Abstract

This study contains the non-destructive testing of resistance spot welds with the ultrasonic multiple pulse echo technique. According to this technique, resistance spot welds are classified by evaluating the echo positions on the time scale and their attenuation characteristics. In this research, in addition to providing the basic guidelines related to the ultrasonic testing of resistance spot welds, different from the previous research in literature, a method has been suggested which makes it possible to predict the shear strengths of the spot welds by using the same technique.

Keywords:
spot welding, spot weld testing, ultrasonic testing

1. Introduction

Resistance spot welding is a method by which two or more pieces of metal can be joined by the application of heat and pressure. Since the production by this process is economical and adaptable to high-speed operation, it is widely used in industry. The common practice for determining spot weld quality is to use some destructive tests, which physically dismantle the welded members to determine presence and size of fused metal nuggets at the spot weld site. Due to the destructive character of these common tests, their not being reproducible and the insufficient correlation of these results to the mass production, for the quality assurance of spot welds in industry, researchers have been trying to develop an appropriate non-destructive testing method since about 1967 [1].

Despite the extensive use of spot welding in industry, very few NDT methods are used -with partial satisfaction- such as visual inspection, radiographic testing, dynamic resistance test, acoustic emission method and various ultrasonic techniques [2,3,4,5]. Another possible NDT method which is also the main subject of this research is the ultrasonic testing of resistance spot welds with the multiple pulse echo technique which was developed especially to detect a typical defect: the stick weld [6]. This method is suitable for the quality assurance of spot welds in the series production line and it significantly reduces testing time and material cost.

2. Basic Principles

In order to be able to make a prediction about the quality of a spot weld, the most important thing that should be known is the three dimensional size of the weld nugget which can be determined by measuring the transit time and the attenuation of the echoes that belong to the ultrasonic wave that is propagated through the weld nugget and directed perpendicularly to the faces of the parent metals.

The transit time of the echoes are used to determine the diameter of the weld nugget. If the spot weld with a diameter greater than or equal to the diameter of the sound beam is considered, only the echoes that belong to the double thickness of the sheets can be seen on the ultrasonic device screen. However, if the diameter of the spot weld is smaller than the diameter of the sound beam, a series of secondary echoes are seen on the ultrasonic testing device screen.

After determining the two dimensions of the nugget the third dimension which is the thickness of the weld nugget can be predicted with the help of the sound attenuation characteristics. The only interface that the ultrasonic wave can be reflected from, are the outer faces of the welded sheets and as the wave is reflected back and forth between the outer faces its amplitude is attenuated. Due to grain scattering as the nugget thickness increases the attenuation also increases. Since the grains scatter some of the ultrasonic energy out of the sound beam, the height of the echo series that come from the outer face of the bottom plate decreases and the rate of this decay strongly depends on the thickness of the weld nugget.

3. Experimental Set-Up

A typical test system for ultrasonic spot weld examination is mainly composed of two elements; a high frequency straight beam probe and a universal ultrasonic instrument with high resolution.

In this research G15MN 8.0 series straight beam probe, with an 8 mm diameter for a material thickness of 3 mm, designed and produced by Krautkramer for spot weld tests, was used (Figure 1). As it is explained in the previous discussions, the theory of ultrasonic spot weld testing depends mainly on the attenuation measurements and since the attenuation -grain scattering- is proportional to the frequency, the transducer's frequency is selected as 15 MHz. As well as the 8 mm diameter probe, a same type of probe again with a 15 MHz frequency but with a smaller diameter of 3.6 was also used.

Fig 1: Probe section (1) Housing, (2) Socket, (3) Cable Feed, (4) Damping Element, (5) Crystal, (6) Water Delay Path, (7) Membrane, (8) Plates.

Fig 2: Ultrasonic probe assembly for spot weld testing. (Krautkramer G15MN 3.6).

The purpose of using this probe is to differentiate the echoes that come from the single plate thickness from the echoes reflected back from porosity or an inclusion. The probe consists of a small encapsulated transducer with an attached transmission medium which is usually the water column delay line (Figure 2).

4. Sample Preparations

St 37 was selected as the test piece material and the material thickness 3 mm was appropriate for the available set of probes -G15MN 8.0 and G15MN 3.6-. The test pieces were prepared by using the Nimak PMP 6-1 spot welding machine in three main steps. The first step was the production of torsional test pieces. The aim of this first step was to determine the appropriate spot welding parameters for the selected material and thickness. A total of 56 test pieces were produced. The second step in preparing the test pieces was the preparation of the first group of tension shear test specimens. 35 test pieces were produced and the dimensions of the test specimens were determined according to the recommendations of 'The American Welding Society' (Figure 3).

Fig 3: Tension-shear test piece.

After the visual and radiographic examination, the test pieces were first tested ultrasonically and then their shear strengths were determined destructively. Finally, according to the results of these tests, a second set of welding parameters from which the necessary strength distribution would be obtained was determined.

After these two group of test specimens, 65 more tension-shear test specimens were produced while the welding force, clamping time, holding time and pulse repetition time were kept constant and the number of pulsations and the machine performance, which directly affects the current, were taken as variables. The same testing steps were applied to the second group of test pieces, but this time before the destructive tests, 25 randomly selected test pieces out of 65 were tested ultrasonically by a second operator in order to determine the operator dependence of this test.

5. Application of the Method

In this section, the applicability of the ultrasonic multiple pulse echo method is examined by evaluating each factor one by one.

5.1 Nugget Diameter
The size of the fused zone is one of the most important parameters that determine spot weld quality and strength. Although it is very difficult to directly measure the nugget diameter, with the help of carefully selected reference specimens, its size can be predicted by comparing the heights of the secondary echoes that come from the single sheet thickness. Figures 4 and 5 give two A-scan views which show the differences between the echo series of a spot weld with a small diameter nugget and one with a large diameter nugget respectively.

Fig 4: A-scan of test piece 2A dia.5.4 mm (G15MN 8.0).

Fig 5: A-scan of test piece 6A dia.8.5 mm (G15MN 8.0).

5.2 Penetration
The second important factor in the determination of spot weld quality is the degree of penetration. The thickness of the nugget can be predicted by comparing the attenuation characteristics of the test pieces and those of the reference specimens. The thicker the nugget, the higher the attenuation factor will be. This attenuation differences can be seen by examining the decay rate of echo heights corresponding to the double sheet thickness. Figures6, 7 and 8 give A-scans of three spot welds with different nugget thicknesses.

Fig 6: A-scan of test piece 4A(G15MN 8.0).

Fig 7: A-scan of test piece 9B(G15MN 8.0).

Fig 8: A-scan of test piece 10F(G15MN 8.0).

5.3 Internal Discontinuities
The most popular non-destructive testing method for the determination of internal discontinuities, such as voids, porosity, internal cracks and inclusions, is radiographic examination. The ultrasonic multiple pulse echo technique also gives very satisfactory results in the determination of internal discontinuties except in the case of cracks. The A-scans of a small weld nugget containing porosity are given in Figures 9 and Figure 10. These A-scans were obtained by using two different probe diameters -8.0 and 3.6 mm-.

Fig 9: A-scan of test piece D42 (G15MN 8.0).

Fig 10: A-scan of test piece D42 (G15MN 3.6).

5.4 Expulsion
Expulsion is another type of defect that can be detected by the ultrasonic multiple pulse echo technique. Basically the detection procedure is similar to the procedure that is used in the determination of external discontinuties. However since a large amount of metal splashes out of the molten metal pool during the welding process, the size of the void is much larger and the internal surface of the void is much more irregular. This size difference and irregularity help the operator to differentiate an expulsion from an internal discontinuity. Although in both the case of an expulsion and internal discontinuity there are echoes nearly corresponding to the top plate thickness, if the A-scan of a spot weld with expulsion is examined, it can easily be seen that the size and the irregularity of the void scatter the ultrasonic waves and as a result prevent the appearance of echo series. Therefore, a spot weld with expulsion gives only one or two echoes corresponding to the top and bottom plate (Figures 11 and 12).

Fig 11: A-scan of test piece D37 (G15MN 8.0).

Fig 12: A-scan of test piece D37 (G15MN 3.6).

6. Shear Strength Prediction of Resistance Spot Welds

This section introduces a method to predict the tensile-shear strengths of resistance spot welds by examining the ultrasonic multiple pulse echo test results. The main idea of this method is to construct a correlation between the attenuation coefficients and the tensile-shear strengths of the resistance spot welds.

The printed A-scans of the spot welds were used in the determination of the attenuation coefficients which can be determined with the help of the decay rate of the echo heights and the equation which shows the relation between the signal amplitude and the attenuation coefficient. The heights of each echo that comes from the double thickness sheet were determined in terms of the percent screen height. Then these amplitude values and corresponding traveled distances were entered to a curve fitting program which uses the Levenberg-Marquart method to solve non-linear regressions.

After the ultrasonic tests, the specimens were destructively tested under tensile load with a tensile testing machine and their shear fracture loads were measured. It was seen that failure occurred along the plane of junction between the two sheets and that the failure region could clearly be differentiated on the sheet surface. The diameter of each welded region was also measured. The results of the destructive and non-destructive tests are summarized in Figure 13 and Figure 14.

Fig 13: Shear fracture load vs nugget diameter.

Fig 14: The shear fracture load vs attenuation coefficient.

Figure 13 shows the relation between the nugget diameter and the shear fracture load of the tensile shear test specimens. It can be concluded from the graph that the shear fracture load of the specimens increases linearly with the increasing diameter, as was expected [7]. The data scatter very little from their expected orientation.

There can be four main reasons for the scatter. The first one comes from the uncertainty in the destructive shear fracture load measurements. The maximum error that is expected at that point is ± 98 N. The second one is related to the diameter determination step. Since the diameters of the welded regions were measured visually, some mistakes in determining the weld area could have occurred in addition to some reading errors from the milimetric scale. The maximum reading error is expected as ± 0.1 mm. The third possible cause for the scatter is related to the effect of internal discontinuities which increase as the welding current increases. The last one is related to the nugget thickness which is the third dimension of a spot weld.

Although the diameters of any two spot weld are the same, since their degrees of penetration are different, their shear fracture loads are also be expected to be different. Figure 14 shows the relation between the attenuation coefficient and the shear fracture load. As can easily be seen from the Figure, the shear fracture load increases with increasing the attenuation coefficient as expected.

Figure 15 compares the main measurements and the attenuation coefficient measurements of the second operator. As can be seen from the Figure, although there are some differences between the two measurements of the same parts, they follow the same trend. It is expected that the differences between the measurements will become smaller as the operator's experience with the test pieces increases.

Fig 15: Comparison of the main attenuation coefficient measurements and the second operator measurements.

7. Discussion and Conclusion

In this study, in addition to providing a basic guideline related to the non-destructive testing of resistance spot welds with the ultrasonic multiple echo technique, a method which makes it possible to predict the shear strengths of the spot welds by using the same technique has also been suggested. In order to apply this method properly, together with a high resolution ultrasonic flaw detector, a high frequency and high resolution probe specially designed according to the required nugget size for the selected material and thickness should be used. For this test to be succesfully performed, it is also necessary that the A-scan data be well-evaluated and that the origins of the secondary echos be determined correctly. This is directly related to the experience and the adequacy of the technical knowledge of the operator.

It is also possible to predict the shear strengths of resistance spot welds with the help of sound attenuation characteristics. However, achieving reliable results depends on the construction of a reliable referance line of shear fracture load vs attenuation coefficient for the selected material type and material thickness. Additionally, as well as the measured attenuation coefficient, the height and the characteristics of the secondary echoes also help to determine whether the shear fracture load of the specimen is lower than the value shown on the reference line or not.

REFERENCES

1. Krautkramer, J., Krautkramer, H., "Ultrasonic Testing of Materials", Springer-Verlag, New York, 1983.
2. Vahaviolos, S.J., Carlos, M.F., Slykhous, S.J., "Adaptive spot weld feedback control loop via acoustic emission.", Materials Evaluation, Vol. 69, 1981, pp. 1057-1060.
3. Rokhlin, S.I., Adler, L., "Ultrasonic method for shear strength prediction of spot welds.", Journal of Applied Physics, Vol. 56, No. 3, 1984, pp. 726-731.
4. Rokhlin, S.I., Mayhan, R.J., Adler, l., "On-line ultrasonic lamb wave monitoring of spot welds.", materials Evaluation, Vol. 43, No. 7, 1985, pp. 879-883.
5. Bendec, F., Perec, M., Rokhlin, S.I., "The ultrasonic lamb waves method for sizing of spot welds.", Ultrasonics, Vol. 22 No. 2, 1984, pp. 78-84.
6. Mansour, T.M., "Ultrasonic inspection of spot welds in thin gage steel.", Materials Evaluation, Vol. 46, 1988, pp. 650-658.
7. Zhang, S.C., "Stress intensities at spot welds.", International Journal of Fracture, Vol. 25, No. 5, 1997, pp. 167-185.