Content uploaded by Daniel Favrat
Author content
All content in this area was uploaded by Daniel Favrat
Content may be subject to copyright.
Scroll expander Organic Rankine Cycle (ORC)
efficiency boost of biogas engines
Kane M.1, Favrat D1, Gay B1., Andres O. 2
Industrial Energy Systems Laboratory (LENI), Ecole Polytechnique Fédérale de
Lausanne, EPFL, Station 9, CH1015 Lausanne, Switzerland, tel +41 21 693 2511, fax
+41 21 693 3502,
Daniel.Favrat@epfl.ch, malick.kane@epfl.ch
2 Rigot+Rieben engineering, Genève, Switzerland, oandres@rigotrieben.ch
ABSTRACT: The most common energy application for biogas is on-site power
generation using gas engine units. The introduction of a bottoming cycle based on
Scroll expander Organic Rankine Cycle (ORC) technology within a biogas plant is
of high interest since the excess heat produced by biogas engines can be used to
produce additional power. This paper describes the retrofit of a plant consisting of
a digester fed with green municipal wastes and two biogas engine units (200kWe
each). The waste heat from the cooling jacket of the first biogas unit is used in
cogeneration for the fermentation process and also for heating the buildings of the
facility. The objective of the retrofit project is, in a first step, to convert the excess
heat from the cooling jacket of the second biogas unit by means of Organic
Rankine cycles. Onsite preliminary tests from the operation of the 7 kWe unit have
been done, allowing performances to be measured over a broad range of
conditions. Those tests confirmed a reasonable behavior of the ORC scroll
expander and the interest of the concept. The measured efficiency of the ORC is
about 7% with a heat source at around 90°C (i.e. 40% exergy efficiency).
Keywords: Scroll Expander, Organic Rankine Cycle, Bottoming Cycle application,
Biogas Engine Cogeneration unit
NOMENCLATURE
ORC = Organic Rankine Cycle
T
E
& = Electrical power of the turbine [kW]
C
E
& = Electrical power of the pump and
others components [kW]
M
& = Mass flow rate [kg/s]
h = Specific enthalpy [MJ/kg]
P = Pressure [Pa]
T = Temperature [K]
v = Specific volume [m3/kg]
s = Superheat vapor
ε
= First Law efficiency
∆
hho = Specific enthalpy difference of the
heating source
1. INTRODUCTION
The valorization of biogas from landfills,
waste water treatment plants or digestors of
green biomass is usually done by internal
1
combustion engines with or without
cogeneration of heat and power. The
efficiency of these engines is, in some
countries like Switzerland, penalized by
tough constraints of emissions (CO, NOx).
Given the fact that catalysts are often
unreliable because of various contaminants
in the biogas itself, the widespread solution
is to use uncatalyzed lean burn engines with
a low compression ratio and therefore a
rather low efficiency. Work is underway to
improve combustion using unscavenged
prechambers allowing an increase of
compression ratio and an efficiency level
close to that of natural gas engines [1].
However the next promising step is to better
use of the exergy potential of the available
heat from the combustion gas and jacket
cooling. This can be done by using Organic
Rankine Cycles (ORC). Figure 1 shows the
efficiency of a broad range of gas engines
installed in Europe [2] together with the
range of improvements, which can be
achieved using unscavenged prechambers
and bottoming ORCs. The latter is the object
of the present paper.
Electricity generation from low-
temperature heat sources (<300 ºC)
generally implies the use of Rankine cycles
equipped with turbines or expanders.
Considering gas or biogas engines of a few
hundreds of kW with a jacket cooling heat
available of the same order of magnitude,
the bottoming ORCs have to be in a range of
electric power of a few tens of kW.
This is typically a power range for
volumetric machines like scroll expanders
[3]. Even if steam is not to be completely
excluded in the future [4], working fluids
mainly considered today are non flammable
organic fluids like HCFC123, HFC134a or
HFC245. Other fluids like toluene,
isobutene, mixtures of siloxane, are also
implemented for geothermal or biomass
fired systems but mainly with dynamic
machines and powers of several hundred of
kW [5]. Smaller systems have not been
economically feasible due to the lack of
turbines in that size range with adequate
efficiencies (particularly when having to
cope with high expansion ratios and variable
operating conditions) and high specific costs
associated with low initial production
quantities. These limitations led to the
innovation of the “ORC scroll turbine”
concept, a feasibility of which has already
been demonstrated in previous studies [3,6].
These turbines are based on the modification
of hermetic scroll compressors widely used
in refrigeration and air-conditioning
applications.
Figure 1 Efficiencies of some installed gas
cogeneration motors in Europe and
perspective for biogas with improved
combustion and ORC.
2. SCROLL EXPANDER ORGANIC
RANKINE CYCLE TECHNOLOGY
Scroll compressors are characterized by
good reliability and high efficiency, with
low specific costs. The concept is basically
simple and was the subject of a patent in
1905 by the French engineer Léon Creux.
The technology only became available from
the eighties, however, due to the previous
lack of high-precision machine tools capable
of achieving the very high tolerances
required for fabrication. Since scroll
compressors offer a number of advantages
in terms of function, operation and
efficiency that far exceed those of
reciprocating machines, their development
represents a technological breakthrough, and
hermetic scroll units are produced
worldwide in large quantities for
2
refrigeration and air-conditioning
applications.
The scroll turbine consists of expansion
chambers delimited by two involute scrolls
(one fixed and one mobile) positioned such
that they form a series of crescent shaped
pockets (Figure 2). The mobile scroll is
mounted on an eccentric drive shaft, which
induces an orbital movement rather than a
simple rotary motion. During expansion,
high-pressure refrigerant gas is introduced
into the center and moves progressively
towards the periphery by increasing the
volume of the pockets. When the pockets
reach the periphery, the gas expands to its
discharge pressure and leaves the expander
through one or two large diameter ports on
the opposite flanks.
Figure 2: Scroll expander expansion process
The main advantages of this type of turbine
compared to conventional volumetric
turbines are: a low number of components
(no suction and discharge valves), few
moving parts, resulting in a high level of
reliability, an ability to successfully work
with a two-phase flow, low mechanical
vibrations and pulsations implying a quasi-
steady mechanical torque, and a quasi
continuous expansion process with an
excellent volumetric efficiency.
The feasibility of the Scroll expander ORC
system has been demonstrated in previous
work for different types of scroll units [3,5],
including a 12-kWe prototype designed to
work at around 150 °C. The latter has been
tested both in the laboratory, using heat
from a thermal oil boiler, and in the field
test using heat from a solar concentrator
system complemented with waste heat from
a 13-kWe reciprocating diesel engine. For
this specific application, two superposed
ORCs were used, each working with a
different fluid, so that they could be
operated independently if required due to
fluctuations in solar conditions and heat
demand requirements. The working fluids
chosen were HCFC 123 for the topping
cycle and HFC 134a for the bottoming
cycle.
In the literature a number of Organic
Rankine Cycles are reported, in particular in
liaison with solar thermal energy conversion
[7,12] but generally not with scroll
expanders.
3. BOTTOMING CYCLE
APPLICATION TO BOOST
EFFICIENCY FOR BIOGAS ENGINES
The production of biogas is an energy-
intensive process that often requires both
electricity and heat. The most common use
of the biogas product is on-site power
generation using conventional gas engines
to provide the energy needed to the process.
A typical plant like in Nant de Chatillon
(Geneva, Switzerland) consists of a digester
fed with green municipal wastes and
producing biogas which is converted in two
biogas engine units (200 kWe each).
Originally the waste heat from the cooling
jacket of the first biogas unit has been used
in cogeneration mode to satisfy the power
needs of the plant as well as the thermal
needs of the fermentation process and of the
heating the buildings of the facility. A
retrofit project of the plant has been
considered by introducing Scroll expander
ORC systems as a Bottoming Cycle
application to boost the efficiency of the
second gas engine unit. The design of the
bottoming cycle concept is a tradeoff
between the efficiency, the expander
characteristics and the control complexity
(reliability, robustness, cost, training
3
requirements, etc.).While both the heat from
the combustion gas and from the jacket
cooling could be used, in a first and
conservative step, it was decided to only
convert the low-temperature (90°C) excess
heat from the cooling jacket to produce
additional power. Available hermetic scroll
expander-generators are limited in size
because they are based on the modified
standard scroll compressor units. Therefore
the proposed concept includes two parallel
ORC units (ORC1-7kWe and ORC2-13
kWe) with the advantage of better load
control by shutting down one of the unit to
match the fluctuations of the available heat.
The present paper discusses the preliminary
operation and tests of the 7 kWe ORC unit
(ORC1). A detailed single-line drawing of
this system is given in figure 3.
The working fluid chosen for the system
is HCFC134a working between 0.6 to 2.2
MPa. The vapor produced in the (plate)
evaporator is either bypassed (during warm-
up) or expanded in the scroll unit. The
discharged vapor is cooled and condensed in
a condenser (plate) heat exchanger. Liquid
HCFC134a is then returned to feed the
evaporator of the cycle by a membrane-
piston pump. The nominal capacity of the
scroll-expander generator is 7 kWe. The hot
source temperature (from the engine water
cooling system) ranged between 80°C and
90°C. A wet cooling tower (not represented
in the drawing) is used to cool the ORC
within an intermediate close-loop water
network of about 20°C to 30ºC depending
on the seasons. This circuit includes a three
way valve regulator that allows the
adaptation of the condensation pressure. The
lubricant oil required by the expander
circulates with the working fluid, and the
separation is made at the end of evaporation
allowing to directly lubricate the bearings
using the available pressure difference. A
sub-cooler component has been added in
order to also level the working fluid
fluctuations. A flexible and robust command
system has been made in place for the
purpose of controlling the working-fluid
mass flow by controlling the frequency of
the pump. Figure 4 shows a picture of the
installation in Chatillon (Genève,
Switzerland). The system has been built and
developed taking into account the
implementation of all safety and technical
requirements. This includes: safety
procedures and automation (operating
procedures, security system and
measurement).
4. ON-SITE TESTING AND RESULTS
A series of tests have been made on-site
to demonstrate the performances of the
scroll expander ORC unit by using a thermal
oil heat source to supply heat to the
evaporator. Onsite preliminary tests from
the operation of the 7 kWe unit have been
done, allowing performances to be measured
over a broad range of conditions.
The objective is to measure the
performances of the cycle for different
conditions and therefore to determine the
operational feasible range of heat supply to
the ORC cycle. The supply temperature as
well as the heat rate from the biogas engine
are directly related to the working
conditions of the biogas engine. The mass
flow rate of the ORC cycle can be adjustable
to change the conditions (pressures and
temperatures) of the working cycle.
Moreover, the ORC cooling water circuit
can be also used to adapt the condensation
pressure of the cycle. The measured data
includes the temperatures and pressures at
the inlet and outlet of the main components
(turbines, pump, heat-exchangers) and the
net electricity output. In addition, flow-
meters and temperature measurements on
the hot and cold streams have been used for
the determination of the energy balance of
the cycles. Figures 5 and 6 show the
variations of the ORC efficiency as well as
the net power output in function of the heat
recovered from the biogas engine.
4
Figure 3: Schematic drawing of the 7 kWe Scroll expander ORC unit
Figure 4: ORC1 installed in Nant de
chatillon
The ORC net electric output is the
difference between the output of the
expander and the electric power consumed
by the others components like pump, fans
and valves, fans divided by the heat
recovered from the engine (equation 1)
ho
CT
ORC hM
EE
∆
−
=&
&&
ε
(1)
ORC1 - R134a
Net Efficiency vs Heat
2,0
3,0
4,0
5,0
6,0
7,0
45 50 55 60 65 70 75 80
Heat from Engine (kW)
ORC Net Efficiency (%)
Figure 5: ORC net energy efficiency for
various recovered heat rate from the engine
Results show that the ORC system can be
operated even at very low power output,
below 20% of its nominal design value. The
cycle efficiency up to 7% was expected for
this low temperature application (up to
90°C).
5
ORC1 - R134a
Net Electric Output vs Heat
1,0
2,0
3,0
4,0
5,0
45 50 55 60 65 70 75 80
Heat from Engine (kW)
Net Electric Output (%)
Figure 6: ORC net electric output for
various heat rate from the engine
All these tests have been realized in
summer when the ORC cooling network
temperature achieved 30°C. The
corresponding Carnot efficiency is of the
order of 18%. Figure 7 shows the variations
of the ORC exergetic efficiency in function
of the heat recovered from the biogas
engine. A net exergetic efficiency of about
40% is achieved for this application.
ORC1 - R134a
Net Exergetic Efficiency vs Heat
10
15
20
25
30
35
40
45 50 55 60 65 70 75 80
Heat from Engine (kW)
Net Exgergetic
Efficiency (%)
Figure 7: Net exergetic efficiencies of the
ORC for various heat rates from the engine
Note that the efficiency decrease at low
heat rates can be explained by the losses
linked to the pressure ratio which is then too
low at the expander of the working cycle.
As shown in figure 8 it can be pointed out
that the expander is used at very part load
(operation with a pressure ratio down to
1.8). This turbine was in fact designed to
work at a pressure ratio of about 3.67
(corresponding to evaporation pressure of
2.2 MPa and a condensation pressure of 0.6
MPa). But in summer operation, as
mentioned above, the ORC cooling network
is at around 30°C and imposes a high
condensation pressure between 0.8 and 1
MPa, which is detrimental to the efficiency.
An alternative is to introduce a variable
speed expander to better adjust the load. For
simplicity of operation and a cheaper
approach for this demonstration project, the
generator is directly connected to the grid
without any inverter.
ORC1 - R134a
Net Efficiency vs Pressure Ratio
2,0
3,0
4,0
5,0
6,0
7,0
1,8 1,9 2 2,1 2,2 2,3 2,4 2,5 2,6
Pressure Ratio (-)
ORC Net Efficiency (%)
Figure 8: Net efficiency of the ORC vs
pressure ratio
Others sources of losses that can
influence the efficiency are related to the oil
mixed with the evaporating working fluid
(HFC134a), whose boiling temperature
strongly increases in the dryout region of the
evaporator. This phenomenon is well known
in heat pumps and is accompanied by a
significant drop in heat transfer coefficient
with a corresponding drop of the
evaporation pressure. One solution would be
to introduce a falling film shell in tube
evaporator instead of the plate evaporator
with an accurate and fine control of the
liquid pump.
Fluctuations of both the amount of heat
recovered from the biogas engine and/or
outside ambient temperature can be coped
with the adaptation of the mass flow rates of
the ORC working fluids and operational
feasible ranges of vapor superheating and
liquid sub-cooling can be maintained.
Figures 9 and 10 show the variations of the
ORC net efficiency in function of
superheating and sub-cooling temperatures.
6
ORC1 - R134a
Net Efficiency vs Surperheating Temperature
2,0
3,0
4,0
5,0
6,0
7,0
15 18 21 24 27 30 33
Superheating temperature (°C)
ORC Net Efficiency (%)
Figure 9: Net efficiencies of the ORC for
various vapor superheating
ORC1 - R134a
Net Efficiency vs Sub-cooling Temperature
2,0
3,0
4,0
5,0
6,0
7,0
3456789101112
Sub-cooling Temperature (°C)
ORC Net Efficiency (%)
Figure 10: Net efficiencies of the ORC for
various liquid sub-cooling
The results show that there is no
advantage to work at relatively high
difference of temperature (superheating
above 20°C and sub-cooling above 10°C). A
high difference of temperature corresponds
in fact to a low mass flow rate that is
accompanied by a significant drop of the
evaporation pressure. This makes the cycles
work at part load, which is detrimental for
the efficiency. A low difference of
temperature (for example superheating
temperature below 15°C or sub-cooling
temperature below 3°C) corresponding to a
high mass flow is not feasible because of the
possibility of refrigerant flow disturbances
at the pump that makes the cycle non-stable.
The limiting range is influenced by the
quantity of refrigerant charged in the circuit.
4. CONCLUSIONS
The concept of bottoming cycle based on
Scroll expander Organic Rankine Cycle
(ORC) system has been applied to boost the
electrical efficiency of a 200kWe biogas
engine in a green waste fermentation plant
in Nant-De-Chatillon (Geneva,
Switzerland). For the retrofit project, two
ORC single cycles (ORC1-7kWe and
ORC2-13kWe) have been built and installed
to convert only the engine cooling energy in
a first approach. The design was optimized
to demonstrate the system taking account of
all safety requirements (safety procedures
and automation). Onsite preliminary tests of
the 7 kWe ORC unit have been conducted
allowing the performances to be measured
for a broad range of conditions. Results
show that the ORC system can be operated
even at very low power output, below 20%
of his nominal design value and a cycle net
efficiency of 7% can be achieved for a low
temperature application (90°C). The field
experience gained is being used to improve
the automatic control of such plants which is
currently underway.
ACKNOWLEDGEMENT
The authors would like to acknowledge the
financial support provided by OCEN and the
contribution of the Swiss Federal Office of
Energy. The authors also thank the
companies who contributed to the
construction and installation on-site.
REFERENCES
[1] Roubaud A, Favrat D Improving
performances of a lean burn
cogeneration biogas engine equipped
with prechambers, Fuel 2005; 84: 2001-
2007.
[2] Kalitventzeff B., Marechal F Optimal
insertion of energy saving technologies
in industrial processes. Applied
Thermal Eng. 2000 ;20 (15-16):1347-
1364.
7
[3] Zanelli R., Favrat D.: Experimental
investigation of a hermetic scroll
expander-generator. Int. Compressor
Eng. Conf. At Purdue, pp. 459-464,
(1994).
[4] Rickli JP, Favrat D, Marechal F,
Demierre J, Can steam play a role in
low to medium power from low
temperature heat. ASME ATI
conference on Energy production,
distribution and conservation, Milan,
p273-279, May 14-18 ,2006
[5] VDI (Verein Deutscher Ingenieure):
ORC-HP-Technology, Working fluid
Problems. VDI-Verlag, (1984).
[6] Kane M., Larrain D., Favrat D., Allani
Y. Small hybrid solar power system.
Energy, 2003; 28(14):1427-1443
[7] Prigmore D. and Barber R.: Cooling
with the Sun’s Heat, design
consideration and test data for Rankine
Cycle Prototype. Solar Energy 1975; 17
(185).
[8] Giampaolo M. and Sukuru M.: Energy
control for a flat plate
collector/Rankine cycle solar power
system. J Solar Energy Engng,
1991;113(2):89-97.
[9] Wolpert J.L. and Riffat S.B.: Solar-
Powered Rankine Engine for Domestic
Applications. Applied Thermal
Engineering, 1996; 16: 281-289.
[10] Yamamoto T., Furuhata T., Arai N. and
Mori K.: Design and Testing of the
Organic Rankine Cycle. Energy 2001;
26: 239-251.
[11] Koai K., Lior N., Yeh H.: Performance
analysis of a Solar-Powered/Fuel-
Assisted Rankine Cycle with a Novel
30HP Turbine, Solar Energy, 1984;
32(6): 753-764.
[12] Lior N., Advanced Energy Conversion
to Power. Energy Conversion Mgmt,
1997; 38, (10-13):941-955.
8
