AC 608 Test Data
Contents
High speed centrifugal compressor
Application Challenge 608 © copyright ERCOFTAC 2004
Test Data
Overview of Tests
The high pressure ratio centrifugal impeller was investigated in a research project funded by a German research consortium known as FVV (Forschungsvereinigung Verbrennungskraftmaschinen e.V.). Performance measurements and laser measurements (L2F) of the flow field upstream, inside and downstream of the rotor have been carried out at the DLR test rig.
NAME 


 








EXP 1 (characteristic) 














 
EXP2 (flow patterns) 







Global  
EXP 1
Π_{21} η_{21} T_{02}/T_{01} Π_{41} η_{41} T_{04}/T_{01} 
char1.dat  
plane  2  1  0  0.5  1 
EXP 2
c, α, M_{rel} β 
L2f_20.dat  L2f_10.dat  L2f_00.dat  L2f_05.dat  L2f_10.dat 
2  4  6  8  10  
L2f_20.dat  L2f_40_l.dat  L2f_60_l.dat  L2f_80_l.dat  L2f_100_l.dat  
L2f_40_r.dat  L2f_60_r.dat  L2f_80_r.dat  L2f_100_r.dat 
Test Case EXP1(Performance Measurements)
Description of Experiment
Test case EXP1 refers to the overall performance of the impeller. At a constant rotor speed (50'000 rpm) the throttle valve was gradually closed and the resulting total pressure ratio π_{imp} monitored as a function of mass flow rate,
The measured quantities, apart from the rotor speed and the mass flow rate , were the total temperature at inlet and exit, the total pressure at inlet and the static pressure at impeller exit. Mass flow was measured with a venturi nozzle located in the suction pipe. At rotor exit 16 pressure taps were circumferentially distributed to determine a reliable mean static pressure. The total pressure at impeller exit P_{02} was derived from the measured total temperature, mass flow and static pressure by using the continuity and Euler equation.
Boundary Data
The rig was not insulated either during the performance or during the laser measurements tests. Performance measurements were however only made after the rig had attained thermal equilibrium.
In view of the high sensitivity of the CFD predictions of high speed compressors to the tip clearance levels, an attempt was made to measure the running tip clearances at inlet and exit with pins of graphite. These pins were attached to the casing and were abraded by the rotating impeller blades such that the remaining length of the pins provides an indication of the running tip clearance. For the design speed (50'000 rpm) the remaining length of the pins indicates clearances that were 0.5 mm at the inlet and 0.3 mm at the exit. It should be noted, however, that when the rig is started from cold, the centrifugal and thermal expansion of the impeller takes place more quickly than the thermal expansion of the casing so that these values indicate the smallest clearance that occurred during the attainment of thermal equilibrium and not necessarily the actual running clearance. Discussions in the FVV have lead to the assumption that the actual running clearance widths were rather 0.5 mm at the inlet and 0.6 mm at the exit.
Measurement Errors
The rig and the test data obtained from it are of very high caliber, and many early impellers tested in this rig have been used as standard test cases for CFD validation (Eckardt impellers and Krain impeller).
In order to determine the total pressure at impeller exit using the measured static pressure, a blockage factor had to be taken into account which in the present case was assumed to be constant for all operating points (17%, according to an initial 3d flow calculation). This value, together with the measured shroud static pressure taken as representative for the whole rotor exit area, influences the result. Compared to a blockage factor of 0%, a value of 17% increases the determined total pressure ratio for the rotor.
Measured Data
The derived total pressure rise, total temperature ratio and efficiency for rotor and stage as a function of mass flow are stored in data file char1.dat.
Test Case EXP2(Laser Velocimetry Measurements)
Description of Experiment
Test case EXP2 refers to laser velocimetry measurements (L2F) within the impeller channel passage. The positions of the laser measurement planes are indicated in Fig. 2.1. The plane numbers are related to 10times the relative distance m/m_{max} along the shroud, where m_{max} is the distance from leading edge to trailing edge. At each plane laser measurements were carried out at several relative span positions z/h (see Tab. EXPC) with a maximum of 16 circumferential positions x/t between the blades. Upstream of the splitter leading edge there were 32 circumferential positions between the two main blades.
Laser Measurement Planes 
 
10 
20 
30 
40 
50 
60 
70 
80 
90 
95  
20 










10 










0 










5 










10 










20 










40 










60 










80 










100 










Table EXPC Overview of measurement locations
Fig. 2.2 Position of Laser Measurement Planes
For the Laser data the mean velocity was found by integration across the exit area. The total temperature was derived from the absolute circumferential velocity component by applying the Euler equation. Adding the diffuser pressure loss to the measured pressure at the diffuser exit delivered the total pressure at rotor exit during Laser measurements.
Boundary Data
Rotor speed and mass flow rate were kept constant with 50'000 rpm, 2.54 kg/s respectively, corresponding to the design point. The laser measurements are triggered circumferentially pointbypoint on surfaces of revolution with respect to the absolute frame of reference. Based on that data, distributions of relative flow motion are derived for presentation on cross flow planes traversed by the laser beam.
The diffuser loss necessary for the determination of the total pressure at rotor exit during the Laser measurements were calculated with the equation of mass conservation and the moment of momentum equation. The surface roughness was taken into account by a constant friction coefficient of 0.024.
No turbulence data are available.
Measurement Errors
The L2F method is only capable of measuring the velocity component perpendicular to the laser beam. Therefore the measured velocity is always equal or smaller than the actual velocity. As the laser beam was positioned perpendicular to the casing, the error increases towards the hub (see for example Fig. 2.2, plane 4).
Measured Data
The experimental relative Mach number was computed from measured velocity and inlet temperature by applying the equation of constant rothalpy. The data for planes 2 to 10 are stored in L2f_*.dat (see Table EXPB).
© copyright ERCOFTAC 2004
Contributors: Beat Ribi; Frank Sieverding  MAN Turbomaschinen AG Schweiz; Sulzer Innotec AG