1. Field of the Invention
The present invention relates generally to resistance welding. More particularly, the present invention is directed to generating spot weld schedules to be implemented by weld controllers.
2. Description of the Related Art
Automobile body panels are spot-welded using a welding apparatus including a robot and a spot welding servo gun coupled to a wrist portion of the robot. The spot welding servo gun has a pair of electrode tips for pressing a work piece therebetween. The electrode tips are electrically connected to a welding transformer. The welding transformer is electrically connected to a welding power source via a controller having a timer function. At least one tip of the spot welding servo gun is movably installed in a servo motor which performs position control or pressure control of the electrode tip.
Many spot welds experience expulsion. That is, molten metal is ejected from the weld decreasing the cleanliness of the mass produced automobile body and compromising the weld integrity. There are five primary causes of expulsion: (i) gun alignment, (ii) condition of the tip of the electrodes, (iii) weld schedule, (iv) misfit parts, and (v) adhesives between the parts.
It is an object of the present invention to reduce the level and frequency of expulsion during the spot welding process. To that end, the inventors studied the interaction of the five causes of expulsion and identified the most important—weld schedules were giving off too much heat and misfit parts.
To reduce the frequency of expulsions, the present invention optimizes existing welding schedules by analyzing an operating window of a weld lobe and uses impulse welding to address, among other things, the misfit parts issue.
A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:
The inventors conducted a Taguchi experiment measuring the effect and interaction of weld time, uses of impulses, part fit, pressure, pulsation, and current. The five factor experiment required 25 permutations. Based on the Taguchi experiment, the inventors concluded that the interaction of current and time were dominant over the other factors. The interaction between pulse and time to address misfit parts, while not as significant as the interaction between current and time, was also found to be significant. Thus, a parameter identified by the inventors for optimization was the number of pulses in the weld time (while operating within the weld lobe). The inventors concluded from the Taguchi experiment that: (i) impulse welding should be implemented due to its effectiveness on expulsion; (ii) the top two traceable factors that affect nugget size and expulsion are current and time. Based on these conclusions, among other things, the inventors developed a methodology for spatter-less welding including optimizing a welding schedule and using impulse welding.
Referring now to the drawings, wherein like reference numerals designate identical or corresponding parts throughout the several views. In developing a methodology for generating spatter-less welds which would be effective where the parts being welded may have a poor fit, the inventors created a weld lobe using a preexisting weld schedule on a perfect part fit and a weld lobe using the preexisting schedule on a poor part fit. The part fits were simulated using coupons. As can be seen in
The overlapping window illustrated in
With conventional single pulse welding, heat greatly dissipates in the metal. In contrast, with multiple impulse welding, heat is focused on the weld reducing metal temperature and expulsion. Heat is a function of the current squared, the weld time, and the pressure.
Qα(I2t)/P
As reflected in
A preferred methodology for creating an optimized weld schedule is next explained in conjunction with
In step 101, a weld lobe for an existing weld schedule is generated. Welding schedules typically include a prescribed current, time, and pressure for the weld. According to an embodiment of the present invention, step 101 includes the following sub-steps. First, taking two metal coupons 201 of the appropriate type and thickness associated with the existing weld schedule and clamping them down to back blocks 203 and 205 as shown in
To generate the data to create the weld lobe, for a first weld, apply the suggested pressure of the existing schedule, apply half the suggested weld time found in the existing schedule, apply 2 kA plus the prescribed current found in the existing schedule, and record expulsion whether an expulsion occurs (Yes/No). Increments of current smaller or larger than 2 kA can be added to the prescribed current dependent on the number of data points wanted.
For the next few welds, hold the current and the pressure constant, vary the prescribed weld time in predetermined increments (e.g., prescribed total weld time/6), and increase the weld time by the calculated interval for each weld until either expulsion occurs or a time cap is reached. According to an embodiment of the invention, the time cap can be established at for example the prescribed weld time plus (6 to 10) predetermined increments. Record expulsion data for each weld. In a preferred embodiment, decrease the current (originally the prescribed current plus 2 kA) by 1 kA, reduce the weld time back to half of the prescribed weld time and repeat the process described in this paragraph until the current being applied is the prescribed current minus 2 kA.
After all of the welds have been created, perform peel tests on the welds to check weld quality. The quality of the welds should be recorded as pass or fail. A plot of the current versus time data should be generated, using a triangle, for example, to represent expulsions, circles, for example, to represent good welds without expulsion, and an “x” or a square, for example, to represent bad welds. See
In step 103, draw the prescribed weld time on the generated weld lobe. See
In step 105, an operating window is identified. As discussed above, the inventors performed a part fit study to investigate the effect of poor part fit on spot welds, and discovered that part fit had a significant impact on the operating window of a spot weld.
According to an embodiment of the present invention, the operating window only includes the region of good welds to the left of the weld time constraint. Measure the current and time range for the operating window. According to a preferred embodiment, the maximum current range should be greater than 2 kA and the maximum time range should be greater than 20 percent of the suggested weld time. If these constraints are met, then go to step 109. Otherwise, go to step 107.
In step 107, if the pressure is being varied for a first time, then increase the pressure force by 200 N or another predetermined increment, recreate the weld lobe, and repeat steps 103 and 105. If the pressure is not being varied for the first time, then judge whether the window improved (i.e., increased in size) or worsened (i.e., decreased in size) based on the maximum current range and maximum time range constraints disclosed above and vary pressure force in the up or down in intervals of 200 N. Recreate the weld lobe, and repeat steps 103 and 105 until an operating window is generated meeting the maximum current range and maximum time range constraints. When this is accomplished, proceed to step 109.
If, after several pressure changes, there is indication that the operating window will not meet the maximum current range and maximum time range constraints, then select the pressure examined with the biggest operating window, and continue to step 111.
The inventors discovered that in a preferred but non-limiting embodiment that an ideal location for an optimized weld schedule is a region on the weld lobe that meets the following conditions:
If an operating window satisfying the maximum current range and maximum time range constraints is identified in step 107 and an ideal window is identified within the window, then in step 111 select a schedule that is at the very right side of the ideal location, because it is the furthest from bad welds. The selected schedule is the optimum schedule. See
In step 113, impulse welding is tested in order to determine a number of cool cycles to include with the pulses during the optimized weld schedule. According to an embodiment of the present invention, in order to test the impulse welding, a part gap is simulated as shown in
When implementing the impulse welding, the total weld time should not increase for the reasons discussed above. Therefore, if cool time is added, then actual weld time must be reduced. See the following formula for detail:
Total Cool Time=(Pulse#−1)*Cool Time
Total Weld Time=Actual Weld Time+Total Cool Time
Note, if cool time is 1 cycle, and there are 3 pulses, then the total cool time would be 2 cycles.
To select the appropriate number of pulses and amount of cool times, the following algorithm can be used:
The following table can be used as a guide for the number of pulses and total cool time combinations. This table should be followed in an ascending manner.
To determine the appropriate individual pulse durations, the following formulas can be used:
Due to most weld controller limitations, pulses after the second pulse repeat the duration of the second pulse. For example:
An example of data obtained by the inventors applying the method of the present invention is provided below:
Base Schedule: 9.6 kA, 23 cycles, 3340N (From step 111)
Incrementally add cool time:
Final Spatter-Less Schedule:
Current: 9.6 kA
Number of Pulses: 3 pulses
Cool times: 2 cool cycles in between pulses, total cool time of 4 cycles
Weld time: total weld time 23 cycles, individual pulse times→7 cycles, 6 cycles, 6 cycles
Pressure: 3340N
Although less practical than a 2D lobe, according to an embodiment of the present invention, a 3D weld lobe (current, pressure, and time) based on the heat equation Qα(I2t)/P can be generated. See e.g.,
Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.
Number | Name | Date | Kind |
---|---|---|---|
2464981 | Leathers et al. | Mar 1949 | A |
2826674 | Peras | Mar 1958 | A |
3569659 | Benton et al. | Mar 1971 | A |
3899653 | Spinnato | Aug 1975 | A |
4024371 | Drake | May 1977 | A |
4230930 | Chang et al. | Oct 1980 | A |
4302653 | Denning et al. | Nov 1981 | A |
4447700 | Cohen | May 1984 | A |
4456810 | Schumacher et al. | Jun 1984 | A |
4678887 | Nagel et al. | Jul 1987 | A |
4724294 | Klein | Feb 1988 | A |
5030814 | Tange et al. | Jul 1991 | A |
5192851 | James et al. | Mar 1993 | A |
5748462 | Moro et al. | May 1998 | A |
5770834 | Davis et al. | Jun 1998 | A |
5786558 | Shimada | Jul 1998 | A |
6208146 | Huang et al. | Mar 2001 | B1 |
6359249 | Brown et al. | Mar 2002 | B1 |
6403913 | Spinella et al. | Jun 2002 | B1 |
20050184031 | Sun et al. | Aug 2005 | A1 |
20050218120 | Shih | Oct 2005 | A1 |
20070119823 | Filev et al. | May 2007 | A1 |
20070170164 | Nadzam | Jul 2007 | A1 |
20100122968 | Buse et al. | May 2010 | A1 |
Number | Date | Country | |
---|---|---|---|
20100122968 A1 | May 2010 | US |