CHAPTER 5 AERODYNAMIC DATA AT ONE CLOCKING POSITION
5.1 DATA SET DEVELOPMENT AND METHODOLOGY
The original entry 1 data set was conceived to provide two major pieces of information: the nominal surface-pressure and heat-flux distributions for the airfoils and flow path of the uncooled turbine stage. Typical design codes use the surface-pressure distribution as input for prediction of heat-flux. This experiment would obtain data that could be used for both design codes and advanced codes under development.
However, an opportunity arose to reconfigure the experiments into two different entries. The first one dealt specifically with aerodynamic data and was designed to look at several different scientific issues; one of which is the clocking data discussed in the next chapter. The second entry that occurred about a year later, dealt specifically with heat-flux data. The reason for this was a very practical one. In addition to getting the first experiments performed earlier, and protecting the relatively fragile heat-flux
instrumentation from over exposure, it became clear that to run a full 3-D data set over a stage and 1/2 machine that each experiment would require data acquisition for many more instruments than even the GTL’s large data system could handle. Extrapolation to the cooling experiments which by nature would require double the heat-flux sensors (two-sided as opposed to single-sided), not to mention the other instrumentation required for cooling diagnostics, would yield experimental conditions that could not be completed in one run. For the upcoming cooling experiments, no alternative was seen but to
56
separate a given experimental setting into at least two different run conditions. In order to determine the best way to do this, this entry was performed in a similar manner.
Traditionally, all the measurements for a particular condition would occur in one experimental run. Both pressure and heat flux measurements were taken together. Run- to-run variation was minimized as best as possible by normalizing the data, but when taking both types of data together at lest one had the measured pressure to go along with the measured heat-flux. Previous investigations at Calspan and at the OSU GTL have shown that pressure measurements with lower noise content could be obtained when the facility was run in a Blowdown mode (Chapter 2). However, the heat-flux measurements should be run in Shock-tube mode to create a suitable temperature differential between the metal and gas, insuring a high quality heat-flux measurement. The engineering question that had to be answered was how well these different modes of operation could be combined to produce a coherent data set. This is the main question that is addressed in appendix. In the following sections, the pressure data will be examined in more detail, and references will only be made to the heat-flux data (see appendix C) when necessary to show the repeatability of the results.
Listed in Table 5.1 are the experimental runs that will be used in this part of the analysis. The run numbers are listed on the left, along with the entry (1 or 2) that it corresponds to. The date of the run and the run type are listed. The approximate inlet pressure (which varies based on when during a run one valuates the conditions) is also listed. For all of these conditions the runs are at 100% corrected speed and Clock position D. The last two columns may require a little more explanation (more detailed discussion is given in appendix C).
First is the patch column. This describes the different physical data acquisitions made in order to acquire more channels of data than the A/D system would allow (about 300 channels). Nominally, if the facility was completely reproducible, there would be no difference from one run to the next, and so if one run had half of the pressure data and the other run the other half, one could easily combine the data. Since we are looking for very small variations due to clocking (one the order of 1 to 2%), it was deemed necessary to keep track of these different “patches”.
57
The last column is the experimental group, which combine the different patches together. For entry 1 data runs 25 and 26 were supposed to be at the same condition as runs 31 and 32, but the change in the position of the exit choke modified these runs a little (see Appendix D) thus they are listed as condition 1 and 1a. Experimental group 2 and 4 were derived from setting the main conditions for experimental groups 3 and 4, and were a little lower than anticipated. They provide data at slightly different Reynolds numbers, but not all the heat-flux and tip pressure data was acquired for these runs since patch 2 was not done for these conditions.
Main
Runs Entry Date Type
Approx.
Pressure Patch
Exp Group
25 1 April 3, 2001 Blowdown 45 NA 1a
26 1 April 4, 2001 Blowdown 45 NA 1a
31 1 April 10, 2001 Blowdown 60 NA 1
32 1 April 10, 2001 Blowdown 60 NA 1
1 2 Feb. 25, 2002 Shock 46-48 1 2
2 2 Feb. 26, 2002 Shock 46-48 1 2
3 2 Feb. 27, 2002 Shock 46-48 1 2
4 2 Feb. 28, 2002 Shock 46-48 1 2
5 2 Feb. 28, 2002 Shock 58-60 1 3
6 2 March 1, 2002 Shock 58-60 1 3
7 2 March 1, 2002 Shock 58-60 1 3
8 2 March 5, 2002 Shock 58-60 2 3
9 2 March 5, 2002 Shock 58-60 2 3
10 2 March 5, 2002 Shock 58-60 2 3
11 2 March 5, 2002 Shock 58-60 2 3
12 2 March 12, 2002 Shock 78-80 2 4
13 2 March 12, 2002 Shock 78-80 2 4
14 2 March 13, 2002 Shock 78-80 2 4
15 2 March 13, 2002 Shock 85-90 2 5
16 2 March 18, 2002 Shock 85-90 2 5
17 2 March 18, 2002 Shock 85-90 2 5
18 2 March 19, 2002 Shock 85-90 1 5
19 2 March 19, 2002 Shock 85-90 1 5
20 2 March 19, 2002 Shock 85-90 1 5
25 2 April 22, 2002 Blowdown 60 All Press. 6
26 2 April 22, 2002 Blowdown 60 " 6
27 2 April 22, 2002 Blowdown 60 " 6
Table 5.1 Experimental Conditions Used for Heat-Flux and Pressure Data
58
The colors in this table also help to compare the different groups. Each color represents a different Reynolds number condition (even though they may occur via different experimental conditions). Note that the red condition has both shock tube and blowdown conditions and occurred at two different entries. The data in this table will be the basis for the analysis in this chapter, with a concentration of the red and light blue Reynolds number cases. The purple and yellow cases will be shown for comparison.
The derivations of the main experimental set points and the variation within each group, as well as the variation between the groups is shown in Appendix C. The overall variation for all experimental groups is on the order of ±1.7% (this is a range, not a standard deviation) over all runs. The Reynolds number range for the experimental points is shown in Figure 5.1. One can see that the Reynolds number comes close to doubling, while the pressure ratio variation between the experimental points is well within the variation at each experimental point.
2.5 106 3 106 3.5 106 4 106 4.5 106 5 106 5.5 106 6 106 6.5 106
-1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
0 1 2 3 4 5 6 7
Experimental Points
(Bars Represent Range of data at point)
Reynolds number
% Variation in Pressure Ratio
Reynolds Number at Vane Inlet % Variation from Average Pressure Ratio
Experimental Group
Figure 5.1 Reynolds Number and Pressure Ratio Variation for Groups The slight downward trend in the pressure ratio in Figure 5.1 is examined further in Appendix C by comparing the pressure ratio with the corrected speed. As shown in Appendix C, the connection between the pressure ratio and the corrected speed is not surprising. The key point is that as the corrected speed changes, the incident angle on the
59
rotor is changing. This has the effect of changing the blockage area for the flow, which changes the pressure ratio. These effects are small (as seen above) since turbines are more forgiving than compressors, but are noticeable. An overall comparison for all the runs listed in Table 5.1 (this includes both shock runs and blowdown runs and runs from two different entries a year apart) for the properties that should be constant is shown in Table 5.2. For all runs, the range in corrected speed is ±1.3% and the range in pressure ratio is ±1.7, which is comparable to the variation seen in the conditions for just the heat- flux experimental conditions (see Appendix C). This is another example that there is no different between the shock and blowdown data sets.
Corrected Rotor speed
Total Pressure
Ratio
Total Inlet pressure/Exit Static Pressure Overall Variation
For values that should be constant for all Experimental groups RPM/K^.5 % of Avg
NCR PTR PUPS
Avg 359.676 100% 5.658
STD 2.550 NA 0.073
% Range/Avg 1.286 1.683 2.202
% STD/Avg 0.709 1.073 1.289