The temperature cycles occurring during welding (temperature-time curve) have a decisive influence on the mechanical properties in the weld metal and in the heat-affected zone. The temperature cycles in turn depend on the welding conditions. Welding conditions are understood to be a multitude of influencing variables such as arc voltage, welding current, welding speed, working temperature, sheet thickness, welding process and seam shape [1].
The temperature-time curve occurring at a defined point during an arc passage consists of a short heating phase and a generally much longer cooling phase. As the arc approaches, the temperature quickly rises to a maximum value and then drops again after the arc has passed, with the cooling rate steadily decreasing. While the same peak temperatures occur everywhere in the weld metal, the different areas of the heat-affected zone are heated to different peak values; their height decreases with increasing distance from the melting zone [2].
The mechanical properties of the weld metal are primarily determined by its chemical composition and the rate at which it cools from the liquid phase. Decisive for the effects of welding temperature cycles on the mechanical properties in the heat-affected zone are the peak temperature reached during welding, the dwell time in the upper austenitic zone and the rate at which cooling from the austenitic zone takes place [2]. Experience shows that high peak temperatures lead to the most unfavourable microstructural conditions and mechanical properties. It is therefore sufficient to consider the temperature cycles with the highest peak temperature, which occur immediately adjacent to the melting line in the coarse grain area of the heat-affected zone. Their peak temperature is at the level of the melting temperature of the respective material. It can therefore be assumed that the mechanical properties in the heat-affected zone are determined by the cooling process after the arcing.
To identify welding temperature cycles, the reciprocal value of the cooling rate is generally chosen instead of the cooling rate, i.e. the time required to pass through a certain temperature interval. The cooling time t8/5 has proven to be a good choice for dealing with material issues. This is the time required during the cooling of a weld bead and its heat-affected zone to pass through the temperature range from 800 °C to 500 °C.
From the general differential equation of heat conduction in solid bodies, equations can be derived which describe the temperature course in the weld seam area as a function of place and time. After appropriate transformation, these equations are suitable for calculating the time t8/5 [3] needed to cool the weld metal to pass through the temperature range of 800 °C to 500 °C.
When calculating the cooling times, a distinction must be made between three- and two-dimensional heat dissipation. When welding relatively thick workpieces, heat dissipation is three-dimensional. The heat introduced via the arc can be dissipated in the workpiece plane and additionally in the direction of the workpiece thickness. This therefore has no effect on the cooling time. With two-dimensional heat dissipation, however, the heat flow occurs exclusively in the plane of the workpiece. In this case, the workpiece thickness is decisive for the cross-sectional area available for heat dissipation and thus has a pronounced influence on the cooling time [4].
When welding relatively thick plates (three-dimensional heat dissipation), the cooling time t8/5 is calculated according to the following equation [5]:
Formula (three-dimensional heat dissipation):
t8/5 = (6700 - 5 * T0) * Q * [( 1 / (500 - T0)) - (1 / (800 - T0))] * F3
with
- Q: Heatinput
- T0: Preheat temperature
- F3: Seam factor for three-dimensional heat dissipation
The cooling time is therefore proportional to the applied heat in the case of three-dimensional heat dissipation and increases with the preheating temperature.
When welding products of relatively small thickness, two-dimensional heat dissipation is present. The cooling time t8/5 is calculated according to the following equation [5]:
Formula (two-dimensional heat dissipation):
t8/5 = (4300 - 4,3 * T0) * 105 * (Q2 / d2) * [( 1 / (500 - T0))2 - (1 / (800 - T0))2] * F2
with
- Q: Heat input
- T0: Preheat temperature
- d: Plate thickness
- F2: Seam factor for two-dimensional heat dissipation
The cooling time for two-dimensional heat dissipation thus increases with the square of the energy per unit length and with the preheating temperature and is inversely proportional to the square of the workpiece thickness.
The heat input Q can be calculated as follows [6], [7].
Q = eta * E = eta * (U * I) / v
with
- Q: Heat input
- E: Energy input
- eta: thermal efficiency
- U: Arc voltage
- I: Arc current
- v: Welding speed
Unless otherwise specified, the thermal efficiency of welding processes (eta) shall be based on the values given in the following table [5].
| Thermal efficiency of welding processes | |
| Process | Faktor eta |
| SAW | 1,0 |
| MAW with stick electrode | 0,8 |
| Metal Active Gas Welding (MAG) | 0,8 |
| Metal Inert Gas Welding (MIG) | 0,8 |
| Tungsten Inert Gas Welding (TIG) | 0,6 |
The number of conceivable types of sutures is so large that a quantitative clarification of the influence of all of them on the cooling time would require an extremely high effort. For this reason, the table below summarizes only the seam factors for the most common seam types with three-dimensional heat dissipation (F3) and two-dimensional heat dissipation (F2) [8]. It can be seen that especially with two-dimensional heat dissipation the cooling times of fillet welds are much shorter than those of application beads. The value of the seam factor depends on the ratio of the energy per unit length to the sheet thickness.
| Seam factors | ||
| Seam type | F3 | F2 |
| Deposit weld | 1,0 | 1,0 |
| 1. and 2. fillet weld on T- or Cross joint | 0,67 | 0,45 bis 0,67 |
| 3. and 4. fillet weld on T- or Cross joint | 0,67 | 0,3 bis 0,67 |
| Fillet weld at corner joint | 0,67 | 0,9 |
| Fillet weld at overlap joint | 0,67 | 0,7 |
| Root pass of V-seams (opening angle 60°, face 3 mm) | 1,0 bis 1,2 | rd. 1,0 |
| Root pass of double V seams (opening angle 50°, face 3 mm) | 0,7 | rd. 1,0 |
| Center layers of V and double V seams | 0,8 bis 1,0 | rd. 1,0 |
| Cover layers of V and double V seams | 0,9 bis 1,0 | 1,0 |
| I-seam, 'position-counter position welding' | - | 1,0 |
If the respective workpiece thickness is close to the transition sheet thickness (see below), the value of the seam factor F2 corresponds to that of F3. The smaller the workpiece thickness compared to the transition sheet thickness, the more clearly F2 and F3 differ [4].
The sheet thickness at the transition from three- to two-dimensional heat dissipation is called transition sheet thickness dü. It is obtained by equating the formulas for calculating the cooling time t8/5 for three- and two-dimensional heat dissipation:
dü = [((4300 - 4,3 * T0) / (6700 - 5 * T0)) * 105 * Q * (( 1 / (500 - T0)) + (1 / (800 - T0)))]0,5
with
- Q: Heat input
- T0: Preheat temperature
When calculating cooling times, it should be noted that the assumptions underlying the equations are often not exactly fulfilled. Calculated values of the cooling time can therefore deviate from the actual values by about 10%. The calculation can be afflicted with a larger error in the transition area from two- to three-dimensional heat dissipation. In critical cases it is recommended to control the cooling time by measurement [5].
Literature:
[1] Degenkolbe, J., Uwer, D., und Wegmann, H. G.:
Kennzeichnung von Schweißtemperaturzyklen hinsichtlich ihrer Auswirkung auf die mechanischen Eigenschaften von Schweißverbindungen durch die Abkühlzeit t8/5 und deren Ermittlung. Thyssen Technische Berichte, Heft 1/85, S. 57 - 73
[2] Uwer, D. und Degenkolbe, J.:
Temperaturzyklen beim Lichtbogenschweißen - Berechnung der Abkühlzeiten. Schweißen und Schneiden, Jahrgang 24 (1972), Heft 12, S. 485 - 489
[3] Uwer, D. und Degenkolbe, J.:
Temperaturzyklen beim Lichtbogenschweißen - Einfluß des Wärmebehandlungszustandes und der chemischen Zusammensetzung von Stählen auf die Abkühlzeit. Schweißen und Schneiden, Jahrgang 27 (1975), Heft 8, S. 303 - 306
[4] Uwer, D.:
Rechnerisches und grafisches Ermitteln von Abkühlzeiten beim Lichtbogenschweißen. Schweißen und Schneiden, Jahrgang 30 (1978), Heft 7, S. 243 - 248
[5] Stahl-Eisen-Werkstoffblatt 088 Beiblatt 2:
Schweißgeeignete Feinkornbaustähle - Richtlinien für die Verarbeitung, besonders für das Schmelzschweißen; Ermittlung der Abkühlzeit t8/5 zur Kennzeichnung von Schweißtemperaturzyklen. 4. Ausgabe, Oktober 1993, Verlag Stahleisen, Düsseldorf
[6] Uwer, D. und Wegmann, H. G.:
Temperaturzyklen beim Lichtbogenschweißen - Einfluß von Schweißverfahren und Nahtart auf die Abkühlzeit. Schweißen und Schneiden, Jahrgang 28 (1976), Heft 4, S. 132 - 136
[7] DIN EN 1011-1: Empfehlung zum Schweißen metallischer Werkstoffe, Teil 1: Allgemeine Anleitungen für Lichtbogenschweißen, April 1998, Beuth Verlag GmbH, Berlin
[8] Uwer, D. und Degenkolbe, J.: Kennzeichnung von Schweißtemperaturzyklen hinsichtlich ihrer Auswirkung auf die mechanischen Eigenschaften von Schweißverbindungen. Stahl und Eisen 97 (1977), Nr. 24, S. 1201 - 1207