AIAA 2009-4631
7th International Energy Conversion Engineering Conference
2 - 5 August 2009, Denver, Colorado
Long-Lived Venus Lander Conceptual Design:
How To Keep It Cool
Rodger W. Dyson∗ and Paul G. Schmitz†
L. Barry Penswick‡
NASA Glenn Research Center, Cleveland, OH 44135
Sest, Inc., Middleburgh Heights, OH 44135
Geoffrey A. Bruder§
Embry-Riddle Aeronautical University, Daytona Beach, FL 32114
Surprisingly little is known about Venus, our neighboring sister planet in the solar
system, due to the challenges of operating in its extremely hot, corrosive, and dense environment. For example, after over two dozen missions to the planet, the longest-lived lander
was the Soviet Venera 13, and it only survived 2 hours on the surface. Several conceptual
Venus mission studies have been formulated in the past two decades proposing lander architectures that potentially extend lander lifetime. Most recently, the Venus Science and
Technology Definition Team (STDT) was commissioned by NASA to study a Venus Flagship Mission potentially launching in the 2020 to 2025 timeframe; the reference lander of
this study is designed to survive for only a few hours more than Venera 13 launched back
in 1981!
This report reviews those studies and recommends a hybrid lander architecture that
can survive for at least 1 Venus day (243 Earth days) by incorporating selective Stirling
multistage active cooling and hybrid thermoacoustic power.
Nomenclature
η1
η2
ηe
Q1,in
Q1,out
Q2,out
Qe,in
Qenv
S1,in
S1,out
S2,in
S2,out
Se,in
Se,out
T1,in
T1,out
Te,in
Tenv
W1,act
W2,act
Percent of Cooler Carnot Efficiency (First Stage)
Percent of Cooler Carnot Efficiency (Second Stage)
Percent of Engine Carnot Efficiency
Stirling Refrigerator (First Stage) Electrical Heat Lifted, W
Stirling Refrigerator (First Stage) Heat Rejected, W
Stirling Refrigerator (Second Stage) Heat Rejected, W
Stirling Engine Heat Input From General Purpose Heat Source, W
Venus Environment Heat Leakage Into Lander, W
Stirling Refrigerator (First Stage) Entropy In
Stirling Refrigerator (First Stage) Entropy Out
Stirling Refrigerator (Second Stage) Entropy In
Stirling Refrigerator (Second Stage) Entropy
Stirling Engine Entropy In
Stirling Engine Entropy In
Stirling Refrigerator (First Stage) Refrigerated Temperature, K
Stirling Refrigerator (First Stage) Reject Temperature, K
Stirling Engine Acceptor Temperature, K
Stirling Engine Rejector Temperature, K
Stirling Refrigerator (First Stage) Work Input, W
Stirling Refrigerator (Second Stage) Work Input, W
∗ Research
Engineer, Thermal Energy Conversion Branch, 21000 Brookpark Road, Mail-Stop 301-2, AIAA Member
Engineer, Electrochemistry Branch, 21000 Brookpark Road, AIAA Member
‡ Thermal Design Engineer, Sest, Inc., AIAA Member
§ Intern, Thermal Energy Conversion Branch, 21000 Brookpark Road, Mail-Stop 301-2
† Research
1
Institute
of protection
Aeronautics
Astronautics
This material is declared a work of the U.S. Government and American
is not subject
to copyright
in theand
United
States.
I.
Introduction
xploration of Venus has been accomplished with flybys, orbiters, balloons, transient descent probes,
E
and short-lived landers as listed in Table 1. All of the past missions and planned future missions have
been constrained to operate for less than 10 hours on the surface-with the one exception being the planned
Russian Venera-D mission in which a longer-term surface lander is intended, but details are not yet available.
Since Cytherean mission planners lack a viable approach to a long-lived surface architecture, specific
scientific objectives outlined in the National Science Foundation Decadel Survey and Venus Exploration
Advisory Group final report cannot be completed. These include: mapping the mineralogy and composition
of the surface on a planetary scale; determining the age of various rock samples on Venus, searching for
evidence of changes in interior dynamics (seismometry) and its impact on climate; and many other key
observations that benefit with time scales of at least a full Venus day (i.e., daylight/night cycle).
Table 1. Missions To or By Way of Venus
Mission Type
Flybys
Orbiters
Descent Probes
Landers
Balloon Probes
Future Missions
Spacecraft Designation
Venera 1 / Mariner 2 / Zond 1 / Venera 2 / Mariner 5,10
Venera 11, 12, 13, 14 / Vega 1,2 / Galileo,
Cassini-Huygens / MESSENGER
Venera 9,10 /Pioneer Venus Orbiter / Venera 15,16 / Magellan / Venus Express
Venera 3,4,5,6 / Pioneer Venus Multiprobe
Venera 7,8,9,10,11,12,13,14 / Vega 1,2
Vega 1,2
PLANET-C (2010) / BepiColombo / Venus Entry Probe (2013) / Venera-D (2016)
Since the last Venus surface mission in 1982 that lasted only 126 minutes (the Soviet Venera 13 spacecraft),
new technology advancements have enabled longer-term operations on the planet. But, very little information
is available in the open literature and hence many mission planners do not consider long-term surface
operations a viable option.1, 2
Numerous studies examining components of a longer-lived lander exist including: Venus Interior Structure
Mission (VISM),3 Glenn Research Center (GRC) duplex study,4 GRC rover study,5, 6 JPL extreme environments technology development survey,7 VEXAG report,8 Embry-Riddle report,9 and the most recent,
STDT report.10
However, they are either inaccessible in the literature, are not integrated with current scientific objectives,
require exotic unavailable materials, or resulted in an overly inefficient and expensive design. For example,
previous studies have proposed that a single-stage refrigeration cycle be used, but given the electrical power
and heat load demands anticipated for a flagship surface mission, the single stage requires too much energy
to operate. As a result, mission planners have assumed phase change material cooling and battery power
as the short-lived alternative. For example, as assumed in the STDT study, Lithium Nitrate Trihydrate
is packaged around the payload and used to cool the electronics to 30◦ C providing for only 5 to 10 hours
survivability.
Instead, a multi-stage refrigeration cycle offers an attractive alternative that rejects the incoming Venusian
surface heat at a thermodynamically more efficient high temperature, while cooling the electronics in a second
stage. This report will expand upon that and other concepts identified in previous studies to provide a viable
Venus lander conceptual design that actively refrigerates the payload for extended surface utility.11–52
II.
Venusian Environment and Mission Requirements
Venus is the only planet in our solar system that very closely matches Earth’s gravity field. This feature
alone is significant for future colonization considerations since non-ideal gravity is the one uncontrollable
environmental variable that poses significant long-term health risks to humans. While perhaps challenging,
especially on Venus, all other environmental hazards can be mitigated.
Note in Fig. 1, that other than its gravity field, its environment is potentially the most inhospitable in
the solar system. Its high surface temperature (460o C) is due to a thick atmosphere of predominantly CO2
that also presents a crushing pressure that is 92 times greater than Earth’s. In addition, at an altitude of
2
American Institute of Aeronautics and Astronautics
approximately 55 km, a cloud layer of sulfuric acid encircles the planet. It is this combination of supercritical CO2 , H2 SO4 , and high temperatures/pressures has made Venus exploration challenging despite its
proximity to Earth.
(a) Venus Atmospheric Conditions
(b) Venus Relative Surface Temperature
Figure 1. Venus Environmental Conditions
Note in Table 1 that since 1964, twelve missions to the surface of Venus have been attempted, mostly by
the former Soviet Union. It took 6 years of development before the first vehicle successfully landed–lasting
only 23 minutes on the surface after descending through the atmosphere for 55 minutes.
As shown in Table 2, the longest-lived lander was Venera 13, launched in 1981, surviving for 127 minutes
before communications halted. This lander utilized passive cooling and batteries for power.
II.A.
Previous Mission Studies
Significant interest in understanding both the interior structure and atmosphere of Venus has been a driving
influence for a long-lived surface mission. Questions about the runaway greenhouse effect, the lack of a magnetic field, and the age of surface features have remained unanswered, largely due to a lack of environmentally
robust technology capable of withstanding the corrosive, high-temperature surface conditions.
II.A.1.
Venus Interior Structure Mission
Several studies have been completed over the past two decades to identify key technologies required for such
an undertaking. A notable first study entitled, Venus Interior Stucture Mission, in 1993, identified Stirling
power and cooling as the best approach. In that study, as shown in Fig. 2, it was assumed the electronics must
be refrigerated to 25o C and only required 20 W of power, with 134 W heat leak in from the environment.
The Stirling convertor had a dual-opposed displacer configuration that drove rotary alternator for power
and a mechanically connected cooler. The Stirling engine construction was proposed to be Molybdenum,
which oxidizes quickly at 760o C, but may react differently in a supercritical CO2 environment. The exposed pneumatic coupling was titanium, the heat source canister was iridium (same as cladding surrounding
plutonium-238), and the multifoil thermal insulation was nickel. The total mass for the power and cooling
system, including radioisotope heat source was projected to be 65.8 kg.
3
American Institute of Aeronautics and Astronautics
Table 2. Venus Lander Mission Log
Launch
Date
1964
1965
1967
1969
1969
1970
1972
1975
1975
1978
1978
1978
1981
1981
1984
Spacecraft
Zond 1
Venera 2&3
Venera 4
Venera 5
Venera 6
Venera 7
Venera 8
Venera 9
Venera 10
Pioneer
Venera 11
Venera 12
Venera 13
Venera 14
VEGA 2
Arch.
Probe
Probe
Probe
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Lander
Temp.
(o C)
80
80
300
320
320
460
460
460
460
460
460
460
460
460
460
Pressure
(bar)
5
5
20
27
27
92
92
92
92
92
92
92
92
92
92
Phase
Entry
Descent
Descent
Descent
Landed
Landed
Landed
Landed
Landed
Landed
Landed
Landed
Landed
Landed
Alt.
(km)
200
25
20
20
0
0
0
0
0
0
0
0
0
0
Duration
(min)
93/0
53/0
51/0
55/23
55/50
55/53
55/65
55/68
60/95
60/110
55/127
55/57
55/56
Note that the high Stirling efficiency of 31% was 66% of Carnot efficiency which has never been achieved
to date, although the percentage of Carnot achievable does increase with the hot-end temperature as discussed later. The cooler coefficient of performance (COP) of .37 was assumed although this may be too high
considering that is 59% of Carnot efficiency whereas only approximately 30% of Carnot has been demonstrated in hardware. However, as the size of the refrigerator increases, so does its effiency since parasitic
losses become less significant.
Under these perhaps overly aggressive performance assumptions and inadequate electronics and heat
leakage assumptions, this design does not appear feasible, but it does provide an overall architecture that
can be successfully employed by incorporating recent technological improvements.
II.A.2.
Other Studies
Several other more recent studies have developed more accurate designs that investigated utilizing a Stirling
duplex that had either an electrical, mechanical, or pneumatical connection between the convertor and
cooler. In Fig. 3(a) is a simple schematic showing the energy flow between the major lander components. In
this study, an electrical requirement of 8 W and a heat leak of 50 W was assumed. Note that the primary
trade-space was duplex connection method (electrical or pneumatic) and alternator temperature (cooled or
exposed). In the best case, it was the cooled alternator with pneumatic connection that provided the highest
specific power/cooling solution. As shown in Fig. 3(b), a total of 10 GPHS modules were assumed necessary.
In another study, a Stirling duplex mechanically connected, shown in Fig. 3(c), was designed. It was
assumed to have 27.48% Stirling convertor efficiency (57% of Carnot), which is an agressive assumption. The
cooler was also assumed to have a COP of .376 (58% of Carnot), with a payload temperature of 200o C, 77
W heat leakage to payload, and 100W electrical. A total of 8 GPHS modules were assumed necessary. The
high COP has not been demonstrated in space coolers to date.
III.
Facility Development and System Testing
It is common to test flight hardware under conditions expected throughout its lifecycle to ensure its
quality and reliability. NASA has facilities at some of its field centers available for such testing, particularly
those at GRC and Ames may be best suited for a Venus mission. As shown in Table 3, facilties exist for
launch, transit, entry, and descent testing. Note the lack of a surface testing environment in this list. There
are small test chambers available as shown in Table 4 suitable for limited materials testing. But for Stirling
duplex and lander testing, a larger facility is required.
4
American Institute of Aeronautics and Astronautics
(a) Mass Breakdown
(b) Power Flow
Figure 2. Venus Interior Structure Mission
Table 3. Existing and Proposed Venus Mission Profile Chambers
Mission Phase
Facility/Center
Launch
Transit
Entry
Entry
Entry
Descent
Surface
SDL/GRC
SPF/GRC
IHF/ARC
HTF/GRC
20g Centrifuge/ARC
Wind Tunnel/ARC
Proposed/GRC
Size
(feet)
10x10
100x122
Coupon
25x20
7.6x5.9
80x120
6x10
Pressure
(bar)
1
1.3e-9
1
.143 thru 1
1
1
100
Temp.
(o C)
20
-195
1649
1893
20
20
510.2
Simulates
Vibration
Solar Radiation
Viscous Heating
High Velocity
Deceleration
Full Vehicle
Pressure & Temp.
Shown in Table 4 is a list of known facilities capable of simulating the Venus surface conditions.
Shown in Fig. 4(a) is a simple, low-cost component testing chamber that could be used for Stirling
duplex component testing, including a two-stage cooling system. For a full-scale lander test, including
entry/deceleration effects, a facility such as shown in Fig. 4(b) is required. Hot Isostatic Presses (HIPs) are
commercially available and can be manufactured with operating conditions of up to 3000 bar and 3000o C
with test chambers up to 78 inches. The major difference is that these vessels operate with strictly inert fill
gases such as Argon. A proposed lander design stands approximately 3 feet wide, while up to as much as 6 feet
high when joined with the parachute and balloons. Premixed bottles of custom gas mixture are commercially
available, enabling a variety of atmospheres to be introduced. Multiple rod-type heating elements placed
evenly around the outer chamber wall would enable homogenous heating within a desired degree.
The main chamber of the vessel will need to be able to withstand not only extreme temperature and
pressure, but also intense corrosive forces. Industry HIP chambers use steel alloy SA723. Titanium allow Ti6Al-4V was used for the Pioneer Venus mission vehicle, and Nickel Alloy 625 was used in the Massachusettes
Institute of Technology (MIT) autoclave and provides the highest resistance to supercritical CO2 corrosion.
It is proposed that Venus test chamber and Stirling duplex development should be completed concurrently.
5
American Institute of Aeronautics and Astronautics
(a) Single-Stage Cooling Energy Flow
(b) Relative Mass and GPHS Required
(c) Mechanical Duplex
Figure 3. Previous Studies
Table 4. Existing Venus-Capable Chambers
Location
Size
Georgia Institute of Technology
University of Iowa
Jet Propulsion Lab
Massachusettes Institute of Technology
Massachusettes Institute of Technology
12 by 12 in.
5 by 12 in.
4 by 54 in.
1 by 48 in.
0.5 by 12 in.
Pressure
(bar)
100
90
92
200
200
6
American Institute of Aeronautics and Astronautics
Temp.
(o C)
343
500
500
700
700
Gas
Environment
Variable
CO2
CO2 , N2 , trace
CO2
CO2
Since most of the other mission profile test chambers are located at GRC, it would be most cost effective to
install this new facility there as well.
(a) Component Test Chamber
(b) Free-Standing Single-Walled Facility
Figure 4. Venus Environment Test Facility Development
IV.
Design Studies and Analysis
A number of power generation and cooling technologies exist that could be considered for Venus operations
under limited circumstances. The original energy source is either solar radiation or radioactive decay heat
that is converted to electrical, mechanical, or pneumatic energy. That energy is in turn applied to a cooling
device (often a power conversion device run in reverse). Note that batteries and phase change materials are
short-term power and cooling options that do not require a continuous power source.
IV.A.
Power Conversion Options
Table 5. Power Options
Approach
Efficiency, %
Thot
1123K
Tcold = 773K
Properties
Free-Piston Stirling
Free-Displacer Stirling
Thermoacoustic Stirling
Brayton/Rankine
Thermoelectric (Segmented)
Solar Array
Beamed Power
Thermionic
Battery
17
15
13
11
3-4
<1
<1
<1
-
Alternator cooling required, forms a pneumatic duplex
Alternator cooling required, forms a pneumatic duplex
Alternator cooling required, forms a pneumatic duplex
High-speed rotation gear reduction required for cooling
Difficult to couple with efficient dynamic cooling
Additional development required for high temperature
Energy dissipates in atmosphere, requires development
Difficult to couple with efficient dynamic cooling
Limited mission duration or requires repeated charging
In contrast to photovoltaic power, Radioisotope Power Systems (RPSs) are capable of providing substantial power levels (hundreds of Watts) at all altitudes. The 87 year half-life of plutonium- 238 makes an RPS
an effectively unlimited source of electrical energy for virtually any conceivable Venus exploration mission.
Table 5 describes the properties and relative advantages of the main RPS options along with some non-RPS
options for contrast. NASA is currently developing two types of RPSs. Both systems convert the radioisotopic decay heat of plutonium-238 to electricity, using either static or dynamic methods. The Multi-Mission
Radioisotope Thermoelectric Generator (MMRTG) is utilizing the Seebeck effect of static thermocouples
7
American Institute of Aeronautics and Astronautics
for heat-to-electric power conversion, and incorporates flight heritage elements from the General Purpose
Heat Source Radioisotope Thermoelectric Generator (GPHS–RTG). The Advanced Stirling Radioisotope
Generator (ASRG) uses a dynamic Stirling convertor to generate power that is not yet space qualified in an
RPS; however, note that Stirling-cycle coolers have been successfully employed in space.
IV.B.
Active Cooling Options
Table 6. Cooling Options
Approach
Free-Piston Stirling
Free-Displacer Stirling
Thermoacoustic/Pulse Tube
Brayton/Rankine
Thermionic
Thermoelectric (Segmented)
Mixed Refrigerant
Phase Change
Efficiency
% of Carnot
28
24
20
18
15
1
-
Properties
Space operations heritage, forms a pneumatic duplex
Less bearings required, forms a pneumatic duplex
Few moving parts, forms a pneumatic duplex
Gear reduction required from power takeoff
Electrons carry heat across vacuum, requires development
Peltier Cooling, Useful for localized cooling
Venus high-temperature applications not developed yet
Limited mission duration, can complement active cooling
The ability to actively refrigerate instruments and electronics fundamentally changes the nature of any
long-lived mission, including landers, low altitude platforms, or independent in situ instruments. Such a
refrigeration system has two main components: a power source and a refrigeration machine that uses the
power source to pump heat from the payload back out into the environment. The radioisotope power is the
only realistic long-lived power source for the surface of Venus. Typically an RPS system would be used to
jointly power the electronic components of the payload as well as the refrigeration system. The options for an
active Venus refrigeration system are briefly summarized in Table 6. The most mature and highest efficiency
options for Venus are the Stirling refrigeration systems. These require either an electrical power input
or directly pneumatic coupling with a Stirling heat engine in what is known as duplex operation. Duplex
operation is schematically illustrated in Figure 7. Long life operation in Stirling machines is achieved through
the absence of sliding mechanical parts. Indeed, life tests of Stirling convertors for the ASRG program have
accumulated in excess of 4 years of operation and are still going. No Stirling machines have yet been built
and tested for the Venus surface environment. However, many Stirling heat engines and refrigerators have
been built and used for both terrestrial and space applications. This experience provides confidence that this
technology can be successfully extrapolated to the Venus refrigerator application with sufficient technology
development resources. There are two main aspects to that extrapolation: first, the Stirling machines must
be adapted for Venus environmental temperatures; second, a duplex Stirling machine must be produced that
integrates the heat engine and refrigerator functions into an integrated, high-efficiency device.
IV.C.
Principles of Stirling Convertor Operation
The Stirling convertor works as shown in Fig. 5. Heat is supplied to the convertor from a GPHS module
producing thermal power from plutonium-238. The heat input to a convertor results in a hot-end operating
temperature. Heat is rejected from the cold end of the convertor. The Stirling closed-cycle system, using
helium as the working fluid, converts the heat from a GPHS module into reciprocating motion with a linear
alternator resulting in an alternating current (AC) electrical power output. An AC/Direct Current (DC)
convertor in the Stirling convertor controller converts the AC power to DC. With proper masses, spring
rates and damping (dynamic/acoustic tuning), the convertor will resonate as a free-piston, free-displacer,
or thermoacoustic Stirling thermodynamic cycle convertor. RPSs based on direct thermoelectric conversion
(i.e., the MMRTG) can easily exceed their 14 year design lifetime, due in part to the use of well-known
materials, rigorous component testing, and a plutonium-238 heat source with an 87.7-year half-life. A major
motivation for using RPS on NASA missions is their ability to produce continuous, reliable electrical power in
remote and often severe environments, with no reliance on sunlight. Some past NASA missions to the outer
8
American Institute of Aeronautics and Astronautics
planets could not have been performed without RPS, and some spacecraft continue to operate far beyond
their original expectations due in large part to the long-life RPS. Since dynamic conversion is about four
times more efficient than static conversion, the ASRG requires about a quarter of the plutonium compared
to the MMRTG, while generating the same amount of electric power and rejecting proportionately less waste
heat.
Figure 5. Basic Stirling Principles
IV.C.1.
Life-Limiting Mechanisms
Historically, Stirling convertors were not considered for space applications due to many wear mechanisms that
could not be serviced. As shown in Fig. 6, there are three categories of Stirling convertors. The kinematic
convertor mechanically controlled the displacer and piston required sliding seals to contain lubrication, and
rolling element bearings. Over time, those elements do degrade. A more recent development, the free-piston
convertor eliminates the wear mechanisms by letting the resonant properties of the convertor determine the
motion of the piston and displacer. This eliminates the need for lubricants since either gas bearings or flexure
springs can provide bearing support. A very recent Stirling convertor based upon thermoacoustic resonance,
replaces the displacer with a high-intensity acoustical wave. This results in one less moving part, particularly
in the hot end with some loss in efficiency. More study and testing is required to determine accurate trades
of these latter two options.
Despite the elimination of wear mechanisms, some other life-limiting mechanisms include material fatigue
due to pressure and piston oscillations, material creep due to relatively high pressure and temperature,
material permeation due to thin walls and large grain size, and radiation effects on magnets and organic
materials. Fortunately, the Venus environment actually offers some protections of the latter mechanisms
since the high- pressure atmosphere reduces hoop stresses on the convertor and blocks most of the radiation
from reaching the surface.
9
American Institute of Aeronautics and Astronautics
Figure 6. Life-Limiting Mechanisms
IV.D.
Stirling Duplex Principle of Operation
There are some examples of duplex Stirling machines that were built for
terrestrial refrigeration applications. However, those devices were rather
exploratory in nature and not close the high efficiency, long-life machines
required for commercial or space applications. In contrast, a lot of work
has been done on Stirling heat engines for electricity production and considerable technical maturity has been obtained. Recent work on Stirling
cycle power convertors for the ASRG program includes long-lived performance at hot end temperatures of 650 and 850o C with efficiencies of up
to 55% of Carnot. This level of performance is suitable for a Venus power
application, although higher hot end operating temperatures approaching
as much as 1200o C are preferred because they would yield a higher specific power and hence lower mass device. Stirling refrigerators have been
built for both terrestrial and space applications. In particular, long-lived,
space-based cryocoolers have been in operation for many years, and they
operate at comparable or greater temperature ratios than are required
for Venus refrigeration. However, these cryocoolers are typically small
devices that pump just a few Watts of heat from very low temperatures,
55 to 80 K. The thermodynamic efficiency of these cryocoolers tend to Figure 8. Two-Stage Energy Flow
be in the range of 10 to 15% of Carnot, although, like most other types of refrigerators, larger Stirling
devices show better efficiencies due to the proportionally reduced effects of parasitic heating, so that 20 to
25% (of Carnot) efficiencies become possible. Adaptation of this cryocooler technology to the Venus surface
temperature environment will require a significant re-design to accommodate the much higher 460o C heat
rejection temperature.
10
American Institute of Aeronautics and Astronautics
Figure 7. Stirling Duplex Principles
(a) Heat Pump Schematic
Figure 9. Staged Refrigeration
11
American Institute of Aeronautics and Astronautics
IV.E.
Thermodynamic Two-Stage Refrigeration
By staging the cooling as shown in Fig. 9, the power requirements drop considerably. Instead of employing
a single-pressure vessel that contains all the sensitive equipment and directly refrigerating by pumping heat
from 30 to 500o C, it is considerably more thermodynamically efficient to utilize two or more pumping stages.
Shown in Fig. 8 is a heat flow diagram in which one power convertor and two coolers are employed. The
work required to pump Q1,in Watts
from
 a temperature of T1,in to a temperature of T1,out with a cooler
efficiency of, η1 , is given by:
T1,out
T1,in
−1
η1
 (Q1,in ). The rejected heat of the first-stage cooler will include
the heat lifted, Q1,in , plus the wasted heat from the cooler itself. This rejected heat will in turn need to be
lifted by the second-stage cooler. In addition, heat leakage from the Venus environment will also need to
be lifted. Notice that either increasing the cooler efficiency or reducing the temperature difference reduces
overall energy needs.
IV.F.
GPHS Requirements and Availability
(a) Hot-End Effects on Efficiency
(b) Potential Plutonium Available
Figure 10. Efficiency and Availability Curve
A Stirling convertor’s performance improves signficantly with higher hot-end temperatures due to an
increase in Carnot efficiency and a higher percentage of Carnot efficiency being achieved. For example, in an
MTI Component Technology Power Convertor (CTPC) scaling study,53 the percentage of Carnot achievable
as a function of temperature ratio is shown in Fig. 10(a). Clearly, a higher hot-end operating temperature
provides significant improvements in performance which is necessary due to the limited Plutonium-238 supply
(Fig. 10(b)).
In addition, the 92 bar surface pressure on Venus provides an opportunity for increasing the internal
working fluid pressure and the operating temperature. The higher operating pressure is possible because
the pressure vessel is constrained on the outside by the atmosphere. The higher temperature is also possible
because creep effects are minized since hoop stresses are similarly reduced.
The combination of refrigeration staging and higher temperature operation can significantly reduce the
plutonium required.
12
American Institute of Aeronautics and Astronautics
IV.F.1.
Two-Stage Cooling
The work required to pump heat from a refrigerated compartment to a higher temperature is a function of the cooling efficiency, η1 , hot-end temperature, T1,out , cold-end
temperature, T1,in . Shown in Fig. 11 is the thermodynamic system for a single-stage
Stirling duplex. The heat from the GPHS modules, Qe,in enters the Stirling convertor at temperature, Te,in and produces work in the form of electrical energy and/or
a pneumatic pressure wave, W1,act , and waste/unused heat energy, Qe,out is released
to the Venus environment at temperature, Tenv . The higher the efficiency of the
convertor, the more work is produced and less heat is rejected.
The refrigerator then utilizes the work energy to act as a heat pump lifting
heat from the coldbay, Q1,r,in , at temperature, T1,r,in , and releasing the heat into
the environment. The total heat into the environment, Q1,r,out , is the sum of the
heat lifted and the work input provided to the refrigerator. This rejected energy is Figure 11. Single-Stage
radiated into the Venus environment at temperature, T1,r,out .
In the best case, both the heat engine and refrigerator operate with an ideal Carnot cycle in which no
irreversible losses are present. An energy balance across the heat engine and refrigerator implies:
W1 = Qe,in − Qe,out
(1)
Q1,in = Q1,out − W1
(2)
For a Carnot cycle, there is no net entropy, ∆S =
∆Q
T ,
creation. Therefore, for the heat engine:
Qe,in
Qe,out
= Sout =
Te,in
Te,out
(3)
Q1,in + Qenv
Q1,out
= S1,out =
T1,in
T1,out
(4)
Se,in =
And for the refrigerator:
S1,in =
Then substitute the heat engine entropy relationship, Eq. 3 into the energy balance Eq. 1 yields:
W1 = Qe,in − Qe,out →
Te,out
W1,act
=1−
Qe,in
Te,in
(5)
Similarly, the refrigeration Carnot performance is given by subsituting Eq. 4 into Eq. 2:
T1,out
W1
=
−1
Q1,in + Qenv
T1,in
(6)
With a coldbay temperature of 30o C, and a Venus environment temperature of 470o C, then Eq. 6 yields:
W1
T1,out
470 + 273.15K
=
−1=
− 1 = 1.45
(7)
Q1,in + Qenv
T1,in
30 + 273.15K
This relationship means it takes 1.45 W of ideal work to pump 1W of heat from the coldbay.
Similarly, from Eq. 5, with a hot-end temperature of, 850o C:
W1
Te,out
470 + 273.15
=1−
=1−
= .338
Qe,in
Te,in
850 + 273.15
(8)
This relationship means .338 W of work is produced with 1 W of heat input. The leftover heat is not useable
and is rejected to the environment.
Therefore, to lift 1 W of heat from the coldbay requires:
Qe,in =
W1
1.45
=
= 4.28W
.338
.338
(9)
the 4.28 W of heat from the GPHS module in the case of ideal Carnot efficiency resulting in a duplex system
efficiency of:
Q1,in + Qenv
ηsystem =
= 23.3%
(10)
Qe,in
13
American Institute of Aeronautics and Astronautics
Now if thermodynamic losses are included in this analysis, we will have a lower overall efficiency. For
example, we will assume 30% of Carnot efficiency on the refrigerator, and 55% on the Stirling heat engine.
A 30% efficiency on the refrigerator means the actual work required to lift 1 W of heat is 3.33 times more
than the Carnot ideal.
1 T1,out
W1,act
=
− 1 = 1.45/.3 = 4.83W
Q1,in + Qenv
ηc T1,in
(11)
This means it takes 4.83 W of actual work input to pump 1 W of heat from the coldbay if a more realistic
refrigerator is used.
Similarly, for the heat engine, a 55% of Carnot efficiency applied to Eq.(8) results in:
W1,act
Te,out
= ηe 1 −
= (.55 ∗ .338) = .185
(12)
Qe,in
Te,in
This means .185 W of work is produced with 1 W of heat input.
The nonideal cycle analysis for the duplex system is now:
Qe,in =
4.83
W1,act
=
= 26W
.185
.185
(13)
This means the system efficiency is 3.8% or approximately 5 times more heat input is required to pump the
heat out of the coldbay compared to the ideal Carnot case.
T
env
env
− 1, R2 = TT1,out
− 1, Re = 1 − TTe,in
, then the total single-stage system efficiency can be
Let R1 = T1,out
1,in
written as:
Te,out
1
−
Te,in
Q1,in + Qenv
R
= ηe ηc e
ηsystem =
= ηe ηc (14)
T1,out
Qe,in
R
1
T1,in − 1
In Fig. 12, three system efficiency curves are shown. The effects of increasing the coldbay temperature, the
hot-end temperature, and the combined convertor/refrigerator efficiency, ηcombined = ηc ∗ ηe , are compared.
Clearly, increasing the coldbay temperature has the greatest influence on overall system efficiency. Notice
the system efficiency of all three curves starts at 3.6%. Increasing the hot-end temperature to the maximum
conceivable level of 1200o C provides up to 5% system efficiency, assuming the convertor and refrigerator
perform at 55 and 25% of Carnot. If the product of convertor and refrigerator efficiency, η = ηc ηe decreases,
the system efficiency will decrease as shown. Clearly, any improvements in instrumentation hardness will
have the most impact on overall GPHS requirements.
By increasing the number of stages to
two, significant benefits are achieved.
The energy balance for the power convertor is:
W1 + W2 = Qe,in − Qe,out
(15)
For the first-stage cooler:
W1 = Q1,out − Q1,in
(16)
For the second-stage cooler:
W2 = Q2,out − Qenv − Q1,out
(17)
Initally assuming isentropic, the engine entropy relationship is:
Se,in =
Qe,in
Qe,out
Qe,out
Tenv
= Se,out =
⇒
=
Te,in
Tenv
Qe,in
Te,in
(18)
Figure 12. Single-Stage Trades for System Efficiency
14
American Institute of Aeronautics and Astronautics
The first-stage cooler relationship is:
S1,in =
Q1,in
Q1,out
Q1,out
T1,out
= S1,out =
⇒
=
T1,in
T1,out
Q1,in
T1,in
(19)
The second-stage cooler relationship is:
S2,in + Senv,in =
Q1,out
Qenv
Q2,out
Q1,out + Qenv
T1,out
+
= S2,out =
⇒
=
T1,out
T1,out
Tenv
Q2,out
Tenv
Some substitution:
W1 + W2
Qe,out
Tenv
=1−
=1−
Qe,in
Qe,in
Te,in
And the actual work relation assuming ηe percent of Carnot efficiency due to losses:
W1,act + W2,act
Tenv
= 1−
ηe
Qe,in
Te,in
(20)
(21)
(22)
An expression for Qe,in is then:
For the Carnot cooler:
W1,act + W2,act
Qe,in = env
1 − TTe,in
ηe
(23)
W1
Q1,out
T1,out
=
−1=
−1
Q1,in
Q1,in
T1,in
(24)
Assuming η1 percent of Carnot efficiency due to losses:
W1,act
T1,out
1
=
−1
Q1,in
T1,in
η1
(25)
An expression for Q1,in is then:
Q1,in = W1,act η1
T1,out
T1,in
−1
(26)
Finally, the second stage Carnot cooler:
W2
Q2,out
Tenv
=
−1=
−1
Q1,out + Qenv
Q1,out + Qenv
T1,out
Assuming η2 percent of Carnot effiency dues to losses:
W2,act
Tenv
1
=
−1
Q1,out + Qenv
T1,out
η2
(27)
(28)
An expression for Q1,out + Qenv is then:
Q1,out + Qenv = Qe,in =
W2,act η2
Tenv
T1,out
−1
(29)
(W1,act + W2,act )
Re ηe
(30)
(W1,act η1 )
R1
(31)
Q1,in =
Q1,out + Qenv =
(W2,act η2 )
R2
15
American Institute of Aeronautics and Astronautics
(32)
Now, an energy balance across the first-stage cooler actual flows:
Q1,out = Q1,in + W1,act =
(W1,act η1 )
η1
+ W1,act = W1,act 1 +
R1
R1
(33)
Let R10 = R1 R2 + R1 + R2 .
Note that the one-stage efficiency can be written as:
η 1 η 2 Re
η 1 η 2 Re
Q1,in
=
=
0
Qe,in
R1 R2 + R1 + R2
R1
(34)
(Q1,in + Qenv )R10
W1,act
=
Re η e
Re ηe η1
(35)
W10 η1
R10
0
=⇒
W
=
(Q
+
Q
)
1,in
env
1
R10
η1
(36)
We found for single stage:
Qe,in =
And,
Q1,in + Qenv =
Similar analysis for two stage:
Qe,in =
h
1
Q1,in R
η1 (1 +
η1 R 2
R 1 ) η2
i
2
+ 1 + Qenv R
η2
Re ηe
(37)
Figure 13. Two-Stage Electrical and Pneumatic Stirling Duplex in Pressure Vessel
A trade study was conducted to quantify the size of refrigerator required for typical Venus surface applications. It dramatically illustrates that the use of multistage refrigeration will greatly reduce the amount of
plutonium required to power the system. In this context, multistage refers to multiple refrigerators that operate in series such that the heat rejected by one refrigerator is collected and pumped to a higher temperature by
16
American Institute of Aeronautics and Astronautics
the next one. Fig. 13 illustrates a two-stage design in which the two refrigerators work from a common Stirling
power source. In this example, 700 W of heat is entering the lander from the environment and an additional
400 W of electrical energy is being dissipated as heat energy by the payload. The first-stage cooler pumps the
heat entering the payload up to the intermediate temperature of 250o C, from where the second-stage cooler
pumps this heat out to the environment, along with the waste heat from operation of the first-stage cooler
and the incoming heat leak from the environment through the insulation, for a total of 3000 W. Use of multiple stages allows for the environmental heat to be intercepted and removed at a higher temperature than the
30o C payload, providing major improvements in thermodynamic efficiency. Fig. 14 shows how this improved
thermodynamic efficiency translates into greatly reduced requirements for plutonium, measured in GPHS
modules, which are the building blocks of the RPSs (A GPHS module houses 0.5 kg of plutonium-238).
The GPHS savings as the number of cooling stages increases
and required number of units for four cases: First, cooling of
a Lander system assuming from 100 W to 500 W of electrical
power dissipation at 30o C and 700 W of environmental cooling
with a cooler efficiency of 20% of Carnot for a single stage and
second, repeat with double stage duplex, third and fourth are
the same except the refrigeration efficiency is increased to 30%
of Carnot. (Power at 55% of Carnot, 1200o C Hot-end) Four
cases are shown in Fig. 14.
When cooling a complete lander with 20% of Carnot efficiency coolers, the number of GPHS modules necessary to
provide the required cooling for a single and two-stage refrigerator ranges from 45 to 156. In the cases when higher efficiency
refrigerators are utilized, the mission requires from 37 to 70
GPHS modules. These are comparable numbers to that of the
Figure 14. GPHS Requirements
planned 2020 Outer Planets Flagship mission for which a total
of approximately 40 GPHS modules are planned (that is, when using 5 MMRTGs).
The benefits of using lower power electronics or for having a reduced set of science instruments is clear.
The reduction in GPHS modules is substantial. Even further reductions in GPHS modules can be achieved
by improving the thermal insulation to reduce the heat leak, or by using high temperature electronics to
raise the payload temperature.
Finally, it should be noted that this analysis makes some aggressive assumptions about the achievable
performance of the Venus machine, particularly a refrigerator efficiency of 30% of Carnot and a heat engine
hot end temperature of 1200o C, both of which are beyond what has been demonstrated in any kind of
experimental device to date. Alternate assumptions based on a less capable Venus refrigerator will lead to
a correspondingly larger number of GPHS modules.
IV.F.2.
Availability
NASAs RPSs under development, that is, MMRTG and ASRG, use plutonium-238 housed in GPHS modules. Plutonium availability was identified as a key issue for enabling future NASA missions in all mission
classes, namely for Flagship, New Frontiers, and Discovery class missions. In response, NASA and the U.S.
Department of Energy is assessing plutonium needs for future NASA missions and making necessary steps
to allocate a sufficient inventory to enable these missions. For the near future the primary driver is the next
Outer Planet Flagship Mission to Europa, planned for a 2020 launch with 5 RPSs on the orbiter. Additional
plutonium needs may arise from Discovery and New Frontiers missions, but at a significantly smaller scale
using one or two RPSs each. A potential long-lived Venus Flagship mission could contribute to further
demands on the plutonium inventory. Therefore, future mission studies on alternative Venus mission architectures should assess plutonium needs and work with NASA Headquarters to be included in plutonium-238
production and allocation plans.
Since plutonium-238 production was ceased in the late 1980s, NASA has relied on stockpiles of the
material and purchased the stockpiles from Russia to fuel its radioisotope thermoelectric generators. Once
the Mars Science Laboratory, the Outer Planets Flagship, and a few other missions needing them are supplied,
no further missions are possible. A recent report by the National Research Council Space Studies Board
made a high-priority recommendation that the Department of Energy’s fiscal 2010 budget include funds to
re-establish production of 5 kg of material per year. The agency has requested $30 million in its fiscal 2010
17
American Institute of Aeronautics and Astronautics
budget proposal.
Note that if this schedule is followed, then as shown in Fig. 10(b), the supply of plutonium will be
sufficient for the proposed Venus surface mission. This assumes the Outer Planets Flagship mission will be
launched in 2018, the next flagship mission will be post 2025, the New Frontiers 3 and solar probe missions
use non-RPS power source, new U.S. production starts in 2015, and Discovery and Flagship 2 are deferred.
V.
Component Technology Development
The Stirling duplex comprises various subsystems that must operate within a specific temperature, pressure, and corrosive environment. Some of these systems include a heat source, high temperature alternator,
and variable conductance heat pipe technology. Each subsystem liability and its effect on overall system
performace is identified.
V.A.
GPHS Limits
Figure 15. GPHS Performance Limits
The GPHS provides approximately 250 W of heat with a mass of 1.61 kg. It contains plutonium-238 that
is cladded with iridium as shown in Fig. 15. The GPHS can be stacked to provide additional heat, but must be
arranged to minimize internal heating. The iridium cladding cannot exceed 1335o C during normal operation.
The half-life of plutonium is 87.74 years and provides heat through radioactive decay with emission of alpha
particles that must be vented. The ventilation paths create gaps in the GPHS module that may insulate the
plutonium’s heat from effectively reaching the aeroshell exterior. By immersing the GPHS in a cover gas,
these vented paths will fill with the cover gas and provide an additional path for heat conduction away from
the plutonium. The higher thermal conductivity of the gas, the higher the safe temperature can be achieved
at the Stirling heater head/GPHS junction since the iridium cladding is effectively cooled. Notice in Fig. 15
18
American Institute of Aeronautics and Astronautics
that of the gases, helium provides the highest allowable temperature, 1266o C.
V.B.
High-Temperature Materials
Stirling RPS systems designed for Venus applications have been proposed since
the 1990s. Stirling hot-end material (MarM-247) is being developed in the Advanced Stirling Converter (ASC)/ ASRG project to operate for 17 years at 650o
C design. A single (previous generation) Stirling convertor has been operated
in 2005 over 300 hours with a 850o C hot-end temperature and 90o C coldend temperature with 38% efficiency and 88 W power output with heat input equivalent to 1 GPHS (and 114 W power output with unlimited heat input). While impressive, these are not yet Venus environmental temperatures.
To be truly validated in the Venus surface environment, the cold end temperature
has to be raised from 90 to 480o C, with an expected decrease in overall thermodynamic conversion efficiency. An increase in conversion efficiency could be achieved by
increasing the hot end temperature beyond 850o C to as much as 1200o C. However,
this will require further development for the hot-end material. Maturation for flight
application is on-going: A 7 to 8 W/kg, 17-year life (i.e., 3 years storage plus 14
16. MarM-247
years operations) ASRG is slated for potential use on the Discovery 13 Mission in Figure
Stirling Heater Head
the 2016 timeframe.
For Venus missions of less than 1 year, the current Venus hot end material,
MarM-247, may be suitable for temperatures of up to as high as 977o C and, with
the addition of a protective coating, up to as high as 1077o C. This is in part due to
the high pressure in Venus environment that will reduce the stresses on the material
during operation. Nevertheless, proper testing will be required to quantify the actual
maximum temperature with the existing materials of construction. For even higher
temperatures, a different class of material would be required. NASA GRC conducted
initial development of advanced materials (refractory metal alloys and ceramics)
specifically for high-temperature Stirling applications. Although not fully mature
at the present time, these advanced materials have the capability of operating at
temperatures in the range of 1100 to 1200o C.
Tradeoffs of maximum operating temperature versus required development and
risk need to be investigated in terms of long-term thermal stability, outgassing, and
synergistic effects, for example, the combined effects of radiation, temperature, and
aging time. Identifying the appropriate size for the RPS is also an important issue,
Figure 17. Refractory
Stirling Heater Head
in light of science goals and exploration objectives. Static landers, for example,
may require more power than aerial platforms, but they are less mass and volume constrained. Aerial
platforms, such as the Venus Mobile Explorer concept, traverses using a metallic bellows system, limiting
the suspended mass for the gondola, which accommodates the power and refrigeration systems. Therefore,
future RPS technology development for a Venus RPS with active refrigeration should reflect science drivers
and related mission architectures.
As shown in Fig. 16, current Stirling hot-end material (MarM-247) is being developed in the ASC/ASRG
project to operate for 17 years at 850o C. For Venus missions of less than 1 year, MarM-247 needs to be
evaluated for potential use at temperatures up to 1000o C. The use temperature may be able to be raised to
as high as 1100o C.
For higher temperatures, a different class of refractory material (Fig. 17) would be required. Although not
fully mature at the present time, these advanced materials have the capability of operating at temperatures
as high as 1200o C. From a stength perspective, molybdenum may be a good candidate with a yield stress
of 360 MPa at 1200o C, if oxidation issues can be mitigated with a protective coating or cover gas.
V.C.
High Temperature Linear Alternator
The production of electrical power from dynamic power conversion can be accomplished via magnetohydrodynamic, piezo-electric, permanent magnet, or induction generators. At the relatively high electrical
power requirement of 400 W, a permanent magnet is the most developed option. In most of the mass efficient
designs, the magnet oscillates linearly and is encased within inner and outer stator laminations as shown
19
American Institute of Aeronautics and Astronautics
in Fig. 18. Unfortunately, these magnets are temperature sensitive and performance is highly dependent
upon material selection and use temperature. The CTPC linear alternator was designed for 275o C. The
best high temperature permanent magnets are samarium-cobalt (Sm-Co). With an increase in temperature,
the magnet strength decreases as measured by the residual induction or residual flux density (Br ). Also,
the magnitude of the demagnitizing field which can be tolerated with no loss of Br also decreases. As Br
drops, the coil current must be increased to maintain power output. An increase in coil current increases
the magnitude of the demagnetizing field which in turn limits the available power density. Note in Fig. 18,
three magnetic positions are shown corresponding to its motion during a Stirling cycle. During the magnet’s
”in-stroke”, the magnetic flux encircles (links) the coil in a counterclockwise direction. In the ”out-stroke”
position, the flux direction reverses. At mid-stroke, the magnet flux is localized, and the coil flux linkage is
zero.
dφ dx
The instantaneous voltage produced is: V = N dφ
dt = N dx dt . The peak voltage occurs as the magnet
crosses the mid-stroke position since both maximum velocity and flux linkage change occurs. The current
flowing in the coil is sinusoidal with the peak occuring at midstroke for a tuned circuit. The flux linking the
coil due to coil current with the magnet at its midstroke position results in a demagnetizing field that the
magnets must be able to withstand.
By using a pneumatic duplex, the alternator only needs to provide power for the instruments. This reduces
the expected coil temperatures due to wire resistance losses. The coil requires insulation and potting organics
that are also temperature sensitive. The coil can be wrapped in polyimide-coated fiberglass. Previous studies
have accepted 320o C as a maximum coil temperature. Matrimid 5218 polyimide adhesive was selected.
Figure 18. Alternator Schematic
The ASC is being developed with a maximum alternator temperature of 130o C for 17 years. The Fission
Surface Power convertor is being developed for 150o C using similar materials as ASC. The CTPC was
developed, as part of the SP-100 program, for 275o C alternator temperature for a 60,000-hour life. Longlife, ceramic-coated coil development is still needed, and tradeoffs of maximum operating temperature vs.
required development/risk. Tradeoffs include:
• Long-term thermal stability
• Outgassing
• Synergistic effects, for example, radiation plus temperature plus aging time
• Selection and validation of high-temperature alternatives, especially for 177o C or higher alternator
Primary limit to long life operation is wire insulation. Current commercial technology wire insulation
temperature not to exceed 250o C for 20,000 hour operation (Thermal Class 250). Known Sm-Co type
magnets may be used potentially up to 300o C. Magnet remanence declines with increasing temperatures.
20
American Institute of Aeronautics and Astronautics
V.D.
Organics/Adhesive and Joining Technology
Organic materials are used throughout Stirling convertors and
coolers. They provide functions such as seals, insulation, adhesion, and lubrication. During duplex operation, the organic
materials can deteriorate if operated in a high-temperature environment. Typically, commercial temperature rating is based
on 20,000 hours of operation. In Fig. 19, an example of the
expected lifetime of Viton(FKM), used in sealing applications,
is shown. Note that though it is reported use temperature is
200o C, it could in fact be utilized at 230o C if the mission duration is less than a year. This trend is similar for all organic
compounds that would be utilized in a duplex. Note that in
Table 7, a list of commonly used organics in Stirling convertors is shown along with their unlimited life temperature. The
Figure 19. Viton Heat Resistance
lower temperature Hysol EA9394 would need to be replaced with a higher temperature substitute or the
second coldbay temperature would need to be reduced.
Notice that nearly all the organics would be located in the first- or second-stage refrigeration compartments. The exception, Xylan, would possibly be exposed to high temperature if a free-piston or free-displacer
Stirling is used in which a moving part would exist in the hot end of the convertor. Since the Xylan material
is typically relied on only during startup for gas-bearing-based devices, and not required for flexure-based
bearings, this issue may not be lifelimiting.
Table 7. Organics Applications for Stirling
Organic
Compounds
Viton(FKM)
Silicon
Hysol EA9394
Loctite 2422
Nomex Paper
Polyamide
Polythermalize
Teflon
Tra-bond
Xylan
Matrimid 5218
V.E.
Temp.
o
C
200
300
177
343
220
240
200
260
190
260
250
Variable Conductance Heat Pipe
The Venus lander vehicle will undergo severe changes in environmental temperatures during its complete
mission lifecycle. The range of pressures and temperatures must be accounted for in the design of the Stirling
duplex and multistage pressure vessel.
During launch and cruise, the instruments do not require refrigeration. The coolers may be inactivated
either by direct control of the loading or by shunting the GPHS heat away from the heater head. If the
duplex is disabled via control, then the excess heat will still need to be dissipated safely.
A recently developed variable conductance heat pipe (VCHP) is available for shunting the heat.54 This
allows for the option of commanded stop and restart of Stirling for GPHS installation and taking sensitive
science data with zero vibration and minimal electromagnetic interference. It also offers the ability to protect
the Stirling heater head in the event of an unexpected Stirling duplex shutdown and allow restart if possible.
The VCHP is designed for up to 1000o C operation, although for short-term use and under a higher
environmental pressure such as on Venus, its use temperature could be extended. It would normally be off,
except during transit when excess GPHS heat is produced.
21
American Institute of Aeronautics and Astronautics
A noncondensable gas normally covers the condenser preventing heat pipe operation. When the duplex
stops, the temperature and alkali-metal vapor pressure increase to uncover the condensor and remove GPHS
heat. It turns on with 30o C temperature rise to not affect normal duplex operation. Also, when coupled with
currently available energy storage technology, enables quiet seismometer and magnetic field measurements.
In Fig. 20, the variable conductance heat pipe is utilized to cool the interior of the vehicle when the
duplex is off. Once the lander vehicle begins to heat up due to Venus entry, the duplex is activated to
provide instrument refrigeration and the heat pipe turns off.
Figure 20. Thermal Control During Cruise
VI.
Lander/Launch Vehicle Integration and Testing
RPSs based on direct thermoelectric conversion (i.e., the MMRTG) can easily exceed their 14- year design
lifetime, due in part to the use of well-known materials, rigorous component testing, and a plutonium-238
heat source with an 87.7-year half-life. A major motivation for using RPS on NASA missions is their ability
to produce continuous, reliable electrical power in remote and often severe environments, with no reliance
on sunlight. Some past NASA missions to the outer planets could not have been performed without RPS,
and some spacecraft continue to operate far beyond their original expectation due in large part to the longlife RPS. Since dynamic conversion is about four times more efficient than static conversion, the ASRG
requires about a quarter of the plutonium-238 compared to the MMRTG, while generating the same amount
of electric power and rejecting proportionately less waste heat. Excess heat can be either a benefit or a
shortcoming depending on the mission in question. For example, the Mars Science Laboratory rover, to be
launched in 2011, will use a single MMRTG. On the surface of Mars it will utilize the waste heat to keep
the Warm Electronic Box (WEB) at a desired temperature during the cold nights. Although this excess
heat is desirable on the surface, during the cruise phase while bottled up inside the aeroshell it needs to be
removed and rejected to space. Therefore, MMRTG-enabled missions require more capable cooling systems
during the cruise phase inside an aeroshell than the ones using ASRGs, since the former requires four times
more plutonium-238 than the latter. This would be particularly important to a future mission to Titan
considering five RPSs are to be carried inside an aeroshell, where the generated heat would be 10,000 W(t)
with MMRTGs and 2500W(t) with ASRGs].
In Fig. 21, three lander concepts previously studied are shown: (a). Single vessel/Dual Stage Cooler, (b).
Dual Vessel/Dual Stage Cooler, and (c). Single Vessel/Single Stage Cooler. The Embry-Riddle concept (b)
is a perhaps the optimal configuration since refrigeration is performed on only a subset of the instruments
while utilizing effient multi-stage cooling. The VISM concept (a) clearly shows the GPHS modules and heat
engine would be exposed to the Venus environment and only the alternator requires refrigeration. The single
vessel concept (c) is the typical approach followed in the past, but it also results in a short-lived mission.
22
American Institute of Aeronautics and Astronautics
(a) Early VISM/MTI Concept
(b) Embry-Riddle Concept
(c) Single Vessel
Figure 21. Proposed Lander Configurations
VII.
Manufacturability and Reliability
Manufacturing the duplex will follow the practices employed for many of the previously developed hightemperature Stirling convertors. Full life-testing is possible due to the less than a year total mission time
expected.
VII.A.
Engines Operating at Venus Temperatures
A number of convertors have been designed and operated at a hot-end temperature suitable for operation
on Venus (> 500o C). The cold-ends have not been tested at the Venus reject temperature of 500o C but the
materials and joining technology are similar to the hot-end. The main issue is keeping the alternator and its
organics below 270o C and this is achieved with active refrigeration. With a Stirling duplex, the alternator
can be directly cooled utilizing one stage of the cooler.
An important first development goal is to demonstrate both the power and cooling components in a
relevant Venus environment. The following sections present existing power systems that have operated at
Venus hot-end temperatures.
VII.A.1.
MTI/Foster-Miller CTPC
During the mid-90s Mechanical Technologies Incorporated developed a high-power Stirling convertor, CTPC,
for operation at 777o C and a cold-end of 252o C. The convertor operated at 70Hz, at 15 MPa, and produced
23
American Institute of Aeronautics and Astronautics
Figure 22. Component Test Power Convertor
12 kW. Shown in Fig. 22 is a cross-section of a derivative of that convertor, including its flow field through
heat exchanger tubing. The alternator heats up in this design due to the high power output and associated
losses. The alternator was intended for operation up to 270o C. Notice that refrigerating the alternator with
a multistage duplex would keep a design such as this successfully operating on Venus.
VII.A.2.
ASC/ASRG
At NASA GRC, a number of convertors have operated at 850o C hotend temperature for well over 1 week. At this hot-end temperature,
the convertors are 38% efficient, have a mass of 1.3 kg, operate at
102 Hz, are charged to a pressure of 3.6 MPa, and provide from 88
W up to 114 W depending upon reject temperature. The ASRG is
designed for a 17-year life and slated for potential use on a Discovery
mission in the 2016 timeframe. In Fig. 23, the ASC and a schematic
of its components are shown.
VII.B.
Thermoacoustic Technology
It is possible to eliminate moving parts in the hot end of the
Figure 23. High Temperature ASRG
convertor that is exposed to the Venus environment by adopting
thermoacoustic technology in which the displacer is replaced with a high amplitude acoustical circuit.
Under two separate efforts, thermoacoustic convertors have been successfully designed, built, and tested. In Fig. 24, a Northrop Grumman and Sunpower thermoacoustic convertor are shown. Both convertors produced about the same power after
adjusting for linear alternator differences. The pressure and frequency was 3.65 MPa,
100 Hz and 5.28 MPa, 125 Hz, respectively. Both designs can be readily adapted to
duplex operation, but their efficiency is 25% less than convertors with a displacer.
The Sunpower design is coaxial with heat exchangers surrounding the thermal buffer
tube. The Northrop Grumman design is circular with a thermal buffer tube adjacent
to the environment. Since the ASC is successfully operating at 850o C with a moving
part (displacer), it may be desirable to maintain high efficiency. Additional studies
Figure 24.
Thermo- and testing are required to contrast the relative advantages.
Acoustic Convertors
24
American Institute of Aeronautics and Astronautics
VIII.
Conclusion
In summary, there is considerable technical maturity in the field of Stirling heat engines and refrigerators
that can serve as the foundation for development of refrigerators for Venus. However, substantial technological development is still required given the extreme temperature environment of the Venus surface. The
most significant technical challenges are:
• To combine a Stirling heat engine and refrigerator into a long-lived duplex machine with at least two
stages of cooling.
• To achieve a high thermodynamic efficiency that will keep the GPHS module requirements at a manageable and affordable amount.
• To create a complete system design with the multistage refrigerator integrated into the Venus platform
(lander, rover, and balloon). Surface and near-surface payload compartments are typically spherical
pressure vessels of minimum diameter to limit the environmental heat leak. Integration of a two-stage
Stirling-based refrigerator into this architecture is a challenge given the need to preserve the thermal
insulating properties of the original pressure vessel.
• To address issues arising from the potential electromagnetic or mechanical vibration byproducts of the
Stirling-convertor-based power source and refrigerator that could interfere with scientific instruments.
In particular, there is a concern that the mechanical vibration of the machine could interfere with
seismometry measurements if the Stirling convertor is not physically decoupled from the seismometer.
The Venus Exploration Advisory Group (VEXAG) report specifically identified three out of the four
flagship mission concepts would require radioisotope active cooling. Specifically, a Venus Geophysical Network for determining the internal structure, monitoring seismic activity of the planet, and for monitoring
the circulation of the atmosphere. This included at least three stations on the surface of Venus that operate
for at least one Earth year.
Second, a Venus Mobile Explorer, to acquire and characterize core samples at multiple sites, determine
composition and isotopic measurements of surface and atmosphere while operating in a Venus surface environment for 90 days.
Third, Venus Surface Sample Return, measure isotopic composition of oxygen in surface rocks, trace
elements to characterize core-and-mantle formation, and determine age of returned rocks.
In conclusion, numerous studies over the past 15 years have indicated the need for duplex Stirling
power/cooling on Venus. Stirling convertors have already operated at the required hot-end temperature,
and crycoolers have flown in space since 1971. With modest technology development, a Stirling duplex can
be built to cool a Venus lander and enable exploration for at least a full Venus day duration.
IX.
Acknowledgements
This work performed in this paper was performed for NASA through the Science Mission Directorate for
the Radioisotope Power System (RPS) Program.
This manuscript was possible because of the assistance provided by staff from Glenn Research Center
(GRC), Foster-Miller (FM), Clever-Fellows (CFIC), Gedeon Associates (GA), Northrop-Grumman (NGST),
Sunpower, Los Alamos National Laboratory (LANL), Venus Exploration Advisory Group (VEXAG) including Randy Bowman-GRC; Steve Geng-GRC; Jan Niedra-GRC/ASRC; Jeff Schreiber-GRC; Eugene ShinGRC/OAI; Roy Tew-GRC; Lanny Thieme-GRC; Wayne Wong-GRC, Scott Backhaus-LANL; Pete ChapmanFM; John Corey-CFIC; David Gedeon-GA; Mike Petach-NGST; Jeff Raab-NGST; Ellen Stofan-VEXAG;
Nick Vitale-FM; Tom Walters-FM; and Gary Wood-Sunpower
25
American Institute of Aeronautics and Astronautics
References
1 Kerzhanovich, V. V., Yavrouian, A. H., Hall, J. L., Cutts, J., Baines, K. H., and Stephens, S. K., “Dual Balloon Concept
For Lifting Payloads From The Surface of Venus,” 2006.
2 Kerzhanovich, V., Balaram, J., Campbell, B., Cutts, J. A., Gershman, R., Greeley, R., Hall, J., Klaasen, K., Zimmerman,
W., and Hansen, D., “Venus Aerobot Multisonde Mission: Atmospheric Relay for Imaging the Surface of Venus,” 2000, IEEE
2000, Paper # 0-7803-5846-5.
3 Stofan, E. R. and Saunders, R. S., “Venus Interior Structure Mission (VISM),” 1992, Concept Number 81, Discovery
Missions Workshop.
4 Schmitz, P. and Penswick, L. B., “Venus Lander Power System,” 2007, Internal GRC Study provided at request of JPL.
5 Mellott, K. D., “Power Conversion with a Stirling Cycle for Venus Surface Mission,” 2004, 2nd AIAA IECEC, Provide,
RI, Paper #: AIAA-2004-5622.
6 Mellott, K. D., “Electronics and Sensor Cooling with a Stirling Cycle for Venus Surface Mission,” 2004, 2nd AIAA
IECEC, Provide, RI.
7 JPL, N., “Extreme Environment Technologies for Venus Exploration,” 2008, http://www.nasa.gov/.
8 Team, V., “Venus Exploration Advisory Group Report,” 2006, NASA sponsored study group formed in 2005.
9 Coburn, A., Daniel, E., Frazier, M., Grimsley, E., O’Such, M., Pepich, M., Reed, A., and Richard, H., Concept of Design
Report: Venus Lander , Ph.D. thesis, Embry-Riddle Aeronautical University, Prescott, AZ, Dec. 2008.
10 STDT, “Science and Technology Definition Team Report,” 2008, NASA sponsored study group formed in 2008.
11 Penswick, L. B., Beale, W. T., and Wood, J., “Free-Piston Stirling Engine Conceptual Design and Technologies for Space
Power,” 1990, pp. 136, NASA CR-182168, Contract NAS3-23885.
12 Gedeon, D. and Penswick, B., “Feasibility of a 2 MWe Free-Piston Stirling Engine Power Unit,” 2003, pp. 136, Bettis
Atomic Power Laboratory, West Mifflin, PA.
13 Brown, A. T., “Space Power Demonstrator Engine,” 1992, pp. 1823–1828, NASA CR-179555, Phase I Final Report.
14 Spelter, S. and Dhar, M., “Space Power Research Engine Power Piston Hydrodynamic Bearing Technology Development,”
1989, NASA CR-182136, NASA Contract NAS3-23883.
15 Abdul-Aziz, A., Bartolotta, P., Tong, M., and Allen, G., “An Experimental and Analytical Investigation of Stirling Space
Power Converter Heater Head,” 1995, NASA TM-107013.
16 Shah, A. R., Halford, G. R., and Korovaichuk, I., “Reliability-Based Life Assessment of Stirling Convertor Heater Head,”
2004, NASA/TM-2004-213077.
17 Cairelli, J., “NASA Advanced Refrigerator/Freezer Technology Development Project Overview,” 1994, 8th International
Cryocooler Conference, NASA/TM-1994-106309.
18 Berchowitz, D., “Miniature Stirling Coolers,” june 1993, NEPCON East ’93, National Electronics Packaging and Production Conference, Boston, MA.
19 Berchowitz, D. and Bessler, W. F., “Progress on Free-Piston Stirling Coolers,” may 1993, 6th International Stirling
Engine Conference and Exhibition.
20 Berchowitz, D., “Free-Piston Stirling Coolers for Intermediate Lift Temperatures,” aug 1992, Intersociety Energy Conversion Engineering Conference, San Diego, CA.
21 Berchowitz, D., “Free-Piston Stirling Coolers,” jul 1992, International Refrigeration Conference-Energy Efficiency and
New Refrigerants, Purdue University.
22 Berchowitz, D. and Unger, R., “Experimental Performance of a Free-Piston Stirling Cycle Cooler for Non-CFC Domestic
Refrigeration Applications,” aug 1991, International Congress of Refrigeration, Montreal, Canada.
23 Wertz, J. and Larson, W. J., Space Mission Analysis and Design, Microcosm Press, 2007.
24 Griffin, M. and French, J. R., Space Vehicle Design, American Institute of Aeronautics and Astronautics, 1991.
25 Penswick, L. B. and Urieli, I., “Duplex Stirling Machines,” 1984, pp. 1823–1828, Paper Number 849045.
26 Durda, D. D., “Rise of the Veiled Planet,” Mercury, 2008, pp. 8.
27 Dyson, R. W., Geng, S. W., Tew, R. C., and Adelino, M., “Towards Multiphysics Analysis of Stirling Convertors,”
Engineering Applications of Computational Fluid Mechanics, 2008.
28 Baines, K., Limaye, S., Zahnle, K., and Atreya, S. K., “Exploring Venus with high-altitude balloons: Science objectives
and mission architectures,” 2008, pp. 161, Paper Number: B06-0018-08.
29 Balint, T. S. and Baines, K. H., “Nuclear Polar VALOR: An ASRG-Enabled Venus Balloon Mission Concept,” 2008,
Bibliographic Code: 2008AGUFM.P33A1439B.
30 Balint, T., Thompson, T., Cutts, J., and Robinson, J., “Dual Balloon Concept For Lifting Payloads From The Surface
of Venus,” 2006, Venus Entry Probe Workshop, Netherlands, Jan. 2006.
31 Pantano, D. R., Dottore, F., Tobery, E. W., Geng, S., Schreiber, J., and Palko, J. L., “Utilizing Radioisotope Power
System Waste Heat for Spacecraft Thermal Management,” Oct. 2005, NASA/TM–2005-213990.
32 Schmitz, P. C., Penswick, L. B., and Shaltens, R. K., “A Design of a Modular GPHS-Stirling Power System for a Lunar
Habitation Module,” Nov. 2005, NASA/TM–2005-213991.
33 Landis, G. A. and Schmitz, P. C., “Solar Power System Design for the Solar Probe+ Mission,” July 2008, 6th AIAA
IECEC.
34 Oleson, S., McGuire, M., Sarver-Verhey, T., Juergens, J., Parkey, T., Dankanich, J., Fiehler, D., Gyekenyesi, J., Gilland,
J., Colozza, T., Packard, T., Nguyen, T., and Schmitz, P., “Radioisotope Electric Propulsion Centaur Orbiter Spacecraft Design
Overview,” July 2008, 44th AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
35 Schmitz, P. C., Penswick, L. B., and Shaltens, R. K., “Stirling Isotope Power Systems for Stationary and Mobile Lunar
Applications,” Nov. 2007, NASA/TM–2007-214426.
36 Foster, A. M., “Analysis of Past, Present, and Potential Venusian Instruments,” 2008, NASA Lercip Intern Rept.
26
American Institute of Aeronautics and Astronautics
37 Pollack,
“Trace Gas Abundances in the Deep Atmosphere of Venus from Near-Infrared Spectroscopy,” 1992.
R., “Cryocoolers for Superconducting Machines,” 2008, Short Course Given to NASA, Nov. 2008.
39 Dhar, M., “Stirling Space Engine Program,” 1999, NASA/CR-1999-209164/VOL1.
40 Dudzinski, L., “Radioisotope Power for NASA’s Space Science Missions,” 2008, Briefing to Outer Planets Advisory
Group.
41 Schock, A., “Integration of Radioisotope Heat Source with Stirling Engine and Cooler for Venus Internal-Structure
Mission,” 1993, 44th Congress of the International Astronautical Federation.
42 Turpin, J. B., “Conceptual Trade Study of General Purpose Heat Source Powered Stirling Convertor Configurations,”
2007, NASA/TM-2007-215132.
43 Mackwell, S. and Baines, K., “VEXAG Report on High-Priority Technology Development Requirements,” 2006, Memo
from Chair, VEXAG chairs, Planetary Formation and Evolution Focus Group & Atmospheric Evolution Focus Group.
44 Gershman, R., Nilsen, E., and Sweetser, T., “Venus Surface Sample Return,” 2000, IEEE 2000.
45 Rodgers, D., Gilmore, M., Sweetser, T., Cameron, J., Chen, G., Cutts, J., Gershman, R., Hall, J., Kerzhanovich, V.,
McRonald, A., Nilsen, E., Petrick, W., Sauer, C., Wilcox, B., Yavrouian, A., and Zimmerman, W., “Venus Sample Return: A
Hot Topic...” 2000, IEEE 2000.
46 Brake, T. and H.J.M., “State-of-the-art review on low-power cryocoolers,” 2002, IECEC19, Grenoble, France, 2002.
47 Ross, R. G., “Aerospace Coolers: a 50-year Quest for Long-life Cryogenic Cooling in Space,” 2005, Cryogenic Engineering
Conference, Keystone, CO.
48 Ross, R. G. and Boyle, R. F., “NASA Space Cryocooler Programs–An Overview,” 2002, International Cyocooler Conference, June 2002.
49 Harrison, R. and Landis, G. A., “Batteries for Venus Surface Operation,” 2008, 6th IECEC AIAA-2008-5796.
50 Taverna, M. A., “Mercury and Venus Sample Returns Eyed,” feb 1999, Aviation Week & Space Technology.
51 Dudenhoefer, J., Thieme, L., Kohout, L., McKissock, B., and Hoffman, “A Stirling Power/Cooler Subsystem for the
Team X Venus Lander Study,” oct 1998, Internal Report.
52 Hall, J. L., private communication, NASA JPL, 2008.
53 Jones, D., “Space Power Free-Piston Stirling Engine Scaling Study,” 1989, pp. 141, NASA CR-182218, Contract NAS325148.
54 Tarau, C., Walker, K. L., and Anderson, W. G., “High Temperature Variable Conductance Heat Pipes for Radioisotope
Stirling Systems,” , pp. 8, SPESIF, Feb. 2009, Huntsville, AL.
38 Radebaugh,
27
American Institute of Aeronautics and Astronautics