Abstract
The evaluation of the Resistance Spot Welding (RSW) that guarantees satisfactory performance of mechanical characteristics without altering physical properties can be reached by modeling the input parameters such as current, welding time, and applied force from which each unit has been built and correlating with digital images of the surface and infrared images that allows to identify variations on the parameters that modify the quality of the welding spot [1]. With this, mechanical and surface characteristics can be detected without the need for a mechanical test that modifies the structure of the unit. The database serves as a comprehensive record of the welding spot process, including the monitor of crucial input parameters such as current and force. The constructions and documentation of the testing platform through the instrumentation of a resistance welding will assess the variability of the input parameters and their impact on the output in surface and thermographic imaging, welding nugget diameter and it's mechanical strength. Additionally, it documents characteristics of the material used as thickness and material type and its output as the mechanical resistance and nugget diameter, along with its corresponding classification. Thus, the database not only captures the details of the welding process, but it also provides a valuable resource for analyzing and evaluating the performance of the welding operation.
Keywords: Spot welding, Infrared images, Surface images, Current, Force, Welding time
Specifications Table
| Subject | Engineering |
| Specific subject area | Resistance Spot Welding |
| Type of data |
Table, Image |
| How the data were acquired | The record of the input parameters was performed by the integration of an Arduino ONE, connected to sensors to monitoring the input parameters such as current, welding time and pressure. Images of the melting point (nugget) were taken with a thermographic camera and digital camera. |
| Data format | Raw, Analyzed |
| Description of data collection | The following inputs parameters were recorded as a tabular data and saved as a csv file:
|
| Data source location | Universidad Autónoma de Querétaro, México |
| Data accessibility | Repository name: Mendeley Data identification number: 10.17632/rwh8kjzdch.2 Direct URL to data: https://data.mendeley.com/datasets/rwh8kjzdch/3 |
1. Value of the Data
-
•
Resistance spot welding is a process used to join two metal sheets by applying high pressure and electric current to metal sheets [2]. This data describes the main parameters of resistance spot welding including:
-
•
Electrode force: The amount of pressure applied to the metal sheets by the electrodes during the welding process [3]. This affects the quality and strength of the weld and is typically measured in newtons (N).
-
•
Welding current: The amount of electrical current flowing through the metal sheets being welded [3]. This affects the amount of heat generated during the welding process and is typically measured in amperes (A).
-
•
Welding time: The time that current is applied to the metal sheets to create the welding spot [3]. This parameter is typically measured in milliseconds (ms).
-
•
Electrode angle: The angle formed by the electrode tips when they contact the workpiece.
-
•
Additionally, data includes information about the material such as material thickness, and material type.
-
•
For the decision attribute, a mechanical test was measured in accordance with industry standards AWS D8–9 [4]. The standardized mechanical test was chosen due to its reliability in accurately measuring the decision attribute, providing a dependable basis for making an informed decision.
-
•
Finally, the use of digital and thermographic images is a key component in the relationship between input parameters for each unit build. This allows to have a better understanding of how input variables affect the output. Digital images were taken front and back of the welding spot surface.
-
•
Table 1 describes the input and output parameters, and the number of samples.
-
•
This data will benefit multiple engineering disciplines such as Artificial Intelligence (AI). By analyzing this data, AI models can be trained to accurately categorize and predict any defect in the welding process, as well as to identify correlations between the variations from the input parameters. Through this process, AI can be used to improve the overall efficiency and quality of welding, leading to greater productivity and improved end products. As a result, the data produced by the resistance spot welding process can have a significant impact in the engineering field.
-
•
This database can be used but is not limited to Machine Learning and Deep Learning models which are two of the most popular supervised AI models, and they both rely on labeled data to create accurate results. It is essential that the data is labeled accurately for the AI models to be successful.
-
•
Supervised Artificial Intelligence models are a powerful tool that are used in a variety of industries. To train these models, labeled information is required to provide the models with the information needed to learn features to make accurate predictions.
-
•
This database is a component of a research study aimed at assessing the efficacy of artificial intelligence techniques in resistance welding spot analysis [5].
Table 1.
Input and output parameters.
| Input / output | Parameter | Number of samples |
|---|---|---|
| Input | Pressure | 495 |
| Input | Welding time | 495 |
| Input | Electrode angle | 495 |
| Input | Electrode force | 495 |
| Input | Welding current | 495 |
| Input | Material thickness | 495 |
| Input | Material type | 495 |
| Output | Pull test force | 495 |
| Output | Nugget diameter | 495 |
| Output | Category | 495 |
| Output | Digital images | 990 |
| Output | Thermographic images | 495 |
2. Objective
Generate a database by obtaining input parameter data through the instrumentation of a small-scale resistance spot welding machine to determine the variability of the input parameters and their influence on the output represented in processed images.
3. Data Description
The data for this experiment were divided into three files.
The first file, a csv, contained 8 attributes described in Table 2.
Table 2.
Description of dataset attributes.
| N. | Attribute | Format | Description |
|---|---|---|---|
| 1 | Sample ID | Numeric | Unique identification number for each sample |
| 2 | Pressure | Numeric | Pressure on pneumatic cylinder |
| 3 | Welding time | Numeric | Welding process time |
| 4 | Electrode angle | Numeric | Angle between the electrodes |
| 5 | Electrode force | Numeric | Force applied to the electrodes |
| 6 | Welding current | Numeric | Current flowing though the metal sheet |
| 7 | Material thickness | Numeric | Thickness of the material |
| 8 | Material type | Categorical | Composition of the material |
| 9 | Pull test force | Numeric | Mechanical resistance of the welding join |
| 10 | Nugget diameter | Numeric | Diameter of welding nugget |
| 11 | Category | Categorical | Category of welding spot (good, bad, explode) |
The second file contained the surface images of each sample taken from the front (F) and back (B) of the welding spot and name according to the same ID as the first file. All surface images of the welding spot were preprocessed to a standard size as shown in Fig. 1, including a scale bar as a reference of its dimension.
Fig. 1.
Example of a surface image of spot welding process standardized to 350 × 350 pixels.
The third file contained infrared images named according to the sample ID. All images were preprocessed to a standard size as shown in Fig. 2, including a scale bar is included to represent the dimension of the spot welding.
Fig. 2.
Sample of infrared image of spot welding process standardized to 300 × 300 pixels.
4. Experimental Design, Materials, and Methods
The instrumentation of the small-scale resistance welding machine Chicago Electric model 61205 consists of an Arduino UNO as a control board for the record of the force and current parameters, as well as determining the cycle time of the process.
A strain gauge and analog-to-digital converter were used to process and record the applied force to the electrodes with a frequency of 10 Samples Per Second (SPS). The specification of the strain gauge is described in Table 3, and the electronic circuit is presented in Fig. 3.
Table 3.
Specification of strain gauge.
| Attribute | Specification |
|---|---|
| Model | PSD-S1 |
| Maximum load | 300kg |
| Material | Metal |
| Type | Weighting |
| Output signal | Analog |
| Protection grade | IP67 |
| Linearity | ±0.02 %F.S. |
| Delaying | ±0.02 %F.S. |
| Repeatability | ±0.02 %F.S. |
| Sensitivity | 2.0 ± 0.004 mv/v |
Fig. 3.
Electronic circuit for strain gauge.
A Current transformer and current meter were used to process and record the applied current flowing through the metal sheets with a frequency of 10 Samples Per Second (SPS). The specification of the Current Transformer is described in Table 4, and the electronic circuit is presented in Fig. 4.
Table 4.
Specification of current transformer.
| Attribute | Specification |
|---|---|
| Model | TS-816 |
| Material | ABS |
| Manufacturer | Taiwan |
| Frequency | 50–60 Hz |
| Class | 1.0 |
| Primary current | 6000A max. |
| Secondary current | 0–5A |
Fig. 4.
Electronic circuit for current transformer.
For the application of the force, a double-acting pneumatic cylinder was used to control and regulate the force applied to the system, the activation of the cylinder is performed by a solenoid valve. The electronic and pneumatic circuit is presented in Fig. 5.
Fig. 5.
Electronic and pneumatic circuit of force.
The welding time cycle was controlled and measured using the Arduino UNO using the function millis() to measure the time (in milliseconds). Cycle time was set from 200 ms to 1500 ms for the different tests.
The system is controlled by a program following the diagram flow presented in Fig. 6 was implemented to register and control the system. The program initiates by conducting calibration of the force and current sensors, utilizing the libraries “HX711_ADC.h” and “ACS712.h” respectively. The program then go to a waiting state until the user presses the start button, sending a signal to one of the digital inputs of the Arduino UNO. After this, the digital output connected to a relay is activated, inducing a change in the state of the solenoid valve and activating the pneumatic cylinder. A waiting period of 1000 milliseconds, representing the compression time, then a digital output of the relay that triggers the resistance welding machine is activated. The program then proceeds to gather and display the measurements recorded by the electrode current and pressure sensors until the predetermined cycle time (welding time) is completed. Subsequently, the digital output connected to the welding machine relay is deactivated, followed by the deactivation of the digital output of the relay connected to the solenoid valve, which shuts down the pneumatic cylinder.
Fig. 6.
Data acquisition diagram flow.
A total of 495 units were welded varying electrode forces, electrode angles, and welding times. Set-up parameters and registered parameters were saved into .csv file.
The specifications of the tip electrodes are described in Table 5.
Table 5.
Specification of tip electrodes.
| Attribute | Specification |
|---|---|
| Material | Cooper Alloy |
| Contact diameter | 1/8″ |
The samples were built according to AWS D8-9 standard, the samples were overlapped and welded together, geometrical dimensions of the testing samples are presented in Fig. 7.
Fig. 7.
Size and configuration of welding samples.
Pull tests were conducted using a Tinus Olsen testing machine with a maximum load capacity of 300KN and resolution of 0.1N. After conducting the test, the nugget diameter was measured with an electronic caliper with a resolution of 0.01 mm and categorized into good, bad, and samples with spatter.
Results for the mechanical test are presented in Fig. 8, presenting a maximum and minimum values of 5806.5 N and 1410.3 N respectively, with a standard deviation of 437.76, This variability indicates that further investigation is necessary to better understand the underlying factors affecting these properties.
Fig. 8.
Results of pull test.
Additionally, the failure mode of the pull testing was categorized into interfacial failure (IF) caused when the breaking point occurs in the welded zone, and pull-out failure (PF) when the breakage occurs in the base material adjacent to the welding spot. An example of each category is presented in Figs. 9, and 10 for infrared images.
Fig. 9.
Classification of welding spot (a) good, (b) bad, (c) spatter.
Fig. 10.
Infrared images of welding spot (a) good, (b) bad, (c) spatter.
Images were taken from a distance of 15 cm for RGB images, and 10 cm for infrared images. Infrared images were captured immediately after the process was performed during the cooling stage, since in process monitor was not possible due to a reflection of the metal. Nonetheless, images show the size, shape, and propagation of the heat affecting zone.
The state of the electrodes was monitored after each sample, and in the case of material splatter on the tip, a file was used to remove the excess material to restore the spot diameter and face as shown in Fig. 11.
Fig. 11.
Electrode with splatter material on the tip (a), and restored to original shape (b).
Ethics Statement
The data resulting from the experiment do not involve human subjects, animal experiments, or information from social media platforms.
Acknowledgments
We are grateful for the financial support provided by CONACYT through the grant for Luis Alonso Domínguez Molina, as well to the Universidad Autónoma de Querétaro for the financial support FONDEC project (20205007071001).
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data Availability
References
- 1.Williams N.T., Parker J. Review of resistance spot welding of steel sheets part 1 modeling and control of weld nugget formation. Int. Mater. Rev. 2004;49:45–75. [Google Scholar]
- 2.Summerville C., Compston P., Doolan M. A comparison of resistance spot weld quality assessment techniques. Proc. Manuf. 2019;29:305–312. y. [Google Scholar]
- 3.Srikunwong D.C. Resistance welding numerical simulation. Rev. Eur. Élém. Finis. 2004;13:313–341. [Google Scholar]
- 4.American Welding Society (ANSI); 2012. Test Methods for Evaluating the Resistance Spot Welding Behavior of Automotive Sheet Steel, D8.9M:2012. [Google Scholar]
- 5.Domínguez L. Universidad Autónoma de Querétaro; 2024. Evaluación Del Proceso De Soldadura Por Resistencia Por Medio De Imágenes Termográficas y No Termográficas. M.S. Thesis. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.