T106A Bar Passing Flat Plate

T106A 5-Blade bar passing cascade

The following is edited from Stieger (2001).

Experimental Apparatus The bar passing cascade facility, shown in Figure 1, is a cascade analogue of a multi- stage turbomachine. The facility has previously been described by Banieghbal et al (1995) and Schulte (1995).

The wakes shed from upstream blade rows in a real machine are simulated by an array of cylindrical bars moving across the inlet plane of the cascade. The bars are held between two nylon belts that run on two sets pulleys. The configuration of the bar passing cascade required that the top and bottom walls of the cascade be slotted to permit the passage of wake generator bars (Figure 2). These slots provide two additional passages to the flow. The slot on the top wall does not present a problem, as it requires the flow to change direction, however, the slot on the bottom wall was more closely aligned with the flow. Moreover, the pressure gradient due to the incidence of the cascade forces flow out of the bottom slot. The result is an effective increase of incidence on the cascade and a non- uniform inlet static pressure. In order to minimise this problem and maintain inlet periodicity, an additional dummy passage was created on the suction side of the cascade as shown in Figure 2. This was achieved by adding a straight flap to the bottom wall. The throat of this dummy passage was then adjusted so that the combined throat of the end wall gap and the dummy passage matched that of one of the cascade passages. The flap of the dummy passage was set while monitoring the inlet static pressure and thereby ensuring periodicity. The periodicity of inlet static pressure was maintained within a 2.5% range of inlet dynamic head.

The unsteady wake-passing flow conditions were chosen to match those of a repeating stage of the T106 profile. The Reynolds number based on chord and time mean isentropic 5 exit velocity was Re2C=1.610 and the flow coefficient was =0.83. The bar pitch matched the cascade pitch so that sb/sc=1 (fr=0.68). The bar diameter of 2.05mm was chosen to match the loss of a representative turbine blade. The cascade incidence was set to 1=37.7.

The static pressure distribution measured on the cascade under wake passing conditions is also shown in Figure 3. It is apparent that in the time mean the wake passing suppresses the suction surface separation bubble. It is an unavoidable consequence of the experimental facility that the wake passing conditions introduce a small amount of incidence on the cascade as shown by the small differences between the steady and unsteady pressure distributions over the front of the blade. This small incidence has no significant impact on the blade surface boundary layers downstream of peak suction. However, it may be necessary to alter the inlet flow angle by 1-3 in order to improve the match between CFD and experiments. This is because of the leakage referred to above. Geometry

An IGES file T106A_cascade_v01.iges is available containing the geometrical definition of the experiment. Table 1 contains further details. Figure 4 shows the position of the bars at time t=0.

Datafiles

Further information

The reader is referred to the accompanying publications by Stieger (2001), Stieger et al (2003), Stieger and Hodson (2003), and Stieger and Hodson (2003).

References and Further Reading

Banieghbal, M.R., Curtis, E.M., Denton, J.D., Hodson, H.P., Huntsman, I., Schulte, V., Harvey, N.W. and Steele, A.B., 1995, “Wake passing in LP Turbine Blades”, presented at the AGARD conference, Derby, UK, 8.5-12.5.

George, W.K., 1975, “Limitations to measuring accuracy inherent in the laser-Doppler signal”, Proc. LDA Symp., Copenhagen

Schulte, V., 1995, “Unsteady Separated Boundary Layers in Axial-flow Turbomachinery”, PhD Dissertation, Cambridge University.

Stieger, RD, 2001, “The Effects of Wakes on Separating Boundary Layers in Low Pressure Turbines”, Ph.D. Thesis, Cambridge University Engineering Department

Stieger, R, and Hodson, HP, 2003, “Unsteady Dissipation Measurements on a Flat Plate Subject to Wake Passing”, 5th European Conf on Turbomachinery Fluid Dynamics and Thermodynamics, Prague, Mar.

Stieger, R, and Hodson, HP, 2003, “The Transition Mechanism of Highly-Loaded LP turbine Blades”, ASME Paper No GT-2003-38304, ASME Turbo Expo, Atlanta, Jun.

Stieger, R, Hollis, D, and Hodson, HP, 2003, “Unsteady Surface Pressures due to Wake Induced Transition in a Laminar Separation Bubble on a LP Turbine Cascade”, ASME Paper No GT-2003-38303, ASME Turbo Expo, Atlanta, Jun. Number of blades 5 Chord [mm] 198 Axial chord [mm] 170 Blade stagger [] 59.3 Pitch [mm] 158 Span [mm] 375 Suction surface length [mm] 264.7 Pressure surface length [mm] 230.0 Inlet flow angle [] 37.7 Design exit flow angle [] 63.2 Bar diameter [mm] 2.05 Bar Pitch [mm] 158 & 316 Axial distance: bars to LE [mm] 70 Flow Coefficient ( =Vx/Ub) 0.83

Reduced frequency (fr) 0.68 Table 1: Specification of T106 cascade. lugs for bars

pulleys

moving belt

cascade

glass window

DC motor

Figure 1: Bar passing cascade facility consisting of wake generator and cascade.

Figure 2: Detail of periodicity flap for bar passing cascade. 1.25

0.75 s i 2 v / v

0.25 sb/sc=1 steady

0 0 0.25 0.5 0.75 1

Figure 3 Isentropic surface velocity distribution measured on the T106A LP turbine cascade.

p la n e o f b a rs

position of bar at t=0

37.7 m m l 0 ia 7 ax

Figure 4 Position of Bars at t=0