltage) where the arc is stable after about 0.2 seconds, although a rapid increase in the arc voltage is recognized at the time of arc ignition. Although not shown, similar arc voltage measurement results were obtained with the torch height of 5, 10, 13 mm using a 6φ test piece shown in FIG. Further, although not shown in the figure, the same arc voltage measurement results were obtained with the torch height of 5, 10, 13 mm using the test piece without holes shown in FIG. A graph (lower hole diameter-voltage data 21) shown in FIG. 7 is obtained by extracting the arc voltage when the time is 0.2 seconds in the total nine measurement results using each test piece.

(Re-ignition data 22)
The re-ignition data 22 (for example, the graph shown in FIG. 8) can be created by using the calibration curve data shown in FIG. Here, the calibration curve data shown in FIG. 11 will be described. FIG. 11 is an enlarged view of FIG. The horizontal axis of the graph shown in FIG. 11 indicates time, and the vertical axis indicates arc voltage. A voltage line 111 shown in FIG. 11 shows a change over time of the arc voltage from 0.02 seconds after the arc is ignited with respect to the data of the voltage line 101 shown in FIG. That is, using a 3φ test piece shown in FIG. 3 (b), an arc is formed by plasma arc welding toward a hole with a diameter of 3 mm already drilled under a predetermined welding condition with a torch height of 5 mm. The result of having measured the time change of the arc voltage when starting is about 0.02 second is shown. Here, the graphs of FIGS. 10 and 11 differ not only in the time axis but also in the scale of the voltage value on the vertical axis. In addition, the voltage value which has been shaken out in the graph of FIG. 10 is stored in the graph of FIG.

  Moreover, the voltage line 112 shown in FIG. 11 shows the time change of the arc voltage from the time of starting the arc to 0.02 seconds for the data of the voltage line 102 shown in FIG. 10 (3φ test piece, torch height 10 mm). Furthermore, the voltage line 113 shown in FIG. 11 shows the time change of the arc voltage from the time of starting the arc to 0.02 seconds with respect to the data of the voltage line 103 shown in FIG. 10 (3φ test piece, torch height 13 mm). In these voltage lines 111, 112, and 113, an arc voltage larger than the stable voltage shown in FIG. 10 (hereinafter referred to as a rising voltage) is recognized for about 5 milliseconds immediately after the arc ignition. Although not shown in the figure, the same rise voltage measurement results with a torch height of 5, 10, 13 mm were obtained using a 6φ test piece shown in FIG. Further, although not shown in the figure, the same rise voltage measurement results were obtained with the torch height of 5, 10, 13 mm using the test piece without holes shown in FIG. FIG. 8 shows a graph obtained by extracting the arc voltage when the time is 5 milliseconds in the total nine measurement results using each test piece.

(Arc length-voltage data 23)
The arc length-voltage data 23 (graph shown in FIG. 6) can be created by using the calibration curve data (bottom hole diameter-voltage data 21) shown in FIG. The graph shown in FIG. 6 will be described below in comparison with FIG. The horizontal axis of the graph shown in FIG. 6 indicates the arc length, and the vertical axis indicates the arc voltage. Here, the arc length is the length of the arc converted from the torch height measured as the calibration curve data, and is therefore distinguished from the arc length converted from the measured arc voltage. .

  The circles in the graph of FIG. 6 indicate three data (triangle, circle, square) of the 3φ test piece (hole diameter 3 mm) in the graph of FIG. Here, the torch height (5, 10, 13 mm) of the 3φ test piece in the graph of FIG. 7 was defined as the arc length in the graph of FIG. Similarly, the square marks in the graph of FIG. 6 indicate three data (triangle, circle, square) of the 6φ test piece (hole diameter 6 mm) in the graph of FIG. Furthermore, the triangle mark in the graph of FIG. 6 shows three data (a triangle, a circle, and a rectangle) of the test piece without a hole (hole diameter 0 mm) in the graph of FIG.

  Therefore, when the line (hole diameter 0 mm) connecting the three data of the triangle mark in the graph of FIG. 6 is viewed from the lower left to the upper right, this line (hereinafter referred to as a calibration curve without holes) is at a torch height of 5 mm. For example, the relationship between the arc length that fluctuates when continuously drilling a workpiece without a hole having a total thickness of 3 mm and the arc voltage value at that time is shown. Specifically, in the graph, when a voltage value (about 30.3 V) indicated by a triangle is detected when the arc length is 5 mm, it is detected that the torch height is 5 mm. In this case, when the arc length is 5 mm in the graph (when the torch height is 5 mm), the workpiece has not penetrated but the digging has progressed, and the graph is above the triangle. It is assumed that a voltage value indicated by a certain circle (about 31.4 V: 3φ data) is detected. At this time, according to the calibration curve without holes, since the detected voltage value corresponds to the arc length of 6 mm (torch height of 6 mm), the workpiece is not penetrated, but the workpiece is dug vertically by the thickness of 1 mm. You can see that it has advanced.

Subsequently, for example, it is assumed that after a workpiece having a total thickness of 3 mm penetrates, the hole diameter of the lower plate of the workpiece is expanded to 6φ to ensure the strength. In this case, when the arc length is 5 mm in the graph (when the torch height is 5 mm), the arc voltage is increased even after the workpiece having a total thickness of 3 mm penetrates, and the arc voltage is larger than the circle in the graph. It is assumed that a voltage value (about 34.1 V: 6φ data) indicated by the square above is detected. When the workpiece penetrates in this way (the workpiece plate thickness is 3 mm in total), if the torch height is 5 mm, the hole diameter of the lower plate of the workpiece is 6φ at this detected voltage value (about 34.1 V: 6φ data). I understand that
Thus, for example, a calibration curve is created with a test piece having a hole diameter changed, such as 3φ or 6φ, and a threshold value for a voltage that guarantees the hole diameter is obtained. If the measured arc voltage is equal to or greater than the threshold value, a predetermined value is obtained. The hole diameter can be guaranteed.

  In the above description, the arc length is estimated from the graph when the arc length is 5 mm (that is, the torch height is 5 mm). However, the arc length may be 10 mm. . Further, by interpolating the data of these arc lengths on the graph, the diameter of the prepared hole can be detected even if it is between 5 and 10 mm or between 10 and 13 mm.

(Penetration detection using arc length-voltage data 23)
Using the arc length-voltage data 23, penetration can be detected by voltage increments as follows. For example, in the graph of FIG. 6, when the arc length is 5 mm as the starting point (when the torch height is 5 mm), when there is no gap between workpieces, the total thickness of 3 mm is dug. Since the necessary arc length is 8 mm (= 5 + 3), according to the calibration curve without holes, when a predetermined voltage value (about 33.7 V) is detected, the penetration is made. Therefore, in this case, when the initial voltage value (about 30.3 V) indicated by the triangle is detected and the voltage increases by about 3.4 V (= 33.7-30.3), the increment discriminating means 313 ( (See FIG. 5) can detect penetration.

  Further, when detecting penetration based on the voltage increase rate, the increase rate calculating means 314 (see FIG. 5) calculates the increase rate without using the arc length-voltage data 23, thereby increasing the rate determining means 315 (FIG. 5). 5) can detect, for example, inflection points shown in the graph of FIG. 2 and detect penetration. When the inflection point is detected when the torch height is 5 mm as in the previous example, the same predetermined voltage value (about 33.7 V) is detected.

  As described above, in the in-line inspection method and the in-line inspection apparatus according to the first embodiment of the present invention, the arc voltage when the through hole is formed by plasma arc welding is monitored and measured after a predetermined time has elapsed since the penetration. By comparing the measured arc voltage with the calibration curve data, the lower hole diameter of the through hole is detected. Further, by using the inline inspection method according to the first embodiment of the present invention, it is possible to detect the length and diameter of the through hole inline without using a new apparatus, and the quality assurance accuracy is improved.

Second Embodiment
In the in-line inspection method according to the second embodiment, the outline of the in-line inspection method and the configuration of the welding system are the same as those in the first embodiment, and thus description thereof is omitted.

[6. Configuration of inline inspection equipment]
A configuration of an inline inspection apparatus for performing the inline inspection method according to the second embodiment will be described with reference to FIG. 12 (refer to FIG. 5 as appropriate). In the inline inspection apparatus 10A shown in FIG. 12, the same components as those of the inline inspection apparatus 10 shown in FIG. The in-line inspection apparatus 10A is different from the in-line inspection apparatus 10 in the penetration detection means 31A and the through-hole diameter detection means 32A of the processing means 13. Below, for the convenience of explanation, 6-1. Configuration of penetration detection means 31A, 6-2. Specific example of operation of penetration detection means 31A, 6-3. Configuration of through-hole diameter detecting means 32A, 6-4. A description will be given in this order by dividing each section of a specific example of the operation of the through-hole diameter detecting means 32A.

(6-1. Configuration of penetration detection means 31A)
As illustrated in FIG. 12, the penetration detection unit 31 </ b> A includes a torch height calculation unit 311, an increase rate calculation unit 314, and an increase rate determination unit 315. In other words, as a method for detecting that the workpiece W has penetrated, only the configuration for detecting using the measured increase rate of the arc voltage is adopted, which is different from the penetration detecting means 31 in FIG.

  The increase rate calculating means 314 calculates the time change rate of the measured arc voltage at a predetermined sampling period. The time change rate (time differentiation) of the measured arc voltage has a plus sign (in the case of an increase rate) and a minus sign (in the case of a decrease rate).

  The increase rate discriminating means 315 discriminates whether or not the measured arc voltage has become an inflection point based on the increase rate calculated by the increase rate calculating means 314. It is determined that the workpiece W has penetrated (lower plate penetration detection step: see FIG. 2). The increase rate determining means 315 determines that an inflection point has been reached when the time derivative of the arc voltage exceeds a predetermined threshold. For example, if the voltage change value per second changes at a relatively small value, the voltage change value suddenly increases at a certain time, exceeds a predetermined threshold value, and then changes so as to decrease thereafter. The moment of exceeding is the turning point. When it is determined that the workpiece has penetrated, the increase rate determination unit 315 notifies the through hole diameter detection unit 32A to that effect.

(6-2. Specific example of operation of penetration detection means 31A)
FIG. 13A shows time series data of the arc voltage measured before and after the penetration of the three-layer workpiece under the predetermined welding conditions in the phase P1 of penetration welding (during plasma arc welding). Here, there was no gap between the workpieces, and the thicknesses of the three workpieces were 2.0 mm, 1.0 mm, and 1.2 mm from the bottom. As welding conditions, the plasma current was 150 A (constant). The plasma gas flow rate was 3 L / min. The sampling period for measuring the arc voltage was 2 milliseconds.

  In the graph of FIG. 13A, the horizontal axis indicates time, and the vertical axis indicates measured arc voltage (raw data). In the voltage waveform 131, a voltage value of about 27 V is observed in the time zone before the time of about 2.87 seconds (time divided by the one-point difference line 132), which is substantially constant compared to the subsequent time zone, Small noise vibrations occur with this constant voltage value as the center of vibration. Further, as shown in FIG. 13 (a), the voltage waveform 131 has a voltage higher than before in the subsequent time zone, and a small noise vibration having a larger amplitude than before occurs, while approximately 28 to A large vibration (swell) having a fluctuation range of 31 V is also generated.

  In the following, regarding the waveforms after the time divided by the one-point difference line 132, the noise vibration is a micro time unit vibration, the waviness is a macro time unit vibration, and the time unit larger than the waviness (about 1 second order). It is also called a global period.

FIG. 13A shows an image of arc leakage light observed from the back surface of the three-layer workpiece at a predetermined time in this case, together with a graph of the measured arc voltage.
When the time is about 2.7 seconds, as shown in the image 133, light starts to leak slightly into the position of the through hole.
When the time is about 2.87 seconds (time divided by the one-point difference line 132), as shown in the image 134, it can be seen that light has leaked to the position of the through hole and has penetrated. However, at this time, the hole diameter is not sufficient.
When the time is about 3.3 seconds, as shown in the image 135, it can be seen that the arc leaks from the lower hole of the through hole, and a sufficient hole diameter is secured.

  When a substantially constant voltage value is observed as the voltage waveform 131 as in the time zone before the time divided by the one-point difference line 132 in the graph of FIG. As shown, the excavation of the work has not reached the bottom plate of the lowest layer. Therefore, the molten metal of the workpiece cannot be melted under the hole. Then, the temperature is increased by the gas pressure (arc pressure) near the surface (upper surface) of the molten pool and the air expands, and the balance is maintained at a substantially constant pressure by this gas pressure and the surface tension of the molten pool. Therefore, a constant voltage of about 27V is observed. Then, for example, plasma hits the molten pool or a shield gas is blown, and the molten metal shakes, causing vibration on the surface of the molten pool. For this reason, noise vibration occurs in the measured arc voltage raw data.

In the time zone after the time divided by the one-point difference line 132 in the graph of FIG. 13A, the excavation of the work has reached the lower plate of the lowest layer, and the hole diameter as shown in the image 134 is enlarged. There is a state in which a sufficient hole diameter is secured as shown in an image 135.
Among these, during the expansion of the hole diameter, the metal around the through-hole melts from the next while the molten metal of the workpiece melts under the hole as the hole diameter increases. Surface vibration becomes intense. At this time, the surface of the molten pool vibrates up and down greatly in micro time units (noise vibration), but gradually moves to a deeper position of the hole as time elapses in macro time units. Swelling occurs. If the arc voltage is monitored with a short sampling period as in the measurement conditions of this measurement example, fine voltage fluctuations can be picked up with high accuracy, but sudden fluctuations due to arc or molten pool instability are also picked up at the same time. Will end up. In the global period from the detection of penetration to the detection of the diameter of the through-hole while greatly oscillating at a higher voltage value than before penetration, a voltage increasing tendency is observed in the voltage waveform. In addition, since the balance between the gas pressure and the surface tension of the molten pool is steadily lost after securing the hole diameter, the measured arc voltage raw data includes vibrations with large micro and macro amplitudes.

  By comparing the arc voltage measured in this way with a predetermined threshold value, penetration detection is possible. However, since the waveform of the measured arc voltage raw data has an amplitude of vertical movement, the comparison result with a predetermined threshold value may vary depending on the timing at which the arc voltage is measured. Therefore, in the second embodiment, the penetration detection unit 31A detects the arc voltage with accuracy based on time differentiation (voltage increase rate).

  FIG. 13B shows time-series data of time differentiation of the arc voltage calculated by the increase rate calculation means 314 of the penetration detection means 31A in the case of the example of FIG. The horizontal axis of the graph in FIG. 13B indicates time, which coincides with the scale on the horizontal axis of the graph in FIG. The vertical axis of the graph of FIG. 13B shows the voltage change (time differential value) per unit time of the measured arc voltage (raw data).

  In the waveform 136 (time differential value) of the time differential value of the voltage, in the time zone before the time of about 2.87 seconds (time divided by the one-point difference line 132), in the case of a plus sign (in the case of an increase rate) Although there is a case of a minus sign (in the case of a decrease rate), a state of almost constant change rate is observed compared to the subsequent time zones, and a change rate of about 0 is observed. Has occurred. In the subsequent time zone, the waveform 136 of the time differential value of the voltage has a vibration with a larger amplitude than before, with the rate of change 0 being the center of vibration. A large voltage increase rate exceeding a voltage change of 50 V / s is observed at the time divided by the one-point difference line 132.

  Therefore, in the case of this example, the increase rate determination unit 315 of the penetration detection unit 31A uses the time threshold value of the arc voltage calculated by the increase rate calculation unit 314 on the basis of a predetermined threshold (for example, 50 V / s). From this, it is possible to accurately detect the penetration of the workpiece without using the raw data of the arc voltage by determining that the voltage-time differential value has reached an inflection point that is a point in time exceeding a predetermined threshold value.

  Moreover, the diameter of the through hole can be detected by comparing the measured arc voltage with a predetermined threshold value. However, the waveform of the arc voltage raw data measured after the penetration detection has a more up-and-down movement amplitude than before the penetration detection, so that it is more difficult to compare with a predetermined threshold value than before the penetration detection. Therefore, in the second embodiment, the through-hole diameter detecting means 32A detects with high accuracy based on the cumulative moving average.

(6-3. Configuration of the through-hole diameter detecting means 32A)
As shown in FIG. 12, the through-hole diameter detecting means 32A includes cumulative moving average calculating means 321 and collating means 322.
The cumulative moving average calculation means 321 calculates a cumulative moving average CA (Cumulative Moving Average) for the measured arc voltage. The cumulative moving average calculating means 321 calculates the cumulative moving average at a predetermined sampling period (for example, 1 to 10 ms).
For example, it is known that the cumulative moving average CA _i of _i measurement values x ₁ ,..., X _i from the first to the i-th is expressed by the following equation (1).

For example, when the (j + 1) -th measurement value x _{j + 1} is input next to the j-th measurement value x _j , the (j + 1) -th (j + 1) -th measurement value x ₁ ,. , x _j, cumulative moving average CA j _{+ 1} of x j _{+ 1} is represented by formula (2) of the recurrence formula using the accumulated moving average CA _j.

  Here, the calculated values when j = 0, 1, 2,... In the equation (2) are the same as the calculated values when i = 1, 2, 3,. In other words, Equation (2) indicates that the cumulative moving average at that time can be calculated while performing the operation of accumulating the measurement values in the measurement sampling period.

The cumulative moving average calculating means 321 detects the cumulative moving average CA _i of the equation (1) by substantially calculating the equation (2) for the measured arc voltage. In the present embodiment, the cumulative moving average calculating means 321 starts the calculation of the cumulative moving average triggered by the detection of penetration by the penetration detecting means 31A. The cumulative moving average calculation means 321 outputs the cumulative moving average data of the detected arc voltage to the matching means 322.

  Similar to the through hole diameter detecting means 32 in FIG. 5, the collating means 322 collates the measured arc voltage with the lower hole diameter-voltage data (calibration curve data) 21 to detect the lower hole diameter of the formed through hole. To do. The collating means 322 does not use the raw data of the measured arc voltage but uses the cumulative moving average data of the arc voltage detected by the cumulative moving average calculating means 321 to prepare the lower hole diameter-voltage data (calibration curve data). 21 is different from the first embodiment in that it is collated. The calibration curve data is also obtained in advance using the accumulated moving average data of the arc voltage, not using the raw data of the measured arc voltage. Since the collation method with the calibration curve data is the same as that in the first embodiment, the description is omitted.

(6-4. Specific example of operation of through-hole diameter detecting means 32A)
Here, taking the specific example of the through welding described in Section 6-2, the operation of the through hole diameter detecting means 32A will be described with reference to FIG. 14 (see FIGS. 12 and 13 as appropriate).
Here, in order to compare with the through-hole diameter detecting means 32A, it is assumed that a moving average SMA (Simple Moving Average) is calculated. Hereinafter, an apparatus for calculating a moving average SMA for a measured arc voltage. Is called a comparative example.

For example, a moving average SMA _M identified by M, n number of measured values _{p M, p M-1,} ..., the moving average SMA _M of p _{M- (n-1),} the following equation (3) It is known that Here, n is predetermined by the sampling period of measurement. In the case of this comparative example, n is assumed to be about 5 to 10. For example, if n is 10, Expression (3) is described as Expression (3a).

Then, as shown in Equation (4), for example, when the moving average SMAM identified by _M is obtained, a new measurement value p _{M + 1} is added, and the oldest measurement value p _{M− (n−} By averaging the measured values of the new n combinations excluding ₁₎ , for example, the moving average SMAT identified by _T can be obtained.

FIG. 14A shows data of the moving average SMA of the measured arc voltage obtained by calculating the equation (3a) from the actually measured voltage value shown in FIG.
The voltage moving average SMA waveform 141 is obtained from the voltage waveform 131 shown in FIG. In FIG. 14, the time (about 2.87 seconds) divided by the one-point difference line 142 is the same as the time divided by the one-point difference line 132 in FIG. Overall, it can be seen that the waveform 141 of the moving average SMA of the voltage has a smaller degree of minute amplitude such as noise than the waveform 131 of the voltage.

  However, in the waveform 141 of the moving average SMA of the voltage, the vibration in the micro time unit is smoothed in the time zone after the time of about 2.87 seconds (the time divided by the one-point difference line 142). Similar to the voltage waveform 131 (raw data) shown in FIG. 13A, a large amplitude vibration (swell) in a macro time unit occurs. In addition, the trend of the entire waveform during the global period is more difficult to understand than the raw data. Therefore, it is considered that there are many cases where the comparison result with the predetermined threshold value differs depending on the arc voltage measurement timing only by using the moving average SMA data of the measured arc voltage as in the comparative example.

On the other hand, the cumulative moving average calculating means 321 of the through hole diameter detecting means 32A shown in FIG. 12 calculates the cumulative moving average CA for the measured arc voltage.
FIG. 14B shows measured voltage values in a time zone 143 indicated by a two-dot chain line in FIG. 14A, that is, a time zone of 2.5 to 3.5 seconds. A voltage waveform 131a shown in FIG. 14B is an enlarged view of a part of the voltage waveform 131 shown in FIG. 13A (a time zone of 2.5 to 3.5 seconds).

The cumulative moving average calculation means 321 starts the calculation of the cumulative moving average at about 2.87 seconds (the time divided by the one-point difference line 142 in FIG. 14A). The waveform 144 of the cumulative moving average of voltage shown in FIG. 14B appears after about 2.87 seconds.
Overall, it can be seen that the noise of the waveform 144 of the cumulative moving average of the voltage is significantly smaller than the waveform 131a of the arc voltage (raw data). Further, since one scale on the vertical axis of the graph of FIG. 14B is about half that of the graph of FIG. 14A, the waveform 144 of the cumulative moving average of the voltage is compared with the waveform 141 of the SMA according to the comparative example. It can be seen that the noise amplitude is approximately halved.

  Further, as shown in FIG. 14B, macro time unit vibration (swell) is almost removed from the waveform 144 of the cumulative moving average of voltage. Furthermore, the waveform 144 of the cumulative moving average of the voltage well reproduces the tendency of the entire waveform of the raw voltage data generated along with the increase in the hole diameter after passing through the workpiece in the global period. That is, according to the waveform 144 of the cumulative moving average of the voltage, the arc voltage tends to increase in the global period after the penetration detection as well as the effect of reducing noise and swell compared to the SMA waveform 141 according to the comparative example. It can be understood that the data after the analysis processing can be retained without losing the information of the observation data that exists in the data.

  Since the overall flow of processing by the inline inspection apparatus 10A according to the second embodiment is the same as that of the first embodiment, description thereof is omitted.

  As described above, according to the inline inspection method according to the second embodiment, the inline inspection apparatus 10A calculates the cumulative moving average data for the measured arc voltage, so from the viewpoint of the moving average processing of the cumulative moving average It is possible to have both the effect of processing the small noise of the voltage raw data and the effect of processing the undulation of the voltage raw data from the viewpoint of the cumulative average processing of the cumulative moving average. Therefore, when the arc voltage is monitored with a short sampling period in in-line inspection, it is assumed that noise and undulation occurred in the waveform by picking up sudden fluctuations due to arc and molten pool instability along with fine voltage fluctuations. However, it is possible to generate data that can be easily compared with the calibration curve data by the cumulative moving averaging of the arc voltage. Therefore, the matching means 322 of the through-hole diameter detecting means 32A compares the detected cumulative moving average data with the calibration curve data without using the raw data of the arc voltage whose amplitude of vertical movement is larger than before the penetration detection. The through-hole diameter can be detected with high accuracy.

(Modification of the second embodiment)
An in-line inspection method in penetration welding when there is a gap between workpieces will be described as a modification. FIG. 15A shows time series data of the arc voltage measured before and after the penetration of a three-layer workpiece under a predetermined welding condition in the phase P1 of penetration welding (during plasma arc welding). Here, the thicknesses of the three workpieces are 2.0 mm, 1.0 mm, and 1.2 mm from the bottom, for example, a gap of 1.0 mm between the first workpiece from the bottom and the workpiece above it. It was assumed that there was a (gap). Except for the presence of a gap, the conditions are the same as the example described with reference to FIG.

  As shown in FIG. 15A, in the voltage waveform 151, the time before the time of about 2.7 seconds (the time divided by the one-dotted line 152) is substantially constant compared to the subsequent time. A voltage value of approximately 28 V is observed, and a minute vibration having the constant voltage value as the center of vibration is generated. Further, as shown in FIG. 15 (a), the voltage waveform 151 has a voltage higher than before in the subsequent time zone, and a slight vibration with a larger amplitude than before occurs, while approximately 29 to 32V. There is also a large swell with a fluctuation range of.

FIG. 15A shows an image of arc leakage light observed from the back surface of the three-layered workpiece at a predetermined time in this case, together with a graph of the measured arc voltage.
When the time is about 2.6 seconds, as shown in the image 153, light begins to leak slightly into the position of the through hole.
When the time is about 2.8 seconds, as shown in an image 154, it can be seen that light has leaked to the position of the through hole and has penetrated.
When the time is about 3.5 seconds, as shown in the image 155, it can be seen that an arc leaks from the lower hole of the through hole, and a sufficient hole diameter is secured.

  As shown in FIG. 15B, the waveform 156 (time differential value) of the voltage time differential value has a time zone before the time of about 2.7 seconds (time divided by the one-point difference line 152). Compared to the subsequent time zones, a state of almost constant change rate 0 is observed, and a minute vibration having the change rate 0 as the center of vibration is generated. In the subsequent time zone, the waveform 156 of the time differential value of the voltage has a vibration with a larger amplitude than before, with the rate of change 0 being the center of vibration. A large voltage increase rate exceeding a voltage change of 50 V / s is observed at the time divided by the one-point difference line 152. Therefore, even in the case of this modification, the increase rate determination means 315 of the penetration detection means 31A can accurately detect penetration of the workpiece.

FIG. 16A shows the moving average SMA data of the measured arc voltage obtained by calculating the above-described equation (3a) from the actually measured voltage value shown in FIG.
The voltage moving average SMA waveform 161 is obtained from the voltage waveform 151 shown in FIG. In FIG. 16, the time (about 2.7 seconds) partitioned by the one-point difference line 162 is the same as the time partitioned by the one-point difference line 152 in FIG. In FIG. 16, the time (about 2.8 seconds) divided by the one-point difference line 163 is the same as the time when the image 154 is observed in FIG. The waveform 161 of the voltage moving average SMA shows a noise reduction effect as a whole, and in the time zone after the time of about 2.7 seconds (the time divided by the one-point difference line 162), it is as large as the raw data. An amplitude vibration has occurred. The appearance of the peak of the waveform 161 at the time of about 2.8 seconds is considered to be due to the gap between the workpieces. If the through-hole diameter is to be detected by moving average SMA, this peak may be erroneously detected.

  FIG. 16B shows measured voltage values in a time zone 164 indicated by a two-dot chain line in FIG. 16A, that is, a time zone of 2.5 to 3.5 seconds. A voltage waveform 151a shown in FIG. 16B is an enlarged view of a part of the voltage waveform 151 shown in FIG. 15A (a time zone of 2.5 to 3.5 seconds).

The cumulative moving average calculating means 321 starts the calculation of the cumulative moving average at a time of about 2.7 seconds (the time divided by the one-point difference line 162 in FIG. 16A). A waveform 165 of the cumulative moving average of voltage shown in FIG. 16B appears after about 2.7 seconds.
Overall, it can be seen that the noise in the cumulative moving average waveform 165 of the voltage is significantly smaller than the waveform 151a of the arc voltage (raw data). It can also be seen that the noise amplitude is approximately halved even when compared with the SMA waveform 161 of the comparative example.

  Also, as shown in FIG. 16B, the influence of the peak appearing in the waveform 165 of the SMA according to the comparative example is removed from the waveform 165 of the cumulative moving average of the voltage. That is, the waveform 165 did not show a conspicuous peak at a time of about 2.8 seconds. And the voltage increase tendency of the whole waveform of the raw data of the voltage which arises with the hole diameter expansion after a workpiece | work penetration is reproduced well. Therefore, even when there is a gap between the workpieces, the in-line inspection method in penetration welding can achieve the same effect as when there is no gap between the workpieces.

Moreover, even when the inline inspection method according to the first embodiment is applied when there is a gap between the workpieces, the same effect as when there is no gap between the workpieces can be obtained.
Further, by using the in-line inspection method according to each embodiment of the present invention for the plasma-MIG welding method in which plasma arc welding and MIG welding are combined, a welding system for performing this plasma-MIG welding method ( For example, in FIG. 4), the quality assurance accuracy can be improved in the same manner as each embodiment.

  As mentioned above, although preferable embodiment of the in-line test | inspection apparatus of this invention was described, this invention is not limited to each above-described embodiment. For example, the in-line inspection apparatus 10 (10A) of each embodiment has been described in the best mode including the through detection means 31, the through hole diameter detection means 32 (32A), and the through hole diameter guarantee means 33, but the present invention is limited to this. Instead of this, the penetration detecting means 31 and the through hole diameter detecting means 32 (32A) may be provided, and arc re-ignition may not be performed. When arc re-ignition is not performed, that is, when the diameter of the through hole of the through hole is not re-detected, the configuration becomes simple, the in-line inspection can be performed quickly, and the manufacturing cost can be reduced.

  On the other hand, there is an advantage that in-line inspection can be performed with high accuracy when arc re-ignition is performed, that is, when the through hole diameter of the through hole is re-detected. Here, the reason why the arc can be re-ignited with high accuracy is obvious from the graphs of FIGS. In the graphs of FIGS. 7 and 8, the horizontal scale is substantially the same, but the vertical scale of the voltage value is greatly different. Specifically, in the graph of FIG. 7, for example, when the torch height is 5 mm, the voltage between the arc voltage value when a 3φ test piece is used and the arc voltage value when a 6φ test piece is used. The change width 701 was about 2V. On the other hand, a similar voltage change width 801 in the graph of FIG. 8 is about 50V. Therefore, the rising voltage collating means 332 of the through hole diameter guaranteeing means 33 can detect the lower hole diameter with higher accuracy than the through hole diameter detecting means 32.

  Moreover, in each embodiment, although the in-line inspection apparatus 10 was provided separately from the welding power supply 5 and the welding control apparatus 7, it is not limited to this, You may incorporate in the welding power supply 5 or the welding control apparatus 7. Of course, they may be incorporated into one unit.

DESCRIPTION OF SYMBOLS 1 Welding system 2 Welding torch 3 Robot 4 Robot control apparatus 5 Welding power supply 6 Filler supply apparatus 7 Welding control apparatus 8 Plasma torch 9 MIG torch 10, 10A Inline inspection apparatus 11 Voltage monitoring means 12 Storage means 13 Processing means 21 Lower hole diameter-Voltage Data (calibration curve data)
22 Re-ignition data (calibration curve data)
23 Arc length-voltage data (calibration curve data)
31, 31A Penetration detection means 311 Torch height calculation means 312 Increment calculation means 313 Increment discrimination means 314 Increase rate calculation means 315 Increase rate discrimination means 32, 32A Through hole diameter detection means 321 Cumulative moving average calculation means 322 Verification means 33 Through hole diameter guarantee Means 331 Arc ignition control means 332 Rising voltage collating means 333 Hole diameter confirmation means 51 Plasma power supply 52 MIG power supply 53 Gas supply device 54 Voltage detector W Workpiece

Claims (6)

An in-line inspection method in an in-line inspection apparatus that inspects through holes in-line when through welding a plurality of stacked workpieces,
The in-line inspection apparatus is
Storage means for storing calibration curve data indicating the relationship between the prepared hole diameter of the through-hole formed in the workpiece and the arc voltage of plasma arc welding, which is obtained in advance for each workpiece thickness and welding condition; With
The processing means includes
A penetration detection step of monitoring the arc voltage during plasma arc welding and detecting that the plurality of superimposed workpieces have penetrated based on the arc voltage;
A through-hole diameter detection step of monitoring the arc voltage and detecting the prepared hole diameter of the through-hole formed by comparing the arc voltage measured after a predetermined time has passed since the penetration of the plurality of workpieces with the calibration curve data. And an in-line inspection method. An in-line inspection method in an in-line inspection apparatus that inspects through holes in-line when through welding a plurality of stacked workpieces,
The in-line inspection apparatus is
Storage means for storing calibration curve data indicating the relationship between the prepared hole diameter of the through-hole formed in the workpiece and the arc voltage of plasma arc welding, which is obtained in advance for each workpiece thickness and welding condition; With
The processing means includes
A penetration detection step of monitoring the arc voltage during plasma arc welding and detecting that the plurality of superimposed workpieces have penetrated based on the arc voltage;
The arc voltage is monitored, the accumulated moving average data of the arc voltage is detected, and the arc voltage accumulated moving average data measured after a predetermined time has passed since the penetration of the plurality of workpieces is collated with the calibration curve data. And an in-line inspection method characterized by executing a through-hole diameter detecting step of detecting a prepared hole diameter of the formed through-hole.   The calibration curve data is determined for each plate thickness and welding condition of the plurality of superimposed workpieces, and the inner diameter of the hole on the back surface side when a through hole of a desired size is formed in the plurality of superimposed workpieces. 3. The lower hole diameter-voltage data indicating a correspondence relationship between a prepared lower hole diameter and a stable voltage indicating an arc voltage when an arc ignited in the through hole is stabilized. Inline inspection method as described in. The storage means of the in-line inspection apparatus uses, as a calibration curve data obtained in advance for each workpiece thickness and welding condition, plasma arc welding to through holes of a desired size already formed in the plurality of superimposed workpieces. Re-ignition data indicating the correspondence between the arc voltage within the predetermined dielectric breakdown time and the diameter of the pilot hole is further stored before the arc at the time of voltage rise when the arc is re-ignited and before the arc is stabilized And
The processing means includes
After detecting the lower hole diameter of the through-hole in the through-hole diameter detecting step, the arc is once cut off, and immediately after that, a re-igniting step of re-igniting the plasma arc welding arc;
Performing a re-detection step of detecting a prepared hole diameter of the through-hole by comparing arc voltage after re-igniting the arc with re-ignition data which is the calibration curve data. The in-line inspection method according to claim 3. An in-line inspection device for in-line inspection of through-holes when through-welding a plurality of stacked workpieces,
Storage means for storing calibration curve data indicating the relationship between the diameter of the through hole formed in the workpiece and the arc voltage of plasma arc welding, which is obtained in advance for each plate thickness and welding condition of the workpiece;
Voltage monitoring means for monitoring the arc voltage during plasma arc welding and measuring the arc voltage at a predetermined frequency;
A penetration detection means for detecting that the plurality of superimposed workpieces have penetrated based on the measured arc voltage;
Based on the measured arc voltage, by comparing the arc voltage measured after a predetermined time has passed since the penetration of the plurality of workpieces with the calibration curve data, the through hole diameter for detecting the lower hole diameter of the formed through hole is detected. Detection means;
An in-line inspection apparatus comprising: A plasma-MIG welding method combining plasma arc welding and MIG welding,
Forming a through-hole by a plasma arc in a plurality of stacked workpieces;
Inspecting the through hole in-line by the in-line inspection method according to any one of claims 1 to 4,
And a step of welding the plurality of stacked workpieces by filling the through-holes by MIG welding.