CS262413B2 - Method for metal elements connection by diffusion welding - Google Patents

Description

The invention relates to a method of joining metal parts by diffusion welding, wherein a gap is formed between opposing limiting surfaces of the welded parts, the parts are heated to a desired temperature at least in the areas adjacent to the joint, pressed together and cooled.

Advantages and drawbacks of diffusion welding are described, for example, in Welding Handbook, 7th Edition, Volume III, p. 312 et seq. An advantage is that joints can be formed whose properties and microstructure are very similar to the base material, and that they can be joined even if the shape of the parts to be joined makes it difficult to use other methods. In addition, joints with minimal deformation can be produced, which means that there is no need for additional machining or forming.

The drawbacks include the very high cost of the device, which limits the size of the items that can be economically diffusion welded. The necessary heat supply and high contact force under defined conditions, for example in a vacuum or a protective atmosphere, create difficult-to-solve problems with the necessary equipment and therefore constitute a significant limitation of the applicability of the process. In addition, it is believed that the method requires very careful and thorough surface treatment and that it takes more time than other conventional methods.

The purpose of the invention is to overcome these drawbacks and to adapt the diffusion welding method to the production of pipelines for the transport of gas and oil offshore. Such pipelines are currently manufactured by welding the pipeline sections together either manually or by automatic arc-welding equipment in CO2 shielding on board the vessel to be laid. As a rule, pipe welding takes place simultaneously in several stations to increase the pipe laying speed. For practical reasons, the welding stations are located in a horizontal line and several welders can work simultaneously in each station. The finished pipeline is lowered from the vessel via the so-called “sting”, which prevents the pipeline from breaking directly behind the vessel and runs in an S-curve down to the seabed. The laid pipeline maintains a certain tension, and therefore the vessel must have anchors to draw it forward gradually during the laying of the pipeline. The anchors must be moved in front of the vessel from time to time and additional auxiliary ships shall be used.

For various reasons, it would be desirable to lower the pipe vertically from the vessel so that it runs in a single arc down to the seabed. This would, inter alia, allow for better regulation of the stresses in the pipeline laid down, and anchors and auxiliary vessels could be eliminated, and instead a dynamic displacement of the vessel could be used. However, it would be very difficult to use more than one welding station in the vertical orientation of the pipeline on board the vessel used for laying the pipeline. Thus, in today's slow welding processes, rapid pipe laying would not be feasible enough to be practically applicable.

The method according to the invention makes it possible to carry out welding of two pipe sections so quickly that the pipe laying speed is sufficiently high even with the pipe vertical orientation on board the vessel, so that all the advantages associated with such orientation can be realized.

The method according to the invention is characterized in that the gap between the limiting opposing faces of the parts to be joined is formed in the form of a cavity whose height increases from circumference to the center and provided with at least one connecting channel. gas until the sealing of the joint is circumferentially closed, then the joint cavity is evacuated through the connecting channel and the joint is compressed at a predetermined rate for further diffusion welding.

Due to the shape of the joint, high surface pressure is applied at the beginning of the welding at the periphery of the joint. As a result, diffusion can begin around the periphery at such low temperatures that surface oxidation does not occur. Purging with inert gas prevents oxygen from entering and thus does not oxidize. If a reducing gas is used, it is possible to remove the oxides which may have formed previously on the joint surfaces. After the circumference of the cavity is closed, which results in the purging gas ceasing to escape, a connection channel can be connected to the vacuum source so that the cavity is evacuated. This removes residual gases and induces forced dissociation of oxides and other impurities to some extent at increasing temperatures. Due to the temperature and the pressure, the joint closes simultaneously and quickly over the entire perimeter. complete diffusion welding takes place for a short time. Rapid compression is advantageous because it exerts a higher pressure at the joint, which is due to the fact that the stress on the creep rupture increases with the rate of deformation.

According to a preferred embodiment of the invention, at least one of the cavity limiting surfaces is notched circumferentially. This ensures a sufficiently rapid escape of the purge gas during the first bonding phases.

It is furthermore very advantageous according to the invention to provide one of the gap limiting surfaces in a concave shape. This creates an advantageous stress distribution during the final compression of the parts and thus ensures perfect welding over the entire joint surface.

A further advantageous embodiment of the invention is characterized in that at least one and preferably both of the confining opposing surfaces are formed in a conical shape. The cone shape is machined much more precisely by machine; when the two surfaces are conical, one concave and one convex, a self-opening roof is created between the two parts when pressed.

With dici effect. Such a shape of the welded surfaces also ensures their complete welding.

In order to increase the contact pressure on the joint surface during compression, according to the invention, the cross-section of the parts in the joint region is reduced. As a result, the contact pressure on the joint surface increases substantially above the uniaxial stress value under the stress at the material flow at the prevailing temperature; the result is faster and more perfect diffusion.

Various inert gases can be used for flushing, but helium is preferred according to the invention, since it is cooled to a temperature that is virtually oxygen-free during production. In addition, reliable helium detectors, which are commercially available, can be used to check that the circuit of the joint is perfectly sealed before evacuation.

When a reducing gas is used for purging, hydrogen is very preferred for this purpose. It not only removes the oxide layers, but also allows visual inspection of the closure of the joint, because the leaking hydrogen on the joint surface burns. Hydrogen has absolutely no harmful effect on the soft structural steels joined by the process of the invention.

According to the invention, it is proposed to provide the joints with an activating alloy, for example a palladium-nickel 40/60 alloy. The alloy may be introduced as a thin strip of material or may be applied to one or both of the welded surfaces by electroplating. This also reduces the time required for diffusion welding. The connection channel for the purging gas supply and for the gas discharge and evacuation can advantageously be provided in one of the parts to be connected, preferably near its periphery. In such an arrangement, the channel can be closed immediately after diffusion welding by deep-welding electrode or by non-melting electrode plasma welding, as long as the temperature of the welded parts is sufficiently high, for example 400 to 600 ° C.

The invention will be explained in connection with the examples shown in the drawing, wherein Figures 1 and 2 show schematically a first embodiment of two parts to be welded at the beginning and after diffusion welding, Fig. 3 and a second embodiment of two parts to be welded at the beginning and after welding. giant.

and 6 shows another example of two welded parts at the beginning of the welding and after it is finished and FIG. 7 shows the parts according to FIG. 6 after further compression.

In the drawing, corresponding parts are denoted by the same reference numerals.

Giant. 1 shows a cross-section of a portion of two parts 1, 2, which may be, for example, ST-52-3 structural steel bolts. Giant. 1 can also be considered as an axial section of a thick-walled pipe, the axis of which is on the left side of the drawing. Between the two parts 1, 2 there is a gap in the form of a cavity 3, which is limited by a concave limiting surface 4 on part 1 and a planar limiting surface 5 on part 2. In part 1 a connecting channel 6 is formed. the end may alternatively be connected to a purge gas source and a vacuum source not shown. The arrows F show the variable contact force, while the bell curve to the left of the parts 1, 2 indicates the temperature distribution in the axial direction of the two parts 1, 2.

When the parts 1, 2 are to be joined, they are first brought into position according to FIG. 1 and subjected to a thrust force F. This force may be exerted by a hydraulic system (not shown), which may for example comprise a clamping ring on each of the two parts 1, 2 the clamping rings are interconnected by hydraulic cylinders. Of course, other embodiments may be used, depending on the shape and cross-section of the parts to be joined. Heating then begins, for example by means of an induction coil (not shown). At the same time, purging gas is supplied through the connecting channel. The flushing gas first leaks around the perimeter 7 of the cavity 3 due to minor irregularities or indentations on the limiting surfaces 4, 5 of the joint. The purpose of the flushing is to remove oxygen from the surfaces to be joined to prevent oxidation thereof during heating and optionally to remove the oxides present. Indeed, irrespective of how the limiting surfaces 4, 5 have been cleaned prior to welding, the short-term action of air oxygen also results in the formation of an oxide layer of 350 to 1000 x 10 ^-4 µm, depending on the temperature and humidity of the air.

When the temperature of the material near the periphery 7 of the cavity 3 rises to 600-800 ° C, and the temperature distribution follows the schematic curve on the left side of Fig. 1, the appropriate pressure force F causes diffusion between the parts 1, 2 so that the cavity 3 closes around circumference 7. This can be monitored in various ways, for example by stopping the gas from escaping from the circuit 7 or by decreasing the pressure in the cavity 3 of the joint.

Once the joint is closed at the periphery 7, the connecting duct 6 is connected to a vacuum source to reduce the pressure in the cavity 3 to about 0.0133 Pa. At the same time, the temperature of the parts 1, 2 increases, as shown schematically in the curve to the left in FIG. 2, to a temperature of about 1350 ° C. With a suitable contact force F, the cavity 3 closes within a few seconds. The result is shown schematically in FIG. 2. Complete diffusion welding occurs within 15 to 30 minutes. Perfect welds were formed even within as short a diffusion time as 8 minutes at a temperature of about 1350 degrees Celsius. Then, the parts 1, 2 are cooled in still air to about 600 ° C, which takes about 4 minutes at a material thickness of 40 mm. At this temperature, the feed channel 6 can be closed, for example by deep-welding electrode or by plasma welding with a non-consumable electrode.

Giant. 3 shows 2 parts 1, 2 of generally the same external shape as in FIG. 1. However, the welding cavity 3 has a different shape and its limiting surfaces 4, 5 are both conical, one convex and one concave. The apex angles of the two cones are unequal so that the cavity 3 has an increasing height toward the center. The distance between the apexes of the confining surfaces 4, 5 of the cavity 3 can be about 10% of the thickness of the parts 1, 2. The conical shape of the confining surfaces 4, 5 causes self-alignment to the axis when the parts 1, 2 are pressed together. The conical shape also prevents insufficient diffusion in the center of the joint, which could easily occur if the lower limiting surface S were planar and the upper limiting surface 4 conically convex.

Otherwise the joining of the parts 1, 2 proceeds in the same manner as described in connection with FIGS. 1 and 2. The final result is shown in FIG.

4.

The pressure during diffusion welding is a very important parameter that is difficult to control. According to the above publication, it is assumed that the pressure cannot be increased beyond the creep stress at the temperature prevailing during welding without the use of molds to prevent the parts from moving together. The use of such molds is difficult and increases the cost of the device, and in some fields of such molds it is not possible to use them at all because the welded areas are inaccessible. Since the diffusion rate generally increases with a square of pressure, increasing the pressure to two to three times the creep stress can reduce the diffusion time to about a quarter to ninth of the corresponding diffusion time at uniaxial pressure. Alternatively, considerably larger amounts of oxides can be admitted on the surfaces to be joined, thereby possibly avoiding purging the cavity with reducing gas.

In order to create such high pressures, it is proposed according to the invention that a triaxial stress is created in the region of the joint. It is known that tensile tests of material in the form of strong test rods cause shrinkage due to shear deformation, with a fracture fracture in the central part. The fracture fracture is due to the fact that the axial stress in the center of the rod is much higher than the yield strength of the material, while the material holds on both sides of the shrinkage due to radial stresses. As a result, the axial stress increases and the deformation decreases when the difference between the axial and radial stresses falls below the yield strength value, as the Trescas principle teaches.

According to the invention, a triaxial state of stress can be created by providing a considerable taper of the cross-section relative to the adjacent material in the joint area 1, 2. An example is shown in Fig. 5, where hydrogen is used as the purge gas, which results in the formation of flames drawn around the perimeter of the joint.

Once the flames have disappeared, the gap is closed so that no detection device may be used to detect this. When the cavity 3 is closed circumferentially, the application of full compression force F causes the entire cavity 3 to close under very difficult stress conditions that are of a dynamic nature. The triaxial stress, acting exactly at the moment of closure, causes stresses that run transversely to the cavity 3 and are 5 to 6 times the creep stress of equally thick parts due to the cross-sectional narrowing and relatively cold surrounding material. This situation is illustrated in FIG. 6. Thereafter, the parts 1, 2 are pressed against each other by a relatively small contact force applied until the constricted area has the same cross section as the other parts of the parts. The two parts 1, 2 then have the same thickness along the entire length as shown in FIG. 7.

Thus, without the use of an external mold, the welded parts 1, 2 have the correct shape, in the important phase of the welding there is a high contact pressure, the oxides that may have been on the welded surfaces decompose into double or multiple surfaces, so the oxide layer is substantially thinner and increased globular conversion and accelerated recrystallization.

When the desired joint thickness is reached, the pressing force can be stopped and the diffusion time and temperature are maintained until the remaining pores are closed by recrystallization. If the joining is to be much faster, the limiting surfaces 4, 5 can be electroplated or a strip of activating alloy, for example Pd / Ni 60/40 alloy, can be interposed therebetween.

Due to the very high stresses that occur in the region of the joint when reducing the cross-section of FIG. 5, it can be assumed that the original shape of the cavity 3 and the pretreatment of its surfaces is less critical than in other cases.

The tests carried out according to the invention in the welding of ST 52-3 structural steel bolts having a diameter of 40 mm produced excellent results, in particular in terms of ductility and tensile strength. More than 60% elongation was measured in the heat-exposed area of the studs that were bent to the hairpin shape. The break surface clearly showed a shear fracture.

It goes without saying that the invention can be used for purposes other than welding large diameter pipes. For example, the method is applicable to welding trusses in the construction of bridges and marine platforms.

As noted, serrations may be formed on one or both of the confinement surfaces of the cavity to provide a purge gas leakage during the first welding phases. In certain cases, it may be advantageous to provide a slight indentation over the entire limiting surface of the joint. Such indentations then echo in later ultrasonic testing when the joint is not fully welded.

Claims (10)

SUBJECT 1. A method of joining metal parts by diffusion welding, wherein a gap is formed between opposing faces of the parts, the parts are heated to a desired temperature at least in the areas adjacent to the joint, joined by pressing together and cooled, characterized in that the joint is shaped. provided with at least one connecting channel, during the heating of the parts, an inert or reducing purging gas is first introduced into the cavity until the joint is closed tightly around the periphery, then the joint cavity is evacuated through the connecting channel, and the joint is compressed at a predetermined speed for further diffusion welding. Method according to claim 1, characterized in that at least one of the gap limiting surfaces is formed in a concave shape. 3. Method according to claim 1, characterized in that at least one of the gap limiting surfaces is formed in a conical shape. 4. The method of claim 3, wherein: INVENTION that both surfaces are given a conical shape, one of which is concave and the other convex. Method according to one of Claims 1 to 4, characterized in that the cavity is purged with helium as an inert gas. Method according to one of Claims 1 to 4, characterized in that the cavity is purged with hydrogen as a reducing gas. Method according to one of Claims 1 to 6, characterized in that the connection channel is provided in one of the parts to be joined, preferably at its periphery. 8. Method according to claims 1 to 7, characterized in that the cross-section of the parts to be joined in the region of the joint is reduced. 9. A method according to claims 1 to 8, characterized in that an activating alloy is introduced into the joint. Method according to one of Claims 1 to 9, characterized in that at least one of the joint limiting surfaces is knurled at least on the circumference.