Radial turbine preliminary design and

performance prediction
Cite as: AIP Conference Proceedings 2191, 020097 (2019); https://doi.org/10.1063/1.5138830
Published Online: 17 December 2019

Angelo Leto

ARTICLES YOU MAY BE INTERESTED IN

Design and CFD modeling of a solar micro gas turbine for rural zones in Sahel
AIP Conference Proceedings 2126, 140006 (2019); https://doi.org/10.1063/1.5117654

DEVELOPMENT OF A NEON CRYOGENIC TURBO-EXPANDER WITH MAGNETIC BEARINGS

AIP Conference Proceedings 1218, 895 (2010); https://doi.org/10.1063/1.3422446

Performance comparison of different thermodynamic cycles for an innovative central receiver

solar power plant
AIP Conference Proceedings 1850, 160024 (2017); https://doi.org/10.1063/1.4984558

AIP Conference Proceedings 2191, 020097 (2019); https://doi.org/10.1063/1.5138830 2191, 020097

© 2019 Author(s).
 Radial Turbine Preliminary Design and Performance
Prediction
Angelo Leto1, a)
1
CIRA. Italian Aerospace Research Center, via Maiorise, 81043 Capua (CE), Italy
a)
a.leto@cira.it

Abstract. This work will present a model for the preliminary optimized design of a radial turbine and an analysis of the
most relevant loss models for efficiency prediction.
A one-dimensional code, Radial Turbine Global Design, was developed in MATLAB environment. The model developed
has the objective to estimate the efficiency and optimization of the parameters related to it, and losses prediction.
The flow that through a turbine is complex and many mechanisms of the flow losses in turbine, have not be known well. The
computational fluid dynamics CFD simulations is based on complex three-dimensional viscous methods, which require
considerable iterations and, therefore time, in order to obtain an improvement of the geometry, through the convergence to an
acceptable solution often not optimal. While, can be used of one-dimensional models able to predict the global design of the
machine based on the combination of empirical and experimental loss models, able to provide a good performance estimate with
reduced times. In this work, the loss models are analyzed and developed to obtain useful thermodynamic and geometrical
information, necessary for an optimization CFD simulation.
The RTGD has proven useful for the geometric design three-dimensional of the rotor blades and was tested by considering
literature data of real radial turbines that use both Aria and Argon as working fluid, in addition it was compared with the NASA
code RIFTUD.
The code has excellent reliability because the results obtained reproduce the turbine geometry with a low error compared to
the literatures case.
Furthermore, RTGD can predict turbine performance through two approaches, the first considering the turbine geometry and
the second considering the enthalpy drop due to losses. The second approach for estimating efficiency is necessary in order to
confirm the one calculated with the first approach, and therefore the good geometric design of the turbine.
The code, after being validated, was used for the radial turbines design for space applications for expander cycle rocket
engines which require a high efficiency and a low expansion ratio.

NOMENCLATURE
b blade height
c Absolute velocity - chord
Dh hydraulic diameter
Lh hydraulic length
M Mach Number
p Pressure
r Radius
T Temperature
u Rotor Tangential Velocity
w Relative Velocity
Ws Specific Work
Δh0 Total Enthalpy Drop
Δhloss Total Enthalpy Loss

74th ATI National Congress

AIP Conf. Proc. 2191, 020097-1–020097-9; https://doi.org/10.1063/1.5138830
Published by AIP Publishing. 978-0-7354-1938-4/$30.00

020097-1
 Δp0 Relative Pressure Loss
α Flow Angle
β1_opt Inlet Optimum Blade Angle

Subscripts
4 Rotor inlet 5 Rotor outlet
m meridional t throat

INTRODUCTION
Radial turbines find nowadays-widespread use in turbochargers for automotive applications and for auxiliary
power units. The main features are compact design, a higher efficiency at high numbers of revolutions, low
expansion ratios and high power-to-weight ratio.
The main elements for radial turbine are in the order: the inlet volute, the stator nozzle, the rotor and, finally, the
exhaust diffuser, as shown in Figure 1.

1 = volute inlet parameter

2 = volute exit and nozzle inlet parameter
3 = nozzle exit parameter
4 = rotor inlet parameter
5 = rotor exit parameter
6 = diffuser exit parameter
FIGURE 1. Radial Turbine Stage Cross Section

In this work will be analyzed the rotor design, that it is the element that most influence the performance of the
entire turbine. Indeed, it is the turbine’s component where the major energy losses occur due to friction between the
flow and the blades, and the transition from laminar to turbulent flow conditions. The geometry of the blade channel
affects the elaborated mass flow. The motion of the flow from the end towards the inside of the blade at the hub
radius determines power generation, and then work exchange. It also decreases the speed of tangential component
that generates a high-pressure gradient along the vane channel. At the output of the rotor, the flow is axial, the radial
velocity component is zero, and the flow angle is equal to 90°.
For radial inflow turbines, the inlet section of the rotor blades is generally radial.
The parameters that influence the rotor geometry are represented by the loading coefficient and the flow
coefficient [1]:

020097-2
 𝛥ℎ0
𝜓= (1)
𝑢42

𝑐𝑚5
𝜑= (2)
𝑢5

Efficiency can be plotted as a function of these two coefficients, it is observed that the maximum efficiency
occurs when the loading coefficient is in the range of 0.9-1.0 and the flow coefficient is between about 0.2-0.3.

FIGURE 2. Efficiency: Loading and Flow Coefficients [1]

Radial turbines have been extensively investigated, both experimentally and numerically, at NASA Lewis
Research Centre, starting in the mid1960s. The first analytical study on the radial turbine performance was carried
out by Rohlik in order to determine the optimum geometry.
In these introductory notes, will be illustrated primarily on the various, one-dimensional computer models that
have been developed in the past. Reference [2] describes a computer code, based on the model reported in [3] that
can be used to predict the off-design performance of a Radial Inflow Turbine. An improved version of the computer
code described in [2] is presented in [4]: it features better loss models than those implemented in [3] and also allows
to account for trailing edge blockage and to compute flow conditions for pressure ratios at or beyond stator and/or
rotor chocking. A parallel development beside the program described in [2, 4], NASA also released the RIFTUD
(Radial Inflow Turbine Design) [5] code. RIFTUD is a design code based on Rohlik’s model [6] that produces an
“optimal” geometry for a prescribed mass flow and torque. RIFTUD features a slightly different flow-model than
the one used in [2, 4], it uses a multiple stream tube analysis at the rotor outlet. This feature has subsequently [7]
been merged into the off-design code of references [2, 4], thus giving rise to the RTOD (Radial Turbine Off-Design)
code [7]. RTOD is still being distributed by NASA today [8], although its use is limited to US citizens.
The purpose of this paper is the development of new methodologies for the optimal design of radial turbines, a
tool called RTGD has been developed, which uses mathematical models to simplify the turbine design considering
the flow losses in the rotor. Regarding loss phenomena have been implemented six different types of losses:
incidence, passage, tip-clearance, trailing edge, windage losses and kinetic-energy loss at the rotor exit.
Contrary to what was expected, not much data on real radial turbines has been found in the literature, therefore
RTGD has been validated considering the only two cases found, concerning a turbine that uses Air [9] as the
working fluid and another that uses Argon [4]. In the case of the turbine Argon, RTGD it was compared with the
RIFTUD code.

020097-3
 RADIAL TURBINE GLOBAL DESIGN CODE

Radial Turbine Global Design is an engineering-based design tool, that has been implemented in MATLAB
environment to allow a preliminary design of radial gas turbines for different applications.
The only limit of MATLAB software is that there are no thermodynamic tools already implemented, therefore it
was necessary to use the Coolprop libraries [10] to calculate the thermodynamic properties of a wide variety of
fluids, such as density, the specific heats, cp and cv, their ratio γ and the viscosity, 𝜇.
RTGD is based on an automated model that allows determining several turbine’s features, including geometry
and performance, considering few simple inputs.

INPUT OUTPUT
- Efficiency
- Stator and Rotor Pressure Discharge
- Inlet Temperature - Stator and Rotor Temperature Discharge
- Inlet pressure - Enthalpy
- Number of revolutions [rpm] - Pressure Ratio
- Power RTGD - Rotor Geometry:
- Mass Flow Rate ▪ Blade Numbers
- Specific Speed ▪ Inlet Rotor Radius,
- Fluid ▪ Exit Shroud Radius
Properties by CoolProp: ▪ Exit Hub Radius
▪ Inlet and outlet Density ▪ Outlet Blade Thickeness
▪ Hot gas characteristics (cp, γ) ▪ Inlet and Exit Blade Height

FIGURE 3. RTGD Diagram

The RTGD [11] it has been organizing in two different parts, in which the second is a consequence of the first
and cannot be a standalone code.
In the first part the turbine geometry and velocity triangles are computed, while in the second part will analyze,
using several loss models and correlations, the efficiency that the geometry already produced in the first can achieve.
The detailed estimation of the rotor geometric parameters is necessary in order to improve the three-dimensional
blade shape through well-defined models, in order to minimize the losses and to maximize the stage efficiency.
Regarding the stator nozzles losses have been provided in terms of the total pressure drop and total temperature.
Running the code using different input parameters allows appreciating their impact on the achievable
performance changes. This feature can be clearly considered one of the most important goals of this work, because it
can provide an important information for a further design strategy.
The main advantage of the RTGD compared to other codes, such as the NASA code RIFTUD [5], used as a
comparison in this work, consists in the request of only seven input parameters unlike RIFTUD which requires 26,
some of which are specific of the stator and rotor geometry. Moreover, RTGD is able to estimate different types of
rotor loss, which will be discussed in detail in the next section, necessary for turbine performance prediction.

Loss Models

In this part, the different types of losses will be analyzed.

The losses in the rotor are associated with many complex phenomena that make the extraction of power from
the rotor degraded with respect to what is prescribed by the Euler equation. The losses analyzed in this work are
estimated in order to determine the radial turbine performance and the effective work done by the rotor blades, and
therefore the developed power. They will be implemented six different types of losses: incidence, passage, tip-
clearance, trailing edge, windage losses and kinetic-energy loss at the rotor exit.
The rotor is the main element to be investigated, which is strongly integrated within the turbine, therefore, the
radial turbine rotor design represents the most complex element to be realized. The cause of this complexity depends
on the three-dimensional fluid-dynamic interactions between the stator and the rotor and between the rotor and
casing, which are not entirely known.

020097-4
 The stator outlet flow enters the rotor with an absolute velocity inclined of the α angle with respect to the
tangential direction of the reference system.
The rotor rotational velocity generates a dragging speed of the inlet blades which has the same direction as the
tangential component of the inlet flow absolute velocity.
The inlet rotor flow relative velocity is a function of the rotor rotation velocity as well as the absolute flow
velocity:
𝑢
𝑐 = 𝑠𝑖𝑛 𝛼 (3)

In the ideal case the rotor inlet flow has zero incidence, therefore the relative velocity is exclusively radial, so
the rotor blades have a radial inlet shape. The zero incidence does not allow the minimum value of losses due to the
fluid-dynamic nature of the relative flow that enters radially into the blades, since a counter-rotating vortex must be
generated which will restore the irrotational conditions of the absolute flow velocity.
Radial blades will be assumed in the implemented model because they have the following advantages, better
response to high centrifugal loads, due to the rotational velocity, and mechanical stresses. Furthermore, radial blades
are very versatile in off-design conditions because they react well to directions of the relative flow different from the
design one.
In the radial blades hypothesis, the incidence will not be null, therefore an unguided component of the flow is
generated, therefore the inlet relative flow will have a radial and tangential component.
The model developed RTGD estimates the optimal incidence necessary to minimize losses. The tangential
velocity component generated by the non-zero incidence is a function of the rotor number of revolutions and blade
numbers, therefore, it is of fundamental calculating the correct number of blades.
Generally, in real applications, many vanes are not used for several reasons, e.g. excessive flow blockage at
rotor exit, a disproportionally large wetted surface causing high friction losses, and because the weight and inertia of
the rotor may become too high.
The optimum number of blades [5] in the RTGD is determined by a relationship that is a function of the rotor
inlet flow angle α1.
The incidence losses refer to the losses that occur at the inlet of the radial inflow turbine rotor blade passages
when the turbine is operating at a non-zero incidence and, therefore, the flow does not enter the passage in the
optimum direction. Losses in the rotor blade passage are mainly due to the occurrence of secondary flows, and these
are considerably affected by the flow deviation from optimum incidence.
The incidence angle is obtained from the difference between the inlet blade angle and inlet optimum blade angle
β1_opt [4]. The experimentally optimum incidence should lie within the range of about -20° to -40° [12].
The term passage losses include a wide spectrum of different phenomena occurring to the fluid crossing the
rotor. In fact, after a rapid acceleration in the flow direction, the fluid is turned in the meridional plane along the
camber line: this creates a complex pattern of secondary and cross-stream flows, which still today are not
completely understood. Moreover, this causes the growth of boundary layers with loss of kinetic energy and
blockage. A fully detailed model that considers separately all these loss sources, as the ones existing for the axial
turbines, has not yet been developed. In fact, in axial turbine cascades, this can be done by a careful set up and
measures, but this is not actually possible for radial turbines, due to the three-dimensionality of the flow pattern,
which does not permit to differentiate the losses.
A passage loss model, namely the CETI model [12], was developed to estimate more realistically the losses due
to the secondary flow and friction in the rotor passages. This loss type is estimated between the inlet and the throat
of the rotor section:

𝑤42 +𝑤𝑡2 𝐿 𝑟 2 𝑐𝑜𝑠 𝛽𝑡

𝛥ℎ𝑝 = 0.11 (𝐷ℎ + 0.68 (1 − (𝑟𝑡 ) ) b𝑡 ) (4)
2 ℎ 4
𝑐

The impeller blades mate up against the turbine housing with a small clearance to avoid mutual contact. In
addition, the fluid pushes on the suction surface of each blade essentially creating a pressure difference between the
suction and pressure blade surfaces i.e. across the blade. This pressure difference gives rise to a flow through the
blade-housing clearance gap. This flow results in pressure dissipation and a consequent loss.
In the radial and axial turbines, the tip clearance is defined as the gap between the rotor blade and the shroud.
The radial clearance contributes to generate training the secondary flow and deviation of exit flow angle. The third

020097-5
loss calculated in the RTGD code is the clearance loss, using the Dambach [13] equations, model that consider the
axial and radial clearances influences.
Trailing edge losses [14] arises due to mixing occurring, as the adjacent rows, of the two-blade surfaces exit the
blade row. Moreover, at supersonic velocities shock losses also contribute. In the fourth step the RTGD code
determine the trailing edge loss as a function of the exit Mach number, the exit pressure and temperature and relative
pressure loss Δp0 [5], as follows:
2 𝛥𝑝0
𝛥ℎ𝑡 = 2 𝛾 (5)
𝛾 𝑀5
𝑤25 )
𝛾−1
𝑝5 (1+
2 𝑇5 𝐶𝑝

Windage losses are frictional losses occurring on the back face of the turbine disk. The windage loss [14] is
calculated as a function of the Reynolds number.
The only external loss that is usually considered in radial turbine modelling is that of disk friction. This occurs
because of the fluid leakage between the rotor disc and the stationary back plate, where the windage flow causes
quite strong friction. Depending on the turbine, this leakage could also be recirculated into the turbine annulus or
taken away.
The back face of the impeller hub is an annular disk that spins with the impeller. Either this disc may mate with
a fixed surface separated from it by a small clearance or it may spin freely far from other fixed surfaces. In either
case, the fluid adjacent to the disc exerts a shear on the disc with result that the fluid does unproductive work.
Finally, the rotor exit kinetic loss is expressed as:
1 2
𝛥ℎ𝑘 = 𝑐
2 5
(6)

In conclusion, the turbine overall efficiency can be determined as a function of the different loss contributions:

Δℎ0 −∑ Δℎloss
𝜂ts = (7)
Δℎ0

RTGD does not require the pressure ratio as input for the rotor exit pressure calculation, but the latter is
calculated more accurately as a function of the efficiency and the total inlet pressure and temperature:
𝛾−1
𝑊𝑠 𝛾
𝑝5 = 𝑝00 (1 − 𝜂 ) (8)
𝑡𝑠 𝑐𝑝 𝑇00

RESULTS PRESENTATION
The RTGD code has been validated considering two cases present in the literature, in order to demonstrate the
effective validity of the implemented models. In the first case it was considered a real T-100 [9] turbine that uses Air
as the working fluid, while in the second case RTGD was validated considering a radial turbine that uses Argon as
working fluid, whose geometry is provided from RIFTUD code [5].
The purpose of RTGD is to simplify the radial turbine preliminary design in the absence of specific references
and to optimize the turbine geometry in order to maximize performance while minimizing losses.
The Table 1 shows the data of both turbines.

TABLE 1. Working Fluid Input Data

T-100 Argon
Rotational Speed [rpm] 106,588 38,500
Inlet Total Temperature [K] 1056.5 1083.33
Inlet Total Pressure [Pa] 580,400 91,011
Mass Flow Rate [kg/s] 0.33 0.2771
Power [kW] 121 22.37
Pressure Ratio 5.73

020097-6
 RIFTUD [5] was used as a reference test case considering the Argon turbine, moreover, it was also used in the
case of the T-100 turbine in order to validate the RTGD tool.
The NASA RIFTUD [5] code turbine design is implemented in Fortran language. RIFTUD requires a detailed
input related to the geometry of both rotor and stator, therefore it is more suitable for the improvement of a real
turbine.
This code is difficult to use in the case of a radial turbine preliminary design, as it requires 26 input parameters,
some of which are not available, since are related to the turbine geometry.

Real Case Comparison Between RTGD and RIFTUD

This section shows the results of the RTGD tool and RIFTUD code, compared with the real T-100 turbine [9].
The results show that RTGD slightly overestimates the turbine geometry, it can be observed that outlet hub and
shroud radius are about 10% higher compared to the design parameters. While, considering the efficiency
calculation has been predicted with the maximum precision.
In the case of the RIFTUD code the turbine overall dimensions were found very similar to the real case results.
Several attempts were necessary, as the code requires the input of both inlet and outlet stator flow angles, as well
as of different diameters ratio:
- Stator Inlet/Rotor Inlet;
- Stator Exit/Rotor Inlet;
- Rotor Exit Shroud/Rotor Inlet;
- Rotor Exit Hub/Rotor Exit Shroud.

TABLE 2. Comparison RTGD and RIFTUD with Real Case

T-100 RTGD RIFTUD
Inlet Radius [m] 0.0582 0.0557 0.0587
Inlet Blade Height [m] 0.00635 0.00590
Inlet Absolute Flow Angle [°] 76.80 77 75.93
Inlet Relative Flow Angle [°] -31.80 -31.67 -31.53
Number Blades 16 15 15
Outlet Hub Radius [°] 0.0152 0.0167 0.0153
Outlet Shroud Radius [°] 0.0368 0.0410 0.0371
Outlet Blade Height [m] 0.0216 0.0243 0.0218
Outlet Absolute Flow Angle [°] 0.03 0.00 0.00
Outlet Relative Flow Angle [°] 57.40 72.59 73.79
Chord [m] 0.0457 0.072
Trailing Edge Thickness [m] 0.00076 0.0011
Total to Static Efficiency 0.86 0.86 0.86
Pressure Ratio 5.69 5.92
Loading Coefficient 0.91 0.95
Flow Coefficient 0.23

RTGD Comparison whit RIFTUD

This part illustrations the second test case between RTGD and RIFTUD. The table shows that both the overall
turbine geometry that the performances are almost similar. The main differences are in the rotor exit geometry
estimation, in particular, RTGD overestimates the outlet hub radius by about 20%, while the outlet shroud radius is
about 3% higher.
The Table 3 shows the validation of the RTGD tool with the Argon turbine.

020097-7
 TABLE 3. Comparison RTGD with RIFTUD
RTGD RIFTUD
Inlet Radius [m] 0.0777 0.0777
Inlet Absolute Flow Angle [°] 72 71.92
Inlet Relative Flow Angle [°] -32.3 -31.5
Number Blades 12 12
Outlet Hub Radius [°] 0.0233 0.0194
Outlet Shroud Radius [°] 0.0572 0.0554
Outlet Blade Height [m] 0.034 0.036
Outlet Absolute Flow Angle [°] 0.00 0.00
Outlet Relative Flow Angle [°] -64.9 -54.9
Total to Static Efficiency 0.83 0.82
Pressure Ratio 1.63 1.61
Loading Coefficient 0.82
Flow Coefficient 0.34

Loss Discussion

RTGD deals with estimating the most significant losses that occur in the rotor. The diagram shows the six losses
reported in the loss models paragraph. It can be observed that the loss that has the greatest effect on the rotor
performance in both cases is passage loss. While, the windage and trailing edge losses are negligible. It has not been
possible to carry out any type of comparison either for the T-100 turbine or for the Argon turbine, since the data
relating to the losses are not available in both cases examined.

Windage Loss

Tip Clearance Loss

Exit Energy Loss

Trailing Edge Loss

Passage Loss

Incidence Loss

0 10 20 30 40 50 60 70

Trailing Edge Tip Clearance

Incidence Loss Passage Loss Exit Energy Loss Windage Loss
Loss Loss
Argon 8,98 46,58 0,003 35,91 8,53 0
T-100 4,59 63,47 0,002 17,55 14,38 0

Argon T-100

FIGURE 4. Losses Contribution %

020097-8
 CONCLUSIONS
This document represents an "RTGD" engineering tool that aims to get a fast design of a radial turbine.
Loss models have been implemented to predict the radial turbine overall performance with a rotor geometry
optimization.
Despite limited data available in the literature, RTGD has been validated using a real radial turbine test case, in
addition a code-to-code comparison has been made.
In the cases examined it is known that the passage loss represents the highest loss.
From the comparison with the NASA RIFTUD code, the RTGD tool has proved easier to use by providing more
complete results.
The generation of the rotor 3-D geometry is currently under development, to carry out a further optimization
through CFD analysis.
In conclusion, the tests carried out were useful to understand the RTGD reliability for a future application in the
aerospace propulsion field. The latter one will be integrated into Concurrent Design Facility (CDF), aimed at design
all different components of the entire rocket-engine.

REFERENCES

1. Baines N.C., “Low Development in Radial Turbine Rotors”. THE AMERICAN SOCIETY OF
MECHANICAL ENGINEERS, 1996.
2. Todd C.A. and Futral S. M. Jr., “A FORTRAN IV program to estimate the off-design performance of
radial inflow turbines”, NASA TN D-5059, (1969).
3. Futral S. M. Jr. and Wasserbauer C. A., “Off-design performance prediction with Experimental Verification
for a Radial Inflow Turbine”, NASA TN D-2621, (1965).
4. Wasserbauer, C. A. and Glassman, A. J., “Fortran program for predicting off-design performance of radial-
inflow turbines”, NASA TN D-8063, (1975).
5. Glassman, A. J., “RIFTUD Radial Inflow Turbine Design”, NASA TN D-8164, (1976)
6. Rohlik, H. E., “Analytical Determination of Radial Inflow Turbine Design Geometry for Maximum
Efficiency,” NASA, Technical Report No. TN D-4384, 1968.
7. Meitner, P. L. and Glassman, A. J. “RTOD Radial Turbine Off-Design”, NASA TP 2199, (1983).
8. RTOD NASA https://software.nasa.gov/software/LEW-14060-1 for U.S. Release Only.
9. Jones, A. C., “Design and Test of a Small, High Pressure Ratio Radial Turbine,” Trans. ASME J.
Turbomach., 118(2), pp. 362–370, 1996.
10. Alexander Ponomarenko, “RPA: Tool for Rocket Propulsion Analysis”, 2015.
11. Leto A., Bonfiglioli A., Preliminary Design of a Radial Turbine for Methane Expander Rocket Engine.
72nd Conference of the Italian Thermal Machines Engineering Association, ATI2017. Energy Procedia Volume
126, Pages 738–745, 2017.
12. H. Moustapha, M.F. Zelesky, N.C. Baines, D. Japikse, Axial and Radial Turbines, Concepts NREC, (2003).
13. Dambach, R.; Huntsman, I.; Hodson, H.P. “An experimental study of tip clearance flow in radial inflow
turbine”. J. Turbomach. 121, 644–650, 1999.
14. Ghosh, S. K., Sahoo, R. K. and Sarangi, S, K., “Mathematical Analysis for Off-Design Performance of
Cryogenic Turboexpander Trans”, ASME Journal of Fluids Eng. 133(3), p, 031001, 2011.

020097-9