Home

Summer Workshops Administration Members Benchmark Problems Other

Centro de Investigacion Industrial
Av. Leandro N. Alem 1067
1001 Buenos Aires, Argentina

Center for Industrial Research 
Alberto Pignotti 
FUDETEC 
Campana, Argentina

BROCHURE

FUDETEC, a non-profit foundation created in Argentina nearly ten years ago by the Techint group of companies, runs a Center for Industrial Research where our work on NDT is performed. The final aim of our activities is the improvement in industrial practices leading to higher standards of productivity and product quality.

Because our Center is located within the premises of SIDERCA, a seamless steel tube manufacturing plant, our work in nondestructive testing has focused on the inspection of steel tubes by magnetic flux leakage (MFL). The analysis of this technique involved the numerical modeling of the leaked field, the measurement of magnetic properties of various steels, laboratory and plant measurements of signals generated by machined and natural flaws, a Monte Carlo simulation of the probability of detection and of occurrence of false alarms, and a model evaluation of the improvement of detection effectiveness through the use of artificial neural networks to process the signal. We also devoted considerable effort to the development of a robust industrial system for digitizing and processing in real time the signals from conventional MFL equipment.

Our laboratory includes a facility for the MFL testing of pipes and an immersion tank for ultrasonic inspection. We use the latter for the detection of macroinclusions in samples of continuous cast steel, for studying ultrasound attenuation, and for analyzing geometries in flaw detection in steel tube walls. 


Center for Industrial Research
Alberto Pignotti 
FUDETEC Campana, Argentina

FUDETEC stands for the Spanish words that denote the Foundation for Technological Development, a non-profit organization that was created by a group of companies belonging to a corporation that is active in steel, engineering, construction, oil, and other related industrial activities. This set of companies, collectively denoted as the Techint Organization, originated in Argentina nearly half a century ago, but by growth and purchase have spread over several countries, including Italy, Mexico, Venezuela and even the United States.

Almost one decade ago this corporation decided to create FUDETEC, the main purpose of which is to run CINI, a Center for Industrial Research where, among other activities, our nondestructive work is performed. The idea was to have an instance where industrial problems could be examined by people not directly involved in the manufacturing work, but in close contact with it. Such people have both stronger scientific background and more time available than those devoted to production, and are therefore in a better position to develop a deeper understanding of manufacturing techniques and of the performance of industrial products in service. The final aim of such activities is the improvement in industrial practices leading to higher standards of productivity and product quality.

Our Center is located within the premises of SIDERCA, a seamless steel tube manufacturing plant in Campana, some 50 miles Northwest of Buenos Aires, Argentina, and because of that reason, we have focused our work in nondestructive testing on the inspection of steel tubes. In more recent times, two similar seamless tube manufacturing plants have been incorporated to the Techint group: one is Dalmine, located near Bergamo, in northern, Italy, and the other one is Tamsa, in Veracruz, Mexico. Thus, the geographical area to which we have potential access spreads from Argentina to North America and Europe. This implies a two-way relation with these manufacturing plants: having access to their problems and providing or implementing in them whatever technical improvement we may develop.

Our methodology is roughly the following: after a problem or issue is raised, a simplified numerical model is formulated to provide guidelines and identify the relevant variables and process parameters, and their interdependence. Laboratory tests, performed either at CINI or outside facilities, are used to determine material properties which are essential ingredients of the models. More elaborate finite-element three-dimensional models are also performed when detailed quantitative results are required. And validation through laboratory and plant tests is the decisive criterion to judge the final applicability of a proposed course of action. Plant implementation is of course performed in a joint effort with production lines. Collaboration with experts from other research institutions is sought whenever a needed expertise is not found in our staff.

Because flaw detection by magnetic flux leakage (MFL) is widely used in nondestructive evaluation of steel pipes, our major efforts have been devoted to this technique. The main problem here is to ensure a high probability of detection of flaws that exceed some specified critical value, while minimizing productivity losses due to the inevitable presence of false alarms. The analysis of this problem involved the numerical modeling of the leaked magnetic field in the presence of defects, the measurement of magnetic properties of various steels, laboratory and plant measurements of signals generated by machined and natural flaws, a Monte Carlo simulation of the probability of detection and of occurrence of false alarms, and a model evaluation of the improvement of detection effectiveness through the use of artificial neural networks to process the signal. Because we aim at the implementation of this or alternative data processing techniques in on-line inspection, we have devoted considerable effort to the development of a robust industrial system for handling in real time the digitized signals from existing conventional MFL equipment. Such a system, developed in collaboration with Siderca's personnel, is now operative in one of the existing Siderca lines, and has also been installed in one of Tarnsa's lines.

We also share with Siderca a laboratory where we have a facility for the MFL testing of pipes and an inunersion tank for ultrasonic inspection. We have used the latter for the detection of macroinclusions in the inspection of samples of continuous cast steel, for studying ultrasound attenuation in steel and for analyzing geometries in flaw detection in steel tube walls.

 As a final remark, it should be mentioned that it is rather uncommon for private industries in Argentina to sponsor centralized R&D activities such as those that we carry out at CINI. Even though our budget is still a negligible fraction of the combined sales of the Techint group of companies, it is steadily increasing, and by no means includes all the R&D money spent by such companies. We expect our growth to continue to the extent that we are able to provide answers to the demands that originate in our industrial environment, which includes not only the companies that created our foundation, but any other one willing to resort to us.

 Publications

  • A. Pignotti and A. Kohan, "Nonlinear Model of Steel Tube Sorting Using Eddy Currents", Proceedings of the XIII-th World Conference on Nondestructive Testing, Sao Paulo, Brazil, Elsevier Publishing Co., pp. 336-340 (1992)

  • A. Pignotti and A. Kohan, "Importance of magnetic Saturation Effects in the Detection of Internal Tube Cracks by Magnetic Flux Leakage", Proceedings of the XIIIth World Conference on Nondestructive Testing, Sao Paulo, Brasil, Elsevier Publishing Co., pp. 456- 460(1992)

  • E. Altschuler and A. Pignotti, "Nonlinear Model of Flaw Detection in Steel Pipes by Magnetic Flux Leakage", NDT&E International, Vol. 28, pp. 35-40 (1995)

  • H. Gavarini, R. P. J. Perazzo, S. L. Reich. E. Altschuler and A. Pignotti, "Neural network classifier of cracks in steel tubes", Insight, 38, pp. 108-111 (1996)

  • E. Altschuler, J. Paiuk and A. Pignotti, "Monte Carlo Simulation of False Alarms and Detection Reliability in MFL Inspection of Steel Tubes", Materials Evaluation 54, pp. 1032-4 (1996)

  • H. Gavarini, R. P. J. Perazzo, S. L. Reich. E. Altschuler and A. Pignotti, "Automatic assessment of the severity of cracks in steel tubes using neural networks", Insight 40, pp.92-100 (1998)

 

Back to Top