CHAPTER 4 DATA DESCRIPTION AND CLOCKING ANALYSIS TECHNIQUES 32
4.3 DATA ANALYSIS/ FACILITY MODELS
4.3.3 Mass-Flow and Temperature
Probably the largest and most important modeling of the data that was done in this dissertation has to do with the mass-flow through the rig and the inlet total temperature.
It has been known that the inlet total temperature rakes do not respond as well as the exit total temperature rakes due to the extremely low Mach numbers at the inlet. The
differences between the rakes are relatively small, but as the experimental accuracy has improved, these differences have become more significant. The rake design used for the measurements reported in this dissertation dates back to the mid 1990’s when the original work was done on this thermocouple configuration, and a decision was made to keep all the sensors the same (forward and aft), even at the expense of some operations at the extreme conditions. Since then, similar rakes have been tested and the recovery factors observed as a function of Mach number have replicated the original results. We know that the recovery factors are less for the lower Mach number cases.
This poses a problem, since temperature is one of the main design variables for setting the corrected speed. Even a 5% change in temperature will change the corrected speed by about 2.25%, which may lead to incorrect matching conditions which become more prevalent as one tries to match blowdown and shock-tunnel runs. The solution to this problem is a three-step procedure. First, the mass flow at the exit of the machine must be calculated. This is based on the exit total pressure and temperature, and the exit choke area. A global discharge coefficient was calculated which accounted for the experimental discharge coefficient of the choke and the recovery factors of the pressure and temperature sensors, by equating the inlet and exit mass flow (since the choke area of
50
the vane is well known from the airfoil design) during a blowdown run when the mass- flows are stable for a lengthy period of time and the flow conditions are benign. This uses the idea that the mass flow in has to be the same as the mass flow out, if there is no leakage into internal passages (that would result in a change in static pressure) during this time. This technique gave remarkably consistent discharge coefficients. Then using the time delays calculated beforehand, the exit mass flow was delayed so that it corresponded to what the mass flow should have been at the inlet. Using the measured inlet total pressure, one can back out the inlet total temperature. Care has to be taken to account for gamma variations. This also has the benefit of creating an unnaturally high frequency response on the implied temperature since the pressure signal has a great deal of fidelity.
The final implication is that the variation that would seem to occur in the calculated inlet mass flow goes away because the measured inlet temperature does not have the fidelity to match the pressure variations, thus creating an implied “noise” in the mass flow.
Reduced noise in the mass flow translates into reduced noise in the Stanton number calculation. The end result is an upstream temperature calculated based on the following parameters.
Tup t fi F mexit t ,GD , PT,U p(t), g Tup t , Pup t ,Vane Choke Area
There are several ways to check the validity of this model. The first is that the global discharge coefficients should be constant. The second is that for blowdown runs and later times, the variation between the measured temperatures and the calculated
temperatures should be less. Finally, during a shock run, the ratio for the measured inlet temperature to the calculated inlet temperature should be equal to the measured recovery factor of the total temperature probes at this Mach number. All of these hypotheses have been verified (see Appendix D and accompanying discussion). The impact of this
“adjustment” to the measured temperature is shown below.
51
Figure 4.7 Examples of Inlet Mass Flow and Temperature Calculations The blue lines in this plot are the calculated values based on these techniques, the black line for the left plot is the measured total temperature, whereas on the right plot it is the calculated mass flow based on the measured inlet total pressure and temperature. Note that the observed noise on the inlet mass flow decreases when the exit mass flow is used, and the calculated inlet temperature has much more frequency content (required to balance the pressure data). As a check, the ratio of the measured to calculated total temperatures are consistently about 0.95 that has been the traditional measured recovery factor for these sensors. As a final check, one can look at the temperatures using the two methods for the entry 1 data (blowdown runs).
350 360 370 380 390 400 410
5 10 15 20 25 30 35 40
Inlet Total Temperature (Entry 1, Blowdown Runs) Time Window "M"
Measured Calculated
Inlet Total Tempersture (K)
Run Number
Figure 4.8 Measured and Calculated Temperatures
Here one can see that the variation between the two methods is relatively small (usually less than 10 deg C) and that sometimes the calculated is higher, other times the measured
52
is higher, illustrating that at least at this time window there is no bias in the answers, which is predicted by the model. For the design point conditions used in this dissertation, the mass-flow and the calculated inlet total temperature will be used for the normalizing properties.
The effect of this increase in the measured temperature is to move the time- window where one would perform the actual measurements in a run, to a point where the corrected conditions are closer to the desired value. A plot of the variation of several design properties from the design condition is shown below.
Time ms
NCR_TUP % Change PTR_DELAY % Change PTUA % Change TUP_COR_NEW_RA % Change MDOT_A_NEW % Change
Push when done viewing 12.00E+0
-28.00E+0 -26.00E+0 -24.00E+0 -22.00E+0 -20.00E+0 -18.00E+0 -16.00E+0 -14.00E+0 -12.00E+0 -10.00E+0 -8.00E+0 -6.00E+0 -4.00E+0 -2.00E+0 0.00E+0 2.00E+0 4.00E+0 6.00E+0 8.00E+0 10.00E+0
105.000
60.000 65.000 70.000 75.000 80.000 85.000 90.000 95.000 100.000
0.
1.
2.
3.
4.
5.
6.
7.
8.
Plot 9
OFF Print at end (T) Data Window used
dDue to Increased Temp
% Change
Target Data Window
Figure 4.9 Effect of Temperature Change on Time-Window
These lines are all design property variations from the design point (in percentage). The main ones are the red line (pressure ratio) and the blue line (corrected speed). Also shown are the total inlet pressure (green), mass flow (black), and total inlet temperature (gray) for a shock run. The window on the left was the area where the original time window was expected to be. Because of the increase in temperature, the time windows were shifted later in the experiment (to the right in Figure 4.9) to get the proper values.