·Home ·Table of Contents ·Methods and Instrumentation

Estimation of Stresses in Pipelines by Magnetic Noise Kuznetsov N., Lyachenkov I., Kuznetsov A., Shaternikov V. Moscow State Academy of Instruments Engineering and Computer Science (Moscow). Contact

The attempts to carry out non-destructive testing of various objects to determine mechanical stresses analyzing Barkhausen magnetic noise have already been made for quite a long time.

The background for such a testing is experimentally established proportional relationship between energetic parameters of magnetic noise (amplitude, electromotive force) and actual mechanical stresses.

The most complex problem during such a testing is determination of quantitative relationship between parameters of magnetic noise and the magnitude of actual stresses. The unsolved state of this problem is caused by the fact that parameters of magnetic noise are influenced not only by stress, but also - and significantly - by physical and mechanical properties of metal of which the object is made, condition of its surface and a number of other factors. It is not always possible to take into account the influence of all these factors. And not taking them into account makes the error of measuring the stress comparable to measured value.

In the present article we submit the materials on determination of absolute stresses in steel pipelines by registrating magnetic noise. The authors have managed to account for all the influencing factors, described above. We have used the results of research which show, that the more is the size of the domain in steel [1], the higher are the energetic parameters of magnetic noise. In its turn, the more is the elastic strain and the magnitude of the applied magnetic field [2], the more is the size of the domain in steel.

These data have allowed to develop a technique of determination of elastic strain in steel pipelines on the basis of measurement of electromotive force (emf) E of noise. To realize the technique PION-01 device was developed with transducer, allowing to cycle reverse magnetization of the metal to the direction between the poles of magnetic core of the transducer. Fig.1 shows the block diagram of PION-01 device, and Fig.2 shows the time diagram of determination of the value of E.

Fig 1: Block diagram of PION-01 device: 1 - pipeline; 2 - magnetic core; 3 - field coil; 4 - sawtooth generator; 5 - pick-up coil; 6 - voltage amplifier; 7 - integrator; 8 - indicator

Fig 2: Time diagram of device operation: H - magnetic field intensity; UMN - pick-up coil output voltage; E - emf of magnetic noise


E is measured during half the cycle of oscillation of magnetic field intensity H in field coil 3. E is determined as:


(1)

where U_MN(t) - variation of pick-up coil 5 output voltage; T_0.5 - half the cycle of magnetization.
According to the developed method it is necessary to determine proportionality coefficient K in a certain zone of an operational pipeline before the beginning of the test.
This determination includes:
1. Choosing a reference section of the pipeline. The distribution of elastic strain in this reference section should be known and the state of metal - the same as that in the test zone. The method of choosing the reference section is described in [3].
2. Taking at least two measurements of emf (E₁₀ and E₂₀) in directions, for which the magnitude of strain is known (e₁₀ ¹ e₂₀).
3. Calculation of proportionality coefficient by the following equation:

   K = (e 10 - e 20)/(E10 - E20),

   (2)

   where e₁₀ and e₂₀ - magnitudes of known elastic strains and E₁₀, E₂₀ - emf values, corresponding to these known elastic strains.
4. Calculation of E₀, corresponding to emf in a zone with zero strain. The calculation is carried-out by approximating E₁₀ and E₂₀, measured in zones with corresponding strains.

Values of K and E₀ are constant for the test zone of the pipeline.
Emf E_i is measured in the test zone at the same operating conditions as in the reference section.

Magnitude of elastic strain e _i in the direction of measurement is determined from the following equation:

ei = (Ei - E0)K.

(3)

Calculation of stresses in the pipeline walls is carried-out on the basis of strain measurement data from known relationships. For example, axial s₁ and lateral s ₂ stresses may be determined as:

s 1 =G(e 1 + n e 2)/(1 - n 2),

(4)

s 2 =G(e 2 + n e 1)/(1 - n 2),

(5)


where e ₁ and e₂ - axial and lateral strains respectively (determined from equation (3)); n- Poisson's ratio; G - Young's modulus.

The test method proposed and PION-01 device are used to estimate state of stress of various pipelines.

For example, during test of a section of Ostrogozhsk-St. Petersburg pipeline (station 119) stresses were measured in several zones of an aerial crossing over a ravine. The measurements were taken by the authors of this paper with participation of representatives of DAO «Orgenergogaz» and RAO «Gazprom».

The pipeline has been in operation for more than 20 years. The pressure of gas in the pipeline is 4.01/44.1MPa. The wall thickness in the test zone is 10.5mm. The outer diameter of pipe is 1020mm. The yield strength of the metal of the pipe is 390MPa according to certificate. The aerial crossing length is 45m. The thickness of soil above the pipe on both sides of the aerial crossing is 1.01/41.3m. The ambient temperature is 18^°C.

It was found out that:

1. Lateral stresses s ₂ in the pipe in the zone of the aerial crossing fall in the range between 190 and 220MPa;
2. Axial stresses in the zone of the aerial crossing change from compressive to tensile; maximal axial tensile stress is observed in the middle of the crossing (the lowest point of the pipe) and amounts to 155MPa.

On the basis of collected data safety assurance factor of the metal of the pipeline was determined and the zone, in which it was necessary to conduct metal-science investigation to estimate the remaining service life of the inspected pipeline section, was selected.

Similar work was carried-out on «Central Asia - Center» gas pipeline (Turkmenia). Stresses were measured on a submerged crossing through Kara-Bogaz-Gol strait. Measurements of magnetic noise were taken during submersion to the depth of up to 3m. It was found out that close to the main run passing-out from the soil into the water bending stresses, caused by buoyant force and water flow (flow velocity was approximately 3m/s), were 260MPa (yield strength of metal of the pipeline was 3801/4390MPa).

Bending stresses reduction work schedule was set-up on the basis of collected data.

Check of correctness of stress determination was carried-out by measuring lateral stresses according to proposed method at zero pressure of gas in the pipeline. Positive results were obtained.

Presently the proposed method and device are being transferred to RAO «Gazprom» which operates gas mains.

References.

1. Kuznetsov N., Investigation into the Relationship of Magnetic Noise Parameters with Variations of Ferromagnetic Steels Structure. Technical Diagnostics and Non-Destructive Testing, 1992, #4, p. 121/418.
2. Bozorth R.M. Ferromagnetism. - d.Van Nostrand Company, Inc. Toronto-New-York-London. - 1951.
3. Patent 2116635 for invention (Russia).

© AIPnD , created by NDT.net

|Home|    |Top|