JPS62108154A - Method for diagnosing weld zone in flash butt welding - Google Patents

Abstract

PURPOSE:To make highly reliable nondestructive diagnoses on weld zone in flash butt welding in real time, by counting acoustic emission (AE) quantities mainly proportional to the quantity of welding pressure and another AE quantities mainly proportional to the quantity of heat input after completing welding and diagnosing the weld zone in accordance with a previously set quality discriminating chart by using both AE quantities. CONSTITUTION:Clamping sections 12a and 12b respectively hold round steel bars 10a and 10b. AE signals detected by AE sensors 13a and 13b provided at the clamping sections 12a and 12b are respectively given to band-pass filters 14a and 14b and only AE signals of a prescribed frequency band are outputted, amplified 15a and 15b, and discriminated 16a and 16b. As a result, only AE signals higher than a threshold level are supplied to a time difference measuring section 17. Then AE producing quantities in the vicinity of a weld zone 11 are counted by means of an AE counter 18 in accordance with the difference in reaching time found by the measuring section 17 and the counted AE producing quantities are given to a quality discriminating circuit 19. At the discriminating circuit 19 the quality of the weld zone 11 is discriminated by checking the given counted AE producing quantities with a quality discriminating chart which is previously stored in a data table 20, and the discriminated results are outputted.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for diagnosing a welded part of spark butt welding.
Especially acoustic emission (Acoustic
The present invention relates to a diagnostic method using AE (hereinafter abbreviated as AE) technology.

[Conventional technology]

In recent years, spark butt welding has been used in aircraft, automobiles, etc. because the weld metal is pushed out from the welding surface during pressure welding, resulting in a clean welded joint, and the welding process is usually completed in a short time, making it extremely efficient. It has been widely put into practical use in the vehicle industry and construction field.

FIG. 5 is for explaining spark butt welding,
In the figure, 1 is a controller, 2 is a transformer, and 3a, 3b and 4a, 4b are each a pair of electrodes.

5 is a member to be welded on the fixed side, and 6 is a member to be welded on the movable side. When welding, the welding surfaces of the two-liquid welding members 5.6 are brought into light contact and a large current is passed through the action of the controller 1 to the roller 1 and the transformer 2, generating electric sparks 7 on the contact surfaces and dissipating the heat. The members to be welded 5 and 6 are heated, and then, with the welding surfaces covered with metal scrap vapor and molten metal, a sudden strong pressure is applied to the movable member to be welded 6 to press them together.

By doing this, welding of the members to be welded 5 and 6 is completed.

(Problems to be Solved by the Invention) By the way, when diagnosing the welded part of the above-mentioned spark butt welding, non-destructive testing methods such as radiographic testing or ultrasonic testing are generally applied. Testing methods cannot diagnose in real time, and diagnosis is sometimes difficult when the shape of the welded part is complex. Because of this, the flash voltage.

The quality of welds was sometimes evaluated by controlling and monitoring various welding processes such as flash current, flash amount, upset energization time, upset 5, etc., but this method was not sufficient to check the presence or absence of weld defects. It was not possible to make a judgment with high reliability. Furthermore, material testing by sampling was also conducted, but this did not improve reliability.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a method for diagnosing a weld in spark butt welding, which can non-destructively diagnose a weld in spark butt welding in real time and with high reliability.

[Means for solving problems]

In order to solve the above problems and achieve the objectives, the present invention has the following features:
AE after spark butt welding is mainly proportional to the amount of pressurization
I and AEI, which is mainly proportional to the amount of heat input, are counted, and the welded part is diagnosed based on a pass/fail identification diagram set in advance using both AEffl.

[Effect]

By taking such measures, the AE fog generated in the welding part during spark butt welding is counted, and the AEE quantity, which is mainly proportional to the amount of pressurization, that is, the upset, and the AE counting color, which is mainly proportional to the amount of heat input, are preset. If the layer area in the pass/fail identification diagram is satisfied, the weld is evaluated to be good, and if it is included in the defective area, it is evaluated to be a weld defect.

〔Example〕

An embodiment of the method of the present invention will be described below with reference to the drawings. In this embodiment, an AE sensor is provided in the vicinity of the weld, and this detects an AE multiplied signal after spark butt welding is completed, and the quality of the weld is determined based on the occurrence of this AE signal.

FIG. 1 shows the AEE generation pattern near the weld during general spark butt welding.

As is clear from the figure, after the end of butt welding at time t○, the AE multiplied signal is generated, and after the AEE signal generation amount reaches the minimum value at time t1, the AE multiplied signal is generated again until time t2. do. Here, time t.

The amount of AE multiplied signal generated between tl and tl is mainly proportional to the amount of pressure (upset amount) during spark butt welding, and the amount of AE multiplied signal generated between time t1 and t2 is mainly proportional to It is proportional to the amount of heat input during spark butt welding. That is, in this embodiment, from this time point 10, t
The weld zone diagnosis is performed based on the AEE yield up to 1 and the AE occurrence M from time t1 to t2.

Note that the above points t1 and t2 have different values depending on the shape (steel bar, pipe, rail, etc.) of the workpiece to be welded, its dimensions, material, etc., but if they are the same shape, same dimensions, and made of the same material, they are almost constant values. Become.

Figure 2 shows the flash voltage, flash current, and flash amount when a round steel bar is selected as the workpiece to be welded.

When spark butt welding was performed with various welding process values such as upset energization time and upset amount, the AE count value (horizontal axis) 10 seconds after the welding completion, that is, from 10 to tl in Fig. 1, , and thereafter, that is, the AE count fl (vertical axis) of 10 in FIG. 1 is shown as a quality recognition diagram. If the welded part is found to be defective due to the presence of defects in the material test results after welding is completed, it will be indicated by a "black circle" in the diagram, and if it is determined to be good, it will be indicated by a "white circle" in the diagram. ”.

As shown in the figure, when the relationship between both AE counts is located in the region below curve A-A, the entire weld part 1 is defective, and the region above curve IB-8 is defective. All welds located in the area are in good condition. Also, the song IA-A
The area between and curve B-B is a boundary area between good and bad.

That is, the amount of AEE generated from the spark butt welding end time tQ to tl and the occurrence of AE generation after the above t1 are counted,
By applying these counts of both AE occurrences to the pass/fail identification diagram, it is possible to diagnose the welded portion in real time and with high reliability using the non-destructive fishing force method. Note that the time corresponding to time t2 in FIG. 1 is at most 300 to 400 seconds, and the diagnostic result can still be obtained quickly.

FIG. 3 is a schematic diagram showing the system configuration for realizing this embodiment. In FIG. 3, 10a.

Reference numeral 10b is a round steel bar serving as a member to be welded, and the welding portion 11 is welded by spark butt welding. Reference numerals 12a and 12b are clamp parts that hold the round bars m10a and 10b, respectively, and AE sensors 13a and 13b are provided on the clamp parts 12a and 12b, respectively. The AE multiplied signals detected by these AE sensors 13a and 13b are applied to band pass filters 14a and 14b, respectively, and only the AE multiplied signals in a predetermined frequency band are outputted to amplifiers i5a and 15b. and,
The AE multiplied signals in a predetermined frequency band amplified by these amplifiers 15a and 15b are identified at a predetermined threshold level by respective identification circuits 16a and 16b, and only signals above the threshold level are sent to the time difference measuring section 17. Supplied. This time difference measuring section 17 operates between the AE sensors 12a and 1 based on internal input signals.
2b and AE multiplied signal arrival time difference, and based on the arrival time difference determined by the time difference measurement unit 17, the AE counter 18 determines the arrival time difference of the AE multiplied signal reaching the welding part 11.
Farms with AE outbreaks in the vicinity are counted.

The count output of this AE counter 18 is given to a pass/fail judgment circuit 19, and the pass/fail judgment circuit 19 calculates the AE count within the time period corresponding to tl from to in Fig. 1 immediately after the completion of welding, and from t1 in Fig. 1. The AE counting amount within the time corresponding to t2 is calculated, and these two AE counting amounts are stored in the data table 20.
The quality of the welded portion 11 is determined by applying it to the quality identification diagram shown in FIG. 2, which is stored in advance, and the determination result is output.

With the system configuration as described above, the round steel bars 10a, 1
The state of the welded portion 11 of 0b can be diagnosed non-destructively in real time. Also, a round bar! j110a, 10b
A clamp portion 12a that holds the.

Since only the AE sensors 13a and 13b are provided on the welding part 12b, the welded portion can be diagnosed regardless of the shape of the welded member.

Note that the present invention is not limited to the above embodiments.

For example, in the above embodiment, welding part diagnosis was explained when spark butt welding was performed on round steel bars as workpieces, but each pass/fail identification line is determined depending on the size and shape of the workpiece (rail, vibrator, etc.) By obtaining the diagram in advance (thereby, welding part diagnosis in spark butt welding of various workpieces can be performed nondestructively and in real time. Also, in Figure 3, AE sensor 13a
, 13b are provided on the clamp parts 12a, 12b, respectively. For example, as shown in FIG. Welding part 11
An AE sensor 13e is provided near the AE sensor 13c,
13d removes noise from the clamp parts 12a and 12b, and the AE sensor 13e removes noise from the welding part 11.
It is also possible to detect a nearby AE multiplied signal.

It goes without saying that various other modifications can be made without departing from the spirit of the invention.

〔Effect of the invention〕

As described in detail above, the present invention counts the amount of AE generated which is mainly proportional to the amount of pressurization after the completion of spark butt welding, and the amount of AE generated which is mainly proportional to the amount of heat input, and calculates the amount of AE generated in advance based on these two amounts of AE generated. The welded part is diagnosed based on the set pass/fail identification diagram.

Therefore, according to the present invention, the amount of AE generated in the welded part during spark butt welding is counted, and the AE counting amount is mainly proportional to the amount of pressurization, and the amount of AE generated is mainly proportional to the amount of heat input.
If the t quantity satisfies the good area in the preset pass/fail identification diagram, the weld is evaluated to be good, and if it is included in the bad area, it is evaluated to be a weld defect. It is possible to provide a weld part diagnosis method for spark butt welding that can nondestructively diagnose a weld part in welding in real time and with high reliability.

[Brief explanation of drawings]

1 to 3 are diagrams for explaining one embodiment of the present invention, in which FIG. 1 is a diagram showing a general AE signal generation pattern, and FIG. 2 is a diagram showing a pass/fail identification diagram. , Fig. 3 is a system diagram showing the system configuration for realizing this embodiment, Fig. 4 is a schematic diagram showing a modified example of the mounting position of the AE sensor, and Fig. 5 is a schematic diagram for explaining spark butt welding. be. 10a, 10b... Round steel bar (member to be welded), 11...
・Welding part, 12a, 12b... Clamp part, 13a~
13e...AE sensor, 14a, 14b...Band pass filter, 15a, 15b...Signal amplifier, 16
a, 16b...Identification circuit, 17...Time difference measuring section,
18... AE counter, 19... Pass/fail judgment circuit, 2
0...Data table. Applicant's agent Patent attorney Takehiko Suzue Time□ frX iFigure u Figure 3

Claims (1)

[Claims] After spark butt welding is completed, the amount of acoustic emission, which is mainly proportional to the amount of pressure applied, and the amount of acoustic emission, which is mainly proportional to the amount of heat input, are counted, and based on a preset pass/fail identification diagram based on these two amounts of acoustic emission. A method for diagnosing a welded part of spark butt welding, characterized in that the welded part is diagnosed using the same method.