Electric heating – Metal heating – For bonding with pressure
Reexamination Certificate
2001-03-30
2003-02-11
Shaw, Clifford C. (Department: 1725)
Electric heating
Metal heating
For bonding with pressure
C706S023000
Reexamination Certificate
active
06518536
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of joining equipment and, more particularly, relates to joining equipment utilizing a neural network trained by results of welding tests through a learning process.
BACKGROUND OF THE INVENTION
Joining equipment for resistance welding and thermo pressure welding is widely used for joining steel plates and other metallic products. Keeping high quality at a welding zone has become increasingly valued in the joining process.
Conventional resistance welding mainly processed a soft steel plate-made work piece, so that a current abnormality was rarely observed. Consistently controlled welding conditions were able to steady the welding quality.
In recent years, however, galvanized steel plates and high-tension steel plates have been used a lot, instead of soft steel plates. It has been difficult to maintain a reliable welding due to such a diversity of materials to be processed.
In joining methods employed in the electrical equipment field, soldering has been gradually given way to welding, and particularly pressure welding for a greater joining-strength and a longer joining-life.
The challenge for such a case is to control the quality at a minute joining portion.
In this shifting climate, a breakthrough—realizing well-controlled joining quality with a higher accuracy—has been awaited.
To wrestle with the challenge, various methods mainly focused on the resistance welding have been developed, for example, (i) a method evaluating whether a joining state is acceptable or not after joining and, (ii) a method controlling joining quality during joining.
In the welding that is the mainstream of joining method, examples of methods that have been developed so far are described below. It will be noted that all the examples below belong to the method either (i) or (ii) described above.
1) calculating resistance between electrode tips from welding current and welding voltage, then evaluating whether a joining state is acceptable or not, according to the changing pattern. The typical example was disclosed in Japanese Patent Laid-Open No. 56-158286.
2) comparing the voltage between electrode tips with predetermined changes in time of a reference voltage to obtain a difference between them. According to whether the difference is in an acceptable range or not, evaluating a joining state that should be acceptable or not. The typical example was disclosed in Japanese Patent Publication No. 59-14312.
Furthermore, according to voltage between electrode tips, extracting the active component that effectively contributes to the exothermic heat at a welding portion, and calculating the integral in time of the active component to evaluate a welding state to be good or not. Such examples were disclosed in Japanese Patent Publication No. 59-40550 and Japanese Patent Laid-Open No. 59-61580.
3) detecting a heat generating temperature and, according to the pattern that indicates changes in temperature, evaluating whether a welding state is good or not. The typical example was disclosed in Japanese Patent Laid-Open No. 1-216246.
4) passing ultrasound between work pieces to obtain the amount of the transmission, from which a welding state is evaluated to be acceptable or not. The typical example was disclosed in Japanese Patent Laid-Open No. 52-94841.
5) utilizing positional displacement of the electrode tip observed during welding to control the welding quality. The typical example was disclosed in Japanese Patent Publication No. 60-40955.
6) detecting welding current flowed during welding to determine the limits, and monitoring the limits to obtain a welding state with consistently good quality.
7) calculating a nugget diameter with a thermal conduction model by a computer.
8) calculating the temperature distribution of a base-metal, from which a nugget diameter is estimated. At the same time, correcting the temperature distribution according to the amount of movement of electrode tip observed during welding. Unlike the methods 1) through 7), this is to directly control a welder. The typical example was disclosed in Japanese Patent Publication No. 7-16791.
Now will be discussed problems to be tackled in the methods 1) to 8).
In the 1) method, deformation at the top of the electrode tip caused by the pressure, non-effective shunt current occurred there, or galvanized steel plate-work piece can cause resistance between the electrode tips having inconsistent changing pattern. This makes difficult to monitor the quality of welding results.
In the 2) method, each time the welding state varies, for example, deformation of electrode tips caused by the pressure or variations in plate thickness, the condition for evaluating welding results has to be adjusted to the changed state. Therefore, for practical use, it is difficult to monitor, with accuracy, the quality of welding results.
The next 3) and 4) methods contain some problems—installation of a heat detector and an ultrasound transmitter/receiver—inapplicable to on-the-spot welding work.
The 5) method also has problems inapplicable to practical use, for example, (i) installation of a displacement-measuring device; (ii) a noise-intrusion problem; (iii) difficulty in measuring minute displacement; and (iv) an individual difference between resistance welders in mechanical strength. These could be obstacles for on-the-spot welding work.
As a potential solution to (i), a method has been under review. According to the method, the positional displacement of the electrode tip is detected from an output of a position-detecting encoder attached to a servomotor in a servomotor-driven pressurized system. In this case, the displacement-measuring device is not required.
With the method, however, a delay in response of an encoder output, and flexure of materials caused in relation to the mechanical strength of a pressuring mechanism could hamper the displacement measuring with accuracy for practical use.
The method described in 6) is economical and easily practicable. It is effective at detecting a power failure or a break occurred in a secondary conductor. As mentioned above, there are problems inherent in welding work—deformation at the tip of the electrode tip caused by the pressure and non-effective shunt current occurred at the electrode tip. Those make current density dropped. Such quality degradation at a welding portion due to the dropped current density cannot be detected by the method.
According to conventional methods discussed so far, monitoring or controlling welding quality need a painstaking preparation work described below, demanding much effort and experience.
a) carrying out an on-the-spot preparatory experiment on a welding material basis. According to the results, the relation between criteria for evaluation and welding quality is predefined.
b) electrode force, welding current, and welding duration are the welding condition-determining elements. Taking the complicate relationship into account, the criteria for evaluation should be defined.
Although the 7) method does not need the preparatory experiment and is widely applicable, it takes much time to solve a thermal conduction equation.
The 8) method, whose concept is extended from the 7) method, employs the amount of movement of electrode for controlling welders. This inconveniently raises the same problems as those in the 5) method. Besides, if the welding position locates at the edge of a work piece or, if work pieces to be welded each other do not fit snugly, the method could not be applied to such situations.
SUMMARY OF THE INVENTION
The present invention therefore aims to provide joining equipment capable of responding to complicated changes in joining states caused by various factors complicatedly involved with each other. The joining equipment of the invention utilizes a neural network to control the joining process.
It is also an object of the invention to provide joining equipment with the learning process for controlling being economical and simple. To realize this, the equipment should be capable of: solving approximately the thermal c
Fujii Koji
Goto Yasuhiro
Matsuyama Kin-ichi
Ryudo Makoto
Wang Jing Bo
Matsushita Electric - Industrial Co., Ltd.
RatnerPrestia
Shaw Clifford C.
LandOfFree
Joining equipment does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Joining equipment, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Joining equipment will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3137424