44th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit AIAA 2008-4825

21 - 23 July 2008, Hartford, CT

In-FEEP Qualification Test Program for LISA Pathfinder

C. Scharlemann1, A. Genovese2, N. Buldrini3, R. Schnitzer4 and M. Tajmar5

Austrian Research Centers GmbH-ARC, 2444 Seibersdorf, Austria

The Laser Interferometer Space Antenna project (LISA) is a co-operative program

between ESA and NASA to detect gravitational waves by measuring distortions in the space-
time fabric. For this task, three satellites will fly in a triangular formation. LISA Pathfinder
is the precursor mission to LISA designed to validate the core technologies intended for
LISA. One of the enabling technologies is the micro-propulsion system necessary to achieve
the uniquely stringent propulsion requirements.

ARC together with Astrium GmbH is jointly developing a flight design for an Indium
FEEP Microthruster-Cluster suitable for LISA Pathfinder. An extensive test program with
three different demonstration units (DM) will provide the verification of the Indium Needle
FEEP technology and the general thruster cluster design. The DMs are flight identical units
in terms of their design and with regard to the Needle FEEP thruster they are equipped
with. Each of the DMs will have to pass through its individual test program. Amongst others,
this includes environmental testing (vibration and thermal tests), off-nominal testing, thrust
balance testing, and a lifetime test phase over several thousand hours.

In particular the lifetime test will provide the most extensive data basis for the system
verification. The lifetime testing will be conducted in the ARC testing facility with a flight
representative unit including an elegant Bread Board unit of the PCU, provided by Galileo
Avionica (Milan plant, Italy), and a neutralizer provided by Thales Alenia Space (Florence
plant, Italy).

In preparation of the upcoming test an assessment of the beam shape with the ARC beam
diagnostic facility has been conducted. The results show clearly that the beam of an
individual LMIS is close to be rotational symmetric. The beam divergence half angle was
evaluated to be equal to 25.5° ±2.5°.

I. Introduction

T he LISA (Laser Interferometer Space Antenna) is a co-operative program between ESA and NASA to detect
gravitational waves by measuring distortions in the space-time fabric. The program consists of two space
missions: LISA Pathfinder, to be launched in 2009, and LISA itself, scheduled to launch in 2013. LISA will consist
of three spacecrafts flying in a triangular formation with a side length of several million kilometers. The position of
each satellite with respect to its two counterparts has to be controlled with an accuracy of 10-9 m to ensure sufficient
accuracy of the scientific measurements. The extreme challenge in position control can only be satisfied with an
ultra precise propulsion system such as an Indium FEEP thruster. LISA will demonstrate for the very first time a
near perfect gravitational free fall to detect gravitational waves.

LISA Pathfinder (LISA PF) is the precursor mission to LISA designed to validate the core technologies intended
for LISA. In general the same challenging propulsion requirements of LISA are also required for LISA PF (see table
1). The micro-propulsion system is one of the enabling technologies for LISA as well as for LISA Pathfinder. In

1
LISA PF Project Manager, Email: carsten.scharlemann@arcs.ac.at, Member AIAA
2
Research Scientist, Email: angelo.genovese@arcs.ac.at
3
Research Scientist, Email: nembo.buldrini @arcs.ac.at
4
Research Scientist, Email: reinhard.schnitzer@arcs.ac.at, Member AIAA
5
Head of Space Propulsion & Advanced Concepts, Email: martin.tajmar@arcs.ac.at, Member AIAA

1
American Institute of Aeronautics and Astronautics

Copyright © 2008 by Carsten Scharlemann. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.
2006 the Space Propulsion & Advanced Concepts Business Division of the Austrian Research Centers (ARC) was
commissioned by the European Space Agency to develop the micro-propulsion system for those missions. The
micro-propulsion system under development is based on a Liquid Metal Ion Source (LMIS) technology which was
developed at ARC over the last three decades. The ARC LMIS technology is already used in different space
application (mass-spectrometry, space-charge compensation) and has logged more than 12,000 hrs of in-space
operation (see table 2).

II. Liquid Metal Ion Source (LMIS)

The Indium Liquid Metal Ion Sources (In-LMIS) have been developed by the Austrian Research Centers over
the last 3 decades. Initially used exclusively for mass spectrometry, ARC soon developed them further for charge
compensation/control on satellites. Up to this date the Indium LMIS technology is the only flight proven liquid
metal ion source and has logged more than 15,000 hrs in space on various missions (see table 1).

Table 1. Space Experience of ARC Indium LMIS.

Experiment Function Spacecraft Nr. of LMIS Operation Time

LOGION Test of LMIS in µ-Gravity MIR 1 24 h (1991)
MIGMAS/A Mass Spectrometer MIR 1 120 h (1991-94)
EFE-IE S/C Potential Control GEOTAIL 8 600 h (1992 -)
PCD S/C Potential Control EQUATOR-S 8 250 h (1998)
ASPOC S/C Potential Control CLUSTER 32 Ariane 5 Launch Failure 1996
Still operational after Crash

ASPOC-II S/C Potential Control CLUSTER-II 32 6515 (2000 -)

COSIMA Mass Spectrometer ROSETTA 2 Launched 2004 (tested in space)
ASPOC/DSP S/C Potential Control DoubleStar 4 4456 h (2004 -)

Realizing the potential of the LMIS for highly accurate propulsion tasks, ARC initiated the development of Field
Emission Electric Propulsion (FEEP) based on the technology developed over several decades. Over the last decade
the development of LMIS based FEEPs has achieved several major development milestones. A large number of
tests1,2 succeeded in demonstrating the technological maturity including the demonstration of key performance
parameters (see table 3). More recent development efforts achieved to prove the sufficient propellant reservoir size
for long term missions, showed the clusterability to achieve thrusts beyond 100 µN and verified the extremely low
thrust noise and the high thrust controllability of this technology. With several tests of durations up to 5000 hrs of
uninterrupted operation, ARC was the first to show experimentally that the In-FEEP technology has also a sufficient
lifetime for missions such as Microscope, LISA and similar missions3,4. Furthermore, in a recently concluded long
duration test (3000 hrs), a LISA PF flight representative cluster unit has been tested and its performance was
assessed. During this test it was found that the ARC In-FEEP matches and even exceeds the rigid LISA PF
requirement.

Based on the achievements described above, ARC was chosen to qualify the In-FEEP technology for application
on LISA Pathfinder and subsequently on LISA (Laser Interferometer Space Antenna)

Table 3. Single LMIS Characteristics.

Parameter Values
Thrust 0.1 – 15 µN / Emitter *
Thrust Resolution < 0.1 µN
Thrust Noise < 0.1 µN/√Hz **
Minimum Impulse Bit < 5 nNs
Total Impulse 490 -1000 Ns / Emitter ***
Specific Impulse 4000 - 8,000 s
Singly Charged Fraction 98%
Electrical Efficiency 95% ****
*
peaks up to 30 µN are possible.
from 10 Hz to 10-4 Hz
**
***
Using the present reservoir size of 15 g, larger sizes are possible, depending
on actual thrust profiles
****
Comparing the current to the emitter with the current in the ion beam (minus
extractor and plume shield current losses)

2
American Institute of Aeronautics and Astronautics
 III. In-FEEP Cluster Assembly
LISA and its precursor mission, LISA Pathfinder, have propulsion requirements unprecedented in the history of
space propulsion. Highly challenging propulsion system requirements (see table 3) in terms of thrust range,
controllability and thrust noise, to name only some of the challenging requirements, make the FEEP technology
basically the only propulsion system capable to satisfy the needs for LISA.

In order to produce the maximum thrust level of 100 µN as required for LISA Pathfinder, nine LMIS are
operated in parallel by a single HV power supply as one FEEP Thruster Cluster Assembly (TCA). The number of 9
emitters is the result of a trade-off between lifetime and available power. By this clustering approach the Indium
FEEP thruster assembly provides an inherent redundancy. In case one LMIS element fails the remaining 8 LMIS
continue to provide the commanded thrust as the cumulative beam current is still equivalent to the selected voltage.
Especially during science mode, when the thrust level is expected to be between 0.1 µN and about 30µN such a
failure of up to 6 emitter would have no significant effect in terms of required thrust.

Table 3. LISA PF Key Thruster Requirements.

Minimum Thrust 0.3 µN (Target 0.1 µN)

Maximum Thrust 100 µN (Target 150 µN)
Total Impulse 2920 Ns (Target 4000 Ns)
Thrust Noise < 0.1 µN/Hz (from 0.01 – 10 Hz)
Thrust Resolution 1 µN
Specific Impulse > 4000 s
Beam Divergence < 35°

In order to transfer the ARC FEEP technology into a flight model, ARC joined forces with Astrium GmbH as
industrial partner covering the experience from electric propulsion flight programs to optimized mechanical/thermal
design and space qualified high voltage experience5. The thruster hardware (shown in Figure 1 and Figure 2) has
successfully passed the Preliminary Design Review (PDR) and is momentarily passing through a series of
qualification tests. The thruster (FCA) consists of 4 thruster sides (TCA). Each TCA consists of 9 LMIS.

IV. In-FEEP qualification tests

The LISA PF propulsion system qualification requires a series of tests. This includes performance evaluations on
LMIS level but also on FCA (thruster) level. Furthermore, direct thrust measurements will establish the accuracy of
the thrust calculation derived from electrical parameters and the directly measured values. Environmental test will
establish the ability of the thruster to operate in the harsh space environment as well as the capability to survive the
launch phase. The former will be established by a thermal vacuum test conducted in a thermal shroud developed
especially for this test (see Figure 3). The vibration test will be conducted on a vibration stand from Astrium GmbH,
Germany. Finally, a so-called off-nominal test will provide insight into the thruster system behavior under
conditions even more extreme than anticipated or required during the LISA PF mission such as thruster temperatures
out of the standard range, significantly increased thrust levels, failure of single thruster components etc.

Finally, the FEEP thruster will be tested for at least 6000 hours. This test will allow investigation of the long
term behavior of the LMIS as well as the general design of the FCA. The test will be run at the ARC test facilities in
one of the large vacuum chambers. Although based on the experience obtained with the recent 3000 hrs test6, no
major difficulties in such a long duration test are expected, special precautions are taken to ensure that the test will
run without any interruptions. This includes a collector plate covered with indium foil to prevent back-sputtering of
collector plate material and subsequent contamination of the FCA, a redundant pump system, and several
independent power grids and UPS.

In order to allow a smooth flow in this extensive test series a total of three identical thruster units are built. Each
of the units will undergo a dedicated series of test. For example while one thruster unit will be utilized for thrust
measurement at Thales Alenia Space, Italy, a second one will be utilized for off-nominal test at ARC while the third
will be utilized for the environmental and lifetime tests.

3
American Institute of Aeronautics and Astronautics
 Figure 1. Complete FCA.

Cluster cold subassy (4x)

Outer housing (cold) (1x)

Isostatic mounts complete (4x)

(incl. thermal isolators)
Cluster hot subassy (4x)

Inner frame (hot)

Baseframe (1x)

Figure 2. FCA main and subassemblies with hot and cold assembly.

4
American Institute of Aeronautics and Astronautics
 Figure 3: Schematic depiction of the assembled DM in the thermal shroud (left) and the thermal shroud (right

One of the first tests in this qualification series is a performance measurement of the thruster. The test will show
the compliance to a rather larger number of mission requirements such as thrust range (0.3 – 100 µN), thrust
resolution (0.3µN), thrust stability and many others. Furthermore the beam divergence will be assessed, both of each
of the 9 individual emitters from a TCA and from all 9 emitters firing together. This test will be conducted with the
ARC beam diagnostic facility6 (see Figure 4). With the beam diagnostic facility, the beam divergence (see and thrust
vector accuracy can be evaluated for the complete TCA as well as for individual emitters. To facilitate the tests of
individual emitters a dedicated control box equipped with HV voltage suitable switches was designed. This box
allows the controlled switch on and off of individual emitters (a feature which is not possible during the LISA PF
mission).

Figure 4: Schematic depiction of the ARC beam diagnostic (left) and its integration into the vacuum chamber
(right)

Figure 5: Schematic depiction of the LMIS beam divergence.

5
American Institute of Aeronautics and Astronautics
 V. In-FEEP Beam Diagnostic
To qualify the beam diagnostic device as well as the testing facility and test software a bread board unit equipped
with three emitters has been tested and assessed with the beam diagnostic (see Figure 6). The three emitters chosen
for this test were actually emitters rejected in the LISA PF emitter manufacturing process at ARC. The rejection was
due to slight imperfections in needle geometry or a mass efficiency too low to satisfy the LISA PF requirements.

Figure 6: Beam diagnostic device (left) and the TCA bread board (right)

Initial tests were performed to investigate if or to what level the beams of individual emitters interfere with each
others. Each scan is repeated three times over a period of roughly 60 sec and the mean of the three measurements is
calculated as demonstrated in Figure 7. Since the wire probe has no biasing, this measurement does not reliably
evaluate the beam current (secondary electrons). The beam diagnostic measurements have only the task to evaluate
the general beam shape and in particular the beam divergence only the relative current values and not the absolute
ones are required. Therefore it is standard practice to normalize the results with the peak value. As shown in Figure
7, the beam of a LMIS has a very distinctive signal. The beam itself is nearly 100% limited to a defined volume with
very sharp boundaries (see sharp rise of the signal on both sides) with the exception of the very left corner. The
reason for this slight asymmetry (a slight offset) is not yet understood. The beam center has a very distinct peak
which is assumed to be an effect of the focusing electrode and the resulting higher current flux in the middle of the
beam.

Individual emitter beam scan

1.2

1
Beam current [a.u.]

0.8

0.6

0.4

0.2

0
0 50 100 150 200 250 300
x-coordinate [mm]

Figure 7: 3 beams scans of an individual emitter and their arithmetical mean (solid line)

To investigate the question if the individual beams of the emitter somehow interfere with each other a scan of the
beam was conducted with all three emitters operating at a total thrust level of 20µN. Each of the emitters contributes
to this total thrust according to its individual characteristic. Following the total scan, scans of individual emitters
were conducted while the other two emitters were switched off. Figure 8 shows very clearly that concerns about a

6
American Institute of Aeronautics and Astronautics
beam distortion or similar effects due to interfering beams of individual emitters are unfounded. The beam has the
same shape independent if all the emitters are running simultaneously or if their individual beam shapes are
superimposed. Furthermore, the signal of the poorly operating emitter 1 shows that a total beam scan can also supply
accurate information about the health/performance of an individual emitter.

Total beam scan Individual emitter beam scan

4 4

3.5 3.5

Superimposed emitter 1 to 3
3 Measured emitter 1 to 3 3

Measured current [a.u.]

Measured current [a.u.]

2.5 2.5

2 2 Emitter 3

1.5 1.5

1 1 Emitter 2

0.5 0.5 Emitter 1

0 0
0 50 100 150 200 250 0 50 100 150 200 250
x-Axis position [mm] x-Axis position [mm]

Figure 8: Comparison between the measured total thruster beam and the one resulting from superimposing
individual emitter beams

While the beam scan depicted in Figure 7 show that the beam is symmetric, with reference to the beam center,
this scan was only done in the x direction. Although by design the beam should be rotational symmetric this
symmetry could be distorted by e.g. a poor emitter alignment. For a full assessment if the beam is rotational
symmetric both scan directions, x and y, are needed. The need for investigation of the beam symmetry is derived
from the requirement that the thrust direction should be deviate only 3° from a nominal thrust direction given by a
line perpendicular to the thruster front face. Figure 9 depicts the result of a x and y scan of an individual emitter
beam. The rotational symmetry of the beam is clearly given in terms of the sharply defined beam boundaries.
Although there are small differences in the beam shape, they have only minor impact in terms of the thrust direction
and can be neglected.

1.8

1.6

1.4
Beam current [a.u]

1.2 x scan

0.8 y scan

0.6

0.4

0.2

0
-150 -100 -50 0 50 100 150
Distance from beam center [mm]

Figure 9: Comparison of the x and y scan of the beam

To evaluate the beam divergence it is necessary to scan the beam at least at two positions downstream of the
thruster. The beam divergence will cause a broadening of the base width of the measured signal. The knowledge of
the degree of broadening and the relative downstream distance where the two measurements have taken place allows
then to evaluate the beam divergence. Figure 10 shows a comparison of beams scans at three downstream locations

7
American Institute of Aeronautics and Astronautics
from the thruster. The broadening of the base is clearly seen and corresponds to a beam divergence half angle of
23°±2.5° independent if the x or y beam scan is used for this assessment.

2.5

2
Beam current [a.u]

1.5

z = 251.8 mm
0.5 z = 231.8 mm
z = 211.8 mm

0
0 50 100 150 200 250 300 350
x axis position [mm]

Figure 10: Beam scans at three different positions downstream of the thruster (z measured from the needle tip)

VI. Conclusion
A first test of the beam shape of emitters operating in a flight representative thruster unit was conducted. The
results show clearly that individual emitter beams do not interfere with each other. Furthermore, the beam has
sharply defined boundaries. The cross section of a beam is circular indicating a high degree of rotational symmetry.
The small differences between the x and y scan indicate a resulting thrust vector deviating only in a minor way from
the ideal line (perpendicular to the thruster face).

Evaluation of the beam divergence has shown that the beam divergence half angle is within the measurement
accuracy below the LISA PF requirement.

A refined analysis of the LISA PF requirements and the degree of compliance respectively will be possible with
an upcoming test with a TCA fully equipped with 9 emitters.

Acknowledgments
We would like to thank the LISA PF team at ESA and ASTRIUM UK for their support and the LISA PF
opportunity.

References
1
Tajmar, M., Genovese, A., and Steiger, W., "Indium FEEP Microthruster Experimental Characterization", AIAA Journal of
Propulsion and Power, Vol. 20, No. 2, 2004, pp. 211-218
2
Genovese, A., Buldrini, N., Tajmar, M., Tamas, E., Vasiljevich, I., Marhold, K., and Ruedenauer, F., "Indium FEEP Cluster
Development", AIAA Joint Propulsion Conference, AIAA-2005-4385, 2005
3
Genovese, A., Tajmar, M., Buldrini, N., and Steiger, W., "2000h Endurance Test of an Indium FEEP Microthruster Cluster",
AIAA Journal of Propulsion and Power, Vol. 20, No. 2, 2004, pp. 219-227
4
Genovese, A., Buldrini, N., Andres, K., and Tajmar, M., "5000h Endurance Test of an Indium FEEP 2x2 Cluster", AIAA
Joint Propulsion Conference, AIAA-2006-4827, 2006
5
M. Tajmar, C. Scharlemann, A. Genovese, N. Buldrini, M. Boss, R. Killinger, “Indium FEEP Micropropulsion Subsystem
for LISA Pathfinder”, AIAA-06-4826, Joint Propulsion Conference, Sacramento, CA, 2006
6
Tajmar, M., Marhold, K., and Kropatschek, S., "Three-Dimensional In-FEEP Plasmadiagnostics", International Electric
Propulsion Conference, IEPC-03-161, 2003

8
American Institute of Aeronautics and Astronautics