·Home ·Table of Contents ·Industrial Plants and Structures

Diagnostics and Forecasting of the Residual Life of Steel Structures After Long-Term Service V.I.Zagrebelny, A.A.Dubov, A.V.Mozgovoj, A.N.Rudakov Contact

Application and development of the calculation and technological methods of improvement of strength and fatigue life of steel structures enables development of new advanced machinery and structures [1]. However, a considerable number of machines and constructions whose design life is almost over, are being operated now.
Long-term service of a large fleet of operating metal structures is impossible without its control based on the norms and criteria which determine the operational reliability of the metal.
One of the urgent problems of technogeneous safety is forecasting the service life of structural elements. The steel structure performance is limited for the reason of accumulation in them of internal damage in service due to a prolonged exposure to mechanical and thermal loads, as well as aggressive media. Under the impact of these factors both the internal structure of the material and its actual mechanical properties are changed. There exist a great number of NDT methods which allow determination and forecasting of the internal damage of the parts. Among the new methods attention should be given to the magnetic methods of determination of the stressed-strained state of the metal. Lately the use of the magnetic methods of non-destructive testing in combination with other methods of non-destructive and destructive testing has allowed improvement of the validity of determination of the residual life of metal structures.
The main areas of technical diagnostics of steel structures after prolonged service for forecasting the safe residual service life are:

Fig 1: Block-diagram of forecasting the residual life of the structure

1. detection of internal and external defects using the methods of non-destructive testing;
2. determination of thickness reduction in the structures;
3. determination of metal properties deterioration, i.e. lowering of ultimate strength, yield point, impact toughness and other mechanical properties of the metal;
4. detection of stressed-strained sections in which development of critical defects is the most probable;
5. assessment of the degree of criticality of the found defects by calculation of the coefficients of stress intensity in a steel structure.
Determination of the residual life of a steel structure is a complex problem, which can be represented by a block-diagram showing the individual problems to be solved (Fig. 1).
Let us consider in greater detail each of the problems presented in the block-diagram. It is obvious that in order to calculate the structure residual life we must have data on its condition, i.e. first the diagnostics of the technical condition of the structure is conducted. The diagnostics of the technical condition of the structure should be started with performance of express-diagnostics of the most critical sections of the structure. Let us consider this stage in detail, namely performance of express-diagnostic to reveal the most critical sections of the structure.
It is known that the main sources of damage initiation in various structures are the zones of stress concentration (SC) due to the impact of the working loads. Therefore, a reliable criterion of equipment reliability is its actual stressed-strained state. An effective method in assessment of the stressed-strained state of the structures is the method of material magnetic memory (MMM) [2, 3]. The method is based on irreversible changes of the magnetic state of ferromagnetic items in the zones of stress and strain concentration. A directional and irreversible reorientation of the domain structure of a magnetostriction nature proceeds in these zones, which is still in place after the working loads are relieved. Investigations showed that the change of residual magnetization in tension, compression, torsion and cyclic loading of ferromagnetic items is unambiguously related to the maximal acting stresses. As to labour consumption in control, MMM belongs to express-methods when specialized instruments with a screen and a recording device allow control to be performed at a speed of 100 m/h without any preliminary preparation of the object of control. The main purpose of MMM is detection of the most critical sections and components in the controlled plant, which are characterized by SC zones. Then, the traditional methods of NDT (UT, X-ray, eddy current inspection, etc.) are used to determine the presence of a particular defect. The use of MMM is the most effective for control performed to estimate the residual life of equipment. Assessment of the actual stressed-strained state of the structure allows its strengthening or replacement of a worn component in it.

Fig 2: Diagram of distribution of magnetic field Hp around the perimeter of a butt welded joint in the zone of residual stress concentration: a - Hp epure around the perimeter of the butt with concentration of residual stresses SC (in zones 1-4; along Hp = Oline); b - Hp epure along the lower generatrix of the pipe in zones 1 and 2 with the maximal stress concentration.


Let us consider the application of this method on a specific example of control of a butt welded joint (Fig. 2) [4]. The operator moves a sensor measuring the residual magnetic field intensity, along the weld over the entire perimeter (separately over the weld metal and the heat-affected zone from both sides of the weld) and then transversely to the weld with the amplitude of deviation from the weld edge for 30 to 50 mm towards the base metal of the pipe element. The second operator records in the log book the data on residual magnetization of the metal, namely magnetic field intensity (H_p, A/m) with the plus or minus sign. An abrupt change of the sign and value of H_p points to a concentration of residual stresses along H_p = 0 line for a specific section of the welded joint. These sections are marked with chalk or paint. The maximal concentration of residual stresses is estimated by stage-by-stage additional measurement of H_p value at the same distance l_k from H_p = 0 line from both sides. Distance l_k can in its turn be divided into several sections D l_k, and a modular value |D H_p| is assigned for each of them. The level of concentration of residual stresses is characterized by the ratio:

K = |D Hp| /D lk.

(1)

Proceeding from the results of determination of K values for different sections with an abrupt change of H_p value, its maximal values and, hence, the maximal concentration of residual stresses, are determined.
In a number of cases, when the residual intensity of the structure magnetic field is distorted (the structure was exposed to external magnetic fields), a magnetic parameter, namely coercive force H_c which is related to the structure stressed-strained state [5] can be used. Fig. 3 presents characteristic dependencies of coercive force H_c at low-cycle fatigue (LCF) of sheets of Vst3sp5 steel, depending on the selected loading modes. At loads below the yield point, damage accumulation actively started after 5x10⁴ cycles and was accompanied by the growth of the coercive force from H_c = 2.3 A/cm up to H_c^crit = 5.6 ± 0.2 A/cm corresponding to the physical fatigue limit of 1x10⁷ cycles at s _a = 20 kg/mm² and 3x10⁵ at s _a = 25 kg/mm². With the increase of the loading amplitude above the yield point s _0.2 = 27 kg/mm² accumulation of damage and strain started with the first loading cycles and reached saturation corresponding to H_c^crit = 5.8 ± 0.2 A/cm after 5x10⁴ cycles, this being a very stringent loading mode by the Rules of Gosgortechnadzor organization, Russia. Thus, having plotted similar dependencies for other steel grades used to make load-carrying metal structures of hoisting mechanisms (HM), it is possible to forecast their residual life directly by the results of evaluation of the value of coercive force H_C.

Fig 3: Magnetic control of fracture resistance at low-cycle fatigue of Vst3sp5 steel sheets.

Detection of defects in the structure.
After performance of express-diagnostics for detection of the most critical sections of the structures, i.e. zones of stress concentration which are marked by chalk or paint, a more through flaw detection is conducted using specific NDT methods to detect structural defects.
As regards revealing the material discontinuities, i.e. defects which can be external, internal or through-thickness, application of non-destructive testing methods allows determination of both the defect presence and its size. The latter enables the defects to be divided into admissible and inadmissible.
One of the effective methods of defect detection in the structure is the application of an ultrasonic computerized P-scan system. This system allows visualization of internal defects in the structure walls, whereas its use in the same controlled sections after certain time intervals also allows tracing the dynamics of defects change. Fig. 4 shows [6] the results of ultrasonic testing of the structure sections, made in 1985 (a) and in 1987 (b). Fig. 2 shows that the number of lamination cracking zones has significantly increased during 2 years, which was also confirmed by the results of metallographic investigations. Determination of the changes of the structure geometrical dimensions caused by mechanical wear, corrosion, etc. is conducted with laser optical instruments allowing sagging and deformations in the structure to be revealed with its rather large dimensions. Measurement of the thickness of the structure walls in the case of access from one side, can be conveniently performed with ultrasonic thickness measurement units, providing the measurement accuracy not lower than ± 0.1 mm. Measurement of the wall thickness is performed in the zones of stress concentration, in the locations where corrosion traces are visible, near crossing points of the welds and the anticipated defects.
Corrosion damage of metal structures is differentiated by the kind of corrosion, namely total, spot corrosion, pitting, intercrystalline corrosion, corrosion fatigue, lamination corrosion, etc. Total corrosion and change of the geometrical parameters of the structures are easy to measure. It is performed similar to the case of mechanical wear. In a number of cases corrosion control of damage can be conducted with magnetic instruments which allow it to be performed under a protective cover [7].
The main mechanical properties of metallic materials are characterized by ultimate tensile strength s _T of the material, yield point s _0.2, impact toughness a_c, et al. These mechanical properties can be approximately determined also by non-destructive testing methods by hardness measurement, as well as by evaluation of the magnetic properties.
The ultimate strength and yield point of steel are approximately determined based on non-destructive testing by hardness measurement, for instance, in keeping with GOST 22761-77 and GOST 18661-73.
The rate of forecasting the drop of ultimate strength can be found from the following equation:

(2)

where s _TO is the initial ultimate strength;
s _Ti is the ultimate strength at the time of measurement t_i;
t is the time interval between the measurements.
During determination of the dynamics of the geometrical dimensions change, let us calculate the rate of change V_s of the structure wall thickness as follows:

(3)

where S₀ is the initial wall thickness;
S_i is the wall thickness at the time of measurement t_i;
t is the time interval between the measurements.
Defects of the type of cracks of metallic surfaces are determined using such non-destructive testing as ultrasonic, magnetic, eddy current and electrical potential methods. In addition to crack detection, its dimensions, including the depth of location, can be determined.
In a number of cases in mechanical testing it is beneficial to apply the acoustic emission method, which permits determination of fatigue fractures and stress corrosion cracking. In bimetal items under loads not exceeding 30% of the breaking load, it is possible to identify poor joints by the emission caused by the start of breaking up of the bond between the layers.
Let us consider forecasting the residual life of the structure for the case of the service life of a gas pipeline after its prolonged service. In keeping with the accepted codes the residual life of a structure can be calculated as follows:

(4)

where V_s is the rate of metal degradation (2);
s₀ is the value of ultimate strength of the pipe metal in keeping with the certificate;
K₁ is the material-dependent coefficient of reliability;
K_H is the coefficient of reliability of the pipeline determined by its purpose;
m is the coefficient of pipeline operational conditions;
P is the working pressure of gas in the pipeline, MPa;
D₃ is the outer diameter of the pipe, mm;
n is the coefficient of reliability determined by internal working pressure;
S_lim is the limit admissible thickness of the pipeline wall specified in the codes;
t ₀ is the length of pipeline service.
The derived value of pipeline residual life is verified proceeding from the limit admissible value S_l, i.e.

(5)


where V_s is given by (3).
The magnetic memory method was applied in practice for diagnostics of DKVP 2.5-12 boilers of the Carpathean thermal power plant. IKN-1M indicator of stress concentration was used to perform the work. Thirty water-wall tubes (51 mm dia x 2.5 mm) of the left and right side water walls were inspected. The welds of the upper and lower drums were controlled.
As a result of control pipe #11 of the right lateral water wall was found to have zones of stress concentration (SC) in which, according to the control procedure, the corrosion and fatigue processes are developed to the utmost degree. In Fig. 5 the epures of distribution of the scatter field along the flame generatrix of pipe #11 of the right lateral water wall are shown and the SC zones are marked. Further inspection by the traditional methods (eddy current and ultrasonic) with application of UT-93P, Bulat-1M instruments revealed a critical thinning of the pipe wall of more than 20% (h = 1.4 mm with the rated thickness of 2.5 mm). The pipe was recommended to be replaced. From the results of cutting-out, a pit of 3 mm diameter 1 mm deep was found in stress concentration zone #1 and a pit of 6 mm diameter more than 1 mm deep was found in stress concentration zone #2.
Based on the results of control of the upper drum welds, zones of stress concentration were found in which ultrasonic diagnostics was performed with UD2-12 instrument. Admissible defects were found here which were classified as grades 2 to 3.
It should be noted that application of the method of metal magnetic memory in examination of a boiler of DKV-2.5-13 type revealed developing defects of the type of pit corrosion in the water wall tube, whose detection by the traditional methods is rather labour-consuming.

References

1. Lobanov L.M., Makhnenko V.I., Trufiakov V.I. Development of the theoretical and technological methods of improvement of the strength, fatigue life and precision of fabrication of welded structures. In.: Coll. of Trans. Of Intern. Conf. "Welding and Allied Technologies in the XXIst century" (November 1998, Kiev), p.137-151.
2. Auth. Cert. 2029263. Patent of Russia and CIS countries. A method of determination of residual stresses in items of ferromagnetic materials / Dubov A.A. // B.I., 1995, #5.
3. Dubov A.A., Chromchenko F.A. Guidelines / Non-destructive magnetic method of diagnostics of welded joints of boiler tubing and power plant piping. RD. 34. 17.437-95. M., SPO "OGRRES", 1995.
4. Dubov A.A. Express-method of welded joint control using magnetic memory of the alloy. Control'. Diagnostika Journal, #2, 1998, p.44-48.
5. Muzhitskii V.F., B.E.Popov, Bezlud'ko G.Ya., et al. Magnetic control of the stressed-strained state and residual life of steel metal structures of cargo cranes. - Defectoscopia, 1996, #2, p.12-19.
6. Girenko V.S., Rabkina M.D., Dyadin V.P., Bernadskii A.V., Davydov E.A., Kuz'min V.V. Some results of engineering diagnostics of vessels and pipelines in petroleum processing industry. - Tekhnicheskaya Diagnostika i Nerazrushaushshij Kontrol' Journal, 1998, #3, p.17-24.
7. Bakunov A.S., et al. Non-destructive testing for corrosion damage of the line gas and oil pipelines under protective coating and measurement of this coating thickness // Defectoscopy. - 1996, #2, p.9-11.

© AIPnD , created by NDT.net

|Home|    |Top|