Laser Velocimeter Measurements in the Pump of an Automotive Torque Converter Part I Effect of Speed Ratio

A torque converter was tested at four turbine/pump rotational speed ratios (0.200, 0.400, 0.600, and 0.800) all with a constant pump rotational speed in order to determine the effect of speed ratio on the torque converter pump flow field. Laser velocimetry was used to measure three components of velocity within the pump and a shaft encoder was employed to record the instantaneous pump angular position. Shaft encoder information was correlated with measured velocities to develop flow field blade-to-blade profiles and vector plots. Measurements were obtained in both the pump mid- and exit planes for all four speed ratios. Results showed large separation regions and jet/wake flows throughout the pump. The midplane flow was found to have strong counter-clockwise secondary components and the exit plane flow had strong clockwise secondary components. Mass flows were calculated from the velocity data and were found to decrease as the speed ratio was increased. Also, the vorticity and slip factors were calculated from the experimental data and are included. The mid-plane slip factors compare favorably to those for conventional centrifugal pumps but less slip was present in the exit plane than the mid-plane. Neither the slip factor nor the vorticity were seen to be strongly affected by the speed ratio. Finally, the torque core-to-shell and blade-to-blade torque distributions are presented for both planes.


INTRODUCTION
The torque converter, commonly used in auto- mobiles, is a hydrodynamic turbomachine which transmits torque smoothly between the engine and the transmission, provides torsional damping, and reduces the shock resulting from sudden changes in operating conditions.The typical torque converter consists of a pump, a turbine, and a stator, and employs oil as the working fluid.The pump is driven by the engine via the crankshaft and energizes the working fluid.The energized fluid Corresponding author.Tel.: (804) 924 6213.Fax: (804) 982 2037.E-mail: rdf@virginia.edu.168 S.B. AINLEY et al.   exits the pump axially and enters the turbine, which extracts energy from the fluid.The stator is the non- rotating component of the torque converter and is located between the turbine exit and pump inlet.The stator is designed to redirect the flow exiting the turbine into the pump with zero incidence angle at the "design" condition.The oil is continually recirculated through the pump, turbine, and the stator.
The torque converter is a very complex turbo- machine for several reasons.First, the flow field changes dramatically over the range of operating conditions.Specifically, incidence angles to the lead- ing edge of the blades for all of the components change from large positive to large negative values over the operating range.Second, the flow is highly three-dimensional as it is turned in the blade passages in two directions.Thus, as in any turbo- machine, blade slip becomes a problem, especially at off-design operating conditions.Also, in addi- tion to turning in the tangential and radial direc- tions, the flow is turned by 180 in the transverse axial direction in both the pump and the turbine in a relatively short distance.Namely, the flow enters these two components in the axial direction, is rapidly turned toward the radial direction, and rapidly turned back into the reverse axial direction.Thus, the torque converter is a very complex mixed- flow variety of hydraulic turbomachine.

Previous Work
A detailed review of early torque converter work was given by Bahr et al. (1990), Gruver et al. (1996)  and Brunet al. (1996) and Brun and Flack (1996).
Thus, for the sake of brevity only some of the more recent publications on torque converters are reviewed herein.Browarzik and Grahl (1992), and Browarzik  (1994) examined the non-steady flow field at the pump and turbine inlet and outlet using hot-film anemometry.A frequency spectrum showing the relative magnitudes of velocities at the turbine exit was presented.Relative velocity magnitude as a function of pump and turbine angular position was also shown.Folchert et al. (1994) presented a mathematical description of the nonlinear behavior of hydrodynamic torque converters.Linear models in the frequency and time domain for each region were established by identification methods and fre- quency response curves for the driving system can then be calculated with high accuracy.An inverse Fourier transformation into the time domain allows the calculation of the dynamic behavior due to general transient excitations.Hedman (1994) presented different ways of implementing the torque converter's non-linear relationships for torque and speeds into a computa- tional model.He presented ways of combining the non-linear torque converter relationships with linear relationships for the rest of the transmission in order to calculate the speeds and torque of all shafts in the entire transmission system.Schultz et al. (1994) described a finite volume method for calculating three-dimensional incompressible viscous flow in a torque converter.Calculations were done in the guide vane, pump rotor, and turbine rotor of a torque converter and the method was shown to be an effective tool for the prediction of turbomachinery flows.and Lakshmi- narayana (1995) used a five-hole probe to acquire pressure data at the turbine exit and stator exit of a torque converter.The flow fields were found to be highly three-dimensional and strong secondary flows were found.A separation region near the shell side of the stator passage was observed.The experi- mental data was in good agreement with the Navier- Stokes computational predictions at the stator exit for both "design" and "off-design" conditions.Wantanabe et al. (1997) used flow visualization and laser velocimetry to study the flow in stator pas- sages for different thicknesses of vanes.They found that the thickness affected the flow fields and other parameters including the flow rate.They also found that the flow was non-uniform and separated re- gions were present regardless of the test conditions.Bahr et al. (1990) measured velocity fields in three dimensions throughout the stator of the torque converter.For the study, a one directional laser velocimeter was used to measure in a machined Plexiglas torque converter.The inlet, one-quarter, mid-, three-quarter, and exit planes were studied using a resolution of 5 by 5 measurement locations for speed ratios of 0.065 and 0.800.At the 0.800 speed ratio a separation region was observed in the mid-plane at the pressure and core sides.For the same speed ratio another separation region was observed in the exit plane at suction and shell sides.The flow rates and torque distributions were found to be strongly dependent on radial position.Gruver et al. (1996) and Brunet al. (1996) used the same experimental facility as Bahr et al. (1990)   to measure steady and unsteady velocities in the pump of the torque converter.Three-dimensional velocity fields were measured in the inlet, mid-, and exit planes of the pump at the speed ratios of 0.065 and 0.800 and pump rotational speed of 800 and 1100 rpm.The flow field in the pump inlet was found to be strongly dependent on the relative position between the pump and the stator.The pump exit velocity field was found to be influenced by the relative position between the pump and the turbine.Secondary flow was found in all three planes with the most significant occurring in the mid-and exit planes at the 0.800 speed ratio.Large separation regions were shown in the mid-and exit planes at the core and suction sides at both speed ratios.The pump inlet and exit flow field were seen to be highly unsteady.
Brunet al. (1996) used the same experimental facility as Bahr et al. (1990) to measure velocity fields in the turbine of the torque converter.The speed ratios of 0.065 and 0.800 were studied for the inlet, one-quarter, mid-, and exit planes of the turbine.
Average flow velocity profiles were obtained for all the measurement planes and transient velocity fields were determined for all the planes except for the exit plane.The relative pump-turbine position exhibited a strong influence on the turbine inlet flow for both speed ratios.Significant separation regions were found in the one-quarter and mid-planes at the 0.065 speed ratio.
By and Lakshminarayana (1995), By et al. (1995), and By (1993) studied a torque converter with the same geometry as Bahr et al. (1990).Steady pressure measurements were taken in both the stator and pump.These measurements were com- pared with both a potential flow analysis and a numerical viscous flow prediction and agreed favorably with the results.The viscous flow predic- tion was used to show that the pump rotation with the radially curved channel precipitated a strong secondary flow.

Motivations and Objectives
Torque converters are used extensively in the automotive industry.However, their internal flow characteristics are not well understood and have received little attention, particularly over a wide range of operating conditions.Theoretical analysis of the torque converter has not developed to a level permitting design solely from computational results.Detailed velocity field experimental data, including velocity distribution, pressures, and tur- bulence intensities, are required for benchmarking and as boundary/initial conditions for computa- tional flow solvers.
The motivation for research in the torque converter pump is three-fold: (i) The current design methods need to be improved in order to advance torque converter technology significantly; namely, velocity data is needed as a benchmark for verifica- tion of flow models.(ii) Since the torque converter operates very inefficiently at off-design conditions, optimization of the internal flow paths will yield an improved torque converter and better automobile fuel efficiency.A greater understanding of the fundamental flow behavior is required to design improved blade passages.(iii) Torque converter pumps are extremely complex mixed-flow turbo- machines because radial, axial, and tangential flow components are present.A basic understanding of its mixed-flow behavior at different operating conditions can be directly related to other mixed flow turbomachines.
A laser velocimeter was used in this study to non- intrusively measure velocity components in the pump of a Plexiglas model automotive torque converter.Velocities were measured in the mid-and exit planes at four speed ratios (turbine rotational speed/pump rotational speed) with a constant pump rotational speed and results are related to this speed ratio parameter.From the velocity data, flow non-uniformities and, in particular, separated regions were determined.Mass flow rates and mass flux distributions in each plane were calculated from the velocity data.Also, slip factors and non- dimensional vorticities were calculated at the mid- and exit planes.Finally, torque flux distribution was calculated for both planes.The use of the obtained experimental results should lead to improved computational modeling of the flow field in the torque converter.
EXPERIMENTAL FACILITY

Torque Converter Rig
The entire experimental apparatus is shown sche- matically in Fig. 1.The torque converter pump was driven by a 18.6kW motor and the rotational energy from the torque converter turbine was absorbed by a 130.5kW eddy current dynamometer.A hydraulic system cooled, lubricated, and pressurized the torque converter rig.The pressures and temperatures throughout the test fixture and support system were monitored and controlled.The pump and turbine rotational speeds were controlled to within rpm.The input and output torque were measured by load cells on the shafts and displayed on the console.The torque converter tested was a standard single stage automotive model with a diameter of 230 mm and was manufactured entirely of Plexiglas for opti- cal accessibility.The torque converter was located in a rectangular containment box also made of Plexiglas as shown in Fig. 2. Both the torque conver- ter and containment box were filled with Shellfiex 212 industrial oil, which has an index of refraction of 1.489 at 25C.This oil was chosen because it matched the index of refraction of the Plexiglas (nominally 1.490), which reduced undesired laser beam refraction, reflection, and scattering.The density of the oil at 25C is 899.1 kg/m and the viscosity is 29 cSt at 25C and 19 cSt at 40C.
The torque converter pump consisted of 27 identical blade passages, the turbine of 29 identical blade passages, and the stator of 19 passages.The pump blades were constructed by milling a solid piece of Plexiglas and thus creating a one piece "nest" with the blades an integral part of the core.The blade nest was then permanently attached to a separately constructed pump shell with a solvent.The same process was used to construct the turbine.The pump and turbine blades were 1.1 mm thick and the shells were 2.67 mm thick.The stator blades and shell were also milled from a single piece of Plexiglas and then glued to the core.In Fig. 3   pump geometry and the two measurement planes are shown.The blade passage geometries and operating conditions were previously fully docu- mented by By and Lakshminarayana (1995), Gruver et al. (1996), and By (1993).Thus, the entire test-facility can be reconstructed from informa- tion in the public domain.Also, the performance curves for the tested torque converter geometry are presented in Fig. 4.

LV System
A one-directional, frequency shifted, back scatter laser velocimeter system was used for this study.
A 2-watt Argon-Ion laser, operated at 50 mW and at a wavelength of 5!4.5 nm, was employed.The laser and the accompanying laser velocimeter optics were mounted on a mill table as shown in Fig. 5 to provide a stable vibration free base.The position of the table and, thus, the measurement location (probe volume) was controlled by 3-axis stepper motors, linear encoders, and a digital read-out.i__J--- The single laser beam from the emission source was split into two equal intensity beams by a beam splitter module.One beam was frequency upshifted using a Bragg Cell in order to eliminate velocity directional ambiguity.Beam spacers, expanders, and a focusing lens (focal length of 250 or 350 mm, depending on the measurement requirements) were employed to first reduce the probe volume size and then focus the laser beams to cross in the test section.To provide for light scattering, the oil was seeded with metallic coated glass particles which have an average diameter of 121am and have a density approximately that of the oil.The scattered light from the particles was focused into a 0.02 mm diameter pin-hole and then converted to a voltage signal by a photomultiplier tube (PMT).The PMT output signal (Doppler signal) was amplified, sent to a burst counter processor, and converted to digital velocity output.The burst counter proces- sor employed bandpass filtering, an adjustable minimum threshold, and a 5/8 peak counter com- parator to validate burst signals.The digital velocity information was read into a data acquisition computer via a high speed digital I/O card.
Simultaneously with each velocity signal, the instantaneous angular positions of the pump and turbine were read in using two 10-bit shaft encoders (1024 circumferential positions) attached to the pump and turbine shafts.
The measurement uncertainty of the velocity was due to the uncertainty of the clock counter in the digital processor, uncertainty in the beam crossing angle, and a the uncertainty of using a finite number of samples to approximate a true distribution; this uncertainty was typically +0.05 m/s with 95% con- fidence.Uncertainties in the angular positions of the two rotating components were +0.35 Due to the small difference of index of refraction of the oil and plastic, a translational uncertainty of the probe volume position was also present and was approxi- mately 0.05 mm.This probe volume position uncer- tainty was statistically added with the uncertainty in the angular position, which resulted in a typical uncertainty in the velocity of 0.07 m/s, when velocity gradients were present.As a consequence, the final total velocity uncertainty was 0.11 m/s.Procedure Measurement plane locations were determined by first using reference points in the pump and the known pump dimensions.The probe volume was then moved to the desired measurement locations in the pump with a total accuracy of the absolute position of 0.05 mm in any direction.In the pump mid-and exit planes, the velocity measurements were taken at 9 evenly spaced core-to-shell positions.Since the pump rotated, the velocity data for each core-to-shell position was resolved into 37 blade-to- blade positions.Therefore, for each pump plane a measurement grid of 9 core-to-shell by 37 blade-to- blade relative positions was obtained.Approximately 150 valid velocity samples were collected for each measurement grid position in the rotating frame, which is sufficient for a high confidence in the average velocities.
The pump was studied in detail for four speed ratios (0.200, 0.400, 0.600, and 0.800) with a cons- tant pump rotational speed of 550 rpm in order to determine the effect of speed ratio on the torque converter flow field.Table I presents the different operating conditions.For reference, Reynolds numbers were also calculated for the pump exit plane for the various operating conditions and are included.The Reynolds number (Re--VDh/U was calculated in two ways: (i) based on the through-flow velocity and (ii) based on the pump tip speed.In both methods of calculation, the hydraulic diameter, Dh, equals 4ALP, where A is the area of the measurement plane and P is the wetted perimeter.The hydraulic diameter was 0.01901 m in the pump exit plane.For the Reynolds number calculation based on the through-flow, V is simply the average through-flow velocity in that plane, and for the calculation based on the tip speed, V is the velocity of the tip of the pump blade (V= rc).

Data Reduction
Average velocities were calculated and organized into blade-to-blade and core-to-shell profiles for the 27 blade passages of the pump using the pump shaft encoder information.The velocity profiles of all blade passages were shown in studies not presented herein to be identical within the mea- surement uncertainties and, hence, could be superimposed and averaged.Blade-to-blade, core- to-shell velocity profiles, and vector plots were plotted using commercially available software.

RESULTS AND DISCUSSION
As previously described, experimental velocities were obtained for two pump planes at four turbine/ pump rotational speed ratios.All velocities presented herein are given with respect to the rotating pump frame.Mass flows were calculated from the experimental through-flow velocity data, where through-flow velocity is defined as the velocity component normal to a given measurement plane.
Secondary flows are also identified and quantified.
Torque values are derived from the angular momentum fluxes and slip factors are determined from the relative tangential velocities.All of the above are presented and the effect of speed ratio on these parameters is discussed in detail.

Pump Mid-plane Velocity Fields
Velocity results for the 0.600 speed ratio mid-plane flow are presented in Figs. 6 and 7. Figure 6 shows a three-dimensional velocity vector plot of the flow field.Highest velocities (jet flow) can be seen to be along the shell side of the mid-plane and at the shell- pressure corner.The lowest velocities (wake flow) Axial Flow Tangential Flow Radial Flw uction FIGURE 6 Pump mid-plane average total velocity vector plot for speed ratio 0.600.
are located directly along the suction and core sides.
The peak through-flow velocities at the shell-pres- sure side are approximately 2.7 times the average through-flow velocity.A significant flow separation region or wake is located near the suction side of the blade passage and extends towards the center of the plane.This jet/wake type flow is the result of a combination of Coriolis, centrifugal, and viscous forces, and was previously identified and described in detail for conventional centrifugal impellers by Eckardt (1975).
Figure 7 shows the secondary flow pattern of the mid-plane at the speed ratio of 0.600.Secondary flow is defined herein as the relative velocities (with respect to the pump frame) normal to the through- flow component.The secondary flows in the mid- plane are seen to have a strong counter-clockwise rotational circulation; average secondary velocities are 1.0 m/s.The region of highest secondary veloc- ities is located at 40% pressure-to-suction and these velocities are approximately 1.2 m/s.Lowest second- ary velocities are located in the core-suction corner region.The mid-plane velocity results for the other speed ratios have qualitatively similar jet/wake and circulatory secondary flow trends, but the Shell Suctlon Pressure 1.5 m/s Core FIGURE 7 Pump mid-plane average secondary velocity vector plot for speed ratio 0.600.magnitudes of the velocities magnitudes vary.Such plots are not presented for the sake of brevity but were documented by Ainley (1994).
For reference similar plots are presented in Figs. 8 and 9 for a speed ratio of 0.800.As can be seen, the general characteristics are the same as for the speed ratio of 0.600.

Pump Exit Plane Velocity Fields
Figure l0 shows the average exit plane vector plot at the 0.600 speed ratio.For this case, highest velocities are again located in the shell-pressure corner region and the lowest velocities are located Axial FIow Tangential Flow Radial Flow Pressure Core Suctlon FIGURE 8 Pump mid-plane average total velocity vector plot for speed ratio 0.800.near the suction side of the plane.The peak through-flow velocities are approximately 2.5 times the average through-flow velocity and are located near the shell-pressure corner.Velocities in the cen- ter of the plane are 1.2 times the average throughflow velocity.Along the core between 10-40% of the pressure-to-suction region, a triangular region of reversed flow is present.Thus, a jet/wake flow is still present in the exit plane.
Figure 11 shows the secondary flow pattern for the exit plane at the 0.600 speed ratio.The second- ary flows in the exit plane have a strong clockwise rotational circulation, which is opposite that of the

Pressure
Pressure FIGURE 9 Pump mid-plane average secondary velocity vector plot for speed ratio 0.800.

FIGURE
Pump exit plane average secondary velocity vector plot for speed ratio 0.600.mid-plane.Hence, the circulatory secondary flows completely reversed their rotational direction between the mid-and exit-planes.Peak secondary flows are approximately 1.2m/s and average secondary velocities in the exit plane are 0.9 m/s.A combined effect of both the Coriolis and centrifugal forces are believed to be responsible for the strong secondary flow in both the mid-and exit planes.The exit plane velocity results for the other speed ratios again have similar jet/wake and secondary flow trends but the magnitudes are different.Again, such plots are not presented for the sake of brevity but were documented previously by Ainley (1994).
For reference similar plots are presented in Figs. 12 and 13 for a speed ratio of 0.800.Once again, the general characteristics are the same as for the speed ratio of 0.600.

Mass Flux
The mass flows were calculated for the mid-and exit planes from the experimental data by integrating the through-flow velocity multiplied by the area and density across the entire measurement plane.The small differences between the mass flows from the mid-plane to the exit plane are well within the calculated mass flow uncertainties.The measured mass flow rate for the two planes and four speed ratios are shown in Fig. 14.The mass flow rate decreases by approximately 40% as the speed ratio is increased from 0.200 to 0.800.
Figure 15 shows the mass flux distribution from pressure-to-suction (blade-to-blade) for both planes at speed ratios of 0.200, 0.600, and 0.800.For this figure the mass flux was averaged from core to shell.For both planes, the highest mass flux is located at 10-15 % ofthe pressure-to-suction surface.Near the suction side of the blade passage the mass flow is minimal and actually negative for high speed ratios in the mid-plane, indicating reverse flow.The core- to-shell mass flux distribution for both the mid-and exit planes at speed ratios of0.200, 0.600, and 0.800 is shown in Fig. 16.Conversely, for this figure the mass flux was averaged from pressure to suction surface.
For the highest speed ratios, near the core side ofthe blade passage the mass flux is approximately zero in the mid-plane and negative in the exit plane; the mass flux increases steadily towards the shell side of the passage in both planes.For the lowest speed ratio, the distribution is more symmetric, especially in the exit plane.For all cases and planes the magnitudes of the mass flux, decrease as the speed ratio increases.

Torque Flux Distribution
The total shaft torque can be calculated from the angular momentum principle and is given by the  The torque flux distribution (N/m) or distribution of torque throughout the plane was determined by calculating the torque contribution of each differ- ential area as follows" Torque Flux where Vo is the absolute tangential velocity, ri is the radius from the shaft center, and V, is the through- flow velocity.The pressure-to-suction (blade-toblade) torque flux distribution for speed ratios of 0.200, 0.600, and 0.800 is shown in Fig. 17 for both the mid-and exit planes and the core-to-shell torque flux distribution is shown in Fig. 18.For these two figures the torque flux was averaged from core to shell and pressure to suction surface, respectively.The highest torque is located at approximately 10% pressure-to-suction and 80% core-to-shell in both the mid-and exit planes, regardless of speed ratio.
Along the suction side and the core side of the blade passage, the torque flux is minimal.In the exit plane near the core surface and in the mid-plane near the suction surface the torque flux is negative for high speed ratios.Low torque flux regions indicate areas where minimal beneficial pump work is being done on the fluid.The magnitudes of the torque flux increase as the speed ratio decreases, as expected when considering the mass flux distribution.Core-to-shell surface torque flux distribution for speed ratios 0.200, 0.600, and 0.800.

Vorticity
The average flow vorticity, , was calculated in the pump using the following equation: where u is the core-to-shell velocity and v is the pressure-to-suction velocity.The values of x and y are the absolute coordinates of the core-to-shell and pressure-to-suction locations, respectively.The vorticity equation in discretized form is -j k,,Xi+l Xi Yi+l X (Xi+I Xi)" (Yi+' Yi) The vorticity, calculated from the above equation and the experimental data, was non-dimensiona- lized by dividing by the average through-flow velocity and multiplying by a reference length.The reference length for this calculation is the core-to- shell length of the pump blade (15.5mm).The non-dimensional vorticity values are shown in Fig. 19 for both planes at the four speed ratios.
A positive vorticity value represents a secondary flow rotation in the counter-clockwise direction when looking directly into the plane from a down- stream position.The highest mid-plane vorticity occurs at a speed ratio of 0.800 and the largest vorticity in the exit plane occurs at a speed ratio of 0.400.In the mid-plane the non-dimensional vortic- ities are seen to be relatively constant.In the exit plane the vorticity is lowest at a speed ratio of 0.800.

Slip Factor
The slip factor, # is defined as the ratio of the actual fluid absolute tangential velocity to the ideal fluid tangential velocity.A slip factor equal to indicates the tangential fluid flow perfectly matches the ideal flow.Since the pump blades in both the mid-and exit planes have a blade angle of approximately zero degrees, the ideal flow in both planes has a relative tangential velocity component equal to zero.The slip factor was calculated using the equation below, where vi is the relative tangential fluid velocity, is the angular velocity in rad/sec, ri is the radius of the measurement point, wi is the through-flow velocity, and wa,g is the average through-flow velocity.The calculated values of slip are weighted by the through-flow velocity, as seen in the above equa- tion.The pump slip factors are shown in Fig. 20 along with the values predicted for a 27-blade conventional centrifugal pump by Stodola (1945), Busemann (1928), and Wiesner (1967).The slip factors varied only slightly between the speed ratios, but the values in the exit plane are significantly higher than in the mid plane.The mid plane values range from 0.866 to 0.887 and the exit plane values range from 0.947 to 0.975.The calculated values of the slip factor from measurements at the torque converter pump mid- plane are similar to the predicted values based on centrifugal pump data.This small difference can be attributed to the fact that, as previously indicated, the torque converter pump operates similar to a centrifugal pump especially in the inlet to mid plane section (the through-flow is in the radial direction).
The values of the slip factor for the exit plane do not, however, agree as well with the predicted values.The measured slip factor values are significantly higher than the predicted values.The additional chord length from the mid-plane to the exit plane (nominal chord length between the inlet and exit plane is 2.2 times greater than from the inlet to the mid-plane) provides additional guidance to the flow and produces a higher slip factor (less slip) than for a normal centrifugal pump.

SUMMARY AND CONCLUSIONS
A laser velocimeter was used to measure the flow field in the mid-and exit planes of an automotive torque converter pump.Four speed ratios (0.200, 0.400, 0.600, and 0.800) were studied at a constant pump speed for both planes in order to determine the effect of speed ratio on the pump flow field.Important conclusions drawn from this investiga- tion are: 1.In both measurement planes at all four speed ratios, the highest velocities (jet flow) are gener- ally located at the pressure and shell sides of the blade passage.In the mid-plane, a separation (wake flow) region is located in the core-suction corner region and in the exit plane, a separation region is located along the core side of the passage.
2. Mass flows were calculated for both planes at the four speed ratios.The calculated mass flows were consistent from the mid-plane to the exit plane.The measured mass flow was found to de- crease from 10.8 kg/s to 6.8 kg/s in the pump exit plane and from 10.6 kg/s to 6.1 kg/s in the pump mid-plane as the speed ratio was increased from 0.200 to 0.800 (with a constant pump rotational speed).Core-to-shell and blade-to-blade mass flux distributions are also presented.In both the mid-and exit planes (all four speed ratios), the mass flux is highest near the pressure and shell sides of the blade passage.3. The torque flux distribution was calculated for the mid-and exit planes in the pump.Torque flux distribution is similar to the mass flux distribution with the highest torque flux located at 10% pressure-to-suction and at 80% core-to- shell.The speed ratio does not have a significant influence on the torque flux distribution trends; however, the torque delivered to the pump increases as the speed ratio is decreased. 4. Strong secondary flow patterns are present in both the mid-and exit planes as is evident from the calculated vorticity values.The secondary flow is counter-clockwise in the mid-plane and clockwise in the exit plane.The circulation is a result of the combination of Coriolis, centri- fugal and viscous forces.The strength of the secondary flow, as seen by the calculated vorti- city, varies with the speed ratio and therefore, is mildly dependent upon the speed ratio.The strongest secondary flow (calculated non-dimen- sional vorticity is 2.42) in the mid-plane is at a speed ratio of 0.800 and the strongest secondary flow (calculated non-dimensional vorticity is -2.58) in exit plane is at a speed ratio of 0.400.
5. The slip factor in the torque converter pump was calculated and compared to predicted values for conventional centrifugal pumps.The calculated values at the mid-plane compared favorably with the predicted values.However, the pump exit plane slip factors were higher than predicted values.The added passage length adds to the fluid control, and thus reduces the amount of slip present at the exit plane.In both the mid- and exit planes, the slip factor was seen to be independent of the speed ratio.
FIGUREExperimental facility.FIGURE 2 Plexiglas torque converter and containment box.
FIGURE 3 Pump passage geometry.

FIGURE 4
FIGURE 4 Torque converter power efficiency and torque ratio.
FIGURE l0Pump exit plane average total velocity vector plot for speed ratio 0.600.
FIGURE 14 Pump mass flow rate as a function of speed ratio.

FIGURE
FIGURE 16Core-to-shell surface mass flux distribution for speed ratios 0.200, 0.600, and 0.800.

E
EN NE ER RG GY Y M MA AT TE ER RI IA AL LS S Materials Science & Engineering for Energy SystemsEconomic and environmental factors are creating ever greater pressures for the efficient generation, transmission and use of energy.Materials developments are crucial to progress in all these areas: to innovation in design; to extending lifetime and maintenance intervals; and to successful operation in more demanding environments.Drawing together the broad community with interests in these areas, Energy Materials addresses materials needs in future energy generation, transmission, utilisation, conservation and storage.The journal covers thermal generation and gas turbines; renewable power (wind, wave, tidal, hydro, solar and geothermal); fuel cells (low and high temperature); materials issues relevant to biomass and biotechnology; nuclear power generation (fission and fusion); hydrogen generation and storage in the context of the 'hydrogen economy'; and the transmission and storage of the energy produced.As well as publishing high-quality peer-reviewed research, Energy Materials promotes discussion of issues common to all sectors, through commissioned reviews and commentaries.The journal includes coverage of energy economics and policy, and broader social issues, since the political and legislative context influence research and investment decisions.S SU UB BS SC CR RI IP PT TI IO ON N I IN NF FO OR RM MA AT TI IO ON N Volume 1 (2006), 4 issues per year Print ISSN: 1748-9237 Online ISSN: 1748-9245 Individual rate: £76.00/US$141.00Institutional rate: £235.00/US$435.00Online-only institutional rate: £199.00/US$367.00For special IOM 3 member rates please email s su ub bs sc cr ri ip pt ti io on ns s@ @m ma an ne ey y. .cco o. .uuk k E ED DI IT TO OR RS S D Dr r F Fu uj ji io o A Ab be e NIMS, Japan D Dr r J Jo oh hn n H Ha al ld d, IPL-MPT, Technical University of Denmark, Denmark D Dr r R R V Vi is sw wa an na at th ha an n, EPRI, USA F Fo or r f fu ur rt th he er r i in nf fo or rm ma at ti io on n p pl le ea as se e c co on nt ta ac ct t: : Maney Publishing UK Tel: +44 (0)113 249 7481 Fax: +44 (0)113 248 6983 Email: subscriptions@maney.co.uk or Maney Publishing North America Tel (toll free): 866 297 5154 Fax: 617 354 6875 Email: maney@maneyusa.comFor further information or to subscribe online please visit w ww ww w. .mma an ne ey y. .cco o. .uuk k C CA AL LL L F FO OR R P PA AP PE ER RS S Contributions to the journal should be submitted online at http://ema.edmgr.comTo view the Notes for Contributors please visit: www.maney.co.uk/journals/notes/emaUpon publication in 2006, this journal will be available via the Ingenta Connect journals service.To view free sample content online visit: w ww ww w. .i in ng ge en nt ta ac co on nn ne ec ct t. .cco om m/ /c co on nt te en nt t/ /m ma an ne ey y and

TABLE Test conditions
Speed ratio Pump speed (rpm) Turbine speed (rpm) Stator speed (rpm) Temperature (C) Re (avg.thru flow) Re (tip speed)