300 150 75 38 19 9 A 47 2 A 212 106 53 27 13 66

SIZE RANGE, p

NORMALIZED 50 ■ POPULATION AQ .

SIZE RANGE, p

NORMALIZED 50 ■ POPULATION AQ .

300 150 75 38 19 9 A A. 7 2 A 212 106 53 ?7 13 6.6 3.3

SIZE RANGE, p

100'

NORMALIZED 90 POPULATION 00 70 •

300 150 75 38 19 9 A A. 7 2 A 212 106 53 ?7 13 6.6 3.3

SIZE RANGE, p

SIZE RANGE, p

100'

NORMALIZED 90 POPULATION 00 70 •

300 150,„,75 38 ,_19 9.4, ,4.7, ,2.4 212 106 53 27 13 6.6 3.3

SIZE RANGE, p

300 150,„,75 38 ,_19 9.4, ,4.7, ,2.4 212 106 53 27 13 6.6 3.3

SIZE RANGE, p

FIGURE 11: MEASURED DISTRIBUTIONS OF SPRAYED POWDERS COMPARED TO AS-RECEIVED POWDER DISTRIBUTIONS

Coating Characteristics:

The characteristics of the as-sprayed TBCs provide a definite reflection of the raw materials employed. It appears that the major powder factors affecting TBC performance are the powder size distribution and the powder's chemical homogeneity and phase structure.

The size distribution of the thermal spray powder is critical because it largely controls the density of the as-sprayed coating. Both the mean particle size and standard deviation of the distribution were found to be significant and of the same order of magnitude in affecting the coating. Figure 9 illustrates microstructures of coatings produces by the three basic powder types. The change in coating porosity with mean powder coating ha3 a distinct effect on TBC thermal cycle life and erosion resistance. These data confirmed that the most dense coatings (made from W's spray dried powder and all those made from fused materials) had the highest erosion resistance and the lowest thermal cycle lives (see Table 4), while the most porous coatings (vendors A, B, and C's powders) had lower erosion resistance and good thermal cycle lives.

To test this hypothesis, the powders were reformulated into distributions which more closely emulated the vendor B powder distribution. The changes in typical powders are shown in Figure 10. The powders were then resprayed and thermal cycle tested. To test the possibility that the spray dried powders tended to decompose explosively in the plasma plume, leaving small, discrete particles to be applied to the coating, the powders were sprayed into a large container, collected and the size distribution measured. For all three spray dried powders, only a slight shift to a smaller distribution possibly attributable to the burn off of the binder and consolidation of the powder was observed. Collected sprayed powder from 'B' sintered powders showed no change in distribution (Figure 11) .

The starting powder's chemical homogene-

produce the tetragonal phase. Retention of the monoclinic or cubic phase may be explained by incomplete melting of the powders during the plasma during the plasma spray process. While the amount of phase transformation upon spraying varied from manufacturer to manufacturer, it may not be possible to produce superior quality TBCs from certain types of powders, regardless of the spray parameters employed.

0 0

Post a comment