Experimental investigation on a centrifugal compressor by means of squealer tips based on exit flow measurements.

Senthil, S. ; Mohan, N. Krishna

Introduction

The flow field in an impeller passage of a centrifugal compressor is highly complex, three-dimensional and unsteady. A designer in designing a centrifugal compressor should have complete understanding with the influencing parameters, there by they can design it for better performance and efficiency. The performance of a compressor is inherently deteriorated by different loses that occur at different sections of the compressor stage. They are as follows: shock loss at an inducer inlet, wall friction loss within impeller channels, tip leakage loss, secondary flow losses, mixing loss at diffuser inlet, wall friction loss within a diffuser, sudden expansion loss at a scroll inlet, wall friction loss within a scroll.

Pampreen (1973) concluded that clearance effects have pronounced influence on the performance of centrifugal and axial compressors compared to Reynolds number effects. Ishida and Senoo(1981)[1] used two entirely different types of centrifugal blowers one with a radial blade impeller and the other with a backward blade impeller, measured the pressure distribution along the shroud at five flow coefficient and seven tip clearances. Senoo and Ishida (1987) observed the deterioration of compressor performance due to tip clearance of centrifugal impeller. They modified their theory on the tip clearance loss of centrifugal impeller to include the variation of slip co-efficient of the impeller due to the tip clearance, by deriving a rational relationship between two empirical parameters in the theory. They have compared experimental data in the literature with prediction, to select corresponding flow rates of a compressor with different values of tip clearance loss.

Heyes et al (1992)[4] observed the effect of blade tip geometry on the tip leakage flow in axial turbine cascades. The investigation includes an examination of the performance of the cascades with a variety of tip geometries. The effects of using plain tips, suction side squealers, and pressure side squealers are reported. Ameriet al (1998)[5] observed the effect of squealer tip on rotor heat transfer and efficiency. Experimental investigations are performed to measure the detailed heat transfer coefficient and static pressure distribution on the squealer tip of a gas turbine blade in the five-blade stationary linear cascade. Results show that the heat transfer coefficient on the cavity surface and rim increases with an increase in tip clearance. The heat transfer coefficient on the rim is higher than the cavity surface. The cavity surface has a higher heat transfer coefficient near the leading edge region than the trailing edge region. The squealer tip blade provides a lower overall heat transfer coefficient when compared to the flat tip blade. Azad et al. (2000)[6] compared the heat transfer and flow on the squealer tip of a gas turbine blade with that on a flat tip. Swamy (2001)[7] conducted tests with partial shrouds, with out partial shrouds and turbulence generator on centrifugal compressor with different values of tip clearance. The configuration with partial shrouds shows higher efficiency and energy coefficients compared to other configuration. Krain Hartmut (2002)[8] observed the effect of unsteady diffuser flow in a transonic centrifugal compressor. Comparative study of unsteady flows in a transonic centrifugal compressor with vaneless and vaned diffusers a was observed by Curi Michael (2005)[9].

Experimental Facility and Instrumentation

The present experimental investigations are carried on a low speed centrifugal compressor set up available in the Thermal Turbomachines Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Madras and Thermal Engineering Laboratory, Department of Mechanical Engineering, Mailam Engineering College, Mailam. A schematic layout of the experimental set up is shown in Figure 1. The experimental set up consists of essentially a centrifugal impeller driven by a 5kW DC motor with a rated speed of 2000 rpm. The DC motor is directly coupled to the shaft carrying the impeller. The main components of the compressor are suction duct, impeller, vaneless diffuser formed by the front and rear walls of casing and volute casing of circular cross section and delivery duct with a throttle outlet and nozzle at the inlet.

The major geometrical details of the impeller are given below.

Pressure rise : 300 mm of WG

Speed of rotation : 2000 rpm

Inducer tip dia : 300 mm

Impeller tip dia : 500mm

Balde angle at Inducer tip : 35 [degrees]

Volume flow rate : 1.2[m.sup.3]/s

No. of blades of the impeller : 16

nducer hub dia : 160 mm

Blade height at the exit :34.74 mm

Blade angle at inducer hub : 53[degrees]

Blade angle at exit a) at hub :75[degrees]b) at mean 90[degrees]c) at tip : 105[degrees]

All the angles are measured with respected to tangential direction.

Spacers

Spacers are used to maintain the tip clearance accurately. Front cover is moved axially to vary the tip clearance. There by the distance between the front and rear cover walls would change. Three spacers are used which are 90[degrees] apart at a radius of 270 mm. The spacers are away from the probe, so that they would not interfere with the flow.

[FIGURE 1 OMITTED]

Squealers

Squealers are used to improve the performance of the compressor. The Squealers are made of mild steel of 3 mm thickness. A squealer tip with 6 mm recess is considered here. Squealers are welded on suction side of the blade, pressure side of the blade and both suction and pressure sides of the blade. Meridional view of impeller with the squealer tips are shown in Figure 2.

[FIGURE 2 OMITTED]

Experimental Programme and Procedure General

The Objective of the present investigations is to explore the possibility of using squealer tips to improve the performance of centrifugal compressor. The flow field in the tip region of centrifugal compressor has been studied for decades. The flow is a three dimensional phenomenon comprised of the complex interactions between the tip leakage vortex, the turbulent end wall boundary layers, and often shock waves distorted by rotational effects.

Experimental Programme

The present experiment programme consists of testing the centrifugal compressors with the following tip configurations at three flow coefficients

i. Suction side squealer

ii. Pressure side squealer

iii. Squealers on both suction and pressure sides

iv. Solid tip by filling up the space between squealers

Performance Characteristics

The performance study of centrifugal compressor is carried out at a constant speed of 2000 rpm by varying throttle positions from high flow rate (near choking) to low rate (in surge region). Under steady state conditions the inlet static pressure across the nozzle, the delivery static pressure across the delivery duct, the input field and armature voltages and input field and armature currents are noted down. The speed of the compressor is mainted at 2000 rpm indicated by a non-contacting type tachomter. From these measurements, energy coefficient and efficiency of the centrifugal compressor are computed and plotted against the flow coefficient. The performance tests are conducted for four configurations.

Shroud Wall Static Pressure Measurements

Static pressure taps were provided on the shroud from the inducer leading edge to the impeller exit. The shroud has its inner contour shape in such a way to match the impeller blade tip from the inlet to the exit of the impeller. The inlet flow gets into the impeller blade passage through the annulus passage between the hub of the impeller and stationary shroud. Hence the static pressure measurements on the shroud give an indications of the static pressure developed in the shroud end of the impeller. There are nine static pressure tappings on the shroud. They are connected through tubes to the twenty channel scanning box. Static pressure on the shroud is measured at the above mentioned flow coefficients and tip clearances.

Exit Flow Measurements

A pre-calibrated five-hole probe is used to measure the flow conditions at exit of the impeller. At the exit, the probe is traversed axially from hub to shroud wall, by means of a traverse mechanism and measurements were taken at 12 axial locations for three flow coefficients ([PHI] = 0.34, 0.28 and 0.18). The twelve axial locations spans the axial distance along blade height, at intervals of 2mm near the walls and 4mm at the center of the walls. For each axial station the five hole probe is rotated in different angular positions. The experiments are carried out for three flow coefficients namely, above the design flow rate [degrees].sub.h = 0.34 near design flow rate, [PHI].sub.d = 0.28 and below design flow rate [PHI].sub.1= 0.18. At each station, pressure readings of five-hole probe and wall static pressure are recorded from digital micromanometer connected via the 20 channel scanning box.

Non Dimensionalization of Flow Parameters

The different parameters of performance used in the analysis of experimental results are defined and determined in the following sections. The following formulae are used to calculate the flow parameters

Flow Coefficients, [??]

The actual volume flow through the compressor is expressed as a nondimensional parameter flow coefficient, defined as the ratio of radial velocity to the peripheral velocity at the impeller exit.

[PHI]=V/[pi]d.sub.2b.sub.2U.sub.2 = [pi].sub.2b.sub.2cr.sub.2/[pi]d.sub.2b.sub.2U.sub.2 = cr.sub.2/U.sub.2

Where V- volume flow rate

d.sub.2 - impeller tip diameter

b.sub.2 - width at the impeller exit

Cr.sub.2 - radial velocity at the impeller exit

U.sub.2 - peripheral velocity at the impeller exit

The actual volume flow rate is obtained from the calibration curve of the inlet nozzle corresponding to the static pressure differential at the nozzle section. In the present investigation, the flow coefficients selected are : [PHI].sub.d = 0.28 (near design) [PHI].sub.1 = 0.18 (below design) and [PHI].sub.h = 0.34 (above design)

Energy Coefficient, [PHI]

The specific work W for the centrifugal compressor is expressed nondimensionally as the energy coefficient, [PHI]

[PSI] = 2W/[U.sup.2.sub.2]

where the specific work W is obtained as the difference of specific total energy between inlet and exit of the impeller. For incompressible flow, the specific work is given by

W = [P.sub.D] - [P.sub.1]/[rho] + [C.sup.2.sub.d]- [C.sup.2.sub.s] + g[DELTA]z

where [P.sub.d] and [P.sub.s] are the average static pressure measured at the exit and inlet sections and [C.sub.d] and [C.sub.s] are air velocities at the exit and inlet sections respectively. [DELTA]Z is the geometric level difference between delivery and suction flanges and [rho] is the density of air. The density of the air is calculated from atmospheric pressure and temperature.

Efficiency, [??]

The efficiency is the ratio of fluid power output to the coupling power input of the compressor.

[eta]=[N.sub.eff]/[N.sub.c]= [rho]VW/[N.sub.c]

[N.sub.eff] = Fluid power, Watts

[rho] = Density of air, kg/[m.sup.3]

V = Volume flow rate, [m.sup.3]/s

W = Specific work, [m.sup.2]/[s.sup.2]

[N.sub.c] = Coupling power, watts = [eta[].sub.m] [[V.sub.a][I.sub.a] + [V.sub.f][I.sub.f]]

[[eta].sub.c] = Efficiency of the motor

[V.sub.a] = Armature voltage, Volt

[I.sub.a] = Armature current, Ampere

[V.sub.f] = Field Voltage, Volt

[I.sub.f] = Field current, Ampere

Wall Static Pressure Coefficient, [??] Ws

Wall static coefficient is defined as

[psi].sub.ws] - [P.sub.ws] - [P.sub.atm]/ 1/2 [rho][U.sup.2.sub.2]

where Pws = Static pressure on the diffuser hub or shroud wall

Total Pressure Coefficient

Total pressure coefficient is define as,

[psi].sub.TP] - [P.sub.0] - [P.sub.atm]/ 1/2 [rho][U.sup.2.sub.2]

where [P.sub.0] = Total pressure

Static Pressure Coefficient, [psi]s

Static pressure coefficient is defined as

[psi].sub.SP] - [P.sub.s] - [P.sub.atm]/ 1/2 [rho][U.sup.2.sub.2]

Where [P.sub.s] = Static pressure

Non Dimensional Velocities

They are defined as follows

Non-dimensional total velocity = C / [U.sub.2]

Non-dimensional radial velocity = [C.sub.r] / [U.sub.2]

Non-dimensional axial velocity = [C.sub.x] / [U.sub.2]

Non-dimensional tangential velocity = [C.sub.u] / [U.sub.2]

Mass Averaged Flow Quantities

The mass averaged total pressure coefficient is defined as follows.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Mass Averaged static pressure coefficient : The mass averaged static pressure oefficient is defined as follows.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Results and Discussion

Centrifugal compressor performance, in terms of energy coefficient across the impeller, derived from inlet nozzle static and delivery static pressures measured for all four configurations is presented. The four configurations are

i. Squealers on suction surface

ii. Squealers on pressure surface

iii. Squealers on both suction and pressure surface

iv. Solid tip by filling up the space between squealers

Performance Characteristics Energy Coefficient and Efficiency

The effect of configuration on the performance of the centrifugal compressor is shown in Figures 3 and 4. Figure 3 shows the performance characteristics in terms of energy coefficients, [psi]vs. flow coefficients, [PHI]. Figure 4 shows the performance characteristics in terms of efficiency, n vs. flow coefficient, [PHI]. From the performance curves; it is found that pressure surface squealer tip configuration show increased energy coefficient and efficiency across the compressor compared to other configurations. Also it is observed that the effect of suction surface squealer tip configuration is comparable to that of pressure surface squealer tip configuration.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Impeller Exit Flow Measurements

Total Pressure Coefficient, [psi]02

The distribution of total pressure coefficient at the impeller exit for four configuration at flow coefficient f = 0.28 is shown in Figure 5. From this figure, it can be clearly seen that configuration with pressure side squealer shows increased total pressure coefficient ([psi].sub.02] compared to remaining configurations for all the flow coefficients. This may be attributed to the pressure side squealers attached to the blades restricting the tip leakage flow and hence increased total pressure obtained. Also it is observed that the effect of suction side squealer tip configuration is comparable. It can also be observed that, the total pressure coefficient is increasing as the flow coefficient decreases.

[FIGURE 5 OMITTED]

Static Pressure Coefficients, [psi]02

The distribution of static pressure coefficient at the impeller exit for four configurations at flow coefficient [PHI] = 0.28 is shown in Figure 6. From this figure, it can be clearly seen that pressure side squealer shows increased static pressure coefficient, ([psi]s2) compared to remaining configurations for all flow coefficient. This may be attributed to the pressure side squealers attached to the blades restricting the tip leakage flow and hence increased static pressure is obtained. Also it is observed that the effect of suction side squealer tip configuration is comparable. It can also be observed that, the static pressure coefficient is increasing as the flow coefficient decreases.

[FIGURE 6 OMITTED]

Absolute Velocity, [C.sub.2]

The axial distribution of absolute velocity at the impeller exit for four configurations at flow coefficient 0.28 is shown in Figure 7. From this figure it can be clearly seen that impeller with pressure side squealer shows increased absolute velocity compared to remaining configurations. This is due to increase in tangential velocity, which means increased energy transfer. A careful study of this figure shows that both suction and pressure side squealer tip configuration gives lower absolute velocity for all the flow coefficients. This lower absolute velocity is due to leakage flow in the flow field, which mixes with the main flow at the exit, resulting in lower values of velocities.

[FIGURE 7 OMITTED]

Radial Velocity, [C.sub.2r]

The distribution of radial velocity at the impeller exit for flow coefficient 0=0.28 and four impeller configurations is shown in Figure 8. It can seen that radial velocity increased for all flow coefficients in the impeller shroud region for the configuration with pressure side squealer, compared to other configurations. It can be observed from the figures that at the shroud region for three values of flow coefficients show decrease in radial velocity for hub region. At shroud region boundary layer thickness increase due to squealers, Therefore higher loading on the blade causes decreases in radial velocity in the hub region.

[FIGURE 8 OMITTED]

Static Pressure Distribution on The Shroud Wall

The static pressure distributions on the front shroud for four configurations at flow coefficient [PHI] = 0.28 is shown in fig.9. On the front shroud, static pressure taps are provided from the inducer leading edge to the impeller exit. The shroud has its inner contour shaped to match the impeller blade tip from the inlet to the exit of the impeller.

The static pressure measurements on the shroud give an indication of variation of the static pressure developed in the tip region of the impeller. From the figure it can be seen that more deceleration pronounced at higher flow coefficients. The pressure initially decreases due to suction and then uniformly increases, indicating that there is no dead zone or eddies near the shroud region inside impeller and of the energy is transferred smoothly to the fluid near the shroud.

[FIGURE 9 OMITTED]

Mass Averaged Flow Performance of Impeller

Mass averaged performance of compressor (Figure.10) shows the variation of mass averaged values of total pressure coefficient [PSI].sub.02]) and static pressure coefficient ([PSI].sub.s2]) with flow coefficient. These values are obtained by mass averaging the flow parameters measured with the five hole probe at three flow coefficient. The mass averaged values of total pressure coefficient ([PSI].sub.O2]) and static pressure coefficient ([Psi].sub.s2]) at the impeller exit for three configurations at three flow coefficients clearly indicate increase in total and static pressure for pressure surface squealer tip configuration compared to other configurations.

[FIGURE 10 OMITTED]

Conclusions

The following conclusions are drawn from the present investigations

1. Pressure surface squealer tip has beneficial effects in increasing energy coefficient and efficiency of the tested compressor.

2. Distribution of total and static pressures at the impeller exit shows higher values for configuration with pressure side squealer.

3. Mass averaged total and static pressures obtained from the probe traverses are higher for configuration with pressure surface squealer

4. Suction surface squealer tip has reasonable beneficial effect on the energy coefficient and efficiency.

Acknowledgments

The research described herein was supported by Thermal Turbo machinery Laboratory, Indian Institute of Technology, Madras, India.

Special thanks is given to Dr. N. Sitaram, Professor, Department of Mechanical Engineering, IIT, Madras for his contributed and support in creating this research work.

References

[1] Shida, M. and Y. Senoo(1981) on the pressure losses due to the tip clearance of centrifugal blower, Trans. of ASME J1. of Engg. for power, 03, 271-278.

[2] Senoo, Y and M. Ishida, (1987) Deterioration of compressor performance due to tip clearance of centrifugal impellers. Trans. of ASME J1. of Turbomachinery, 109, 55-61.

[3] Krain, H., (1987) Experimental and theoretical analysis of centrifugal impeller flow. Proc. Inst. Of Mechanical Engineers conf., Paper C270/87, 199-209.

[4] Heyes, F.J. G., H.P. Hodson and G.M. Dailey(1992) The effect of blade tip geometry on the tip leakage flow in axial turbine cascade. Trans of ASME J1. of Turbomachinery 114, 643-651.

[5] Ameri, A.A., D.L. Rigby and Steinthorsson, E., (1998) Effect of Squealer Tip on Rotor Heat Transfer and efficiency Trans. of ASME J1. of Turbomachinery, 120, 753-759.

[6] Azad, Gm. S.J-C. Han and R.J.Boyle (2000) Heat Transfer and flow on

the Squealer tip of a gas turbine blade. Trans. of ASME J1. of Turbomachinery, 122, 725-732.

[7] Sitraram, N. and T.N. Shridhara (2000) Review : Recent investigations on tip clearance effects in centrifugal compressors. Int. J1 of Jet Engines and Turbomachines, 42 309-315.

[8] Krain Harmut (2002), Unsteady diffuser flow in a transonic centrifugal compressor. International Journal of Rotating machinery, 86, 112-123.

[9] Cui Michael M, (2005), Comparative study of unsteady flows in a transonic centrifugal compressor with vaneless and vaned diffusers. International Journal Rotating machinery, 187, 86 - 98.

[10] Biba Yuri (2004), Inverse design of centrifugal compressor stages using a meanline approach. International Journal Rotating machinery, 135, 146 157.