Energy Conversion and Management 156 (2018) 224–234

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Compact design of planar stepped micro combustor for portable T

thermoelectric power generation
⁎
B. Aravinda, , Gannena K.S. Raghuramb, V. Ratna Kishoreb, Sudarshan Kumara
a
Department of Aerospace Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India
b
Department of Mechanical Engineering, Amrita University, Coimbatore, India

A R T I C L E I N F O A B S T R A C T

Keywords: An efficient prototype of a micro power generator with integrated micro combustor has been developed in the
Thermoelectric generator present study. The proposed design of the integrated micro-combustor provides high surface temperature with
Micro combustors superior temperature uniformity and enhanced flame stability limits, a prerequisite for a thermoelectric power
Flame stability generation system. This novel micro combustor configuration consists of three backward facing steps with a
Conversion efficiency
recirculation hole fabricated in a rectangular heating medium of aluminium material. Parametric studies are
carried out by varying the mixture inlet velocity, equivalence ratio and coolant flow rate to obtain the optimized
operating conditions for maximum power generation. Two thermoelectric modules are mounted on the system
operating with liquefied petroleum gas as fuel. A maximum conversion efficiency of 3.3% is obtained at ϕ =
0.95 with a mixture velocity of 7.5 m/s and a load resistance of 4 Ω across the thermoelectric generator. The
effect of porous media is investigated to enhance the flame stability limits in the micro combustor. Porous media
significantly enhances the upper flame stability limits and maximum conversion efficiencies (3.8%, 4.03%, and
3.73% at ϕ = 1, 0.9 and 0.8 at 10 m/s). A significantly higher power density (∼50% higher than existing
systems) of 0.12 mW/mm3 of system volume is achieved. A compact design of the prototype system with high
conversion efficiency shows the possibility of its application for various systems requiring portable power for
remote, stand-alone, military and aerospace applications.

1. Introduction Catalytic combustion [11–13], Porous media combustion [14–17] and

Combustion driven micro devices are receiving increased attention
and interest from the research community due to their intense demand
for the development of small-scale portable power sources. Combustion
based power devices can be considered as a competitive alternative to
conventional electrochemical batteries because of their high energy
density, high power density, compact size, small recharging time and
long working time, [1–3]. These devices have wide applications in
micro satellite thrusters, MEMS, chemical sensing, and micro air ve-
hicles (MAV) [4,5]. High precision fabrication techniques such as
stereo-lithography and rapid prototyping helped in the faster develop-
ment of micro and nano scale devices. Thermal and radical quenching
are two key issues associated with flame stability in micro combustors
due to high surface area to volume ratio [6]. Proper thermochemical
management techniques have been adopted to minimise the quenching
problems and significantly improve the flame stability limits [7,8].
Many researchers have come up with innovative techniques to solve the
flame stability issues through excess enthalpy combustion [9,10].
use of stepped combustors [18–20] are some of the useful strategies
proposed and successfully implemented to circumvent the flame
quenching issues in small scale combustion devices. Hydrocarbon fuel
based systems with a conversion efficiency of ∼5%, would result in 6
times higher power density than a conventional high efficiency elec-
trochemical battery, because of its high energy density [10] as shown in
Fig. 1. Fig. 1 shows that combustion based systems have significantly
(50–100 times) higher power densities as compared to various ad-
vanced electrochemical battery concepts. Such power generation sys-
tems with higher conversion efficiencies are yet to be explored in detail.
One of the earlier breakthrough in the field of micro combustion
and micro power generation was reported by MIT gas turbine labora-
tory, USA, which provided a clear insight into the micro power gen-
eration methods [21]. Moving components in these types of conversion
devices are frequently subjected to maintenance problems, resulting in
very low conversion efficiencies. Therefore, many researchers focussed
on direct conversion devices such as thermophotovoltaic (TPV) based
devices [22–28], thermo electric generators (TEG) [29–34] and fuel

⁎
Corresponding author.
E-mail address: aravindbpillai69@gmail.com (B. Aravind).

https://doi.org/10.1016/j.enconman.2017.11.021
Received 13 September 2017; Received in revised form 17 October 2017; Accepted 8 November 2017
Available online 11 November 2017
0196-8904/ © 2017 Elsevier Ltd. All rights reserved.
B. Aravind et al. Energy Conversion and Management 156 (2018) 224–234

Nomenclature Ω unit of electric resistance

SJ uncertainty
Am surface area of TEG (m2) TC cold side temperature (K)
As bare surface area of the combustor (m2) TH hot side temperature (K)
hc wall heat transfer coefficient (W/m2 K) T∞ ambient temperature (K)
I electric current (A) Tf combustor wall temperature (K)
Km thermal conductivity of TEG (W/m U uncertainty (%)
Lm width of the TEG (m) Uin mixture velocity (m/s)
m mass flow rate (kg/s) VL load voltage (V)
N number of TEGs Vo open circuit voltage (V)
Ri internal resistance (Ω) Xi measured data
RL load resistance (Ω) X mean value of measured data
S seebeck coefficient (V/K)
SD standard deviation Abbreviations

Greek symbols CV calorific value (kJ /kg)

TEG thermoelectric generator
ϕ equivalence ratio

Fig. 1. Comparison of the energy density of hydro-

carbon fuels with conventional electrochemical bat-
teries.

cells [35,36] integrated to the micro combustors.
Yang et al. [23] proposed a thermophotovoltaic (TPV) based micro
power generation system by integrating it with a backward facing step
combustor. They observed that channel diameter and step size are
crucial factors affecting the performance of the system. The increase in
step height enhanced the heat recirculation near the combustor wall,
thereby facilitating complete and stable combustion. Jiang et al. [26]
fabricated a planar combustor suitable for microTPV, which facilitated
a higher wall temperature (above 1300 K) with superior uniformity in
the temperature profile and high combustion efficiency. Lee at al. [28]
developed a micro combustor with a heat recuperator to extract electric
power using TPV. They used a blend of ammonia and hydrogen as fuel
which generated a secondary flame at the micro emitter due to its low
temperature. They reported an overall conversion efficiency of 2.1%.
Lei et al. [27] recently developed a model to estimate the power gen-
eration using micro TPV integrated to an annular micro combustor.
They achieved a maximum conversion efficiency of 3.15% for a 1 mm
annular micro channel. Further various numerical efforts on different
microcombustor configurations have been made to enhance the uni-
formity and surface temperature suitable for thermophotovoltaic
application [37–39]. One of the earliest development in TEM by
Schaevitz et al. [29] reported MEMS based thermo electric power
generation system using catalytic combustion. This generator was
stable up to 773 K and achieved a conversion efficiency of 2% with an
output voltage of 7 V. Similar work has been reported by Yoshida et al.
[30], with a conversion efficiency of 2.8% and a power output of
184 mW. Jiang et al. [31] fabricated and tested a micro plate flame
combustor for power generation with DME as fuel. Shimokuri et al. [32]
prototyped a meso scale vortex combustor for thermo electric power
generation. Vortex flow helped enhance the heat transfer from burnt
gas to the combustion chamber. This led to a significant improvement
in the conversion efficiency to 0.7%. Yadav et al. [34] developed a
micro power generation system using a rearward facing stepped micro
combustor with heat recirculation cup. A conversion efficiency of
4.56% is reported with four modules at 6.5 m/s mixture velocity. Qiu
et al. [40] developed a power generation system by cascading the TEM
(Thermoelectric modules) and TPV (Thermophotovoltaic modules) to
increase the system efficiency. Their investigations successfully proved
that higher efficiencies from a TE-TPV integrated system can be
achieved over a TPV or TE power generator alone. Walther et al. [41]

225
B. Aravind et al. Energy Conversion and Management 156 (2018) 224–234

reviewed the different types of fuel cells and their impact on micro scale 2. Conceptual design of a micro power generator
power generation methods in a detailed manner. Ahn et al. [35] de-
veloped a power generation device based on single chamber solid oxide Fig. 2a shows the basic working principle of a combustion driven
fuel cell in conjunction with a Swiss-roll combustor. thermoelectric power generator for various portable electric power
Aravind et al. [45] and Singh et al. [46,47] present numerical in- generation platforms. This system essentially consists of three parts: a
vestigations to understand the effect of wall heat transfer conditions on heat source or micro combustor, a heat sink and a thermoelectric
flame shape transition in microtubes. These studies helped understand generator (TEG). The heat source and heat sink creates a temperature
the flame stabilization behaviour for conditions similar to those of ex- gradient across the TEGs, and provides heat energy to the system. The
periments in present work. The role of heat recirculation through solid thermoelectric generator converts a part of the thermal energy from the
walls helped understand flame anchoring and its stabilization in micro micro combustor to electrical energy through Seeback effect. Fig. 2b
combustors. This information was used for the development of in- shows a typical energy flow diagram for the system. It can be seen from
tegrated combustor in the present work. the energy flow diagram (Fig. 2b) that a small amount of thermal en-
Various studies aimed at micropower generation using thermo- ergy is converted into electrical energy due to various heat losses and
electric and thermophotovoltaic based concepts have been reported in low figure of merit of the existing thermoelectric module technology.
the literature at discussed earlier. Various aspects of these studies have Therefore, the present work focuses on minimising the heat losses
been summarized in Table 1. The aim of present work is to develop a and enhancing the overall conversion efficiency of the combustion
prototype of an integrated micro combustor for power generation with based micro thermoelectric power generation systems.
high conversion efficiency, better flame stability limits and improved
power output for various remote, stand-alone millitary and aerospace
3. Experimental setup
applications. The present study reports the development of an in-
tegrated micro combustor based system for obtaining high-power
3.1. Details of micro combustor
density and higher conversion efficiencies for practical applications.
In the present study, a compact three step planar micro combustor
Fig. 3a shows the dimensional details of the micro combustor em-
has been developed for portable power generation [18,19,33,48].
ployed in the present work. A cylindrical three backward facing step
Thermal characteristics and flame stability of the proposed micro
micro combustor with detailed dimensions referred in previous works
combustor are analysed first and thermoelectric generators (TEG) were
[18,19,48] is fabricated inside a rectangular heating medium of
integrated to the combustor along with a multipass cooling jacket to
0.032 × 0.032 × 0.010 m3 overall size with aluminium as a fabrication
improve the performance of the micropower generator. A parametric
material. The dimensions of the heating medium are chosen in such a
study was carried out by varying the mixture velocity, mixture
way that they match exactly with the dimensions of the TEGs. The steps
equivalence ratio and flow rate of coolant to understand the optimum
provided in the micro combustor create a sudden expansion to the flow
operating conditions for power generation. The effect of porous media
and induces local recirculation near the steps. This increases the re-
on the flame stability and power characteristics were also investigated.
sidence time and enhances the flame stability limits [49]. In addition to
The micro-combustor developed in this study has notable ad-
the backward facing steps, two 0.004 m diameter holes are provided as
vantages when compared to the existing similar micro power generators
shown in Fig. 3b to enhance the heat recirculation from hot combustion
reported in the literature as summarized in Table 1. Firstly, the pro-
products (exhaust gases) to solid walls of the combustor. Detailed in-
posed system size is compatible with a conventional power source such
vestigations have been carried out to understand the effect of heat re-
as electrochemical battery. Secondly, it is capable of generating a power
circulation on flame stability limits as discussed in the following sec-
of ∼4 W with comparatively higher conversion efficiency of 4.03%
tions. The mixture is supplied through a 0.002 m mixture delivery tube.
which is one of the highest in the existing devices with a power density
A premixed fuel-air mixture is ignited in the combustion zone as shown
of 0.12 mW/mm3. The system compactness and the output power with
in Fig. 3c. A comparison of the present combustor with a real scale is
high conversion efficiency shows the possibility of its application in
shown for comparison purpose in Fig. 3c.
various portable scale power generators for remote, stand-alone, mili-
tary and aerospace applications. The power density (mW/mm3) of this
novel concept based micropower generator is ∼50% higher than the 3.2. Integration of micro combustor and thermoelectric generator
existing similar stand-alone micropower generation systems.
Fig. 4a shows the complete model of the combustor and thermo-
electric generator assembly employed in the present study. Two ther-
moelectric generators (TEG) are mounted on either sides of the

Table 1
Comparison of some of the typical stand-alone TEG or TPV power generation systems.

Method/Fuel Combustor Vsystem (×10−6) (m3) Power density (×106) (W/m3) Ptotal (W) ηconversion (%)

Micro TPV [28] Micro emitter 270 0.02 5.2 2.1

(NH3-H2)
Micro TEG [31] Plate type micro burner 28.4 0.07 2 1.25
(DME/Air)
Micro TEG [33] Stepped combustor with recirculation cup 108 0.02 2.56 4.56
(C3H8/Air)
Micro TPV [42] Stepped micro combustor 35.3 0.03 1.02 0.72
(H2/Air)
Micro TEG [43] Catalytic combustor 144 0.07 9.86 2.35
(C3H8/Air)
Micro TEG [44] Vortex combustor 168 0.05 8.1 2.85
(C3H8/Air)
Micro TEG Stepped micro combustor 32.4 0.12 3.89 4.03
(LPG/Air)
(Present study)

226
B. Aravind et al.
Fig. 2. (a) Schematic diagram of the working principle (b) energy flow diagram for a typical micro combustor –thermoelectric power generation system.
combustor along with the cooling jackets. Thermoelectric generator of
size 0.03 × 0.03 × 0.003 m3 from Nippon India Ltd. is used. These
thermoelectric generators consist of 64 p-type and n-type doped semi-
conductor material junctions (Bismuth telluride) connected thermally
in parallel and electrically in series as shown in Fig. 4b. The electrical
interconnects of these thermoelectric generators are mounted between
two ceramic wafers. The thickness of the ceramic wafer is 0.0008 m.
This TEG has a capacity of delivering a maximum power of 4.8 W with a
hot and cold side temperature of 250 °C and 30 °C respectively.
3.3. Details of experimental setup
Fig. 5 shows the schematic arrangement of the experimental setup.
It has fuel and air feed systems, consisting of high pressure air, fuel
tank, precision electric mass flow controllers (MFC), pipeline and the
integrated micro combustor-thermoelectric generator system with
power measuring equipments. LPG (liquefied petroleum gas) and air
(b)
Fig. 3. (a) Schematic diagram (b) Combustor model (c) Real images of micro combustor with a scale.
227
Energy Conversion and Management 156 (2018) 224–234
are supplied from high pressure storage tanks through the pressure
reducing valves. Predetermined mass flow rates of fuel and air were
supplied using different flow range electronic mass flow controllers
(AALBORG-GFC) controlled through a computer to which the command
module is connected. The accuracy of the measured mass flow rates
is ± 1% of the full scale. MFC units of 0 – 0.5 LPM and 0–2 LPM were
used for LPG and air respectively with necessary calibration. In order to
measure the temperatures on the outer walls of the heating medium, K-
type thermocouples of 0.0005 mm bead diameter with an accuracy
of ± 2% were used. Thermoelectric generators of size
0.03 × 0.03 × 0.003 m3 from Nippon India Ltd. are used in the present
study. Two thermoelectric generators are sandwiched between the he-
ated combustor plate and the water cooled copper jackets on the top
and bottom sides of the combustor as shown in Fig. 6a and b. The as-
sembly was compressively loaded to maximize the heat transfer across
the interface. The voltage developed by each TEG is measured using
independent multimeters. The premixed LPG-air mixture is ignited at
(a)
Combustion zone
Recirculation path
(c)
B. Aravind et al.
Ignition port
(a)
the outlet by opening the ignition port as shown in Fig. 6a. Once the
flame is stabilized in combustor, ignition port is closed using a steel
plug (Fig. 6b) which ensures recirculation of the heat from hot com-
bustion products (exhaust gases) resulting in a self-sustained flame
within the combustor.
3.4. Experimental uncertainty and energy balance analysis
All the measuring equipments were calibrated and ensured to be
accurate before commencing the experiment. The relative uncertainty
of all the measuring equipments was calculated [50] and summarized in
Table 2. It can be seen that the relative uncertainty due to the equip-
ments is less than 1.5% for all the experiments reported in this work.
To ensure the repeatability of the experiment, uncertainties in
measuring temperature, flow rate, voltage and power were calculated
based on the following equations (Eqs. 15). For instance, experiments
are performed at least three times and the uncertainties in measuring
hot and cold side temperatures, voltage and current are noted at a fixed
mixture equivalence ratio of ϕ = 1 and mixture velocity of 6 m/s. The
data are summarized in Table 3. Results reveal that the experiments in
the current study are reliable since the uncertainty of measurements is
below 1.5%.
n that, although the heat loss from all part of the combustor increases
1
X =
n
∑ Xi
i=1 (1)
m 1/2
⎡ ⎤
Sj = ⎢∑ (Sj )i2 ⎥
⎣ i=1 ⎦ (2)
228
Energy Conversion and Management 156 (2018) 224–234
Fig. 4. (a) Schematic diagram of micro combustor-
thermoelectric power generation system and (b)
thermoelectric generator (TEG) employed in the
present study.
(b)
1/2
n
⎡ 1 ⎤
SD = ⎢
n−1
∑ (Xi −X )2⎥
⎣ i=1 ⎦ (3)
SD ⎞
U% = ⎛ × 100
⎝X ⎠ (4)
δPmax / Pmax = [(δVL/ VL )2 + (δI / I )2]1/2 = [0.912 + 0.982]1/2 = 1.34% (5)
A complete energy balance is obtained using the following equa-
tions (Eqs. 69) for three intermediate mixture velocities of 3 m/s, 5 m/s
and 7 m/s at ϕ = 1.
The total heat input to the system is the chemical energy possessed
by the fuel, calculated using the corresponding fuel flow rate multiplied
with its calorific value (Eq. (6)). The heat loss from TEG to heat sink is
determined using the Peltier effect, heat conduction and the Jules
heating principle (Eq. (7)). The combustor temperature is assumed to be
uniform to calculate the convective heat loss through the side surfaces
of the heating medium where TEGs are not mounted (Eq. (8)). It is clear
from Table 4 that, more than 60% of the heat is lost through exhaust
gases (Qexhaust) (Eq. (9)) which contributes as the major part of the
energy balance. Heat rejected by the thermoelectric generators to the
sink (Qsink) is around 20 – 30% and it increases with increase in the
power output from thermoelectric generators. It is interesting to note
with the increase in fuel input, the overall contribution of the heat loss
in the heat balance decreases (2.26 to 1.36%). This might be due to the
significant dominance of fuel input over the total heat loss. It is iden-
tified that the energy conversion efficiency is a strong function of
mixture flow rate, equivalence ratio and the load resistance in the
electrical circuit.
Fig. 5. Schematic diagram of experimental setup.
B. Aravind et al. Energy Conversion and Management 156 (2018) 224–234

Fig. 6. Direct photography of micro combustor-thermo-

electric power generation system (a)without and (b) with
recirculation.

Table 2 K A I 2Ri ⎤
A list of equipments and its uncertainty. Qsink = ⎡SIT c + m m (T H −T C ) + xN
⎢
⎣ Lm 2 ⎦ ⎥ (7)
Equipment uncertainty
Qloss = As hc (Tf −T∞) (8)
Equipment Operating/measuring Resolution Relative
range Uncertainty Qexhuast = Qinput −(Qsink + Qloss ) (9)
S (T H − T C )
K type thermocouple Up to 1260 °C 1 °C 0.5% where, I = Ri + RL
and N is the total number of thermoelectric gen-
Voltmeter 0– 60 V 0.01 V 1% erators connected.
Ammeter 0–2 A 0.01 A 1%
MFC- LPG Up to 0.5 LPM 1 mL PM 1.5%
MFC- Air Up to 2 LPM 4 mL PM 1.5% 4. Results and discussion

4.1. Thermal characteristics and flame stability limit of the micro combustor
Table 3
Details of experimental uncertainty at 6 m/s mixture velocity and ϕ = 1. Fig. 7 shows the direct photographs of the stabilized flame in the
combustor for various mixture conditions without heat recirculation
Experiment uncertainty
case. The flame is observed to stabilize at the second step for mixture
Variable Values measured at steady Mean Standard Uncertainty velocities up to 1.5 m/s for all the equivalence ratios studied and then
state condition deviation anchors in the third step of the combustor as shown in Fig. 7a and b
respectively. The flame continued to stabilize at the third step for
I II III
mixture velocities up to 4.6 m/s. For higher mixture velocities, the
Flow rate 1.575 1.583 1.557 1.57 0.01332 0.85% flame eventually moves out of the combustor and stabilises at the lip of
(L/ the combustor as a partially premixed flame for mixture velocities up to
min)
5.1 m/s as shown in Fig. 7c. Further increase in the mixture velocity
TH 130 126 128 128 2.00000 1.56%
TC 31 30 31 30.67 0.57735 1.88%
leads to a flame blow-off resulting in the upper flame stability limit. At
Vo 3.772 3.756 3.728 3.75 0.02227 0.59% very small flow rates, the flame is mostly stabilized in the first step and
VL 1.795 1.828 1.814 1.81 0.01656 0.91% flame quenching is governed by heat-loss from flame front to channel
I 0.446 0.438 0.445 0.44 0.00436 0.98% walls and subsequently to surroundings.
Fig. 8a shows the flame stability limits of the present micro-com-
bustor with the premixed LPG-air mixture. Effect of heat recirculation
Table 4
Summary of energy balance for various operating conditions: Qin-Total thermal input,
on the flame stability is investigated by closing the ignition port. This
Qexhaust -Heat loss through exhaust, Qloss -Heat loss through various uncovered sides of the diverts the direction of the hot combustion products (exhaust gases)
system, Qsink -Heat loss through sink, Pelectric-Electric power generated. The numbers in towards upstream side through the two holes provided in the heating
parentheses show respective percentage contribution of various terms. medium. The flame stability limits for the case of heat recirculation are
much wider than for the case of without heat recirculation. There is a
(W) Uin = 3 m/s Uin = 5 m/s Uin = 7 m/s
slight enhancement in the lower flammability limits with heat re-
Qinput 32.14 53.57 75.0 circulation. In both these cases, lower flammability limits show nearly a
(100%) (100%) (100%) straight-line trend and a minimum value is observed for nearly stoi-
Qsink 8.45 15.50 23.54
chiometric mixtures. Upper flammability limits for without recircula-
(26.28%) (28.94%) (31.39%)
Qloss 0.73 0.86 1.02 tion case follow an increasing trend with mixture equivalence ratio.
(2.26%) (1.61%) (1.36%) Upper flame stability limit is not observed within the bounds of the
Qexhaust 22.52 36.0 48.16 experimental limits for the heat recirculation case (explored up to
(70.07%) (67.20%) (64.22%) 10 m/s in present studies). Heat recirculation preheats the incoming
Pelectric 0.447 1.209 2.277
(1.39%) (2.26%) (3.04%)
mixture thereby increasing the flame stability limits. Due to increased
enthalpy of the mixture, the flame remains stabilized either in the third
step or near the exit of combustor at higher velocities due to higher
Qinput = m ·CV (6) overall heat release and preheating of the incoming fresh mixture. This
helps in preventing the flame blow-off even for very higher mixture
flowrate conditions.
Fig. 8b shows the variation of average surface temperature on the

229
B. Aravind et al.
outer side of the combustor at ϕ = 0.9 and 1.0 with mixture velocities.
The average temperature is calculated by considering equidistant points
along the central line on the outer surface of the combustor. It is ob-
served that the average surface temperature reaches around 400 °C for
mixture velocities of around 6 m/s. However, this temperature drops
significantly to safe operating limits of the TEGs (< 250 °C), when the
TEG is mounted on the combustor even at higher mixture velocities.
The high temperature and increased temperature uniformity of the
proposed micro combustor even at very low thermal input conditions
shows the possibility of using this combustor system for power gen-
eration using high efficiency micro thermoelectric modules. Ad-
ditionally, the present design is very compact and easy to fabricate
especially for mass production.
4.2. Effect of cooling water flow rate on power characteristics
The influence of flow rate of the coolant on the performance of the
proposed system is studied in detail. Fig. 9a shows the cross-sectional
view of the cooling jacket employed in the present work. It consists of 3
passes, which imparts sufficient cooling by ensuring increased re-
sidence time of the coolant and high flow velocity of the coolant inside
the cooling jacket [50]. The cooling jacket is 32 × 32 × 10 mm3 and
made up of copper due to its high thermal conductivity. The water
temperature is maintained at 30 °C throughout the experiment. The
variation in the hot side temperature and open-circuit voltage for in-
termediate mixture velocities is shown in Fig. 9b for three different
coolant flow rates of 0.5, 1.0 and 1.5 l/min with same heat input to the
system. It is expected that the output voltage would increase with the
coolant flow rate because increased coolant flow rate is expected to
reduce the temperature of the cold side of the TEG. However, the
temperature difference across hot and cold sides of modules remains
nearly unchanged for the range of flow rates explored in the present
work indicating that the lowest flow rate would be sufficient to obtain
maximum performance from the system. An average percentage de-
viation of less than 2% in the hot-side temperature and open-circuit
voltage is observed for all the three flow rates, which is negligible. Si-
milar observations were reported experimentally by researchers both
experimentally [50] and theoretically [51]. Therefore, all experiments
are carried out for fixed flow rate of 1.5 litres per min.
Fig. 8. (a) Flame stability diagram of the planar micro combustor and (b) variation of average surface temperature of the combustor with mixture velocity for no TEG conditions.
230
Energy Conversion and Management 156 (2018) 224–234
Fig. 7. Direct photography of stabilized
flame for (a) 1.5 m/s, (b) 3 m/s and (c)
5 m/s at different mixture conditions
without heat recirculation case.
4.3. Determination of maximum power output
The open circuit voltage of both the thermoelectric generators has
been calculated prior to the investigation of power characteristics of the
TEGs. Fig. 10 a shows the comparison of open circuit voltage developed
across the top and bottom TEGs at different mixture velocities at a
mixture equivalence ratio of 1.0. The cold side temperature is fixed at
30 °C using a heat sink and hot side temperature varies by increasing
the mixture velocity thereby thermal input to the combustor. It can be
seen that open-circuit voltage increases linearly with the mixture ve-
locity as the temperature difference across the TEG increases. The open-
circuit voltage developed by both the TEGs is found to be almost equal
at all mixture velocities investigated. This is due to the compact size of
the heating medium, which results in uniform and equal hot side
temperature on both top and the bottom surface of the heating medium.
Further, 0 – 20 Ω range rheostat is connected in series with the
thermoelectric generators to obtain the voltage-current characteristics
for the given conditions and apply the maximum power point tracking
(MPPT) for extracting maximum power from the TEGs. The load re-
sistance in the circuit is varied in the step of 0.2 Ω and the corre-
sponding voltage is measured using a voltmeter. Fig. 10b shows the
variation of power with mixture velocity for different load resistances.
It can be seen that, the power output peaks towards 4 Ω resistance and
decreases on both sides of the load resistance. A similar trend has been
observed for all mixture velocities considered in the present work as
shown in the figure. Maximum power is achieved when the internal
resistance of the TEG matches with the external load [34,52].
4.4. Effect of mixture equivalence ratio on power generation
Fig. 11a and b show the variation of power output and overall
conversion efficiency respectively with mixture velocity for equivalence
ratios of ϕ = 0.95 and 1. From the maximum power point tracking
(MPPT) method, a load resistance of 4 Ω is obtained and this resistance
is connected to the circuit to extract the maximum power from the
TEGs. It can be seen that, power obtained for both the equivalence
ratios is almost similar and follows a linear trend. A maximum power of
2.54 W and 2.58 W is achieved for equivalence ratios of ϕ = 0.95 and 1
respectively.
Despite the variation in fuel input for both the equivalence ratios,
B. Aravind et al. Energy Conversion and Management 156 (2018) 224–234

Fig. 9. (a) Flow patterns of the cooling jacket and (b) variation of hot side temperature and open circuit of the bottom TEG with mixture velocity at ϕ = 1.0.

Fig. 10. (a)Variation of open circuit voltage for the top and bottom TEGs with mixture velocity and (b) power with load resistance for different flow velocities at ϕ = 1.0.

Fig. 11. Variation of (a) maximum power and (b) conversion efficiency with mixture velocity at ϕ = 0.95 and 1.0.

enhanced power for ϕ = 0.95 may be perhaps due to the position of the
stabilized flame at a downstream location of the combustor as com-
pared to ϕ = 1 conditions. The stabilization of the flame at a down-
stream location perhaps results in a higher temperature due to in-
creased heat recirculation through the reverse flow configuration of the
planar micro combustor. Further, a conversion efficiency of 3.3% is
achieved at ϕ = 0.95 owing to the enhanced power output and lower
fuel input. These experimental investigations are limited to a mixture
velocity of 7.5 m/s even though the upper flame stability limits of the
current combustor are higher. However, as the TEGs and cooling jackets
are mounted on the combustor, they extract a significant amount of
heat from the combustor surface and reduces the surface temperature of
the heating medium. This results in reduced preheating of the fresh
incoming mixture. This leads to the quenching of the flame at mixture
velocities higher than 7.5 m/s in the present case.
4.5. Effect of porous media on power generation
As discussed in the previous section, the stabilized flame in the
micro combustor quenches for mixture velocities greater than 7.5 m/s.
To further enhance the flame stability limits, an efficient heat re-
circulation porous media (ceramic wool) of 2 mm thickness is in-
troduced in the third step of the combustor. Detailed experiments are
conducted for lean and stoichiometric mixtures to analyse its effect on
flame stability limits and power generation characteristics at higher
mixture velocities.

231
B. Aravind et al.
Fig. 12. Direct photography flame in a micro combustor-thermoelectric power generation
setup with porous media.
Fig. 13. Comparison of power input and power output of micro power generator for
different mixture flow rates at ϕ = 1.0.
The upper flammability limit was found to be enhanced sig-
nificantly with the use of porous media. The ceramic wool significantly
helps sustain the flame in the combustor through heat recirculation
[53,54]. The thin fibres of the ceramic wool helps increased heat re-
circulation to upstream through conduction and helps preheat the in-
coming fresh mixture to a higher temperature. During this period, it
becomes red hot and helps in significant heat recirculation through
conduction and radiation. A photograph if the stabilized flame with
Fig. 14. Variation of maximum power (a) and conversion efficiency (b) with mixture velocity at ϕ = 0.8, 0.9 and 1.0 by stuffing ceramic wool of 2 mm thickness at the third step.
232
Energy Conversion and Management 156 (2018) 224–234
porous media in the combustor is shown in Fig. 12. A comparison of
power provided and power obtained as a function of mixture flow rate
at stoichiometric condition is depicted in Fig. 13. The output power is
calculated up to mixture flow rate of 2.94 × 10−5 kg/s for without
porous media condition since the upper flame stability was limited
7.5 m/s. The increasing trend of power curve provides an insight to the
further increase of mixture flow rate for obtaining higher power output.
Thus, use of porous media helped to augment the flame stability limit
and maximum output power of 4.03 W is achieved for mixture flow rate
of 3.92 × 10−5 kg/s. Inclusion of porous media results in significant
improvement in flame stability limits, which helps control the power
output based on the demand from a given practical system.
Fig. 14a shows the variation of output power with mixture velocity
for different equivalence ratios. It is observed that almost similar power
is achieved for ϕ = 0.9 and 1.0 mixture conditions. A significant re-
duction in electric power output for ϕ = 0.8 conditions can be ob-
served. This might be due significantly reduced thermal input, which
results in reduced power output from the TEGs. Fig. 14b shows the
variation of conversion efficiency with the mixture velocity for different
equivalence ratios. Maximum conversion efficiencies of 3.8%, 4.03%,
and 3.73% are achieved for ϕ = 1, 0.9 and 0.8 respectively at a mixture
velocity of 10 m/s. This is much higher compared to the conversion
efficiencies reported earlier for two TEG configuration in the literature
[33,34].
The novelty of the present system lies in the design and performance
of the micro combustor and power generator system. Even though
many researchers have suggested different micro combustors config-
urations suitable for small-scale power applications
[28,31,33,42–44,55–57], however, no system in the literature has
shown to achieve an overall conversion efficiency greater than 4% with
a high power density (> 0.08 mW/mm3) and a power output of greater
than 4 W as clear from Fig. 15.
5. Conclusions
In the present study, a novel design of a prototype of a planar in-
tegrated micro combustor based micro power generator using thermo-
electric generators has been presented and experimentally investigated.
Detailed parametric studies have been carried out to understand the
flame stability and thermal characteristics of the proposed system. The
effect of coolant flow rate, equivalence ratio and mixture velocity on
power output has been investigated. The coolant flow rate has
minimum influence on the variation of hot side temperature and open-
circuit voltage of the thermoelectric generators. A maximum power of
2.54 W is achieved at ϕ = 0.95. An overall conversion efficiency of
3.3% is achieved at ϕ = 0.95. To further improve the system perfor-
mance, a porous media (ceramic wool) of 2 mm thickness is introduced
in the third step of the combustor. The upper flame stability limits are
B. Aravind et al.
Fig. 15. Performance comparison of various micropower generators.
significantly enhanced due to the preheating of fresh mixture through
heat recirculation in the porous media. A maximum conversion effi-
ciency of 4.03% along with a maximum power output of 3.89 W was
achieved at ϕ = 0.9 at a mixture velocity of 10 m/s. The major ad-
vantages of the present system are (i) Improved flame stability limits
with better thermal characteristics of the integrated micro combustor
(ii) High power density (∼50% higher than existing systems) (iii)
Higher conversion efficiency (4.03%) (iv) A compact and integrated
combustor design for power generation. Currently, the peak tempera-
ture of the combustor walls is slightly lower than the maximum per-
missible temperature (250 °C) of the thermoelectric modules. This
suggests the possibility of further improvement in the performance of
the system.
Acknowledgement
Authors acknowledge the financial support for this work from
Department of Science and Technology, Govt. of India wide grant no.
SB/S3/COMB-001/2014.
References
[1] Dunn-Rankin D, Leal EM, Walther DC. Personal power systems. Prog Energy
Combust Sci 2005;31:422–65.
[2] Kaisare NS, Vlachos DG. A review on microcombustion: fundamentals, devices and
applications. Prog Energy Combust Sci 2012;38:321–59.
[3] Epstein AH, Jacobson SA, Protz JM, Frechette L.G. Shirtbutton-sized gas turbines:
The engineering challenges of micro high speed rotating machinery. in:
Proceedings, 8th Int’l Symposium on Transport Phenomena and Dynamics of
Rotating Machinery, Honolulu, HI2000.
[4] Chigier N. Gemci T. A review of micro propulsion technology. in: 41st Aerospace
Sciences Meeting and Exhibit 2003:6–9.
[5] Yetter R, Yang V, Wang Z, Wang Y, Milius D, Peluse M, et al. Development of meso
and micro scale liquid propellant thrusters. AIAA Paper 2003;676.
[6] Fernandez-Pello AC. Micropower generation using combustion: issues and ap-
proaches. Proc Combust Inst 2002;29:883–99.
[7] Maruta K, Parc J, Oh K, Fujimori T, Minaev S, Fursenko R. Characteristics of mi-
croscale combustion in a narrow heated channel. Combust, Explos Shock Waves
2004;40:516–23.
[8] Lee K, Hong Y, Kim K, Kwon O. Stability limits of premixed microflames at elevated
temperatures for portable fuel processing devices. Int J Hydrogen Energy
2008;33:232–9.
[9] Sitzki L. Borer K. Schuster E. Ronney PD. Wussow S. Combustion in microscale heat-
recirculating burners. in: The Third Asia-Pacific Conference on Combustion. Seoul,
Korea 2001;11–4.
[10] Ju Y, Maruta K. Microscale combustion: technology development and fundamental
research. Prog Energy Combust Sci 2011;37:669–715.
[11] Chao Y-C, Chen G-B, Hsu C-J, Leu T-S, Wu C-Y, Cheng T-S. Operational char-
acteristics of catalytic combustion in a platinum microtube. Combust Sci Technol
2004;176:1755–77.
[12] Maruta K. Takeda K. Sitzki L. Borer K. Ronney PD. Wussow S. et al. Catalytic
combustion in microchannel for MEMS power generation. Third Asia-Pacific
Conference on Combustion, Seoul, Korea 2001:1–4.
[13] Chen J, Song W, Xu D. Optimal combustor dimensions for the catalytic combustion
233
Energy Conversion and Management 156 (2018) 224–234
of methane-air mixtures in micro-channels. Energy Convers Manage
2017;134:197–207.
[14] Marbach T, Agrawal A. Heat-recirculating combustor using porous inert media for
mesoscale applications. J Propul Power 2006;22:145–50.
[15] Li J, Li Q, Wang Y, Guo Z, Liu X. Fundamental flame characteristics of premixed H
2–air combustion in a planar porous micro-combustor. Chem Eng J
2016;283:1187–96.
[16] Li J, Li Q, Shi J, Liu X, Guo Z. Numerical study on heat recirculation in a porous
micro-combustor. Combust Flame 2016;171:152–61.
[17] Chou S, Yang W, Li J, Li Z. Porous media combustion for micro thermophotovoltaic
system applications. Appl Energy 2010;87:2862–7.
[18] Taywade UW, Deshpande AA, Kumar S. Thermal performance of a micro combustor
with heat recirculation. Fuel Process Technol 2013;109:179–88.
[19] Khandelwal B, Deshpande AA, Kumar S. Experimental studies on flame stabilization
in a three step rearward facing configuration based micro channel combustor. Appl
Therm Eng 2013;58:363–8.
[20] Yang W, Chou S, Shu C, Li Z, Xue H. Combustion in micro-cylindrical combustors
with and without a backward facing step. Appl Therm Eng 2002;22:1777–87.
[21] Epstein AH. Senturia SD. Anathasuresh G. Ayon A. Breuer K. Chen K.-S. Power
MEMS and microengines. Solid State Sensors and Actuators, 1997
TRANSDUCERS'97 Chicago, 1997 International Conference on. IEEE1997:753–756.
[22] Yang W, Chou S, Shu C, Li Z, Xue H. A prototype microthermophotovoltaic power
generator. Appl Phys Lett 2004;84:3864–6.
[23] Yang W, Chou S, Shu C, Li Z, Xue H. Experimental study of micro-thermo-
photovoltaic systems with different combustor configurations. Energy Convers
Manage 2007;48:1238–44.
[24] Park J, Lee S, Wu H, Kwon O. Thermophotovoltaic power conversion from a heat-
recirculating micro-emitter. Int J Heat Mass Transf 2012;55:4878–85.
[25] Yang W, Chua K, Pan J, Jiang D, An H. Development of micro-thermophotovoltaic
power generator with heat recuperation. Energy Convers Manage 2014;78:81–7.
[26] Jiang D, Yang W, Tang A. Development of a high-temperature and high-uniformity
micro planar combustor for thermophotovoltaics application. Energy Convers
Manage 2015;103:359–65.
[27] Lei Y, Chen W, Lei J. Combustion and direct energy conversion inside a micro-
combustor. Appl Therm Eng 2016;100:348–55.
[28] Lee S, Um D, Kwon O. Performance of a micro-thermophotovoltaic power system
using an ammonia-hydrogen blend-fueled micro-emitter. Int J Hydrogen Energy
2013;38:9330–42.
[29] Schaevitz SB. Franz AJ. Jensen K.FSchmidt MA. A combustion-based MEMS ther-
moelectric power generator. Transducers’ 01 Eurosensors XV. Springer 2001:30–33.
[30] Yoshida K, Tanaka S, Tomonari S, Satoh D, Esashi M. High-energy density miniature
thermoelectric generator using catalytic combustion. J Microelectromech Syst
2006;15:195–203.
[31] Jiang L, Zhao D, Guo C, Wang X. Experimental study of a plat-flame micro com-
bustor burning DME for thermoelectric power generation. Energy Convers Manage
2011;52:596–602.
[32] Shimokuri D. Hara T. Ishizuka S. Development of a portable power system with
meso-scale vortex combustor and thermo-electric device. Journal of Physics:
Conference Series. IOP Publishing 2014:012117.
[33] Yadav S, Yamasani P, Kumar S. Experimental studies on a micro power generator
using thermo-electric modules mounted on a micro-combustor. Energy Convers
Manage 2015;99:1–7.
[34] Yadav S, Sharma P, Yamasani P, Minaev S, Kumar S. A prototype micro-thermo-
electric power generator for micro-electromechanical systems. Appl Phys Lett
2014;104:123903.
[35] Ahn J, Ronney PD, Shao Z, Haile SM. A thermally self-sustaining miniature solid
oxide fuel cell. J Fuel Cell Sci Technol 2009;6:041004.
[36] Ryi S-K, Park J-S, Choi S-H, Cho S-H, Kim S-H. Novel micro fuel processor for
PEMFCs with heat generation by catalytic combustion. Chem Eng J
2005;113:47–53.
[37] Zhang L, Zhu J, Yan Y, Guo H, Yang Z. Numerical investigation on the combustion
characteristics of methane/air in a micro-combustor with a hollow hemispherical
bluff body. Energy Convers Manage 2015;94:293–9.
[38] Jiaqiang E, Peng Q, Liu X, Zuo W, Zhao X, Liu H. Numerical investigation on hy-
drogen/air non-premixed combustion in a three-dimensional micro combustor.
Energy Convers Manage 2016;124:427–38.
[39] Su Y, Song J, Chai J, Cheng Q, Luo Z, Lou C, et al. Numerical investigation of a
novel micro combustor with double-cavity for micro-thermophotovoltaic system.
Energy Convers Manage 2015;106:173–80.
[40] Qiu K, Hayden A. Development of a novel cascading TPV and TE power generation
system. Appl Energy 2012;91:304–8.
[41] Walther DC, Ahn J. Advances and challenges in the development of power-gen-
eration systems at small scales. Prog Energy Combust Sci 2011;37:583–610.
[42] Yang W, Chou S, Shu C, Li Z, Xue H. Development of microthermophotovoltaic
system. Appl Phys Lett 2002;81:5255–7.
[43] Merotto L, Fanciulli C, Dondè R, De Iuliis S. Study of a thermoelectric generator
based on a catalytic premixed meso-scale combustor. Appl Energy
2016;162:346–53.
[44] Shimokuri D, Hara T, Matsumoto R. Development of a small-scale power system
with meso-scale vortex combustor and thermo-electric device. J Micromech
Microeng 2015;25:104004.
[45] Aravind B, Velamati RK, Singh AP, Yoon Y, Minaev S, Kumar S. Investigations on
flame dynamics of premixed H 2–air mixtures in microscale tubes. RSC Advances.
2016;6:50358–67.
[46] Singh AP, RatnaKishore V, Minaev S, Kumar S. Numerical investigations of un-
steady flame propagation in stepped microtubes. RSC Adv 2015;5:100879–90.
B. Aravind et al.
[47] Singh AP, Kishore V, Yoon Y, Minaev S, Kumar S. Effect of wall thermal boundary
conditions on flame dynamics of ch4-air and h2-air mixtures in straight microtubes.
Combust Sci Technol 2017;189:150–68.
[48] Anil AD, Sudarshan K. On the formation of spinning flames and combustion com-
pleteness for premixed fuel-air mixtures in stepped tube combustors. Appl Therm
Eng 2013;51:91–101.
[49] Khandelwal B, Sahota GPS, Kumar S. Investigations into the flame stability limits in
a backward step micro scale combustor with premixed methane–air mixtures. J
Micromech Microeng 2010;20:095030.
[50] Chen W-H, Liao C-Y, Hung C-I, Huang W-L. Experimental study on thermoelectric
modules for power generation at various operating conditions. Energy
2012;45:874–81.
[51] Esarte J, Min G, Rowe D. Modelling heat exchangers for thermoelectric generators.
J Power Sources 2001;93:72–6.
[52] Angeline AA, Jayakumar J, Asirvatham LG, Marshal JJ, Wongwises S. Power gen-
eration enhancement with hybrid thermoelectric generator using biomass waste
234
Energy Conversion and Management 156 (2018) 224–234
heat energy. Exp Thermal Fluid Sci 2017;85:1–12.
[53] Ning D, Liu Y, Xiang Y, Fan A. Experimental investigation on non-premixed me-
thane/air combustion in Y-shaped meso-scale combustors with/without fibrous
porous media. Energy Convers Manage 2017;138:22–9.
[54] Liu Y, Fan A, Yao H, Liu W. A numerical investigation on the effect of wall thermal
conductivity on flame stability and combustion efficiency in a mesoscale channel
filled with fibrous porous medium. Appl Therm Eng 2016;101:239–46.
[55] Singh T, Marsh R, Min G. Development and investigation of a non-catalytic self-
aspirating meso-scale premixed burner integrated thermoelectric power generator.
Energy Convers Manage 2016;117:431–41.
[56] Vican J, Gajdeczko B, Dryer F, Milius D, Aksay I, Yetter R. Development of a mi-
croreactor as a thermal source for microelectromechanical systems power genera-
tion. Proc Combust Inst 2002;29:909–16.
[57] Li YH, Li HY, Dunn-Rankin D, Chao YC. Enhancing thermal, electrical efficiencies of
a miniature combustion-driven thermophotovoltaic system. Prog Photovoltaics Res
Appl 2009;17:502–12.