See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/42339032

Collection of arc welding process data

Article  in  Journal of Achievements of Materials and Manufacturing Engineering · April 2006

Source: DOAJ

CITATIONS READS

14 3,651

2 authors:

Krzysztof Luksa Zbigniew Rymarski

Silesian University of Technology Silesian University of Technology
12 PUBLICATIONS   74 CITATIONS    45 PUBLICATIONS   225 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Green Railway Crossing – study of the zero emission system raising the safety standards of traffic users in the railway-road crossing areas View project

Methodology for the selection of magnetic materials for elements of electrical power conversion systems based on measurements of their actual parameters View
project

All content following this page was uploaded by Zbigniew Rymarski on 22 May 2014.

The user has requested enhancement of the downloaded file.

 VOLUME 17
ISSUE 1-2
of Achievements in Materials July-August
and Manufacturing Engineering 2006

Collection of arc welding process data

K. Luksa a,*, Z. Rymarski b
a Welding Department, Silesian University of Technology,
ul. Konarskiego 18a, 44-100 Gliwice, Poland
b Institute of Electronics, Silesian University of Technology,
ul. Akademicka 16, 44-100 Gliwice, Poland
* Corresponding author: E-mail address: : [email protected]
Received 15.03.2006; accepted in revised form 30.04.2006

Manufacturing and processing

Abstract
Purpose: The aim of the research was to examine the possibility of detecting welding imperfections by recording
the instant values of welding parameters. The microprocessor controlled system for real-time collection and
display of welding parameters was designed, implemented and tested.
Design/methodology/approach: The system records up to 4 digital or analog signals collected from welding
process and displays their run on the LCD display. To disturb the welding process artificial disturbances were
introduced.
Findings: The occurrence of some welding imperfections is followed by changes of the welding parameters. In
this case they can be revealed by the analysis of the instant values of the welding parameters.
Research limitations/implications: In the paper results of monitoring manual metal arc welding and gas metal
arc welding are presented.
Practical implications: Monitoring of gas metal arc welding is a good tool for evaluation of the quality of weld.
All introduced, artificial disturbances of the welding process destabilize the welding arc and produce changes in
the instant values of the welding parameters.
Originality/value: The paper presents a modern microprocessor controlled system for real-time collection and
display of welding parameters. Results of tests show that simple statistical approach to welding parameters can
help in evaluation of weld quality.
Keywords: Quality assessment; Welding; MAG; Monitoring

1. Introduction
1. Introduction process affect usually the stability of welding arc [10,11]. Some of
them produce momentary instability of welding arc and other make
the arc unstable as long as the disturbance affects the welding arc.
Welding process emits sound, light and electromagnetic
Speed of analogue-digital conversion is a compromise
radiation to surrounded space. Sound emission and light emission between quantity and quality of recorded data. A great many of
were exploit as a source of information about the process [1,2,3,4] recorded data enlarge time of computation but limited number of
but till now welding parameters are counted as a main source of data can be a reason of missing essential information. Previous
data for welding process monitoring which can be essential part of work in monitoring of arc welding processes were emphasized on
online evaluation of weld quality [5,6]. Parameters of welding are finding the best way for analyzing recorded signals and
controlled by electronic control units [7,8] but their modulation by application the statistical quality control in monitoring of welding
welding process can be exploited for monitoring of welding process [12,13]. The aim was to distinguish sound weld and areas
process. The design of the monitoring system should predict the of weld were imperfection are expected. It is possible to find
high electromagnetic interferences during the welding process on faulty area of weld and find the reason of process disturbance by
the microprocessor-controlled system [9]. Disturbances of welding using statistical method or computational intelligence [14,15].

© Copyright by International OCSCO World Press. All rights reserved. 2006 Short paper 377
 Journal of Achievements in Materials and Manufacturing Engineering Volume 17 Issue 1-2 July-August 2006

Condition of data recording have been neglected in those research development of the device is possible owing to the PLD-based
because laboratory recorders can work in wide range of “chipset” with in-circuit programming, the “open” architecture
parameters. Presented in this paper monitoring device intended with external digital and analog buses and the “oversized”
for industry use has fixed recording parameters and input circuit interrupt controller.
dedicated for welding processes.

2. Description of the measuring 2.2. External

2.2. External digital transmission
digital and analog
transmission and
2. Description
system of the measuring system buses of the system
analog buses of the system

The system has an external digital bus that is used for

2.1. Spread
2.1. Spread architecture
architecture of the design
of the system connecting controllers of amplification ratios for the analog
system design inputs. It has 3 serial ports – one for external data transmission,
The architecture of the measuring device, Fig. 1, is based on the second for testing and maintenance of the device, the third
the modern “spread” architecture – there are four 8-bit slave- port, with all the modem lines included, is prepared for (not
controllers that perform the local tasks. The main processor used yet) transmission options IrDA or USB converter. There is
measures and converts input data, its storage in memory and a standard parallel port for printer because the report with
reads it during the external transmission and supervises the welding parameters is required to confirm the high quality of the
supply with a backup. The external transmission is controlled by produced welds. The analog bus with the normalized input
a specialized processor. Basic parameters of each welding range 0-2.5 V is designed for 8 analog inputs converted with 12-
process are computed locally. The device can measure AC (true bit resolution, however, only current meter, voltage and gas
RMS) and DC voltage (0 - 60 V) and AC or DC current (0 - flow sensors have analog outputs; the wire speed is measured by
1000 A). The design of the device is prepared for low-level the encoder that gives 3 series of pulses counted in the motion
technology manufacturing (the device preferably contains PLCC controller. Two inputs measure the voltage of the memory
and DIP devices and only a few SMD devices). The future battery and the main supply battery.

Fig. 1. The block diagram of the measuring system

378 Short paper K. Luksa, Z. Rymarski

 Manufacturing and processing

2.3. Data
2.3. Datamemory
memorystructure
structure 3. Experiments
3. 
Experiments
The sufficient frequency of measurements is 128 The aim of the research was to examine the possibility of
measurements of each parameter per second, every consecutive detecting welding imperfections by recording the instant values of
8 measurements are summed and averaged and the mean value is welding parameters. Disturbances were implemented in three
stored in the memory (16 mean values per second). areas 25 mm in length. These areas were separated by clean areas
With 4MB SRAM, 12-bit data converters, predicting 2 bytes 50 mm in length. The beginning and the end of weld were also
space for the each stored data, and with 4 input data channels, the free of disturbances.
system is able to measure, collect and store data through out the Welds on plates were produced manually. Two types of
whole work day – about 32000 s. The memory organization covered electrode: rutile-acid electrode and basic electrodes of
allows storage of records of various length. The allocation table in diameters 3.15 and 4.0 mm were used for MMA welding.
the memory describes the structure of the memory. For MAG welding the wire SG3S1 of diameter 1,2 mm and
gas mixture M21 were used.
Beads on plate were executed on S235 steel sample tests.
2.4. Local
2.4. Localdata
datavisualization
visualization Sample tests were 250 mm long, 100 mm wide and 6-8 mm thick.
The research covered the area of MMA welding in the range
The measurements are presented on the graphic LCD display – of current from 130 A to 180 and MAG welding in the range of
the average values for each 0.25 second are displayed. The typical current from 160 A to 310 A. Three levels of welding current
time 60 s of welding one pass can be visualized on one screen. All the were used to cover the recommended range of parameters for the
stored records can be independently looked up, and the basic data of consumables that were used.
the particular welding process can be printed locally in the measuring To disturb the MMA welding process three types of
system as a report. After measurements the device can transmit the disturbances were introduced: a layer of paint, a layer of grease
stored data to a PC where the supervisory program analyses the whole and a layer of oil. A lack of shielding gas was introduced as the
process and estimates in detail the welding parameter runs. The forth disturbance during MAG welding.
measured duration time of the welding process, and the real time of Welding current and welding voltage were recorded during
the welding beginning is given in this report. Data collecting can be MMA welding and welding current, welding voltage, wire speed
started by pressing a key on the keyboard or automatically in the case and gas flow were recorded during MAG welding.
when the welding current exceeds the 50 A level.

2.5. Power
2.5. Powersupply
supply 4. Results andand
4. Results discussion
discussion
The system works with an external battery supply and the All test coupons were visually and radiographically examined.
memory has an additional backup supply from a small battery located In the areas where disturbances were implemented macroscopic
on the main board. The system cannot be switched off if the data examination of the weld cross section were performed. Results of
storage is not completed. If the system would be reset during data these examination were compared with the runs of welding
collection, the data set would not be damaged. In this case the user parameters.
can use the option of the automatic recalculation of data. The main There are three types of welding process disturbances. The
processor holds a “soft” switch on till the end of the current action. first influences the stability of welding arc, the second influences
welding parameters and the third influences the metallurgical
reaction in the arc and welding pool. It seems that three applied
2.6. EMC
2.6.  of thethe
EMC of system
system disturbances affect weld metal in the same way, all of them are
able to produce gas pores. It is necessary to note that welding arc
Measuring currents up to 1000 A and voltages to 60 V, deliquesces grease and oil and blow them out from the welding
requires special care for the system EMC [9] and requires the area. Only a part of fumes and vapors is absorbed by the arc
implementation of some extra data and address buffers designed column. A layer of paint burns in the welding arc and most of
into the PLD technology. An isolation amplifier (in the voltage products of its disintegration are absorbed by arc.
measuring channel) and amplifiers with high input over-voltage Monitoring devices can only reveal disturbances and
protection are used as well the most serious problem with noises imperfections that influence the welding parameters, mainly
appears when measuring parameters of the TIG welding, in which welding voltage and welding current. Both of these parameters
the air gap between the electrode and welded material is ionized depend on potential of ionization, which depends on the chemical
with almost 10 kV pulses before firing the arc. The graphics composition of welding arc stream. Changes in the potential of
display with its own controller can easily be disturbed, and in ionization influence the temperature of the welding arc, voltage
some cases it can block the action of the whole system. The drops in the arc stream, cathode and anode region and arc current
problem is to design a high efficiency low-pass filter dumping and density.
cutting the spikes on the isolated voltage input, and EMI filters on Visual examination and X-ray examination of MMA welds
the other inputs. However, in the other methods of welding, the shows that this method of welding is to a minor degree sensitive
described system measures parameters perfectly. to applied disturbance. Only a layer of paint develops minor gas

Collection of arc welding process data 379

 Journal of Achievements in Materials and Manufacturing Engineering Volume 17 Issue 1-2 July-August 2006

pores and changes of parameters in this area are not clearly

visible. A layer of grease and a layer of oil did not cause visible
References
References
changes in the radiograms and plots of runs of parameters.
Minor indications in the plots were caused by the silica borders of [1] K. Luksa, M. WĊglowski, An experimental study on the
contaminated area. relationship between GMA shielding gas composition and
During MIG welding shielding gas, silicon and manganese intensity of arc light, APE, Warszawa, 2004, 59-69.
compounds protect the arc and welding pool from oxygen and [2] C.D. Yoo, Y.S. Yoo, H-K. Sunwoo, Investigation on Arc
nitrogen. There is no possibility to neutralize chemical compounds Light Intensity in Gas Metal Arc Welding, Part 1,
which penetrate the arc. All disturbances influence metallurgical Relationship Between Arc Light Intensity and Arc Length,
processes in the welding arc and the weld pool also influence the Proceedings of the Institution of Mechanical Engineers, Part
stability of welding arc. In all cases disturbances produce B, Journal of Engineering Manufacture, 211, Part B, 1997,
imperfections at the surface of the weld and also produce internal gas 345-353.
pores. It is seen that the intensity of welding parameters change is [3] K. Luksa, Investigation of arc sound in gas metal arc
directly proportional to intensity of disturbance. This also happened welding, APE, Warszawa, 2004, 79-86.
when the gas shield of the arc was penetrated by oxygen and nitrogen [4] D. Saini, S. Floyd, An investigation of Gas Metal Arc
from the air, Fig. 2. Welding Sound Signature for On-Line Quality Control,
DISTANS [mm] Welding Journal, 4 (1998) 172s-179s.
0 20 40 60 80 100 120 140 160 180 200 220 [5] D. Barborak, C. Conrardy, B. Madigan, T. Paskell, Through-
400 20
arc” process monitoring: techniques for control of automated
350 gas metal arc welding, IEEE International Conference on
SHIELDING GAS FLAW Q [l/min]

16
Robotics and Automation, Proceedings on Robotics and
WELDING CURRENT I [A]

300
Automation, 10-15 May, 1999, v. 4, 3053 - 3058.
250 12 [6] S. Adolfsson, A. Bahrami, G. Bolmsjo, I. Claesson I., On-
200 Line Quality Monitoring in Short-Circuit Gas Metal Arc
150 8 Welding, Welding Journal, 2, (1999), 59-73.
[7] P. Verdelho, M. Pio Silva, E. Margato, J. Esteves, An
100
4 electronic welder control circuit, Industrial Electronics
50
I
Q
Society, IECON '98. Proceedings of the 24th Annual
0 0
Conference of the IEEE, Aug.-Sept., 1998, v. 2, 612 – 617.
0 5 10 15 20 25 30 35 40 45 [8] J. Schupp, W. Fischer, H. Mecke, Welding arc control with
TIME OF WELDING[s] power electronics, Eighth International Conference on
Power Electronics and Variable Speed Drives, 18-19 Sept.,
2000, 443 – 450.
[9] Zhang Xiaodong; Cui Xiaoyan; Wen Yinghong, Research
on EMC of inverter welder and welding arc, 3rd
Fig. 2. Instant value of welding current and welding voltage International Symposium on Electromagnetic Compatibility,
during GMA welding. Lack of gas shield as a disturbance of the 21-24 May 2002, 450 – 453.
welding process is seen in the plots and the weld [10] K. Luksa, Influence of weld imperfection on short circuit
GMA welding arc stability, Proceedings of 12 International
Scientific Conference “Achievements in Mechanical and
Materials Engineering” ‘AMME’2003, Gliwice-Zakopane,
5. Conclusions
5. Conclusions (2003), 617-622.
[11] M.I. Hermans, G. den Ouden, A stability criterion for short
The measuring system for arc welding processes, has circuiting gas metal are welding. Proceedings of the
a modern, easy to expand structure. It measures and records the International Conference on the Joining of Materials,
most important welding parameters. It enables online evaluation Helsingor, 1997, 112-119.
of welding data and a detailed post-weld analysis of data on a PC. [12] K. Luksa, Diagnosing of short circuit gas metal arc welding
Tests of the device performed during manual metal arc welding process, Proceedings of 3 Scientific Conference ‘M3E 2005’
and gas metal active welding revealed that the occurrence of some Gliwice-Wisáa, (2005), 977-984.
welding imperfections is followed by changes of the welding [13] U. Dilthey, T. Reichel, W. Scheller, Statistical process
parameters. In this case they can be revealed by the analysis of the parameter surveillance in GMA welding. Proceedings of
instant values of the welding parameters. Experiments show that International Conference on the Joining of Materials,
monitoring gas metal arc welding is a good tool for evaluation of JOM 7, Helsingor, 410-422.
the quality of weld. All introduced, artificial disturbances of the [14] U. Dilthey, J. Heidrich, T. Reichel, On-line quality control
welding process destabilize the welding arc burning and produce in gas-shielded metal arc welding using artificial neural
changes in the instant values of the welding parameters. In the networks. Schweissen & Scheiden, 49 (1997), 2, E22-E24.
case of manual metal arc welding changes of instant values of the [15] C.S. Wu, T. Polte, D. Rehfeldt, A Fuzzy Logic System for
welding parameters are barely noticeable and can be missed Process Monitoring and Quality Evaluation in GMAW.
during online study. Welding Journal (2001), February, 33s-38s.

380 Short paper READING DIRECT: www.journalamme.org

View publication stats