Paper describes methods suitable for crack detection on low pressure steam turbine blades, not including the blade root. These cracks, which are initiated in the corrosion pitting, can cause serious damage to the steam turbine blade leading to its breakaway. Therefore, the detection of these cracks on the early stages of blade fracture is very important. Several methods for detecting of surface cracks (ultrasonic Rayleigh waves, eddy current with flexible array probe, etc.) has been tested on artificial flaws, which were manufactured into turbine blades. The comparison of all these methods is described as well as the evaluation of their advantages and disadvantages. Simulations of ultrasonic testing is also presented in this paper.

Pitting corrosion is a known process in metals. Pitting, observed in turbine blades, is caused by corrosion mechanisms in operation environment that may contain aggressive chemical compounds. Environment, material composition and operating conditions determine the corrosion process causing a cavity formation and their critical growth. Crack initiation can be a following mechanism. Thus the critical cavity size is the important parameter and its precise estimation can avoid a turbine blade failure. Cavity shape as well as corrosion products inside cavities should be taken into account. 3D scanning technique performed during turbine shutdowns is shown to be a suitable method for measurement of critical cavity size. In this study, the accuracy of 3D technique was confirmed by standard sample preparation methods followed by microscopy focused on corrosion products observation. Cross-sectioning of blades was performed close to the cavities, sections were grinded to show the maximum cavity dimensions and finally polished. Cavity size was measured in few steps by Light Optical Microscopy during grinding to find the maximum dimensions and compare to 3D measurements. Local corrosion in cavities was evaluated by Scanning Electron Microscopy using Backscattered Electrons and Energy Dispersive X-ray Spectroscopy. Non-uniform Fe, Cr, Si rich oxides with dimensions from tens up to hundreds of micrometres were analysed in cavities. No significant effect on oxide removing was observed after using sandblasting technology except possible effect in larger pits. Cracks and pores were detected close to the vicinity of trailing edge of blade.

With high operating parameters of the medium in medium-pressure and high-pressure steam pipelines of fossil power plants, creep damage occurs, especially in welded joints leading to complete rupture of the pipe wall in the last phase. Detection of creep damage at an early stage before major cracks may occur can prevent these accidents. For these purposes, phased array ultrasonic testing was performed using a high frequency probe. This testing was performed on real welded joints cut from the power plant. To verify the possibility of detection, metallographic analysis on the tested locations was performed. When comparing the results, it is possible to refine the interpretation of the ultrasonic data and the metallographic results can also be used in the eventual qualification of non-destructive testing.

No abstract

Tubes of super-heaters and heaters used in conventional power plants are exposed to coolant and high temperatures. The growing oxide layer on the inner surface reacts over time as a heat insulator on the water side and reduces heat transfer through the wall of the tube. A relatively thin oxide layer already contributes to the boiler efficiency and causes a permanent overheating of the tube wall. As a result of overheating at the site, the intercrystalline cracks leading to the bursting of the tube are developing. The secondary problem of the growth of oxide layer thickness is so-called exfoliation. For non-destructive evaluation of the thickness of the oxide layer directly at the power station, ultrasonic method (UT) can be used with a high frequency probe. In order to verify the accuracy of the measurement and the qualification of the ultrasonic testing methodology, light and Scanning Electron Microscopy (SEM) was used on specimens that were removed from the super-heater after the UT measurement. The standard longitudinal cut surface imaging in BackScatter Electrons (BSE) and Energy Dispersive Spectroscopy (EDS) analysis for accurate thickness determination with the chemical composition of the layer confirmed the accuracy of the UT measurement.