The flaws in turbine discs are of great significance; since they represent potential sites for the nucleation of cracks, they threaten the integrity of the component and reduce the fatigue life. For this reason, much emphasis is placed on using NDT (where possible on-site non-destructive-testing) to characterise the dimensions of the flaws, both during the manufacturing of the turbine disc – i.e. before the turbine disc enters service for the first time – to ensure that it is of acceptable quality, and then whenever the disc is removed from the engine, prior to a decision being made to place it back in service. Obviously, the first situation arises regardless of the lifing method employed, the second only when the damage-tolerant method is being used. You can learn more at SalemNDT testing systems.
To place turbine discs back in service needs the issuing of Aircraft Form 1 Release Certificates. The various NDT methods which are suitable include (i) liquid penetrant testing, most usually dye penetrant inspection, which fluoresces under ultra-violet light, thus providing a measure of the length of each crack exposed to the surface; (ii) eddy current inspection, which enables a quantitative estimate of the crack depth, but only for cracks located at or near the surface; (iii) X-ray radiography, which has the power to penetrate relatively thick sections, but which cannot usually resolve the finest flaws that are present; and (iv) ultrasonic testing – the use of high frequency sound waves for the detection of inclusions, coarse grains, shrinkage pipe, tears, seams and laps. In practice, for turbine disc applications, this final method is of the greatest importance. Then turbine discs can be back in service with the issuing of Aircraft Form 1 Release Certificates.
To illustrate the use of ultrasonic inspection for on-site non-destructive-testing of turbine discs in this context, consider its application to a forged and heat-treated disc, which will have been machined to a pre-specified geometry known as the ‘sonic shape’ or ‘condition of supply’. To enable ease of ultrasonic inspection, the features of the disc (for example, the blade root sockets, cover plate and flanges) are not introduced by the machining process as yet, so that, at this stage, it contains only a few, flat surfaces which intersect at 90° angles. To detect the smallest flaws, care is taken also with the surface finish. Production and measurement of the ultrasonic waves is accomplished via the use of piezoelectric transducers and receivers – these rely upon crystals, such as quartz or barium titanate, which display a strong inverse piezoelectric effect. Since ultrasonic waves are attenuated strongly by air, the disc and transducer are either kept immersed in water during inspection, or else a thin layer of coupling fluid is maintained between the two; the waves are attenuated only very weakly by the material itself, so that very thick sections can be analysed.
In practice, the waves are sent through the component in at least two different ways. First, the so-called ‘straightbeam top inspection’ is carried out to detect planar flaws lying parallel to surfaces whose normal lies parallel to the axis of the forging; for this purpose, a longitudinal ultrasonic beam is directed parallel to the normal to these same surfaces, using a single pulse-echo transducer, 13-40 mm in diameter, at a frequency in the range 1-5 MHz. Calibration is carried out using standard 50 mm diameter cylindrical reference blocks of varying heights, into which flat-bottomed holes are machined to simulate artificial flaws of different sizes located at different depths.
Detection sensitivity is normally set to 10% of the size of the flaw machined into the calibration block. A second set of tests are conducted using shear waves, to detect flaws that have an axial radial orientation; for this, an angle-beam transducer is used to introduce a beam at 45° to the surface normal, and inspection is performed by scanning in the circumferential direction, both clockwise and anti-clockwise, around the periphery of the forging. Calibration notches -typically 25 mm long and either V -shaped or rectangular- are cut axially on the inner and outer surfaces of the forging, with a width not exceeding twice the depth. The sensitivity of the detection is established by first adjusting the instrument controls to obtain a minimum 13 mm sweep-to-peak signal from the calibration notch on the outer surface, followed by a measurement of the response from the notch on the inside surface; the peaks corresponding to the two notches are connected thus establishing a reference line.