JPS60249035A - Test piece for fluidized bed thermal fatigue test - Google Patents

Abstract

PURPOSE:To evaluate exactly a resistance to thermal fatigue of a material by generating a thermal distortion which is equal to an actual machine, by forming a surface extending from a thick-walled part of the center to a thin part of the outside periphery of an abacus counter-shaped test piece for a fluidized bed thermal fatigue test, to curved surface of a recessed type. CONSTITUTION:A part extending from a center part 10 of a test piece to a tip part 11 is varied like a curve, and a cooling speed of only the tip part 11 has been increased without changing a cooling speed of the thick-walled part 10. Therefore, in the test piece, its isothermal diagram becomes denser, the temperature gradient of the center part and the tip part becomes large, and a thermal distortion which is equal to an actual machine is generated.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a method for evaluating thermal fatigue properties of heat-resistant superalloys for gas turbines, and in particular, a method for evaluating thermal fatigue properties of gas turbine heat-resistant superalloys, which can closely simulate the operating conditions of an actual gas turbine, and can be performed easily and accurately. This invention relates to a test piece that can evaluate the thermal fatigue resistance of heat-resistant superalloys for gas turbines.

[Background of the invention]

Thermal fatigue resistance of heat-resistant superalloys for gas turbines is generally evaluated by fluidized bed thermal fatigue testing. A schematic diagram of the fluidized bed test machine is shown in Fig. 1. The test machine consists of a high temperature furnace 1 and a low temperature furnace 2, and the low temperature furnace is equipped with an electric heater 5. The high temperature furnace is maintained at a predetermined temperature by combustion of city gas 7. The test is generally carried out under temperature conditions of 150 DEG C. to 300 DEG C. in a low temperature furnace and 850 DEG C. to 1100 DEG C. in a high temperature furnace. For this reason, the test piece undergoes rapid heating and cooling, and cracks occur in the test piece due to repeated thermal cycles. The thermal fatigue resistance of a material is evaluated based on the number of repetitions of thermal cycles at which cracks occur.

The test pieces conventionally used in fluidized bed tests are shown in Figure 2 (a
) was of the knife type, and (b) was of the disk type. These are all from the center thick part 8°10 to the thin part 9
, 11 change linearly, and the difference in temperature change rate between the center portion 10 and the tip portion 11 creates a temperature gradient in the test piece, generating thermal stress and thermal strain. These test pieces have a linear change from the center 10 to the tip 11, so there is little temperature gradient in the test pieces, and the temperature is 850°C, which simulates the operating conditions of an actual gas turbine nozzle. Under the temperature condition of 300°C, the thermal strain generated is smaller than in the actual machine.

For this reason, the thermal fatigue resistance of heat-resistant superalloys has been overestimated, making it extremely difficult to properly judge whether they can withstand actual use in actual machines.

[Purpose of the invention]

The purpose of the present invention is to generate thermal distortion equivalent to that of an actual machine even under relatively low temperature conditions such as those assumed for gas turbine nozzles.
The object of the present invention is to provide a test piece that allows the thermal fatigue resistance of a material to be fairly evaluated.

[Summary of the invention]

Figure 3 shows a conventional disk-shaped test piece with a temperature of 850°C to 300°C.
A temperature distribution diagram is shown at the time when the temperature difference between the test center part 10 and the tip part 11 becomes maximum during the cooling process. This distribution map was obtained as a result of temperature distribution analysis using the finite element method, and the isothermal lines are arranged in &9 lines from the cross-sectional center 10 toward the tip 11 at 10°C intervals, and the tip 11 It was found that because it was thin, it cooled quickly and had a low temperature. In order to increase the temperature gradient between the center part 10 and the tip part 11, it is necessary to make a large change in shape from the center part 10 to the tip part 11, and to heat and cool the tip part 11 as quickly as possible than the center part 10. It is. The cross-sectional shape of the test piece determines the temperature distribution generated in the test piece. The cross-sectional shape is determined by appropriately selecting a, b, C, and the radius R of the tip, as shown in FIG. In order to increase the shape change, for example, it is conceivable to increase the wall thickness, increase c/a, and decrease R. However, with the above-mentioned fist-like device, the thick wall portion 10 becomes large, so the heating and cooling rate of the entire test piece becomes slow, and the heating and cooling rate of the tip portion 11 is rather greatly influenced by the center thick portion 10, resulting in a large thickness at the tip. The heating and cooling rates of the parts become slower and the temperature gradient that occurs becomes smaller.

In order to increase the temperature gradient generated in the test piece, c/
It is better to make the a ratio small, but if C is made small, the heating and cooling rate of the center part 10 becomes faster, making it unsuitable for use as a test piece. Since the piece itself becomes large, this poses a problem in manufacturing the test piece.

The optimum size of the test piece is as shown in Figure 2, and there is a limit to changing the c/a ratio. Conventional a,
It is necessary to devise a method to increase the cooling rate of the tip portion than before without changing the values of b, c, and R. FIG. 5 shows a cross-sectional view and a temperature distribution diagram of a test piece according to the present invention. In the present invention, it is possible to increase the cooling rate of only the tip 11 without changing the cooling rate of the thick portion 10 by changing the distance from the center 10 to the tip 11 in a curved line. It can be seen that the test piece according to the present invention has a denser isothermal diagram and a larger temperature gradient between the center and the tip.

FIG. 7 shows a comparison of the maximum strain that occurs at the tip of the test piece of the present invention and the conventional test. Since the test piece according to the present invention has a large temperature gradient, the total amount of strain generated during heating and compression is approximately 0.2% larger, exceeding 1%. In the nozzle of an actual gas turbine, it is estimated that the amount of strain that occurs during the cycle of constant start-up and constant operation-stop is approximately 1.0 to 1.2% per cycle. Enables thermal fatigue testing.

[Embodiments of the invention]

Two types of Go-based superalloys, A and B, used for gas turbine nozzles were melted in an Ar atmosphere.

Precision casting was carried out into the shape shown in FIGS. 3 and 5.

After precision casting, the test piece was machined to the specified dimensions. The chemical components of Co-based superalloys A and H are shown in Table 1. Using these test pieces, a fluidized bed thermal fatigue test was conducted in a heating/cooling cycle of 850° C. and 4,300° C., which simulated the metal temperature of 850° C. and cooling air temperature of 300° C. during steady operation of an actual gas turbine.

Figure 6 shows the number of thermal cycles until crack initiation in both test specimens. In the test piece according to the present invention, cracking occurred after 180 cycles in material A and 450 cycles in material B, whereas cracks did not occur in both materials even after 600 cycles in the conventional test piece. It was not possible to determine the superiority of materials A and B. In actual gas turbine nozzles, periodic inspections are performed once a year, and up to that point the thermal cycles due to start-stop operations are about 300 times.

Material A had many thermal fatigue cracks during periodic inspection, and
It has been confirmed that cracks occur within 30 minutes. It has become clear that the evaluation of thermal fatigue resistance using the test piece according to the present invention can be performed under conditions closer to those of the actual machine, and provides a valid evaluation.

By conducting a fluidized bed thermal fatigue test using the test piece of the present invention, the thermal fatigue resistance of heat-resistant superalloys can be fairly evaluated even under relatively low temperature conditions of 850''C; In addition, due to the large amount of strain generated, cracking of the test piece occurs early, and the test time is shorter than that of a test using a conventional test piece. It can be shortened to 1/2 to 1/3.

[Brief explanation of the drawing]

Figure 1 is a schematic diagram of the fluidized bed tester, Figure 2 (,) is a diagram of the shape of a conventional knife test piece, Figure 2 (b) is a diagram of the shape of a conventional disc test piece, and Figure 3 is a diagram of the disc test. Figure 4 is an isothermal diagram of the specimen; Figure 4 is a diagram showing sections A-A' and BB' in Figure 2; FIG. 6 is a diagram showing the number of thermal cycles until cracking occurs in specimens A and B, and FIG. 7 is a diagram showing a comparison of the maximum amount of strain occurring at the tip. 1...High temperature furnace, 2...Low temperature furnace, 3...Movement device,
4... Test piece, 5... Heating heater, 6... Air, 7... City gas, 8... Center thick part, 9... Tip thin wall part, 10... Middle part 2m ((1) (b)

Claims (1)

[Claims] 1. A fluidized bed characterized in that the surface from the center thick part to the outer thin wall part is a concave curved surface in an abacus bead-shaped fluid thermal fatigue test specimen consisting of a central thick wall part and a peripheral thin wall part. Test piece for thermal fatigue testing.