Just deep enough - automatic penetration control for TIG welding
TWI Bulletin, May/June 1990
With a degree in chemical engineering and after two years teaching experience, David Harvey joined The Welding Institute in 1986. He is now a Head of Section in the Arc Welding Department with responsibility for research in the gas-shielded processes.
Since joining TWI, David has looked at a number of difficult areas relating to MIG/MAG welding, ranging from materials problems such as porosity in aluminium alloys to process problems such as contact tip performance, welding wire condition and arc stability. More recently, he has been involved in the application of magnetic arc oscillation to the orbital TIG process and development of a backface penetration control.
Mechanised TIG welding is widely used to produce high-quality components for the power generation, nuclear and chemical plant, and aeroengine industries. One of the problems experienced in its application is achieving consistent, complete penetration along the entire length of the weld. Inconsistent weld bead penetration is caused by several factors including variations in component dimensions, welding parameters, material composition and electrode wear.
In an attempt to provide real-time adjustment of the welding parameters, several backface penetration control systems have been developed, featuring photosensitive devices which detect the radiation level from underneath the weld pool. This level is converted into a voltage signal and used to control the weld pool size. The major disadvantages of such systems are the need to align the photosensitive device accurately with the joint, and the different surface emissivities of different materials.
To overcome the problem of changes in emissivity, a control system was developed at TWI in which a video camera and a microcomputer were used to measure the weld pool size rather than the general radiation emitted from the underside of the weld. The camera captures the images of the weld pool and the control system measures the weld pool area and controls the welding parameters to maintain the desired weld pool size. Welding progresses by applying a high current level and, when the weld pool penetrates to form a pool of a predetermined size, the current is switched to a lower background level to allow the weld pool to solidify. Welding continues in a series of overlapping spot welds as in more conventional (fixed duration) pulsed welding. However, use of the micro computer-based system was restricted because of its sensitivity to HF arc ignition.
This limitation has now been overcome by the development of a hard-wired control unit at TWI, which effectively replaces the microcomputer. The basic system comprises a commercially-available video camera; an optional fibre optic cable with associated object lens, optical filter and image guide adaptor; a power source; and the hard-wired control unit ( Figure 1).
Fig. 1. TWI penetration control system: a) Equipment set up
How it works
The control unit comprises two input channels, one dedicated to the camera signal and the other to a reference signal. An additional circuit generates the pulse current waveform.
The camera signal is amplified and fed to a level detector whose output is compared with the reference signal level to determine the required degree of penetration. The output from the comparator is fed to the pulse generator circuit, which comprises four circuits controlling cycle time, background current, pulse current, and maximum period of the pulse current.
Fig. 2. TWI penetration control unit
The comparator output controls the pulse time within the maximum value according to the required penetration. The pulse generator then provides the controlled pulse current waveform signal to drive the power source.
The control unit has five multi-turn potentiometer controls ( Figure 2): pulse current, background current, cycle time, pulse time, and threshold.
The unit has been interfaced to both a transistor-controlled and a thyristor-controlled power supply so that the pulse current, background current, cycle time and pulse time directly override the equivalent power supply controls.
When the amplified camera signal exceeds the threshold value set on the penetration control unit, the pulse is automatically chopped. The values for pulse current, background current and cycle time remain constant. Since the cycle time is fixed, if the pulse length is chopped, the background time is increased to complete the cycle time.
The unit features a weld On/Off switch which also overrides the equivalent switch on the power supply. The control unit has four principal input/outputs: power input, camera signal input, interface to the power supply, and power supply signal monitor (not essential to operation of the control system).
Applications
The performance of the system has been evaluated on stainless steel plate ranging in thickness from 1.0 to 3.2mm, under five test conditions:
- plate of constant thickness;
- plate of gradually varying thickness;
- plate with a step change in thickness;
- material with poor weld penetration characteristics;
- welds requiring only partial penetration.
Constant plate thickness
The control system is capable of adjusting the pulse time automatically so that a constant weld pool size is maintained throughout, resulting in consistent and uniform weld penetration over the weld length for a range of plate thicknesses between 1.0 and 3.2mm ( Figure 3).
Fig. 3. Bead-on-plate tests with penetration control on a range of plate thicknesses - weld underbeads: Fig.3a) 1.0mm
Gradually varying plate thickness
The control system is capable of maintaining a constant weld pool size on material of gradually varying thickness ( Figure 4). The pulse time is automatically reduced as the plate thickness decreases. However, if the control system is not used, and the current is constant throughout, the penetration increases as the plate thickness decreases ( Figure 5).
Fig. 4. Weld underbead achieved with penetration control, with test plate decreasing in thickness gradually from 3.2 to 1.0mm
Fig. 5. Weld underbead achieved without penetration control, with test plate decreasing in thickness gradually from 3.2 to 1.0mm
Step variation in plate thickness
For a step change in thickness from 2.0 to 1.0mm the control system adjusts the pulse time automatically (Figure 6). The penetration bead is essentially consistent and uniform throughout the test, even at the step transition.
Fig. 6. Weld underbead showing uniform penetration achieved by the control system with test plate thickness decreasing in one step from 2.0 to 1.0mm
Material variation
The control unit has also been evaluated by welding material with poor weld penetration characteristics. Two casts of type 316L stainless steel were selected:
3D156 - good penetration characteristics;
3D157 - poor penetration characteristics.
The variation in the penetration characteristics is highlighted in Figure 7 where, despite the same fixed duration pulse, penetration has not been achieved in Cast 3D157. When the control unit was used, almost identical penetration was achieved in both casts. Furthermore, since the weld pool is under feedback control, it is possible to increase the welding speed and maintain a satisfactory weld bead profile even in the poor cast of material ( Figure 8).
Fig. 7. Weld underbeads produced with identical fixed pulse duration in two casts of stainless steel of the same thickness: a) Cast 3D157 - poor penetration
b) Cast 3D156 - good penetration
Fig. 8. Weld underbead produced by the control system in two casts of stainless steel at higher speed, 160mm/min: a) Cast 3D157 - 'poor' cast
b) Cast 3D156 - 'good' cast
Partial penetration
The system has also been used to produce welds of deliberately partial penetration. Although the camera is not seeing a fully penetrating weld pool, sufficient light is emitted from the underside of the weld for the penetration control system to operate on lower threshold settings. This is currently being used for welding in an application where post-weld machining of the underbead is undesirable. Weld penetration has been controlled at levels between 65% and 95% in creep-resistant ferrous and nonferrous materials. The system is currently undergoing pre-production trials at a major UK manufacturer for this particular application.
Welding performance
The performance of the control system has been assessed for welding butt joints between plates of equal thicknesses and of dissimilar thicknesses. As described previously, the system was able to maintain a constant weld pool size in a joint with plates of similar thickness ( Figure 9). A butt joint in dissimilar plate 1.6 and 2.6mm thick was equally successful ( Figure 10).
Fig. 9. Weld underbead in square butt joint using 2.6mm thickness plates and control system
Fig. 10. Weld underbead resulting from plates of unequal thickness of 1.6 and 2.6mm
Conclusions
TWI's backface penetration control system is ideal for a wide range of applications, including:
- components requiring controlled full penetration;
- components requiring controlled partial penetration;
- components of variable thickness;
- materials with poor penetration characteristics.
Furthermore, it has applications covering a wide range of material including C-Mn steel, stainless steel and non-ferrous alloys. Further information on the backface control system can be obtained from David Harvey in the Arc Welding Department.
