August 8, 2007
With the advent of transistorized controls in resistance welding, power supplies are available in which feedback can be used to control current, voltage, or power delivered. The use and benefits of these control modes are not well understood, leading to underutilization of the technology. It is important to understand the fundamentals of the control modes as applicable to DC and inverter power supplies and the special situations that you might encounter in their application.
Resistance welding is a unique welding process in which direct feedback from the weld is available to measure electrical parameters such as current and voltage. The feedback can be used to react and adapt to the changing conditions in the weld. With the advent of transistorized controls, power supplies are available in which feedback can be used to control current, voltage, or power delivered. However, the use and benefits of these control modes are not well understood, leading to underutilization of the technology. It is important to understand the fundamentals of the control modes as applicable to DC and inverter power supplies and the special situations that you might encounter in their application.
The physics of resistance welding is governed by a combination of Ohm's law and the heat generation equation, as follows:
V= I * R
Where: V = Voltage
I = Current
R = Resistance
H = I2 * R * t
Where: H = Heat Generated
I = Current
R = Resistance
t = time
Combining the two equations, the heat generated can be written as:
H = V2 * t / R
(Volt2 * Time/Resistance)
= V * I * t (Volts * Current x Time)
= P * t ( Power * Time)
= E (Joules)
The equation I2Rt (current2 * resistance x time) is familiar to most engineers, but alternative forms are quite useful to think of when using different modes of control.
The resistance welding process would be very simple if R (resistance at the weld) were known. Unfortunately, this is where it gets complicated, because R is not only unknown, but also changes in the following ways:
All of these changes affect weld resistance, the amount of heat generated, and, consequently, the weld strength. For analysis of mode selection discussed in this section, all the resistance components that directly or indirectly contribute heat to the weld interface are combined into a single resistance (Rw) (see Figure 1). The variable Rw includes both contact/interface and bulk resistance values.
Figure 2 shows changes in power dissipated at the weld resistance, Rw, for the three control modes. Welding modes are in the first column, and each row shows the effect of weld resistance on voltage (V), current (I), and power (P). The first row shows nominal values for Rw = 1. Voltage and current are also set at a nominal value of 1.000 to give power dissipation at Rw of 1.000 as well. The following rows show the effect of a 10 percent increase in weld resistance on voltage, current, and power dissipated at Rw in each of the modes. Power dissipated at the weld, Pw, is a measure of heat generated at the weld and is compared between different modes. For each of the modes, parameters that are held constant are shaded. Interpretation of the numbers is discussed in following sections.
In this mode, the power delivered to the weld is held constant (1.000), and both voltage and current are allowed to fluctuate as required to compensate for any changes in resistance (see Figure 2). Since most weld profiles are programmed to run for a fixed amount of time, the power mode also can be thought of as the energy mode, because the product of power and time is energy (see equations discussed earlier).
In theory, the power mode provides a fixed amount of energy to the weld, regardless of weld resistance variation. Power mode is intermediate between the current and voltage modes in terms of compensation, as shown in the last column in Figure 2. Power mode usually works well when the current and voltage modes are either over- or undercompensating for changes in resistance. Power mode has been shown to be very effective when part orientation is not consistent, such as a wire of rectangular section that can randomly orient on either of the sides.
In this mode the power supply controls the amount of current supplied to the weld. This mode plugs right into the basic equation for heat generation and could be one of the reasons for its popularity. Given that current and time are programmed into the controller, the only factor that changes heat generation is R. Consequently, any increase or decrease in R will lead to a corresponding increase or decrease in heat generation, which will directly affect weld strength.
As seen in Figure 2, a 10 percent increase in resistance results in a 10 percent increase in power dissipated and heat generated at the weld. This mode is best-suited when the parts are consistent and there are no issues with part alignment. Any significant misalignment of the parts can lead to higher resistance at the part-to-part or part-to-electrode interface, leading to a greater amount of heat generated at the weld, resulting in blowouts. However, current control is quite robust for small changes in surface oxidation, because greater levels of oxides are removed by higher voltages that are required to drive the current.
In voltage control mode, voltage drop across the voltage pickup points is programmed while the current output is allowed to vary so that Ohm's law is satisfied. The voltage control mode can be utilized when the operator wants to avoid excessive heating if the resistance is too high, perhaps due to misalignment of parts, thus preventing any damage to the electrodes.
As shown in Figure 2, as the resistance increases by 10 percent, the voltage mode compensates by decreasing current by 10 percent and correspondingly decreasing the power dissipated at the weld by 10 percent. Even though reduced power will reduce the likelihood of damage to the parts and electrodes, the weld itself may be too weak and might not meet strength requirements.
Voltage mode is used for welding with hot-tip electrodes such as tungsten and molybdenum, in which tip temperature is compensated. At the beginning of the shift, the electrodes are cold and have lower resistance; voltage mode compensates by sending more current and generating sufficient heat at the weld. As the tips warm up over a number of welds, their resistance increases and the voltage mode reduces current flow to compensate. Welding with heated tips requires less current because the tips themselves contribute some of the heat to the weld, plus they block any heat conduction away from the weld.
The logic discussed previously is based on the assumption that resistance values in the stack are in close enough proximity to the weld interface that the heat generated at any of the resistances contributes to the heating of the weld interface. However, in practice sometimes one of the parts is much bigger than the other, or one of the part-to-electrode interfaces is far enough away from the weld interface that it does not contribute directly to the interface heating.
Because the voltage pickup leads are mounted beyond the electrodes (typically on electrode holders), these distant resistance values are also measured by the voltage signal and are essentially a source of steady noise. With consistent noise in the signal, simplified analysis of control modes presented earlier may not be applicable and will have to be refined to suit the circumstances.
Figures 3 and 4 show modified power dissipation analysis in which there are other resistances in the circuit, combined together as Ro, that are distant from the weld resistance, Rw, and are not directly affecting or contributing to the welding heat. The schematic on the right in Figure 1 shows Ro as distinct from Rw. Even though Ro is not contributing to heat generated at Rw, because Ro and Rw are in the same circuit and the voltage measured across the electrodes is the sum of voltage drop across Ro and Rw, any variation in either Ro or Rw will affect power dissipated at the weld, Pw.
The effect of a 10 percent increase in either resistances on the power dissipated at the weld in the three control modes is presented in Figures 3 and 4. Figure 3 shows the effect of a 10 percent increase in Rw with no change in Ro, while Figure 4 shows the effect of a 10 percent increase in Ro with no change in Rw. In both figures, the first row is a set of nominal values with both Ro and Rw at 1.000; corresponding V and I are set such that power dissipated at the weld Pw is still 1.000 for direct comparison with Figure 2.
When Rw increases by 10 percent (see Figure 3), the heat dissipated at the weld in the power mode goes up by about 5 percent. Heat dissipated at the weld in current mode goes up by 10 percent while heat dissipated in the voltage mode remains the same as nominal.
When Ro increases by 10 percent, the heat dissipated at the weld in the power mode goes down by 5 percent. Heat dissipated at the weld in current mode remains the same, while voltage mode dissipates 10 percent less heat at the weld.
Comparison of results in Figures 2, 3, and 4 indicate that, in the presence of other resistances in the circuit that do not directly contribute any heat energy to the weld, compensation provided by the three control modes could be different than expected based on the simple analysis in Figure 2. The mode not affected at all by the presence of Ro is the current mode, because it does not depend on the voltage feedback across the electrodes. The mode that is most affected by the presence of Ro is the voltage mode. In the simplified analysis presented in Figure 2, a 10 percent increase in Rw results in a 10 percent drop in power dissipated at the weld. However, in the presence of Ro, a 10 percent increase in Rw results in practically no change in heat dissipated at Rw. Such a result directly contradicts expectations from Figure 2 and could produce confounding results. On the other hand, a 10 percent increase in Ro would result in a drop in power dissipated at the weld in voltage mode (see Figure 4) by 10 percent; again an unexpected result when no change was expected for a steady value of Rw at 1.000 (Figure 2).
As was the case in Figure 2, power mode provides a compensation that is intermediate between voltage and current mode but is still affected by the presence of Ro, as is evident by comparing Pw in power mode for Figures 2, 3, and 4. Such a response from the power mode is to be expected because power is the product of current and voltage, thus balancing out some of the extreme responses.
The analysis demonstrates the importance of implementing a suitable control mode, especially in the presence of Ro, which to some extent is present in practically every resistance welding configuration. Individual results in specific applications will vary depending on the ratio of Ro and Rw, which was set at a nominal value of 1.0 for Figures 3 and 4. Given that Ro and Rw are dynamic and change rapidly during the weld, making measurements to gauge the ratio and absolute values of Ro and Rw can be very difficult and, in many instances, may not be practical. If a significant contribution from Ro is expected, the best option would be to try all control modes and select the one that gives the most robust welding results.
The FABRICATOR® is North America's leading magazine for the metal forming and fabricating industry. The magazine delivers the news, technical articles, and case histories that enable fabricators to do their jobs more efficiently. The FABRICATOR has served the industry since 1971. Print subscriptions are free to qualified persons in North America involved in metal forming and fabricating.