JPS5828534B2 - Dempo-Koukanyou-Setseigiyo-Souchi - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an apparatus for detecting discoloration of a welded part with high precision and controlling welding conditions during welding of electric resistance welded steel pipes.

There is a close relationship between the conditions of the weld zone (discoloration, temperature, bead shape, etc.) and the quality of the weld zone during welding of ERW steel pipes, and the welding conditions are normally controlled depending on the condition of the weld zone. It is necessary to quantitatively understand the situation in the department.

One way to understand the temperature of the weld zone is to measure the absolute value of the amount of radiated infrared rays, but in this case, the measured value is affected by the influence of the weld zone shape, fluctuations in infrared radiation efficiency, and the influence of the surrounding atmosphere. It is not practical because it has the problem of fluctuating and low accuracy.

As a measurement method that overcomes the disadvantages of absolute value measurement as mentioned above, there is a method using a two-color thermometer that relatively captures the light intensity of each wavelength component of color loss as a wavelength spectrum, but in this case, it is possible to measure only two specific wavelengths. Because it is a relative ratio, it is susceptible to noise.
Also, the sensitivity is low depending on the operating temperature.

In other words, in the case of a two-color thermometer, t is the noise, α is the detection signal (signal plus noise), T1 and T2, and only the signal component is 8.
1, S2, the temperature T becomes, and there is no means to remove the noise α (if T2−
If T1=S2-S, the same sensitivity problem as infrared absolute value measurement occurs).

Therefore, a two-color thermometer is inferior to a naked eye thermometer or an infrared thermometer, and malfunction can be fatal, especially when using a thermometer.

In addition, high frequency high power (e.g. 4
In the field where 50 IG (z, 300 KW) is used, a certain amount of noise cannot be erased no matter how much noise is used, and it is impossible to separate the signal and noise with a two-color thermometer that does not have signal redundancy. It is.

On the other hand, there is a method of performing a destructive test and an ultrasonic test of the welded part after welding, but this method cannot be applied to automatic control during welding because the feedback information is slow.

For this reason, conventionally, the welding conditions during welding of ERW steel pipes were visually checked by the welding machine operator, and the welding conditions were manually controlled by observing discoloration and bead shape. There was a problem with a large error.

The present invention eliminates these drawbacks, detects discoloration of the welded part with high precision during welding of electric resistance welded steel pipes, and makes it possible to control welding conditions.

The reason why the operator of the welding machine can estimate the temperature by looking at the discoloration is that the discoloration of the weld zone is similar to blackbody radiation, and the temperature is uniquely determined by the wavelength characteristics (this is based on this).

Therefore, by measuring the spectral characteristics of color loss, the temperature of the weld zone can be determined.

Based on this idea, the device of the present invention separates the color loss into spectra, electrically scans the spectrum at a very high speed and with high precision using a group of photoelectric conversion elements, thereby obtaining the wavelength pattern of the spectrum, and based on this wavelength pattern. This is to control welding conditions.

In other words, since conventional spectrometers are used in laboratories, emphasis is placed on increasing wavelength resolution, and wavelength scanning is performed mechanically at extremely low speed, making it difficult to scan the entire visible light range once. It took about 10 minutes.

In this case, there is almost no hope for application to processes.

The present invention performs l m5ec by performing electrical scanning of photoelectric conversion elements instead of mechanical scanning.
This system scans the visible range in about 1/1000 seconds, making it possible to use a spectrometer in the process.

An embodiment of the present invention will be described below with reference to the drawings.

Reference numeral 7 indicates an electric resistance welded steel pipe that is being formed by welding, and is squeezed while being fed in the direction of the arrow by a squeeze roll 6.

Reference numeral 5 denotes a welded part, and the discolored light of the welded part is collected by a condensing device 1 and sent to a spectroscope 3 through a fiberscope 2.

The spectrometer is fixed and uses, for example, a grating or a prism to wavelength-resolve the collected light.

The entire wavelength range of the wavelength-resolved spectrum then enters the photoelectric conversion element group 4 simultaneously.

The photoelectric conversion element group 4 is composed of a large number (512 in this example) of photoelectric conversion elements, and uses, for example, a photodiode linear array, an image detector, or an ITV camera.

The photoelectric conversion element group 4 electrically scans the spectrum at high speed, and its output becomes an electric signal with a wavelength pattern and enters the signal processing circuit 8.

The signal processing circuit 8 removes the noise signal, obtains the intensity distribution of each wavelength of the spectrum, that is, the true wavelength pattern, and converts it into a temperature signal corresponding to color loss.

For example, to remove noise, the noise level can be determined by extracting a wavelength range close to ultraviolet light, where the amount of light is almost zero and consists only of noise, and the noise level can be subtracted from the entire wavelength range.

Next, the electrical signal is normalized based on the peak value of the wavelength pattern, and the signal is
The relative values of 20 types of blackbody radiation wavelength patterns (standard patterns) between 0°C are compared, and two types of standard patterns close to the determined wavelength pattern are selected.

There is an actually measured temperature between the temperatures corresponding to these two types of standard patterns (note that the 20 types of standard patterns are determined in advance by correcting them by actual measurements based on theoretical values).

In this way, mild wavelength pattern abnormalities are corrected based on the standard pattern, and in the case of severe wavelength pattern abnormalities, the data is rejected and the process moves on to the next data processing.

That is, the wavelength with the highest sensitivity is selected between these two types of standard patterns, and the detailed temperature is determined by proportionally distributing this amount.

Then, the output signal from the signal processing circuit 8 enters the signal display device 9 and the welding condition control circuit 10, and the welding condition control circuit 10 outputs welding condition control signals (welding current, voltage, welding speed, etc.) for welding. It is something that is controlled.

As a specific example of welding control, as shown in Figure 2, the difference between the detected temperature and the target temperature is determined, the plate voltage (Ep) of the transmitter is changed accordingly, and the transmitter is matched according to this voltage Ep. and other factors (eg, grid voltage, current) depending on the tube operating characteristics and load circuit characteristics.

As described above, the present invention has a very simple configuration, and it is possible to separate noise even in an environment with a lot of noise, water vapor, dust, etc., such as an ERW steel pipe factory, and it is also possible to perform scanning with high precision, which was difficult to do in the past. This makes it possible to control welding by discoloring the welded part.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing one embodiment of the present invention. FIG. 2 is an explanatory diagram of the welding control operation. 1... Concentrator, 2... Fiberscope, 3...
... Spectrometer, 4... Photoelectric conversion device, 5... Welding part,
6... Squeeze roll, γ... ERW steel pipe, 8... Signal processing circuit, 9... Display device, 10... Welding condition control circuit.

Claims (1)

[Claims] 1. A light focusing device that focuses the discoloration of the welded portion of the ERW steel pipe, a fixed spectrometer that is connected to the light focusing device through a fiber scope and separates the focused light, and a fixed spectrometer that separates the focused light. A group of photoelectric conversion elements that simultaneously receive light in a wavelength range and electrically scan at high speed to generate a wavelength pattern output of the spectrum; A signal processing circuit that calculates the discoloration temperature by normalizing the subtraction result and comparing it with a standard pattern group of black body radiation in a predetermined temperature range, and a welding condition control signal that receives the output of the signal processing circuit. An ERW steel pipe welding control circuit comprising a welding condition control circuit that outputs.