Proceedings of the 9th International and 49th National Conference on Fluid Mechanics and Fluid Power (FMFP)

December 14-16, 2022, IIT Roorkee, Roorkee-247667, Uttarakhand, India

FMFP2022-594
Performance enhancement of a 100 watts class Tesla turbine

Arindam Mandal1 , Rajosik Adak1 , and Sandeep Saha1

1
Department of Aerospace Engineering, IIT Kharagpur, Kharagpur 721302, India

ABSTRACT Tesla turbines are an attractive but less This turbine was conceptualized by Nicola Tesla [39]
explored area in low-power applications. This article in the year 1907. Initially, the device was unable to attract
presents an experimental investigation of a centimeter market attention because of the no potential requirement of
arXiv:2301.01483v1 [[Link]-dyn] 4 Jan 2023

scale Tesla turbine in bi and uni-directional outlet low-power harvesting technologies. After that, till the twen-
configuration with compressed air at 6 bar. The turbine’s tieth century, a considerable amount of thrust was given to
performance is enhanced by ≈ 38% for a uni-directional the possible design modifications [27], [28], [12], improved
outlet configuration. Furthermore, we also investigate the seal designs [2] and loss analysis [6], [30] of the turbine
electrical power of the turbine in a bi-directional outlet and its ancillary components for enhancing performance. In
configuration by coupling the turbine with a generator. addition, a few simplified theoretical models were developed
Despite achieving higher performance in uni-directional [1], [21] to get an insight into the influence of the parameters
outlet configuration, we observe substantial losses at the associated to the turbine.
inlet we use for the experiment. To illustrate and improve
the losses, we numerically investigate the turbine inlet
at a total pressure and temperature difference of 2 bar
and 50◦ C, respectively. Subsequently, we design two
more nozzles and compare their performance with the
nozzle we used in our experiment. Our findings suggest
that nozzle 3 performs the best in delivering the highest
Mach no and uniformity across the slits. This observation
would help optimize the nozzle suitable for the Tesla turbine.

Keywords: Tesla turbine, Friction turbine, Low power

application, Slit nozzle,

I. INTRODUCTION
Small scale low microturbines are crucial since there is a
growing need for them in numerous applications. Examples Figure 1: Schematic of the exploded view of the turbine.
of notable applications include waste heat recovery, pico- a: Casing, b: Rotors, c: Inlet configurations (i,ii and iii),
hydro power, organic Rankine cycle technologies, biomass, d: Shaft, e. Bearing. (f) shows the fluid flow between two
waste heat recovery, micro GT, and many more. The increas- consecutive corotating discs
ing need for energy harvesting at these scales poses a variety
of challenges to the performance and manufacturability of More recently, there has been a surge in interest in
the turbine due to their compact sizes, high rotational speeds, exploring this technology due to its several advantages
and increased viscous losses. An alternative expansion de- over conventional energy harvesting technologies. Due
vice addressing these challenges can be highly beneficial to its simple design and flow mechanics, the turbine can
regarding its techno-economic feasibility. The tesla turbine handle particle-laden fluids and two-phase expansion with
is one of its kind, which has a uniquely simple design and minimal damage. Numerical simulations of the gas flow
a unique mechanism of momentum transfer. The turbine and mass transfer between two coaxially co-rotating discs
rotor consists of several co-axial discs closely packed with were performed by Sandilya et al. [32] where they found a
each other. Each of the rotor discs has outlet ports near the satisfactory match between the numerical and experimental
center. A casing covering the rotors helps guide the flow value of mass transfer coefficient. Using numerical
from the inlet. The turbine shaft is attached to the rotor- simulations, Ladino [18] presented the load coefficient
casing configuration with the help of two bearings. Figure 1 curve, efficiency, and degree of reaction variation after
shows the different components of the turbine. After being maintaining a constant rotational speed. Upon developing
injected through the inlet system of the turbine, the fluid an analytical model, Deam et al. [7] scaled down the turbine
flows spirally inward and exits through the ports located in millimeter size to achieve better efficiency. Lemma et al.
near the shaft in the axial direction. The momentum transfer [22] investigated the Parasitic, viscous, and other dissipative
from the fluid to the rotor takes place using the adhesion losses in the bearings and end walls to mitigate their effect
and viscosity of the fluid. This mechanism makes the turbine on decaying the performance.
attractive at small scales where the dominance of the viscous
force becomes significant. An explicit description of a flexible test rig was de-

1
veloped by Hoya et al. [14] where they calculated the
low torque at high RPM using the angular acceleration
method. A comparative study in the efficiency offered by
Tesla and small bladed microturbine in a micro power plant
was addressed by Lampart et al. [20], [19] to establish
the competitiveness of a Tesla turbine. To understand the
transport phenomena inside the turbine rotor, a number
of numerical and analytical models solving the Navier-
Stokes equations using different approaches are present in
the literature repository [15], [9], [36], [10], [31], [35], [34],
[5]. Aside from that, flow diagnostics using particle tracking (a) (b)
velocimetry by Schosser et al. [33] provides an insight into
the component-wise velocity profiles inside the rotor gaps at Figure 2: (a) Fabricated turbine and (b) the inlet system
different radial locations. In recent years, A wide range of
application based studies of Tesla turbine related to Micro-
air vehicles [23], Organic Rankine cycle [37], [38], [8], ter to validate the turbine’s RPM. We conduct our experiment
[25], [24], Combined heat plant [3], Pico-hydro applications at 6 bar of inlet pressure for uni-and bi-directional outlet
[13], [4], [16], [17], Ammonia synthesys [11] have been configuration. The detail of the experimental setup can be
investigated or under investigation [26], [29]. seen in the figure 3.
The recent resurgence in interest indicates the impor-
tance of investigating the turbine further to mitigate the
losses due to the nonuniformity and disturbances associated
with the inflow to rotor and rotor to outflow interaction.
This article presents the experimental investigation of a
centimeter-scaled Tesla turbine in uni and bi-directional
outlet configuration with compressed air. We compare the
performance in terms of Mechanical power output for the
two configurations. Furthermore, we investigate the losses in
the inlet due to the sharp divergent and the slit configuration (a) (b)
at the inlet rotor junction. Finally, we compare the nozzles’
performance by looking at the peak discharge Mach number Figure 3: Details of the experimental set up. 1- Compres-
and channel-wise disparity in flow injection to the rotor. Our sor discharge, 2- Tachometer, 3- Turbine, 4- Arduino
numerical results can be beneficial in coming up with better based Tachometer, 5- Pressure gauge, 6- PC for data
inlet configurations for extracting maximum energy from the acquisition, 7- Camera, 8- Generator, 9- Multimeters,
fluid, which leaves a further scope for further investigation. 10- Rheostat, 11- Electrical circuit

II. METHODOLOGY AND EXPERIMENT

A lab scale turbine prototype (in fig. 2a) is fabricated III. RESULTS AND DISCUSSION
for experimentation in the Aerospace Engineering depart- A. Performance characteristics
ment, IIT Kharagpur. The turbine consists of 10 consecutive We tabulate the RPM with time with the help of a serial
corotating discs having an outer diameter of 10 cm and a monitor and calculate the angular acceleration as a function
thickness of 2 mm. The gap between the consecutive rotating of RPM. Once the turbine reaches its stable RPM, we shut
discs is 2 mm, and the distance between two extreme discs to off the compressed air sypply and continue to perform data
the adjacent casing wall is 1 mm. The turbine has four outlet acquisition until the turbine becomes stationary. We continue
holes near the shaft, having a center distance of 2 cm from the similar process for a uni-directional outlet case. We
the shaft center. The shaft and the exhaust ports’ diameters observe from figure 4 that the turbine with a uni-directional
are 1.5 cm and 1 cm, respectively. The distance between the outlet accelerates faster than the turbine with a bi-directional
shroud and the disc’s edge is 1 mm. The inlet system of the outlet configuration. In addition, for a supply pressure of
turbine is designed to connect the compressor outlet port of 6 bar, we achieve a maximum RPM of 13019 and 11124
6mm dia to the turbine having an inlet of 4 cm thickness. for uni and bi-directional outlet configurations, respectively.
The area of the inlet is 0.4 cm2 , which is segregated using Figure 4 shows the power variation due to accelerating
slit configurations to guide air to each gap with minimum and braking torque. There is an increment of ≈ 38% in
interaction with the peripheral walls of the rotor. power due to accelerating torque for a uni-directional outlet
The turbine inlet is connected to the compressor by configuration.
Polyeurathane pneumatic pipes with an installed pres- We integrate the turbine with a 100 W class Fedus
sure gauge. The RPM of the turbine is measured using RS-775 DC electric motor to measure the electrical power
an Arduino-based tachometer. The angular acceleration- output. However, the motor is used as a generator to measure
deceleration approach computes the net accelerating and de- the output voltage and current. The electrical circuit is
celerating [Link] experiment also uses a digital tachome- attached to a rheostat, and the experiment is performed with

2
 15000 120

Power (Watts)
10000 80
RPM

5000 40

0 0
0 15 30 45 0 4500 9000 13500
t (sec) RPM

(a) (b)
Figure 4: Distribution of (a) RPM with t; (b) Power
due to accelerating torque with RPM for bi-directional Figure 6: Schematic diagram of the nozzles. The dotted,
(dashed line) and uni-directional (solid line) outlet. Dot- dashed and solid lines represent nozzle 1, nozzle 2 and
ted line represents the power due to braking torque. nozzle 3, respectively.

the 5-ohm load resistance. The figure 5 illustrates the motor’s variation ≤ 2%, We conduct the subsequent numerical
voltage and power variation at various RPM. The turbine- simulations using grids with ≈ 1M elements. The inlet
generator can produce a maximum of 78 watts of electrical
power at 7800 RPM for bi-directional outlet configuration.
Table 1: Grid sensitivity of the three inlet configurations
80 80 Grid type Elements Outlet area averaged Mach no
Nozzle 1 Nozzle 2 Nozzle 3
Coarse ≈ 0.5 M .5093 .4993 .5296
Voltage (Volts)
PowerE (Watts)

Medium ≈1M .5089 .4979 .5289

Fine ≈2M .5080 .4971 .5284
40 40

design presented in this article is based on a two-pronged

objective. (a) To Maximize the Mach number peak (b) To
minimize the disparity in the Mach number peaks through
0 0 every slit. To understand the loss mechanism and the flow
0 4000 8000
behavior, we conduct a series of numerical simulations
RPM
Where Nozzle 1 is the replica of the inlet nozzle considered
Figure 5: Experimental results of voltage (dashed line) for the experiment. Due to the abrupt divergent section and
and electrical power (solid line) at different RPM the presence of a large recirculation zone enclosed by a
strong shear layer, we detect significant losses in Mach no.
The dividing streamlines seen in figure 7 (a) are considered
B. Inlet design when designing the following two inlet systems. The second
Designing the inlet is the most crucial part of the turbine nozzle we design is to eliminate the effect of recirculation.
where the maximum loss occurs. Notably, the compressor We gradually increase the inlet nozzle area until we reach
and back pressure at the inlet rotor connection point regu- the section where the flow bends in the previous observation.
lates the flow across the nozzle. In the present article, we Despite the improvement in the average peak Mach number
only consider the inlet section for analysis. The details of observed from figure 7 (b), the disparity in the discharge
the nozzle dimensions are in figure 6. Mach number through the slits increased. We consider
1) Numerical methodology and grid Grid sensitivity: designing the third nozzle as a converging-diverging type
The Inlet section of the nozzle is a pressure inlet boundary nozzle where we place the throat section at x/L ≈ 0.6 to
where the total pressure is at 6 bar. We consider the outlet provide sufficient scope for flow to bend. Figure 8 represents
of the nozzle is at 4 bar. The temperature at the compressor the peak discharge Mach number through the nozzles and
discharge and the inlet-rotor junction are 450K and 400K, the associated % disparity from the mean peak Mach no
respectively. The surface of the nozzle is a no-slip type through each slit. The comparison shows that the third nozzle
boundary. Considering these boundary conditions, we solve offers maximum peak Mach number along with minimum
governing compressible Reynolds-averaged Navier-Stokes % deviation from the mean peak Mach no. It is necessary
equations using the K − ω SST turbulence model in the to note that the differential pressure between the upstream
commercial CFD package Fluent 2021 R2. The numerical and downstream sections, the fluid, and the fluid’s thermo-
domain is discretized using the body-fitted tetrahedral grids, physical characteristics significantly influence the nozzle
maintaining a wall y + ≈ O(1). The table 1 below presents design. The nozzle’s optimal size and shape might differ
the grid sensitivity study. As the desired output shows a depending on these conditions.

3
 0.8

M
0.4

(a) 0
-0.02 -0.01 0 0.01 0.02
(a)
20
20 5
10
10

%M'
0 0

%M/

'
%M
0 -10 -5
-10 -20
-10
-20 -30
-15

(b) (c) (d)

(b) Figure 8: (a) Comparison of peak Mach number across

the slits. The dotted, dashed and solid lines represent
First, second and third nozzle, respectively. (b, c, d) The
% disparity in peak Mach no at discharge for three
nozzles.

RPM.
4) While assessing the inlet, we observe that nozzle 2
offers better peak Mach number than nozzle 1, but
due to the losses accounted by the sharp bend, the
(c) disparity in % deviation of the peak mach numbers
from mean peak Mach no is nearly 30%.
5) Nozzle 3 performs the best among all three nozzles.
It offers a maximum peak Mach no with minimum
Figure 7: Distribution of the Mach number for (a) nozzle % deviation from mean peak Mach no is reduced
1 (b) nozzle 2 (c) nozzle 3. to 12%.
The turbine’s efficiency depends on several factors, e.g.,
IV. CONCLUSIONS RPM, disc gap, rotor radius, rotor to casing clearance,
The article presents a preliminary experimental investiga- outlet configurations, fluid properties, and many more.
tion of a centimeter scaled Tesla turbine using compressed Designing an efficient Tesla turbine could bring down the
air as a working medium and suggests an improved inlet cost of a turbine substantially as compared to the other
design that could deliver higher achievable power compared existing technologies because of its simplistic design. The
to the inlet nozzle considered for this experiment. The key above findings presented in this article could be helpful in
findings of the article are as follows; the design improvement, thus leaving a plausible scope for
further investigation.
1) Experimental investigation of the turbine conducted
for uni and bi-directional outlet configuration at 6 NOMENCLATURE
bar inlet pressure.
2) uni-directional outlet configuration offered a peak t time [sec]
power output of ≈ 110 watts at ≈ 7000 RPM. x/L Non-dimensional-axial location [–]
Whereas, bi-directional outlet configuration a peak k Turbulent kinetic energy [J/kg]
output of 80 watts at 6500 RPM. ω Specific dissipation rate [1/sec]
3) The turbine-generator configuration produced a M Mach no –
peak electrical power output of 78 watts at 7800 M′ % Deviation from peak mean Mach no –

4
R EFERENCES low-boiling medium, Polish Maritime Research 16 (2009), no. Spe-
[1] James Hal Armstrong, An investigation of the performance of a cial, 28–33.
modified tesla turbine, Ph.D. thesis, Georgia Institute of Technology, [21] Michael John Lawn, An investigation of multiple-disk turbine perfor-
1952. mance parameters., Master’s thesis, Arizona State University, 1972.
[2] John Spencer Caldwell, The efficiency of a viscous flow compressor, [22] Engida Lemma, RT Deam, D Toncich, and R Collins, Characterisa-
Ph.D. thesis, Georgia Institute of Technology, 1973. tion of a small viscous flow turbine, Experimental Thermal and Fluid
Science 33 (2008), no. 1, 96–105.
[3] Van P Carey, Assessment of tesla turbine performance for small scale
rankine combined heat and power systems, Journal of Engineering [23] Arindam Mandal and Sandeep Saha, Performance analysis of a
for Gas Turbines and Power 132 (2010), no. 12, 122301. centimeter scale tesla turbine for micro-air vehicles, Electronics,
Communication and Aerospace Technology (ICECA), 2017 Inter-
[4] Tan Wee Choon, AA Rahman, Foo Shy Jer, and Lim Eng Aik, national conference of, vol. 1, IEEE, 2017, pp. 62–67.
Optimization of tesla turbine using computational fluid dynamics
approach, Industrial Electronics and Applications (ISIEA), 2011 [24] G Manfrida, L Pacini, and L Talluri, An upgraded tesla turbine
IEEE Symposium on, IEEE, 2011, pp. 477–480. concept for orc applications, Energy 158 (2018), 33–40.
[5] L Ciappi, D Fiaschi, PH Niknam, and L Talluri, Computational [25] Giampaolo Manfrida, Leonardo Pacini, and Lorenzo Talluri, A re-
investigation of the flow inside a tesla turbine rotor, Energy 173 vised tesla turbine concept for orc applications, Energy Procedia
(2019), 207–217. 129 (2017), 1055–1062.
[26] Pouriya H Niknam, Lorenzo Talluri, Lorenzo Ciappi, and Daniele
[6] ME Crawford and W Rice, Calculated design data for the multiple-
Fiaschi, Numerical assessment of a two-phase tesla turbine: Para-
disk pump using incompressible fluid, ASME J. Eng. Power 96
metric analysis, Applied Thermal Engineering 197 (2021), 117364.
(1974), no. 3, 274–282.
[27] Robert A Oklejas and Eli Oklejas Jr, Gas regeneration tesla-type
[7] RT Deam, Engida Lemma, B Mace, and R Collins, On scaling down
turbine, (1975), US Patent 3,899,875.
turbines to millimeter size, Journal of Engineering for Gas Turbines
and Power 130 (2008), no. 5, 052301. [28] , Tesla-type turbine with alternating spaces on the rotor of
cooling air and combustion gases, December 28 1976, US Patent
[8] Olivier Dumont, Lorenzo Talluri, D Fiaschi, G Manfrida, and Vincent 3,999,377.
Lemort, Comparison of a scroll, a screw, a roots, a piston expander
and a tesla turbine for small-scale organic rankine cycle, ORC [29] Leonardo Pacini, Lorenzo Ciappi, Lorenzo Talluri, Daniele Fiaschi,
conference 2019, 2019. Giampaolo Manfrida, and Jacek Smolka, Computational investigation
of partial admission effects on the flow field of a tesla turbine for
[9] Abhijit Guha and Sayantan Sengupta, The fluid dynamics of the orc applications, Energy 212 (2020), 118687.
rotating flow in a tesla disc turbine, European Journal of Mechanics-
B/Fluids 37 (2013), 112–123. [30] Warren Rice, Tesla turbomachinery, MECHANICAL
ENGINEERING-NEW YORK AND BASEL-MARCEL DEKKER-
[10] , Similitude and scaling laws for the rotating flow between (2003), 861–874.
concentric discs, Proceedings of the Institution of Mechanical En-
gineers, Part A: Journal of Power and Energy 228 (2014), no. 4, [31] Vincent D Romanin, Theory and performance of tesla turbines,
429–439. (2012).
[32] P Sandilya, G Biswas, DP Rao, and A Sharma, Numerical simulation
[11] Kai Han, Jianjun Luo, Jian Chen, Baodong Chen, Liang Xu, Yawei
of the gas flow and mass transfer between two coaxially rotating
Feng, Wei Tang, and Zhong Lin Wang, Self-powered ammonia
disks, Numerical Heat Transfer: Part A: Applications 39 (2001),
synthesis under ambient conditions via n2 discharge driven by tesla
no. 3, 285–305.
turbine triboelectric nanogenerators, Microsystems & nanoengineer-
ing 7 (2021), no. 1, 1–8. [33] Constantin Schosser, Thomas Fuchs, Rainer Hain, Stefan Lecheler,
and C Kahler, Three–dimensional particle tracking velocimetry in a
[12] Leo Entrican Harold Jr, Tesla turbine, December 5 2002, US Patent
tesla turbine rotor using a non–intrusive calibration method, 18th
20,020,182,054.
International Symposium on the Application of Laser and Imaging
[13] Bryan P Ho-Yan, Tesla turbine for pico hydro applications, Guelph Techniques to Fluid Mechanics, 2016.
Engineering Journal 4 (2011), 1–8. [34] Constantin Schosser, Stefan Lecheler, and Michael Pfitzner, Ana-
[14] GP Hoya and A Guha, The design of a test rig and study of the lytical and numerical solutions of the rotor flow in tesla turbines,
performance and efficiency of a tesla disc turbine, Proceedings of Periodica Polytechnica Mechanical Engineering 61 (2017), no. 1,
the Institution of Mechanical Engineers, Part A: Journal of Power 12–22.
and Energy 223 (2009), no. 4, 451–465. [35] Constantin Schosser and Michael Pfitzner, A numerical study of the
[15] DP Kavenuke, E Massawe, and OD Makinde, Modeling laminar flow three-dimensional incompressible rotor airflow within a tesla turbine,
between a fixed impermeable disk and a porous rotating disk, African Conference of Modelling Fluid Flow CMFF, 2015, pp. 1–4.
Journal of Mathematics and Computer Science Research 2 (2009), [36] Sayantan Sengupta and Abhijit Guha, A theory of tesla disc turbines,
no. 7, 157–162. Proceedings of the Institution of Mechanical Engineers, Part A:
[16] Vedavalli G Krishnan, Zohora Iqbal, and Michel M Maharbiz, A Journal of Power and Energy 226 (2012), no. 5, 650–663.
micro tesla turbine for power generation from low pressure heads [37] L Talluri, D Fiaschi, G Neri, and L Ciappi, Design and optimization
and evaporation driven flows, Solid-State Sensors, Actuators and Mi- of a tesla turbine for orc applications, Applied energy 226 (2018),
crosystems Conference (TRANSDUCERS), 2011 16th International, 300–319.
IEEE, 2011, pp. 1851–1854.
[38] Lorenzo Talluri, Olivier Dumont, Giampaolo Manfrida, Vincent
[17] Vedavalli Gomatam Krishnan, Design and fabrication of cm-scale Lemort, and Daniele Fiaschi, Experimental investigation of an or-
tesla turbines, (2015). ganic rankine cycle tesla turbine working with r1233zd (e), Applied
[18] Andrés Felipe Rey Ladino, Numerical simulation of the flow field in a Thermal Engineering 174 (2020), 115293.
friction-type turbine (tesla turbine), Ph.D. thesis, National University [39] Nikola Tesla, Turbine., May 6 1913, US Patent 1,061,206.
of Colombia, 2004.
[19] Piotr Lampart and Łukasz J˛edrzejewski, Investigations of aerody-
namics of tesla bladeless microturbines, Journal of Theoretical and
Applied Mechanics 49 (2011), 477–499.
[20] Piotr Lampart, Krzysztof Kosowski, Marian Piwowarski, and Łukasz
J˛edrzejewski, Design analysis of tesla micro-turbine operating on a