RADIATION INSPECTION OF HIGH-PRESSURE VESSELS USING A COMPACT BETATRON

V. A. Berdonosov

(Presented by the scientific seminar of the Research Institute of Nuclear Physics, Electronics, and Automation)

In connection with the development of modern technology and the steady increase in the requirements imposed on the quality and reliability of manufactured products, the role of flaw detection in industry is growing.

Methods for quality control of materials and products without destroying them have become the most widespread. Such methods provide complete inspection of materials and products, in many cases lend themselves to automation, and make it possible to substantially improve the quality of manufactured products.

Among other methods of flaw detection, inspection by means of radiography is of the greatest interest. Researchers are attracted by the clarity and simplicity of this method, and by the possibility of obtaining high detectability of parts and high productivity. It is sufficient to say that quality control of materials and products by radiography using radiation from X-ray apparatuses, isotopes, and accelerators accounts for approximately 70–80% of the total volume of all applications of nondestructive testing [1, 4, 5].

The need to inspect, more rapidly and reliably, products of great thickness and materials with a large attenuation coefficient, while maintaining acceptable detectability of defects and inspection productivity, has led to the introduction into radiation flaw detection of radiation sources with energies of several mega-electron-volts.

The advantages of accelerators, in particular betatrons, as compared with X-ray apparatuses and radioactive isotopes include a high specific activity favorable for radiographing materials, a radiation-energy range and the possibility of varying it within broad limits in order to establish the optimum value.

In addition, the small dimensions of the focal spot of betatrons, amounting to fractions of a square millimeter, make it possible to obtain radiographs with good resolving power [2, 3].

With the aid of the hard radiation of betatrons, it is possible, by a contactless method, to determine the presence of inclusions in materials of great thickness when these inclusions have a different density or a different effective atomic number. The sources of information in this case may be the changes, arising from the presence of defects, in the spectral composition and in the magnitude of the flux of direct or reflected radiation. In the shadow method of flaw detection, the dimensions of defects extended along the path of the ray are determined from the change in the magnitude of the radiation flux caused by the difference in the radiation attenuation coefficients of the mate-

alongside the defect and the base material. The transverse dimensions are determined from the dimensions of its shadow on the screen, constructed according to the laws of geometrical optics.

Below are the results of experimental studies on industrial testing of a procedure for radiographic inspection of rolled high-pressure vessels with wall thicknesses up to 250–280 mm. Radiography was carried out using a compact PMB-6 betatron with a maximum radiation energy of 6 MeV and an exposure-dose rate at a distance of 1 m from the betatron target of 0.3–0.4 roentgen/min under the conditions of the Uralkhimmash plant [2]. The radiographic procedure was developed at the Research Institute of Electron Introscopy.

The objects of inspection were rolled high-pressure vessels. Three types of vessel specimens with a thickness of 200 mm and an X-shaped weld-profile preparation were presented:

a) filling of the weld profile with deposited metal to 25% of the weld-profile height;

b) filling of the weld profile with deposited metal to 50% of the weld-profile height;

c) filling of the weld profile with deposited metal to 100% of the weld-profile height.

The central tube was made of steel grade Kh18N10T; steel 10G2S served as the material for the roll. Thus, both fully and partially welded seams with thicknesses of 50, 100, and 200 mm were inspected. In addition, at the Uralkhimmash plant radiography was carried out on specimens with weld thicknesses of 80, 150, and 280 mm.

In the course of the experiments, the material and type of intensifying screens, films, and focal distance were selected, and the main parameters of radiation inspection—sensitivity and productivity—were also investigated.

Since all manufactured X-ray films have maximum sensitivity at a radiation quantum energy of 80 keV, it was of interest to determine the principal characteristics of the available X-ray films (sensitivity and contrast of the emulsion layer) for a radiation energy of 6 MeV. As the studies showed, the best of the domestic films in terms of sensitivity and contrast for the indicated betatron radiation energy is type RT-2 film (screen film); close to it in properties is RT-1 (RM-1) film.

The geometry of the experiment was determined by the dimensions of the inspected specimens and the required inspection productivity. We carried out studies to select the optimum focal distance when penetrating various thicknesses of steel absorber. In choosing the focal distance it must be remembered that increasing the focal distance leads to an increase in the area penetrated per exposure and, at the same time, to an increase in exposure time, since the radiation intensity decreases inversely proportional to the square of the focal distance. The focal distance was selected on the basis of obtaining the required inspection productivity, which depends on the total time for penetrating the seam and the length of the section of circumferential seam inspected in one exposure. The results of the studies made it possible to choose a focal distance equal to 0.6 m, which satisfies the above requirements.

For hard X-ray radiation, the image in the film photoemulsion is produced by the action of electrons formed as a result of the interaction of this radiation with the medium surrounding the film. By placing next to the film a substance with a high atomic number, it is possible to significantly increase the number of electrons reaching

onto the film per unit time, i.e., to increase the blackening of the film. The intensifying action of metal foils depends on the foil material, its thickness, the wavelength of the incident radiation, or the radiation energy. In selecting metal intensifying screens, two main tasks were posed:

1) selecting the optimum thickness of the intensifying screens from the standpoint of their maximum intensifying effect;
2) obtaining the best detectability of defects.

In all experiments on selecting the optimum thickness of the intensifying screens, double screens were used: front and back. Lead was chosen as the screen material.

We carried out a series of experiments to determine the most effective thickness of the front intensifying screen. The thickness of the back screen was in all cases taken to be 2.0 mm, since the intensification effect and improvement in image quality for this screen depend little on its thickness.

The influence of the intensifying action of lead screens was studied for a steel absorber with thicknesses of 45, 75, 105, 145, and 200 mm.

Domestic films of the RT-1 and RM-1 types were used. As the measurement results showed, the strongest blackening is produced by the field corresponding to a screen 0.1 mm thick.

However, satisfactory image quality is achieved only at a thickness of 1 mm and greater, which can be explained by filtration by the screen of electrons and γ-quanta scattered in the absorber, which blur the image.

In the course of the experiments, fluorescent intensifying screens were also studied. These screens were used for thicknesses greater than 150 mm, since with such an absorber layer the exposures when using only lead screens become very large, which is unacceptable under production conditions. Crystalline phosphors of the CaWO₄ type were used as the active layer in such screens; under the action of X-rays or γ-rays they give the greatest light output.

Composite screens were also studied (a combination of lead and fluorescent intensifying screens). The use of such screens substantially increases the sharpness of the defect image, although the exposures increase by a factor of 1.5–2 in comparison with fluorescent screens. The influence of composite screens on radiogram quality was investigated at thicknesses of 150, 200, and 280 mm. The exposure time with the screen combination 1, 2FPF2 was respectively 18, 26, and 150 min, which is 3–3.5 times less than when only lead screens are used. The detectability of defects proved to be somewhat worse than when the charging schemes 1, 2P2 were used. The data given above are valid for UFD-type fluorescent screens.

The intensifying action of fine-grained fluorescent screens of the “Perlux” type was also studied. These screens provide high resolving power, but the exposure time in this case is twice as long as with UFD screens.

The main parameters of radiation inspection are the radiographic productivity and the detectability of defects. A study was carried out of these principal factors as functions of the thickness of the inspected steel specimens of coiled vessels.

The exposure time when radiographing a steel absorber depends on the thickness of the inspected object, the exposure dose rate and the energy of the generated radiation, the focal distance, the sensitivity of the X-ray film, and the type of intensifying screens [4, 5]. In addition, the exposure time depends substantially on the following fac-

factors: constancy of development time, which for the indicated film grades was 8 min, constancy of the developer temperature, etc. The films exposed behind different absorber thicknesses were then photometered on an MF-4 microphotometer, and exposure curves were plotted from the photometry results.

Figure 1 gives exposure curves plotted for RT-1 and RM-1 films at absorber thicknesses of 40–300 mm and a focal distance of 0.6 m. At thicknesses of 40–200 mm, the charging circuit \(1,2\text{PP}2,0\) was used; at thicknesses of 150–280 mm, the circuit \(1,2\text{FPF}2,0\) was used. The mean blackening density of the radiographs was 1.6 optical-density units.

To reduce the exposure time when radiographing large steel thicknesses (from 200 mm and above), the method of double films can be recommended (charging circuits \(1,2\text{PP}2,0\) and \(1,2\text{FPF}2,0\)). After exposure and processing, the films are placed one on top of the other according to previously established marks and are viewed. The disadvantage of the method is the double consumption of film, but this is compensated for by the following advantages:

a) the exposure time is reduced by approximately a factor of 2 in comparison with exposure of a single film;

b) an improvement in image contrast is observed, which promotes good detectability of extended defects.

One of the principal parameters determining the field of application of nondestructive testing of materials and products, along with productivity, is defect detectability.

Defect detectability was investigated using reference defectometers with grooves (GOST 7512—55), owing to the absence of defectometers regulated by GOST 7512—69. In the experiments, the defectometer was placed in the most unfavorable position from the standpoint of detectability: on the side of the product facing the radiation source. The exposed films were dried and examined on an MF-4 microphotometer. The blackening densities of the grooves and of the fields (background) between the grooves were measured; for convenience of reporting, the densities of the fields (background) were reduced to a single value. From the measurement results, curves of absolute and relative detectability were plotted (Fig. 2).

On the basis of the investigations carried out, the following conclusions may be drawn:

1. The productivity of inspection of welded seams up to 250–280 mm thick by the radiographic method, using a PMB-6 betatron as the radiation source, is acceptable for production conditions.

2. The minimum exposure time at a thickness of 45 mm with the charging circuit \(1,2\text{P}2,0\) is 3.5 min; the exposure time with the charging circuit \(1,2\text{FPF}2,0\) at a thickness of 280 mm is 150 min; with the charging circuit \(1,2\text{FPPF}2,0\), it is 80 min.

3. The detectability obtained on the radiographs over the entire range of thicknesses studied is no worse than 1.2%.

4. Inspection of seams in the thickness range 80–160 mm is the most favorable, since in this interval the relative detectability proves to be the best.

5. The defect detectability obtained at the indicated thicknesses is significantly higher than when x-ray and isotope radiation sources are used.

Fig. 1. Exposure charts for radiographing steel with bremsstrahlung radiation from the PMB-6 betatron.

Visible labels in the figure:

• Vertical axis: Exposure time, min
• Horizontal axis: Steel thickness, mm
• Inset:
• Radiation source — PMB-6 betatron
• \(W_0 = 6\ \text{MeV}\)
• \(P_0 = 18\ \text{R/hour}\cdot\text{m}\)
• \(F = 50\ \text{cm}\)

Curve labels visible in the graph:

• RM-1, RT-1, KT-2
• RM-1
• 12BP2 (RT-1)
• 12BP2 (KT-2)
• RM-1, 12BP2 (TsFL)
• RM-1, 12BP2 (TsFL)

Fig. 2. Dependence of radiographic sensitivity on the thickness of steel articles

REFERENCES

1. L. M. Anan'ev, A. A. Vorob'ev, V. I. Gorbunov. An induction electron accelerator—the betatron. Moscow, Atomizdat, 1961.

2. L. M. Anan'ev, V. A. Chakhov, M. M. Shtein, Yu. P. Yarushkin. “Small-size betatrons and their application in industry.” Defektoskopiya, 1968, No. 5.

3. Betatron gamma flaw detection of steel. Moscow, Academy of Sciences of the USSR, 1961.

4. A. A. Vorob'ev, V. I. Gorbunov, V. A. Vorob'ev, G. V. Titov. Betatron flaw detection of materials and products. Moscow, Atomizdat, 1965.

5. Industrial Radiography. Edited by U. J. Whitechair. Moscow, Atomizdat, 1960.