Space Sci Rev (2007) 131: 247?338
DOI 10.1007/s11214-007-9266-3
The Mercury Dual Imaging System on the MESSENGER
Spacecraft
S. Edward Hawkins, III ? John D. Boldt ? Edward H. Darlington ? Raymond Espiritu ?
Robert E. Gold ? Bruce Gotwols ? Matthew P. Grey ? Christopher D. Hash ?
John R. Hayes ? Steven E. Jaskulek ? Charles J. Kardian, Jr. ? Mary R. Keller ?
Erick R. Malaret ? Scott L. Murchie ? Patricia K. Murphy ? Keith Peacock ?
Louise M. Prockter ? R. Alan Reiter ? Mark S. Robinson ? Edward D. Schaefer ?
Richard G. Shelton ? Raymond E. Sterner, II ? Howard W. Taylor ?
Thomas R. Watters ? Bruce D. Williams
Received: 24 July 2006 / Accepted: 10 August 2007 / Published online: 23 October 2007
? Springer Science+Business Media B.V. 2007
Abstract The Mercury Dual Imaging System (MDIS) on the MESSENGER spacecraft will
provide critical measurements tracing Mercury?s origin and evolution. MDIS consists of a
monochrome narrow-angle camera (NAC) and a multispectral wide-angle camera (WAC).
The NAC is a 1.5? field-of-view (FOV) off-axis reflector, coaligned with the WAC, a four-
element refractor with a 10.5? FOV and 12-color filter wheel. The focal plane electronics
of each camera are identical and use a 1,024 ? 1,024 Atmel (Thomson) TH7888A charge-
coupled device detector. Only one camera operates at a time, allowing them to share a com-
mon set of control electronics. The NAC and the WAC are mounted on a pivoting platform
that provides a 90? field-of-regard, extending 40? sunward and 50? anti-sunward from the
spacecraft +Z-axis?the boresight direction of most of MESSENGER?s instruments. On-
board data compression provides capabilities for pixel binning, remapping of 12-bit data
into 8 bits, and lossless or lossy compression. MDIS will acquire four main data sets at Mer-
cury during three flybys and the two-Mercury-solar-day nominal mission: a monochrome
S.E. Hawkins, III (

) ? J.D. Boldt ? E.H. Darlington ? R.E. Gold ? B. Gotwols ? M.P. Grey ? J.R. Hayes ?
S.E. Jaskulek ? C.J. Kardian, Jr. ? M.R. Keller ? S.L. Murchie ? P.K. Murphy ? K. Peacock ?
L.M. Prockter ? R.A. Reiter ? E.D. Schaefer ? R.G. Shelton ? R.E. Sterner, II ? H.W. Taylor ?
B.D. Williams
The Johns Hopkins University Applied Physics Laboratory, Laurel, MD 20723, USA
e-mail: ed.hawkins@jhuapl.edu
R. Espiritu ? C.D. Hash ? E.R. Malaret
Applied Coherent Technology, Herndon, VA 20170, USA
M.S. Robinson
School of Earth and Space Exploration, Arizona State University, Box 871404, Tempe, AZ 85287-1404,
USA
T.R. Watters
Center for Earth and Planetary Studies, National Air and Space Museum, Smithsonian Institution,
Washington, DC 20013, USA
248 S.E. Hawkins et al.
global image mosaic at near-zero emission angles and moderate incidence angles, a stereo-
complement map at off-nadir geometry and near-identical lighting, multicolor images at
low incidence angles, and targeted high-resolution images of key surface features. These
data will be used to construct a global image base map, a digital terrain model, global maps
of color properties, and mosaics of high-resolution image strips. Analysis of these data will
provide information on Mercury?s impact history, tectonic processes, the composition and
emplacement history of volcanic materials, and the thickness distribution and compositional
variations of crustal materials. This paper summarizes MDIS?s science objectives and tech-
nical design, including the common payload design of the MDIS data processing units, as
well as detailed results from ground and early flight calibrations and plans for Mercury
image products to be generated from MDIS data.
Keywords MESSENGER ? Mercury ? Imaging ? Camera ? Imager ? CCD ? Heat pipe ? Wax
pack ? Photometry ? Stereo
1 Introduction
Mariner 10, from its three flybys of Mercury in 1974?1975, provided a reconnaissance view
of one hemisphere and measured the planet?s magnetic field and interaction with the space
environment. No spacecraft has returned in the intervening 30 years, however, and our
knowledge of Mercury?s composition, origin, and evolution is therefore limited. From its
high bulk density, Mariner 10 observations (Murray 1975), and Earth-based remote sensing,
Mercury is known to have a high metal-to-silicate ratio, a crust low in FeO (Rava and Hapke
1987; Vilas 1988; Blewett et al. 1997; Robinson and Taylor 2001), and an exosphere with
such species as Na and K (Potter and Morgan, 1985, 1986). Even though our understanding
of Mercury?s bulk composition is limited, some constraints on models of planetary formation
and evolution are possible. If Mercury condensed from the inner refractory portion of a hot
early nebula, it should be strongly deficient in volatiles and FeO (e.g., Lewis 1972, 1974).
The possibility that Mercury?s semimajor axis experienced large excursion during growth
of the inner planets (Wetherill 1994) is permissive of Mercury having greater fractions of
volatiles and FeO.
Mariner 10 images showed a heavily cratered surface grossly similar to that of the Earth?s
Moon (Murray et al. 1975; Spudis and Guest 1988). One of the more distinctive morphologic
features discovered by Mariner 10 is a class of tectonic features known as lobate scarps,
interpreted to reflect large-scale contractional deformation of Mercury?s crust. Lobate scarps
are thought to be the surface expression of thrust faults formed as the planet?s interior cooled
and contracted, possibly during a period in which tidal despinning was also occurring (Strom
et al. 1975; Cordell and Strom 1977; Melosh and Dzurisin 1978; Pechmann and Melosh
1979; Melosh and McKinnon 1988; Watters et al. 2004). Another distinguishing feature of
Mercury is the smooth plains. Smooth plains are comparable in morphology to lunar mare
deposits, but they lack the distinctive low albedo of their lunar counterparts, because of
the very low FeO content (Trask and Guest 1975; Strom 1977; Kiefer and Murray 1987;
Rava and Hapke 1987; Spudis and Guest 1988). Whether the smooth plains are volcanic
or impact deposits is still debated (Wilhelms 1976; Kiefer and Murray 1987; Robinson and
Lucey 1997).
Earth-based radar images of Mercury rival those obtained by Mariner 10 (Harmon et al.
2001). More importantly, radar led to the discovery of an anomalous class of materials inside
permanently shadowed crater interiors in both polar regions. These materials exhibit high
The Mercury Dual Imaging System on the MESSENGER Spacecraft 249
radar reflectivity and a circular polarity inversion consistent with a volume scatterer (Slade
et al. 1992; Harmon and Slade 1992). Water ice remains the leading candidate material to
explain the shadowed deposits, but many unanswered issues remain and final resolution
must await orbital observations (Harmon et al. 2001).
The MErcury Surface, Space ENvironment, GEochmistry, and Ranging (MESSENGER)
spacecraft was conceived, designed, and built to address six fundamental science questions
regarding the formation and evolution of Mercury (Solomon et al. 2001). (1) What planetary
formational processes led to the planet?s high metal-to-silicate ratio? (2) What is Mercury?s
geological history? (3) What are the nature and origin of Mercury?s magnetic field? (4) What
are the structure and state of Mercury?s core? (5) What are the radar-reflective materials at
Mercury?s poles? (6) What are the important volatile species and their sources and sinks on
and near Mercury?
The process of selecting the scientific instrumentation to investigate these diverse ques-
tions balanced the available mission resources for mass, power, mechanical accommodation,
schedule, and money. For MESSENGER, the mass and mechanical accommodation issues
were very significant design constraints. The payload mass was limited because of the large
amount of propellant needed for orbital insertion. The mechanical accommodation was dif-
ficult because of the unique thermal constraints faced in the mission. Taking into account all
these factors, MESSENGER carries seven miniaturized instruments (Gold et al. 2003): the
Mercury Dual Imaging System (MDIS), Gamma-Ray and Neutron Spectrometer (GRNS),
X-Ray Spectrometer (XRS), Mercury Laser Altimeter (MLA), Magnetometer (MAG), Mer-
cury Atmospheric and Surface Composition Spectrometer (MASCS), and Energetic Particle
and Plasma Spectrometer (EPPS). Additionally, a radio science investigation will address
key measurements such as Mercury?s physical libration and gravity field.
The MESSENGER spacecraft was launched from Cape Canaveral on August 3, 2004, in
a spectacular nighttime launch. On August 1, 2005, the spacecraft successfully completed
an Earth gravity assist to slow the spacecraft and redirect it toward the inner solar system. En
route to its primary mission at Mercury, MESSENGER experiences two Venus flybys and
three Mercury flybys. The first Venus flyby occurred on October 24, 2006, and the second
occurred on June 5, 2007. The three Mercury flybys will take place on January 14, 2008,
October 6, 2008, and September 29, 2009, during which regions unexplored by Mariner 10
will be imaged by MDIS. Mercury orbit insertion will occur on March 18, 2011, and the
spacecraft will begin the orbital phase of its mission, which is one Earth year in duration.
The orbital mission is slightly longer than two Mercury solar days.
2 MDIS Measurement Objectives and Design Implementation
MDIS consists of two cameras, a monochrome narrow-angle camera (NAC) and a multi-
spectral wide-angle camera (WAC), coaligned on a common pivot platform. The passively
cooled detectors in each camera are thermally tied to its complex thermal system. This
arrangement allows the detectors to be maintained within their operating temperature, even
during the hottest portion of the orbit at Mercury. The pivot platform provides an added de-
gree of freedom to point the dual cameras with minimal impact on the spacecraft. The full
design details of the instrument are given in Sect. 3. Specifications of the two cameras, given
in Table 1, are tailored to the orbit and imaging requirements of the MESSENGER mission.
MESSENGER will be placed in a highly eccentric orbit with a periapsis altitude of
200 km, a periapsis latitude of ?60?N, and an apoapsis altitude of 15,200 km. The orbit
has a 12-hour period, is inclined 80? to the planet?s equatorial plane, and is not Sun synchro-
nous. During one Mercury solar day (noon to noon), the planet completes three full rotations
250 S.E. Hawkins et al.
Table 1 MDIS camera specifications
Narrow angle Wide angle
Field of view 1.5? ? 1.5? 10.5? ? 10.5?
Pivot range ?40? to +50?
(observational) (Sunward) (Planetward)
Exposure time 1 ms to ?10 s
Frame transfer time 3.84 ms
Image readout timea 1 s
Spectral filters 1 12 positions
Spectral range 700?800 nm 395?1,040 nm
Focal length 550 mm 78 mm
Collecting area 462 mm2 48 mm2
Detector-TH7888A CCD 1024 ? 1024, 14-?m pixels
IFOV 25 ?rad 179 ?rad
Pixel FOV 5.1 m at 200-km altitude 35.8 m at 200-km altitude
Quantization 12 bits per pixel
Hardware compression Lossless, multi-resolution lossy, 12-to-n bits
MDIS Assembly MDIS DPU-A or -B
Mass 7.8 kg 1.5 kg
Powerb 7.6 W 12.3 W
Footprint 398 ? 270 ? 318 mm 157 ? 117 ? 104 mm
Data rate 16 Mbps (to DPU) 3 Mbps (to SSR)
aTransfer to DPU
bNominal power configuration (DISE + NAC or WAC; DPU + MDIS motor + resolvers)
relative to the spacecraft orbital plane. At times the ground track is near the terminator (the
?dawn?dusk orbit?); 22 days later it passes over the subsolar point (the ?noon?midnight?
orbit).
The two primary imaging objectives during the flybys are (1) acquisition of near-global
coverage at ?500 m/pixel, and (2) multispectral mapping at ?2 km/pixel. During the flyby
departures, large portions of the planet will be viewed at uniform low phase angles.
From orbit, gaps in color imaging acquired during the flybys will be filled with images
taken at a wide variety of lighting geometries. Total flyby coverage will exclude only the
polar regions and two narrow longitudinal bands. The flybys each have one of two basic
geometries (Table 2), and similar observation strategies will be used for each (Table 3).
During the flyby phase, 85% of the planet will be imaged in monochrome at a resolution
averaging ?500 m/pixel, and greater than 60% will be imaged in color at about 2 km/pixel.
Half of the planet will be covered in color at ?1 km/pixel. High-resolution NAC swaths will
contain monochrome images at better than 125 m/pixel.
During the orbital phase of the mission the MDIS observation strategy will shift to acqui-
sition of four key data sets: (1) a nadir-looking monochrome (750-nm) global photomosaic
at moderate solar incidence angles (55??75?) and 250 m/pixel or better sampling; (2) a
25?-off-nadir mosaic to complement the nadir-looking mosaic for stereo; (3) completion of
the multispectral mapping begun during the flybys; and (4) high-resolution (20?50 m/pixel)
image strips across features representative of major geologic units and structures.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 251
Table 2 Key parameters describing the three MESSENGER Mercury flybys
Date CA CA Inbound Outbound Illuminated Key features
altitude lon lon lon lon
1/14/08 200 40? 308? 132? 276? to 96? Caloris, EUH
10/06/08 200 230? 136? 324? 94? to 274? Kuiper, WUH, MG
09/29/09 200 212? 108? 315? 90? to 270? Kuiper, WUH, MG
All closest approach (CA) latitudes are near-equatorial, and range is listed in kilometers; closest approach
occurs on the nightside of the planet for all three flybys. The columns ?Inbound lon? and ?Outbound lon?
indicate subspacecraft longitudes at 20,000 km range during the inbound and outbound legs of the respective
flyby. Comments indicate the portion of Mercury imaged during each flyby (Caloris = Caloris basin, EUH =
eastern half of hemisphere unseen by Mariner 10, Kuiper = Kuiper crater, WUH = western half of hemi-
sphere unseen by Mariner 10, MG = Mariner 10 gore). All longitudes are positive east. During the three
Mariner 10 flybys Mercury was illuminated from 350?E to 170?E
Both the nadir and off-nadir image mosaics will be acquired with the NAC for southern
latitudes when altitude is high and with the WAC at lower altitudes over the northern hemi-
sphere. This two-camera strategy results in near-uniform global coverage with an average
spatial resolution of 140 m/pixel. The off-nadir mosaic will be acquired under nearly identi-
cal lighting geometries to the nadir map to facilitate automated stereo matching. The global
digital elevation model derived from stereo imaging will have a spatial resolution of 1?4 km
horizontally and 100?500 m vertically, depending on latitude. MDIS stereo imaging will
be the main source of surface elevation mapping for the southern hemisphere, as MLA?s
1,000-km slant range (Cavanaugh et al. 2007) largely limits laser altimetry to the north-
ern hemisphere. Filling gaps in color coverage is a relatively simple matter of pointing at
and imaging a particular location during times of favorable lighting, except at low altitudes
over high northern latitudes. At northern mid-latitudes, low spacecraft altitudes will limit
viewing opportunities and probably require gap-filling images to be taken in long strips.
At the time of writing, the strategy for gap-filling of flyby color mapping is still being de-
fined. High-resolution NAC imaging is effectively limited by ground motion smear to about
20 m/pixel in the along-track direction; accurate postprocessing correction for electronics
artifacts (Sect. 4.3) requires exposure times of ?7 ms or longer, equivalent to <20 m of
along-track motion smear.
2.1 Science Traceability
The flowdown of requirements from science objectives to instrument design and data acqui-
sition strategy are summarized in Table 4, which also compares the instrument requirements
to as-built performance. Key constraints on the design of the MDIS investigation are the
spacecraft?s thermal environment, its highly eccentric orbit, the vertical accuracy required
of stereo imaging, the low downlink rates, and the need to support optical navigation (opnav)
prior to each Mercury flyby.
At Mercury, the intensity of the solar radiation varies from about 7 to 10 times the to-
tal irradiance falling on the Earth. A large sunshade protects MESSENGER from this in-
tense solar illumination but constrains the spacecraft pointing ability (Leary et al. 2007).
To support optical navigation during the flybys, Mercury will be imaged against the star
background; MDIS thus had to be designed to image at phase angles as high as 140?. At
the other extreme, flyby science observations require imaging at phase angles of 32?. To
compensate for the restricted pointing capability of the spacecraft imposed by the sunshade,
252 S.E. Hawkins et al.
Ta
bl
e
3
M
er
cu
ry
fly
by
im
ag
in
g
pl
an
D
es
cr
ip
tio
n
km
/
Fi
lte
rs
Pi
xe
l
A
pp
ro
x.
B
its
/
W
av
el
et
Cu
m
.
Cu
m
.G
ib
,
Cu
m
.G
ib
,
Cu
m
.G
ib
,
pi
x
bi
nn
in
g
ha
rd
w
ar
e
pi
xe
l
co
m
p.
im
ag
es
u
n
co
m
-
ha
rd
w
ar
e
co
m
-
co
m
p.
ra
tio
pr
es
se
d
co
m
-
pr
es
se
db
ra
tio
pr
es
se
d
A
pp
ro
ac
h
o
pn
av
s
1
?
1.
3
12
1
6
0.
07
0
0.
05
4
0.
05
4
A
pp
ro
ac
h
m
o
v
ie
1
?
2.
5
8
12
36
0.
30
5
0.
14
8
0.
06
2
A
pp
ro
ac
h
co
lo
ri
m
ag
e
5
11
?
1.
3
12
2
47
0.
43
4
0.
24
7
0.
11
1
A
pp
ro
ac
h
m
o
sa
ic
0.
5
1
?
2.
5
8
8
87
0.
74
6
0.
37
2
0.
12
7
H
ig
h-
re
so
lu
tio
n
m
o
sa
ic
1
0.
06
1
?
2.
5
8
8
14
7
1.
21
5
0.
55
9
0.
15
1
H
ig
h-
re
so
lu
tio
n
co
lo
rm
o
sa
ic
1
11
?
1.
3
12
2
24
6
2.
37
5
1.
45
2
0.
59
7
Co
lo
rp
ho
to
m
et
ry
se
qu
en
ce
a
?
11
2
?
2
1.
3
12
2
34
5
2.
66
5
1.
67
5
0.
70
8
H
ig
h-
re
so
lu
tio
n
m
o
sa
ic
2
0.
2
1
?
2.
5
8
8
49
9
3.
86
8
2.
15
6
0.
76
8
D
ep
ar
tu
re
co
lo
rm
o
sa
ic
2
11
?
1.
3
12
2
59
8
5.
02
8
3.
04
9
1.
21
5
1s
td
ep
ar
tu
re
m
o
sa
ic
0.
4
1
?
2.
5
8
6
69
7
5.
80
2
3.
35
8
1.
26
6
2n
d
de
pa
rtu
re
m
o
sa
ic
0.
5
1
?
2.
5
8
6
79
6
6.
57
5
3.
66
7
1.
31
8
D
ep
ar
tu
re
co
lo
ri
m
ag
e
5
11
?
1.
3
12
2
80
7
6.
70
4
3.
76
7
1.
36
7
3r
d
(st
er
eo
)m
o
sa
ic
0.
7
1
?
2.
5
8
6
84
2
6.
97
8
3.
87
6
1.
38
6
4t
h
(st
er
eo
)m
o
sa
ic
0.
8
1
?
2.
5
8
6
87
7
7.
25
1
3.
98
5
1.
40
4
D
ep
ar
tu
re
o
pn
av
s
1
?
1.
3
12
1
88
3
7.
32
1
4.
03
9
1.
45
8
D
ep
ar
tu
re
m
o
v
ie
1
?
2.
5
8
12
91
3
7.
55
6
4.
13
3
1.
46
6
a
Sa
m
e
sp
ot
at
ph
as
e
an
gl
es
50
??
13
0?
in
10
?i
nc
re
m
en
ts
b T
o
ta
ls
pa
ce
av
ai
la
bl
e
o
n
so
lid
-s
ta
te
re
co
rd
er
is
8
G
ib
The Mercury Dual Imaging System on the MESSENGER Spacecraft 253
Ta
bl
e
4a
D
er
iv
ed
M
D
IS
re
qu
ire
m
en
ts
an
d
as
-b
u
ilt
pe
rfo
rm
an
ce
fo
rfi
el
d-
of
-v
ie
w
,
po
in
tin
g,
an
d
sp
at
ia
lr
es
o
lu
tio
n
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
f
v
er
ifi
ca
tio
n
Fl
yb
y
n
ea
r
gl
ob
al
m
o
n
o
ch
ro
m
e
m
ap
at
50
0
m
/p
ix
el
Im
ag
e
M
er
cu
ry
o
u
tb
ou
nd
o
n
al
lt
hr
ee
fly
by
s;
m
ax
im
iz
e
ar
ea
im
ag
ed
in
o
rb
it
at
n
ea
r-
0?
em
iss
io
n
an
gl
e
Po
in
tf
ro
m
n
ad
ir
to
>
40
?a
n
tis
un
w
ar
d,
>
50
?s
u
n
w
ar
d
m
ai
nt
ai
ni
ng
u
n
o
bs
tru
ct
ed
FO
V
U
no
bs
tru
ct
ed
FO
V
to
64
?a
n
tis
un
w
ar
d,
52
?
su
n
w
ar
d
u
sin
g
pi
v
o
ta
n
d
sp
ac
ec
ra
ft
sle
w
B
y
de
sig
n
Pr
ov
id
e
o
pt
ic
al
n
av
ig
at
io
n
su
pp
or
t
fo
rM
er
cu
ry
fly
by
s
O
pn
av
s
ea
rli
er
th
an
en
co
u
n
te
r(
E)
?
2.
5
da
ys
at
fly
by
s
Av
ai
la
bi
lit
y
o
fi
m
ag
in
g
at
31
??
14
2?
ph
as
e
an
gl
es
at
ce
n
te
ro
fF
OV
,
fro
m
co
m
bi
na
tio
n
o
f
pi
v
o
tin
g
an
d
sp
ac
ec
ra
ft
sle
w
in
g,
pr
ov
id
es
:
?
N
A
C
im
ag
in
g
at
E-
6.
0
da
ys
@
fly
by
1,
E-
3.
7
da
ys
@
fly
by
2,
E-
7.
2
da
ys
@
fly
by
3
?
W
A
C
im
ag
in
g
at
E-
7.
0
da
ys
@
fly
by
1,
E-
4.
7
da
ys
@
fly
by
2,
E-
8.
0
da
ys
@
fly
by
3
Tr
aje
cto
ry
an
al
ys
is
N
ea
r-n
ad
ir
gl
ob
al
m
o
n
o
ch
ro
m
e
m
ap
at
25
0
m
/p
ix
el
FO
V
w
id
e
en
o
u
gh
fo
rc
ro
ss
-t
ra
ck
co
n
tin
ui
ty
,
av
er
ag
e
da
ys
id
e
al
tit
ud
e
>
9.
3?
W
A
C
FO
V
fo
rb
as
el
in
e
o
rb
it
(>
16
0
?r
ad
pi
xe
l)
10
.5
?W
A
C
FO
V
(17
9?
ra
d
pi
xe
l)
G
eo
m
et
ric
ca
lib
ra
tio
n
Cr
os
s-
tra
ck
co
n
tin
ui
ty
m
ai
nt
ai
ne
d
at
lo
w
es
ta
lti
tu
de
su
sin
g
pi
v
o
tin
g
o
fF
OV
1.
1?
/s
pi
v
o
tr
at
e
to
st
ag
ge
rW
A
C
FO
V
st
o
pr
od
uc
e
15
?w
id
ec
o
m
po
sit
e
im
ag
e
st
rip
s
1.
1?
/s
Pi
v
o
tt
es
tin
g
A
cq
ui
re
im
ag
e
st
rip
sw
ith
ou
tc
ro
ss
-
tr
ac
k
go
re
so
r
ex
ce
ss
iv
e
cr
o
ss
-t
ra
ck
o
v
er
la
p
A
cq
ui
re
re
ct
an
gu
la
rs
u
bf
ra
m
es
at
u
p
to
1
H
z
to
m
an
ag
e
do
w
n
lin
k
v
o
lu
m
e
A
rb
itr
ar
ily
de
fin
ed
re
ct
an
gu
la
rs
u
bf
ra
m
es
av
ai
la
bl
e
at
1
H
z
Sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
<
25
0
m
/p
ix
el
sa
m
pl
in
g,
gl
ob
al
ly
,
w
ith
lo
w
em
iss
io
n
an
gl
e
W
A
C
an
d
N
A
C
IF
OV
s
pr
o
v
id
e
<
25
0
m
/p
ix
el
av
er
ag
e
sa
m
pl
in
g
fo
r
n
o
m
in
al
o
rb
it
14
0
m
/p
ix
el
av
er
ag
e
sa
m
pl
in
g
o
fs
u
rfa
ce
at
o
pt
im
al
lig
ht
in
g
Co
v
er
ag
e
sim
ul
at
io
ns
W
A
C
an
d
N
A
C
PS
F
su
pp
or
ts
am
pl
in
g
at
<
2?
av
er
ag
e
re
qu
ire
m
en
t
Pr
oje
cte
dF
W
H
M
o
fN
A
C
PS
F
?
70
0
m
in
so
u
th
er
n
hi
gh
la
tit
ud
es
;o
n
gr
ou
nd
PS
F
m
ea
-
su
re
m
en
tt
o
be
u
se
d
fo
ri
m
ag
e
re
st
or
at
io
n
R
ad
io
m
et
ric
ca
lib
ra
tio
n
an
d
an
al
yt
ic
al
m
o
de
lin
g
M
ai
n
pr
oc
es
so
ra
n
d
do
w
n
lin
k
v
o
lu
m
e
al
lo
w
av
er
ag
e
sp
at
ia
lr
es
o
lu
tio
n
o
fb
in
ne
d
pi
xe
ls
<
25
0
m
/p
ix
el
w
ith
m
o
de
ra
te
co
m
pr
es
sio
n
25
.9
G
b
M
D
IS
do
w
n
lin
k
(d/
l)
al
lo
ca
tio
n
al
lo
w
s
d/
lo
f1
40
m
/p
ix
el
u
sin
g
8:
1
co
m
-
pr
es
sio
n;
M
P
lo
ad
in
g
re
qu
ire
so
n
-c
hi
p
2
?
2
bi
nn
in
g
fo
rp
or
tio
n
o
fd
at
a,
de
gr
ad
in
g
re
so
lu
tio
n
by
u
p
to
2?
Co
v
er
ag
e
an
d
sp
ac
ec
ra
ft
pe
rfo
rm
an
ce
sim
ul
at
io
ns
Po
in
tw
ith
su
ffi
ci
en
ta
cc
u
ra
cy
to
al
lo
w
m
o
sa
ic
ki
ng
w
ith
<
10
%
im
ag
e
o
v
er
la
p
<
0.
15
?p
oi
nt
in
g
ac
cu
ra
cy
fo
rs
pa
ce
cr
af
tt
o
m
ai
nt
ai
n
<
10
%
N
A
im
ag
e
o
v
er
la
p
0.
02
?p
oi
nt
in
g
ac
cu
ra
cy
fo
rs
pa
ce
cr
af
t;
0.
03
?i
nc
lu
di
ng
1-
?
u
n
ce
rt
ai
nt
y
Si
m
ul
at
io
n
ba
se
d
o
n
u
n
ce
rt
ai
nt
ie
sf
ro
m
co
m
po
ne
nt
te
st
s
<
0.
15
?s
te
p
fo
rM
D
IS
pi
v
o
tm
ec
ha
ni
sm
0.
01
?s
te
p
fo
rM
D
IS
pi
v
o
tm
ec
ha
ni
sm
B
y
de
sig
n
U
nd
er
lin
ed
ite
m
si
nc
ur
ad
di
tio
na
ld
at
a
pr
o
ce
ss
in
g
o
r
re
v
isi
on
st
o
th
e
da
ta
ac
qu
isi
tio
n
st
ra
te
gy
to
m
ee
tr
eq
ui
re
m
en
ts,
as
in
di
ca
te
d
254 S.E. Hawkins et al.
Ta
bl
e
4b
D
er
iv
ed
M
D
IS
re
qu
ire
m
en
ts
an
d
as
-b
u
ilt
pe
rfo
rm
an
ce
fo
rs
pa
ce
cr
af
ts
le
w
in
g,
st
ab
ili
ty
,
m
em
o
ry
,
an
d
do
w
n
lin
k
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
f
v
er
ifi
ca
tio
n
N
ea
r-n
ad
ir
gl
ob
al
m
o
n
o
ch
ro
m
e
m
ap
at
25
0
m
/p
ix
el
R
ol
ls
pa
ce
cr
af
ts
u
ffi
ci
en
tly
ra
pi
dl
y
to
tr
ac
k
n
ad
ir
at
th
e
m
in
im
um
20
0-
km
o
rb
ita
la
lti
tu
de
1.
35
m
ra
d/
s
1.
7
m
ra
d/
s
Sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
A
bi
lit
y
to
tr
ac
k
n
ad
ir
fro
m
an
y
o
rb
it
Pi
v
o
ta
n
d
sp
ac
ec
ra
ft
ro
ll
co
o
rd
in
at
e
to
po
in
t
M
D
IS
bo
re
sig
ht
to
lo
ca
ln
ad
ir
Sp
ac
ec
ra
ft
ro
ll
to
ke
ep
n
ad
ir
in
pi
v
o
tp
la
ne
,
sp
ac
ec
ra
ft
su
pp
lie
sM
D
IS
w
ith
pi
v
o
t
at
tit
ud
e
Sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
Im
ag
e
sm
ea
r
by
sp
ac
ec
ra
ft
jitt
er
sm
al
l
co
m
pa
re
d
w
ith
im
ag
e
sm
ea
r
du
e
to
do
w
n
tr
ac
k
m
o
tio
n
Jit
te
r<
25
?r
ad
in
10
0
m
s
(<
1
N
A
C
pi
x
el
,
ty
pi
ca
le
x
po
su
re
)
<
1
?r
ad
in
10
0
m
s
A
na
ly
tic
al
m
o
de
lin
g
H
ol
d
o
rb
ita
ld
at
a
in
re
co
rd
er
H
ol
d
v
o
lu
m
e
o
fD
PU
-c
om
pr
es
se
d
im
ag
es
B
as
el
in
e
m
iss
io
n
re
qu
ire
ss
to
rin
g
<
2.
3
G
b;
u
sa
bl
e
re
co
rd
er
v
o
l.
6.
2
G
b
D
at
a
co
v
er
ag
e
sim
ul
at
io
n
H
ol
d
n
u
m
be
ro
fM
P-
co
m
pr
es
se
d
im
ag
es
B
as
el
in
e
m
iss
io
n
re
qu
ire
ss
to
rin
g
<
25
00
co
m
pr
es
se
d
im
ag
es
;>
80
00
av
ai
la
bl
e
Fl
yb
y
n
ea
r
gl
ob
al
m
o
n
o
ch
ro
m
e
m
ap
at
50
0
m
/p
ix
el
H
ol
d
fly
by
da
ta
in
re
co
rd
er
H
ol
d
v
o
lu
m
e
o
fD
PU
-c
om
pr
es
se
d
im
ag
es
N
o
m
in
al
fly
by
sc
en
ar
io
5.
4
G
b;
u
sa
bl
e
re
co
rd
er
v
o
lu
m
e
o
f6
.2
G
b
H
ol
d
n
u
m
be
ro
fM
P-
co
m
pr
es
se
d
im
ag
es
N
o
m
in
al
fly
by
sc
en
ar
io
91
3
im
ag
es
;
u
n
co
m
pr
es
se
d
im
ag
e
di
re
ct
or
y
ho
ld
s2
04
6
M
ul
tis
pe
ct
ra
lm
ap
at
2
km
/p
ix
el
D
ow
n
lin
k
co
m
pr
es
se
d
fly
by
da
ta
to
gr
ou
nd
>
2.
3
G
b
do
w
n
lin
k
al
lo
ca
tio
n
to
M
D
IS
fo
llo
w
in
g
ea
ch
M
er
cu
ry
fly
by
(co
re
da
ta
)
Cu
rre
nt
do
w
n
lin
k
>
6
G
b
af
te
re
ac
h
fly
by
Ta
rg
et
ed
hi
gh
-re
so
lu
tio
n
im
ag
in
g
Tr
an
sf
er
da
ta
to
SS
R
at
ra
te
gr
ea
te
r
th
an
im
ag
e
ac
qu
isi
tio
n
H
ig
he
st
im
ag
e
ac
qu
isi
tio
n
ra
te
is
1
H
z
(3.
2
M
b/
sf
or
im
ag
e
bi
nn
ed
2
?
2
o
n
-c
hi
p)
3.
2
M
b/
sl
in
k
sp
ee
d
In
st
ru
m
en
ta
n
d
sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
The Mercury Dual Imaging System on the MESSENGER Spacecraft 255
Ta
bl
e
4b
(C
on
tin
ue
d)
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
f
v
er
ifi
ca
tio
n
N
ea
r-n
ad
ir
gl
ob
al
m
o
n
o
ch
ro
m
e
m
ap
at
25
0
m
/p
ix
el
D
ow
n
lin
k
co
m
pr
es
se
d
o
rb
ita
ld
at
a
to
gr
ou
nd
5.
6
G
b/
yr
m
in
im
um
al
lo
ca
tio
n
to
M
D
IS
du
rin
g
o
rb
ita
lm
iss
io
n
as
pe
rC
o
n
ce
pt
St
ud
y
?
25
.9
G
b/
yr
al
lo
w
s
im
pr
ov
em
en
ti
n
av
er
ag
e
ba
se
m
ap
sa
m
pl
in
g
fro
m
25
0
to
14
0
m
/p
ix
el
,c
o
lo
rt
o
1
km
/p
ix
el
D
at
a
co
v
er
ag
e
sim
ul
at
io
n
G
lo
ba
l
m
u
lti
sp
ec
tra
lm
ap
at
2
km
/p
ix
el
Pi
xe
ls
2
?
2
or
4
?
4
w
ith
ou
to
v
er
flo
w
at
7
m
s
ex
po
su
re
2
?
2
bi
nn
in
g
o
n
-c
hi
p
su
m
s
sig
na
l,
do
es
n
o
t
o
v
er
flo
w
at
m
in
.s
o
la
rd
ist
an
ce
an
d
ph
as
e
an
gl
e;
2
?
2,
4
?
4
bi
nn
in
g
in
M
P
av
er
ag
es
sig
na
l
R
ad
io
m
et
ric
ca
lib
ra
tio
n;
in
str
um
en
ta
n
d
sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
Lo
ss
y
co
m
pr
es
sio
n,
>
6
:
1
w
ith
ar
tif
ac
ts
n
ea
r
n
o
ise
W
av
el
et
co
m
pr
es
sio
n
in
M
P;
ar
tif
ac
ts
n
ea
r
n
o
ise
at
8
:
1
Si
m
ul
at
io
n
u
sin
g
N
EA
R
16
-b
it
im
ag
es
Pr
ov
id
e
o
pt
ic
al
n
av
ig
at
io
n
su
pp
or
t
fo
rM
er
cu
ry
fly
by
s
Pr
om
pt
ly
do
w
n
lin
k
st
ar
-
pl
an
et
-s
ta
r
tr
ip
le
ts
R
et
ur
n
jai
lba
rs
fro
m
M
er
cu
ry
im
ag
es
Ja
ilb
ar
sa
tr
eg
ul
ar
pe
rio
d
in
im
ag
e
In
st
ru
m
en
ta
n
d
sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
R
et
ur
n
?
3
su
bf
ra
m
es
o
fs
ta
rs
pe
ri
m
ag
e
(to
14
0
?
14
0
pi
xe
ls
fo
rN
A
,a
llo
w
in
g
fo
r0
.
1?
po
in
tin
g
co
n
tr
ol
)
U
p
to
5
su
bf
ra
m
es
pe
r
im
ag
e,
siz
ed
u
p
to
fu
ll
im
ag
e
Lo
ca
te
M
er
cu
ry
ag
ai
ns
ts
ta
r
ba
ck
gr
ou
nd
D
et
ec
t?
3
st
ar
s
in
FO
V
at
m
ax
ex
po
su
re
s
M
et
fo
r
W
A
C
at
lo
ng
es
te
x
po
su
re
;m
ar
gi
na
l
in
N
A
C
R
ad
io
m
et
ric
ca
lib
ra
tio
n
an
d
an
al
yt
ic
al
m
o
de
lin
g
Ta
ke
u
n
sm
ea
re
d
st
ar
fra
m
es
Jit
te
r<
25
?r
ad
in
10
s
(<
1
N
A
C
pi
x
el
)
<
25
?r
ad
in
10
s
>
99
%
o
ft
im
e
A
n
al
yt
ic
al
m
o
de
lin
g
256 S.E. Hawkins et al.
Ta
bl
e
4c
M
D
IS
re
qu
ire
m
en
ts
an
d
as
-b
u
ilt
pe
rfo
rm
an
ce
fo
rm
u
lti
sp
ec
tra
li
m
ag
in
g
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
fv
er
ifi
ca
tio
n
M
ul
tis
pe
ct
ra
lm
ap
at
2
km
/p
ix
el
M
ov
e
fil
te
rw
he
el
to
al
lo
w
11
-c
ol
or
im
ag
in
g
st
rip
fro
m
lo
w
es
ta
lti
tu
de
o
v
er
da
y
sid
e
W
o
rs
t-
ca
se
FO
V
m
o
tio
n
by
1
fo
ot
pr
in
ti
n
15
.5
s
re
qu
ire
s1
4
s
pe
rc
yc
le
(al
low
in
g
10
%
o
v
er
la
p),
1
H
z
im
ag
in
g
in
11
fil
te
rs
+
3
s
fo
rr
ep
os
iti
on
1
H
z
im
ag
in
g
su
pp
or
te
d
fo
r
ex
po
su
re
s
<
50
0
m
s;
2
s
fo
r
re
po
sit
io
n;
13
s
to
ta
l
In
st
ru
m
en
tf
un
ct
io
na
l
te
st
in
g
<
2
km
/p
ix
el
sp
at
ia
ls
am
pl
in
g
<
2
km
/p
ix
el
du
rin
g
fly
by
,
fro
m
o
rb
it
to
fil
l
ga
ps
in
fly
by
co
v
er
ag
e
2
km
/p
ix
el
re
so
lu
tio
n
at
4.
5
pl
an
et
ra
di
i,
o
r
w
ith
in
10
,0
00
km
du
rin
g
o
rb
it
G
eo
m
et
ric
ca
lib
ra
tio
n,
tr
aje
cto
ry
an
al
ys
is
Sp
ec
tra
lfi
lte
rs
fo
rm
ap
pi
ng
o
fo
liv
in
e,
py
ro
x
en
e,
gl
as
s,
o
pa
qu
es
7
W
A
C
sp
ec
tra
lfi
lte
rs
,v
io
le
tt
o
10
50
n
m
11
W
A
C
sp
ec
tra
lfi
lte
rs
to
m
ea
su
re
ke
y
fe
at
ur
es
,p
lu
sc
le
ar
fil
te
r
R
ad
io
m
et
ric
ca
lib
ra
tio
n
M
ap
ab
u
n
da
nc
e
v
ar
ia
tio
ns
in
o
pa
qu
e
an
d
m
afi
c
m
in
er
al
s:
75
0/
41
5
n
m
,
75
0/
65
0
n
m
,
an
d
75
0/
95
0
n
m
ra
tio
s
m
ea
su
re
d
to
1%
pr
ec
isi
on
M
TF
>
0.
62
at
1
cy
cl
e/
8
pi
xe
ls
to
pr
es
er
ve
v
ar
ia
tio
ns
ab
ov
e
n
o
ise
le
v
el
M
TF
?
0.
75
1
cy
cl
e/
8
pi
xe
ls
fo
rt
yp
ic
al
fil
te
r
A
n
al
ys
is
an
d
ra
di
om
et
ric
ca
lib
ra
tio
n
M
od
el
da
rk
cu
rr
en
tt
o

sy
ste
m
n
o
ise
(
1
D
N
)
D
ar
k
cu
rr
en
tm
o
de
lr
es
id
ua
ls
at
re
ad
-n
oi
se
le
v
el
at
<
5?
C,
co
n
sis
te
nt
w
ith
m
o
de
l
ac
cu
ra
cy

n
o
ise
R
ad
io
m
et
ric
ca
lib
ra
tio
n
R
es
po
ns
e
lin
ea
rit
y
to
1%
D
ep
ar
tu
re
so
f?
2%
at
lo
w
D
N
le
v
el
s
co
rr
ec
ta
bl
e
du
rin
g
gr
o
u
n
d
pr
oc
es
sin
g
R
ad
io
m
et
ric
ca
lib
ra
tio
n
Ch
ar
ac
te
riz
e
re
sp
on
siv
ity
to
5%
ab
so
lu
te
,
2%
re
la
tiv
e
be
tw
ee
n
fil
te
rs
O
n-
gr
ou
nd
ca
lib
ra
tio
n
at
?
34
,?
30
,+
24
?C
in
su
ffi
ci
en
tf
o
r
in
te
rp
ol
at
io
n
to
hi
gh
er
(?
10
?C
)C
CD
o
pe
ra
tin
g
te
m
pe
ra
tu
re
s;
n
ee
d
fli
gh
tm
ea
su
re
m
en
to
fM
er
cu
ry
-il
lu
m
in
at
ed
ca
lib
ra
tio
n
ta
rg
et
R
ad
io
m
et
ric
ca
lib
ra
tio
n
M
od
el
fla
tfi
el
d
to
0.
1%
pr
ec
isi
on
(0.
4o
f
n
o
ise
at
fu
ll
w
el
l)
Co
m
po
sit
e
o
f1
00
im
ag
es
pr
ov
id
es
0.
04
%
pr
ec
isi
on
;fl
ig
ht
m
ea
su
re
m
en
ts
n
ee
de
d
du
e
to
m
o
bi
lit
y
o
fd
us
td
on
ut
s
R
ad
io
m
et
ric
ca
lib
ra
tio
n
Pr
ot
ec
to
pt
ic
sf
ro
m
co
n
ta
m
in
at
io
n
th
at
in
cr
ea
se
ss
ca
tte
re
d
lig
ht
St
ow
M
D
IS
in
po
sit
io
n
w
ith
o
pt
ic
s
pr
ot
ec
te
d
fro
m
co
n
ta
m
in
an
ts
du
rin
g
bu
rn
s,
th
er
m
al
sp
ik
es
o
n
su
n
sh
ad
e
Fr
on
to
pt
ic
st
ow
ab
le
in
in
str
um
en
tb
as
e
In
st
ru
m
en
tf
un
ct
io
na
l
te
st
U
nd
er
lin
ed
ite
m
si
nc
ur
ad
di
tio
na
ld
at
a
pr
o
ce
ss
in
g
o
r
re
v
isi
on
st
o
th
e
da
ta
ac
qu
isi
tio
n
st
ra
te
gy
to
m
ee
tr
eq
ui
re
m
en
ts,
as
in
di
ca
te
d
The Mercury Dual Imaging System on the MESSENGER Spacecraft 257
Ta
bl
e
4d
M
D
IS
re
qu
ire
m
en
ts
an
d
as
-b
u
ilt
pe
rfo
rm
an
ce
fo
rh
ig
h-
re
so
lu
tio
n
an
d
st
er
eo
im
ag
in
g
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
fv
er
ifi
ca
tio
n
Ta
rg
et
ed
hi
gh
-re
so
lu
tio
n
im
ag
in
g
A
cq
ui
re
co
n
tin
uo
us
st
rip
o
f
im
ag
es
w
ith
N
A
C
fro
m
m
in
im
um
o
rb
ita
la
lti
tu
de
A
tm
in
im
um
al
tit
ud
e
o
f2
00
km
,N
A
fo
o
tp
rin
t
m
o
v
es
1
FO
V
in
1.
4
s.
R
eq
ui
re
s1
H
z
im
ag
in
g.
1
H
z
im
ag
in
g
su
pp
or
te
d
u
sin
g
2
?
2
pi
xe
lb
in
ni
ng
o
r
qu
ar
te
rf
ra
m
es
In
st
ru
m
en
tf
un
ct
io
na
l
te
st
in
g
A
cq
ui
re
<
20
m
/p
ix
el
N
A
C
fra
m
es
fro
m
lo
w
es
ta
lti
tu
de
ex
pe
ct
ed
o
v
er
da
y
sid
e
(28
0k
m
)w
ith
<
20
m
sm
ea
r
du
e
to
sp
ac
ec
ra
ft
m
o
tio
n
N
A
C
IF
OV
<
71
?r
ad
N
A
C
IF
OV
20
?r
ad
G
eo
m
et
ric
an
d
ra
di
om
et
ric
ca
lib
ra
tio
n
M
in
im
iz
e
tim
e
fo
r
CC
D
fra
m
e
tr
an
sf
er
(m
ini
mu
m
u
se
fu
le
x
po
su
re
tim
e
?
2?
fra
m
e
tr
an
sf
er
tim
e)
3.
7
m
s
fra
m
e
tr
an
sf
er
yi
el
ds
?
7
m
s
ex
po
su
re
.A
t3
.3
km
/s,
18
m
lin
ea
r
sm
ea
r.
Pi
xe
lf
oo
tp
rin
t1
2
m
/p
ix
el
af
te
r
2?
2
bi
nn
in
g
re
qu
ire
d
fo
r
1
H
z
im
ag
in
g
N
o
sa
tu
ra
tio
n
at
m
in
im
um
ex
po
su
re
tim
e
N
o
sa
tu
ra
tio
n
in
7
m
s
N
A
C
im
ag
e
at
33
?p
ha
se
,
m
in
im
um
so
la
rd
ist
an
ce
,2
?
2
pi
xe
lb
in
ni
ng
W
o
rs
t-
ca
se
sa
tu
ra
tio
n
tim
e
16
m
s
R
ad
io
m
et
ric
ca
lib
ra
tio
n
an
d
ph
ot
om
et
ric
m
o
de
l
O
ff-
na
di
ri
m
ag
in
g
to
co
m
pl
em
en
tn
ad
ir
ge
om
et
ry
fo
rs
te
re
o
Im
ag
e
ar
ea
s
bo
th
at
n
ad
ir
an
d
o
ff-
na
di
rw
ith
sim
ila
rl
ig
ht
in
g
Co
ns
ist
en
to
bs
er
va
tio
n
st
ra
te
gy
m
ee
tin
g
o
th
er
sp
ac
ec
ra
ft
co
n
st
ra
in
ts
N
he
m
isp
he
re
n
ad
ir
fir
st
so
la
rd
ay
,
o
ff-
na
di
rs
ec
o
n
d
so
la
rd
ay
;
S
he
m
isp
he
re
n
ad
ir
an
d
o
ff-
na
di
ro
n
ad
jac
en
to
rb
its
fir
st
so
la
rd
ay
.
M
ee
ts
sp
ac
ec
ra
ft
co
n
st
ra
in
ts
as
gi
v
en
in
Ta
bl
e
4b
.
Co
v
er
ag
e
sim
ul
at
io
n
A
bi
lit
y
to
ta
ke
o
ff-
na
di
r
im
ag
es
an
ar
bi
tra
ry
o
rb
it
Pi
v
o
ta
n
d
sp
ac
ec
ra
ft
ro
ll
u
sa
bl
e
to
ge
th
er
to
po
in
tM
D
IS
bo
re
sig
ht
o
ffs
et
al
on
g-
tra
ck
fro
m
lo
ca
ln
ad
ir
R
ol
ls
pa
ce
cr
af
tt
o
ke
ep
n
ad
ir
o
ffs
et
fro
m
pi
v
o
tp
la
ne
,s
u
pp
ly
M
D
IS
w
ith
pi
v
o
ta
tti
tu
de
Sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g
St
er
eo
v
er
tic
al
ac
cu
ra
cy
go
al
is
?
2
km
fro
m
60
00
-k
m
al
tit
ud
e
(ro
ot
su
m
sq
ua
re
d
po
in
tin
g
kn
ow
le
dg
e
24
0
?r
ad
)
18
0
?r
ad
bo
re
sig
ht
kn
ow
le
dg
e
w
ith
in
-fl
ig
ht
ca
lib
ra
tio
ns
(1.
4?
w
o
rs
e
th
an
N
EA
R
w
ith
in
-fl
ig
ht
po
in
tin
g
ca
ls.
u
sin
g
st
ar
im
ag
es
;
al
lo
w
s
fo
rs
ca
n
pl
an
e
v
s.
fix
ed
po
in
tin
g)
A
bi
lit
y
to
ex
tr
ac
tm
u
lti
pl
e
su
bf
ra
m
es
fro
m
st
ar
fie
ld
im
ag
es
fo
rp
oi
nt
in
g
ca
lib
ra
tio
ns
w
ith
in
do
w
n
lin
k
20
0
?r
ad
ex
pe
ct
ed
po
in
tin
g
kn
ow
le
dg
e
In
st
ru
m
en
ta
n
d
sp
ac
ec
ra
ft
fu
n
ct
io
na
l
te
st
in
g;
fli
gh
t
ca
lib
ra
tio
n
re
qu
ire
d
fo
r
po
in
tin
g
kn
ow
le
dg
e
258 S.E. Hawkins et al.
Ta
bl
e
4d
(C
on
tin
ue
d)
M
ea
su
re
m
en
t
o
bje
cti
v
e
M
ea
su
re
m
en
tr
eq
ui
re
m
en
t
In
st
ru
m
en
t/
sp
ac
ec
ra
ft
re
qu
ire
m
en
t
A
s-
bu
ilt
pe
rfo
rm
an
ce
M
et
ho
d
o
fv
er
ifi
ca
tio
n
15
0
?r
ad
kn
ow
le
dg
e
o
fp
iv
o
tp
os
iti
on
85
?r
ad
Pi
v
o
tp
os
iti
on
ca
lib
ra
tio
n
K
no
w
le
dg
e
o
fc
en
te
ro
fM
D
IS
ex
po
su
re
to
?
10
m
s
<
1
m
s
u
n
ce
rt
ai
nt
y
A
na
ly
sis
o
fi
m
ag
e
ac
qu
isi
tio
n
tim
in
g
u
sin
g
ex
te
rn
al
st
ro
be
an
d
U
n
iv
er
sa
lT
im
e
cl
oc
k
A
cq
ui
re
im
ag
es
fro
m
o
rb
its
w
ith
m
in
im
al
th
er
m
al
di
stu
rb
an
ce
st
o
in
str
um
en
td
ec
k
N
ad
ir
an
d
o
ff-
na
di
r
co
v
er
ag
e
pl
an
ne
d
fro
m
n
ea
r-
te
rm
in
at
or
o
rb
it
w
ith
le
ss
er
th
er
m
al
v
ar
ia
tio
n
B
y
de
sig
n
The Mercury Dual Imaging System on the MESSENGER Spacecraft 259
Fig. 1 Photograph of the
Mercury Dual Imaging System
(MDIS) instrument just prior to
integration with the spacecraft
(S/C). Redundant Data
Processing Units (DPUs, not
shown) connect to MDIS through
the DPU Interface Switching
Electronics (DISE). Red-tag
covers were used to protect
apertures during handling and
were removed before flight.
Some thermal blankets are not
shown to reveal structure
the dual cameras of MDIS are able to pivot about a common axis (Fig. 1). The nominal
operational scan range of the platform is 40? in the sunward direction and 50? anti-sunward.
With spacecraft slewing, phase angle coverage of 26??142? is possible in both cameras at
the center of each field of view (FOV). Imaging is available in the WAC 5.3? farther in each
direction because of the wide FOV; in the NAC imaging is possible 0.75? farther. During
launch and key orbital maneuvers, the camera can be placed in a stowed position, providing
contamination protection for the optics (Fig. 2).
The thermal environment poses challenges for MDIS performance, because the cameras
must view the hot surface (>400?C) on some orbits for ?120 minutes. Although this ther-
mal environment presents issues for all parts of the instrument, the most stressing case is
maintaining nominal temperature of the charge-coupled device (CCD). Wide swings in de-
tector temperature potentially degrade signal-to-noise ratio (SNR), calibration accuracy, and
the value of the acquired images.
The thermal environment also poses a challenge for stereo imaging. Stereo provides mea-
surement of both relief (the elevation difference between stereo resolution cells, about 5 ? 5
to 7 ? 7 pixels in size) and elevation relative to mean planetary radius. In the southern hemi-
sphere, beyond the range of the MLA, the primary knowledge of elevation to mean planetary
radius will be from photogrammetric analysis of MDIS images (plus occultations of radio
signals from the spacecraft to Earth). Accuracy in an elevation determination from stereo is
proportional to h?/ tan(e), where h is the orbital altitude, ? is the uncertainty in pointing
knowledge between image pairs, and e is the emission angle of the off-nadir image. The
goal for elevation accuracy is ?2 km. Assuming that accuracy can be improved by a factor
of two using photogrammetric techniques at the corners of four stereo pairs, the required
accuracy is ?4 km. For a 6,000-km orbital altitude, which is appropriate to southern high
latitudes, and a 25? emission angle of the off-nadir images, the required pointing knowledge
is ?240 ?rad. This requirement is budgeted between uncertainty in the knowledge of im-
age acquisition time (which translates into downtrack position error), uncertainty in pivot
260 S.E. Hawkins et al.
Fig. 2 Range of motion of the MDIS pivot platform. Operational range is ?40? sunward to +50? antisun-
ward (planetward). When stowed, the sensitive first optic of each telescope is protected
position within its plane of motion, and variation in orientation of the pivot plane relative
to thermal distortion of the spacecraft. The largest term is due to variation in pivot plane
orientation relative to the star camera (Table 4d). On the Near Earth Asteroid Rendezvous
(NEAR) mission, orientation of its fixed-pointed camera was modeled to ?130 ?rad as
a function of temperature of the spacecraft structure, using star-field images to calibrate
pointing. Allowing for the fact that MDIS moves in a plane, and that the plane may shift in
orientation with temperature, the budget is increased by
?
2 to 180 ?rad, leaving 140 ?rad for
uncertainty in position within the pivot plane and along-track errors. To facilitate pointing
The Mercury Dual Imaging System on the MESSENGER Spacecraft 261
calibrations with MESSENGER?s limited downlink, the spacecraft main processor provides
a subframing capability that allows up to five subframes per image encompassing stars.
The eccentric orbit results in challenges associated with strongly variable spacecraft
range and velocity. Spacecraft altitude above the surface ranges between 300 and 15,000 km
on the dayside. This variation, combined with the requirement for imaging at near-nadir
geometries, drove the selection of camera optics. The WAC?s 10.5? FOV is sufficient that
overlap occurs between nadir-pointed image strips taken on adjacent orbits, even at northern
mid-latitudes where low altitudes occur. The NAC?s 1.5? FOV is sufficiently narrow that
375 m/pixel sampling is attained at 15,000 km altitude. Low emission-angle geometries are
available each solar day for all parts of the planet from altitudes of less than 10,000 km,
providing 250 m/pixel sampling or better. The low altitude at periapsis also drives the speed
of image acquisition. For multispectral imaging, acquisition of 11-color data from the mini-
mum dayside altitude (?300 km) requires the WAC to take images at 1-Hz cadence. Along-
track continuity of high-resolution imaging also requires 1-Hz imaging. In neither case is
full resolution of either the WAC or NAC required; 2 ? 2 pixel binning on-chip meets the
spatial sampling requirements both for WAC color and for NAC high-resolution imaging
from low altitude. Meeting low-altitude imaging requirements thus drives the speed of the
WAC filter wheel (1-Hz imaging in adjacent filters) and link speed from either camera to the
recorder (12-bit, 512 ? 512 frames at 1 Hz) as described in Tables 4c and 4d.
The MESSENGER mission requires compression to meet its science objectives within
the available downlink. Figure 3 summarizes the compression options available to MDIS
at the instrument level using the spacecraft main processor (MP). At the focal plane, 2 ? 2
binning is available on-chip to reduce the 1,024 ? 1,024 images to 512 ? 512 format, 12-bit
data number (DN) levels can be converted to 8 bits, and data can be compressed losslessly.
The strategy for image compression is to acquire all monochrome data in 8-bit mode, and
color data in 12-bit mode, and to compress all data losslessly to conserve recorder space.
After data are written to the recorder, they can be uncompressed and recompressed by the
MP more aggressively using any of several options: additional pixel-binning, subframing,
and lossy compression using an integer wavelet transform. The strategy for MP compression
is that all data except flyby color imaging will be wavelet compressed, typically 8:1 for
monochrome data and to a lower ratio (?4 : 1) for orbital color data. Color imaging but not
monochrome imaging may be further pixel-binned. For the special case of optical navigation
images, there is a ?jailbar? option that saves selected lines of an image at a fixed interval for
optical navigation images of Mercury during flyby approaches.
Compression performance was extensively modeled prior to launch. The 12-to-8 bit look-
up tables have been designed to retain preferentially information at low, medium, or high
12-bit DN values, for a nominal detector bias or for one that has decreased with time (Fig. 4).
Compression ratios to be used for flight have been based on a study of the magnitude and
spatial coherence of compression artifacts using NEAR images (Fig. 5). For expected load-
ing of the main processor, simulations have shown that the MP can compress the equivalent
of 82 full 1,024 ? 1,024 images per day (or 330 512 ? 512 images per day). The actual
number of images has also been simulated, based on orbital trajectory simulations and the
imaging plan described in the following. The MP image compression capabilities are con-
sistent with the mission-average number of images per day. However, on days when lighting
is favorable for global mapping, a peak of ?260 images per day may be expected, requiring
on-chip binning of most of the data on peak days. The full implications for average imaging
resolution are still being assessed.
The final set of requirements involves the responsivity required for mapping Mercury and
for optical navigation. For WAC spectral filters, passband widths were selected to provide
262 S.E. Hawkins et al.
Fig. 3 Hardware (H/W) compression options available for MDIS images in the instrument DPU and software
compression options available in the spacecraft main processor (MP). Note: (1) If 2 ? 2 binned mode is
selected as the camera mode, further binning options are not available in the MP. (2) A compression ratio of
0 results in a losslessly transformed image (the resulting image size actually grows due to transforming from
12- to 16-bit representations of each pixel)
required SNR in exposure times sufficiently short to prevent linear smear by along-track
motion, yet sufficiently long (>7 ms) to avoid excessive artifacts from removal of frame
transfer smear during ground processing. SNR is not an issue, as sufficient light is available
for SNRs >200, but saturation is a concern at low phase angles. At the same time, both cam-
eras must be sufficiently sensitive to provide star images for optical navigation and display
adequate rejection of stray light. A sequence of long and short exposures will be used for
optical navigation. A short exposure in which the planet is not saturated will permit determi-
The Mercury Dual Imaging System on the MESSENGER Spacecraft 263
Fig. 4 Mapping of 12 bits to 8 bits will be accomplished using onboard look-up tables. The tables are
designed to preserve preferentially information at different DN ranges, and they can accommodate a nominal
detector dark level as well as one that has changed with time. (1) Low noise, high bias, SNR proportional.
Usage: Typical imaging with varied brightness. (2) Low noise, high bias, DN-weighted, SNR proportional.
Usage: Faint-object imaging. (3) High noise, high bias, DN-weighted, SNR proportional. Usage: Black and
White (B/W), low brightnesses. (4) Low noise, medium bias, SNR proportional. (5) Low noise, medium bias,
DN-weighted, SNR proportional. Usage: Faint objects. (6) High noise, medium bias, DN-weighted, SNR
proportional. Usage: B/W, mostly low brightness. (7) Zero-bias, SNR proportional. Usage: Typical imaging,
varied brightness. (8) Linear. Usage: High-brightness mapping, preserves high-DN information
nation of Mercury?s location within the image using centroiding techniques. When imaging
a saturated Mercury against a star background (as will be the case for long exposures), at
least three stars must be visible per image at ?7? noise. For the WAC this requirement was
easily met (Table 5a) using a clear filter. For the NAC, its single filter was designed with a
first priority of not saturating on bright crater ejecta while imaging Mercury at low phase
angles using pixel binning. As a consequence, sensitivity to stars is limited. Detection of
three stars per frame for a typical patch of sky is only marginal (Table 5b).
2.2 Common Payload Design
In order to satisfy the science requirements and the design constraints of the MESSENGER
mission, many aspects of the science payload were implemented in common. The need to
share resources among the instruments played a significant part in the design implementa-
tion. The complexity of the MDIS instrument required a separate electronics box, shown in
264 S.E. Hawkins et al.
Fig. 5 Effects of compression to different ratios using the MESSENGER integer wavelet transform.
Well-exposed 12-bit (DN peaks near 3,000 out of 4,095) NEAR MSI images simulate well the properties
of MDIS raw data. The left column shows the image prior to compression, and the middle column after com-
pression and decompression. The right column shows the ratio of the decompressed image to the original; the
standard deviation of the ratio is a measure of the artifacts for typical illumination
Table 5a WAC sensitivity to stars
Mv 10 s exposure, 10 s exposure, 10 s exposure, # stars of ? mag in
no pixel sum, 2 ? 2 pixel sum, no pixel sum, WA FOV
ensquared energy ensquared energy ensquared energy
= 70% = 90% = 22%
DN DN DN
0 21,000 27,000 6,500 0.06
1 8,200 11,000 2,600 0.16
2 3,300 4,200 1,000 0.37
3 1,300 1,700 410 0.9
4 520 670 160 2.2
5 210 270 65 5.2
6 82 110 26 12
7 33 42 10 30
8 13 17 4.1 71
9 5.2 6.7 1.6 180
10 2.1 2.7 0.66 420
11 0.84 1.1 0.26 1,000
Light gray boxes represent visual magnitude (Mv ) values with <3 stars in FOV; dark gray boxes represent
DN levels below 7? read noise or 7 DN for the WAC. White boxes show the range of Mv values meeting
optical navigation requirements (?3 stars per FOV at 7? read noise)
Fig. 6, a Data Processing Unit (DPU) for image processing, motor control, and commanding.
By adding some additional boards to the DPU to support the other payload science instru-
ments, a significant savings was achieved in spacecraft resources, design effort, and testing.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 265
Table 5b NAC sensitivity to stars
Mv 10 s exposure, 10 s exposure, 10 s exposure, # stars of ? mag in
no pixel sum, 2 ? 2 pixel sum, no pixel sum, WA FOV
ensquared energy ensquared energy ensquared energy
= 43% = 70% = 30%
DN DN DN
0 19,700 32,000 14,000 0
1 8,000 13,000 5,600 0
2 3,100 5,100 2,200 0.01
3 1,300 2,000 890 0.01
4 500 820 350 0.03
5 200 320 140 0.08
6 80 130 56 0.2
7 31.33 51 22 0.48
8 12.29 20 8.8 1.2
9 5.04 8.2 3.5 2.8
10 2.03 3.3 1.4 6.7
Light gray boxes represent visual magnitude (Mv ) values with <3 stars in FOV; dark gray boxes represent
DN levels below 7? read noise or 14 DN for the NAC. White boxes show the range of Mv values meeting
optical navigation requirements (?3 stars per FOV at 7? read noise)
Fig. 6 Redundant Data Processing Units (DPUs) support the entire science payload. Photograph shows
DPU-A and DPU-B with their flight harnesses configured for thermal-vacuum qualification tests
266 S.E. Hawkins et al.
Fig. 7 High-level flow diagram showing relationship between the MDIS DPUs, spacecraft IEMs, and other
science payload instruments (see Leary et al. 2007, and references therein). The redundant DPUs provide
all processing and power (pwr) interfaces for MDIS, as well as the spacecraft interface for the other instru-
ments. Note: SAX = Solar Assembly for X-rays; MXU = Mercury X-ray Unit; FIPS = Fast Imaging Plasma
Spectrometer; EPS = Energetic Particle Spectrometer; GRS = Gamma-Ray Spectrometer; NS = Neutron
Spectrometer
To increase the reliability of the payload, redundant DPUs were created to buffer all data
interfaces between the payload elements and the spacecraft; one DPU is powered when-
ever a payload element is active, while the other DPU is maintained unpowered as a cold
spare. The DPUs communicate with the spacecraft processors via the spacecraft MIL-STD-
1553 busses (Leary et al. 2007), but they communicate with the instruments via separate
dedicated RS-422 Universal Asynchronous Receiver Transmitter (UART) interfaces. The
DPUs greatly simplified the spacecraft-to-payload interface issues, allowing payload devel-
opment and testing separate from the rest of the spacecraft. Figure 7 provides a high-level
flow diagram showing the relationship between the MDIS DPUs, the spacecraft Integrated
Electronics Modules or IEMs (Leary et al. 2007), and the other science payload instruments.
Payload power and mass limitations impacted the design and operational constraints of
the science instruments. The most limited power period for MESSENGER occurred during
early cruise, when the solar arrays generated their lowest power, restricting the size of in-
strument heaters that could be used. In contrast, during the orbital phase of the mission, the
solar arrays generate ample power, but during eclipse the battery power is still very limited.
The instrument designs were limited by their ability to dissipate heat to the spacecraft deck
or to the space environment.
The payload employs both distributed power and data processing for each of the instru-
ments. Each instrument (other than MDIS) has its own power supply and microprocessor,
thereby greatly reducing the risk of noise or software problems that could have impacted the
The Mercury Dual Imaging System on the MESSENGER Spacecraft 267
spacecraft integration schedule. This distributed interface architecture provided a balanced
tradeoff between payload reliability, power, mass, and cost constraints.
Although each instrument has individual power supplies and processors, all but one of
the instruments use a common power supply board and processor board. This common-
ality facilitated the use of identical power and data interfaces for the science payload. It
also allowed common software modules that would handle a number of common tasks,
including command and telemetry processing, timing, macro tools, data compression, and
inter-integrated circuit (I2C) bus mastering. This common software set greatly reduced the
amount of code development required for each instrument and reduced risks for payload
integration. Having these common power and data interfaces also allowed development of a
common set of ground support equipment (GSE) hardware and software that emulated the
DPU and power bus interfaces and provided a graphical environment for command menus
and telemetry displays. The detailed description of the common low-voltage power supply
(LVPS) and event processing unit (EPU) electronics is given in Sect. 3.4.3; the common
software description is provided in Sect. 3.5.1.
3 Instrument Design
The full MDIS instrument includes the pivoting dual camera system as well as the two
redundant external DPUs. The dual camera assembly without the DPUs is usually simply
referred to as ?MDIS.? The overall design and look of the MDIS, shown in Fig. 1, was driven
by mass limitations, the severe thermal environment, and the requirement for a large field-
of-regard needed for optical navigation and off-nadir pointing. A functional block diagram
of MDIS is shown in Fig. 8.
Fig. 8 MDIS does not communicate directly with the spacecraft, other than for spacecraft temperature sen-
sors and heaters. All power, control signals, and data are cross-strapped through the redundant DPUs, as
indicated in this functional block diagram for MDIS
268 S.E. Hawkins et al.
Fig. 9 Science instruments located within the payload adapter ring. The white calibration target was used
during in-flight radiometric calibrations
On the pivot platform are the multispectral WAC and the monochrome NAC. A passive
thermal design maintains the CCD detectors in the WAC and NAC within their operating
temperature range of ?45?C to ?10?C. Only one DPU may be active at a time, and due
to thermal constraints only one camera will operate at a time; however, observations with
the two cameras can be interleaved at 5-s intervals. A separate electronics assembly ac-
commodates switching between the various modes of operating with the redundant DPUs.
The pivot platform has a large range of motion (?240?) to allow the cameras to be ?tucked
away? to protect the optics from contamination. The pivot motor drive-train provides pre-
cision rotation over the 90? operational range of motion (Fig. 2) about the spacecraft +Z
axis.
A spectral calibration target was mounted on the inside of the payload adapter ring. Early
in the MESSENGER mission it was possible to tilt the spacecraft in order to provide solar il-
lumination on the calibration target. The large range of motion of the pivot assembly enables
either camera to point at the target, permitting an absolute in-flight radiometric calibration
and flat-field measurement. Figure 9 shows the calibration target, along with the four instru-
ments mounted inside the adapter ring. This picture, taken shortly before integration with the
launch vehicle, shows the final blanket configuration of the instruments with MDIS pivoted
for NAC observations of the target.
3.1 Optical Design
The WAC consists of a four-element refractive telescope having a focal length of 78 mm
and a collecting area of 48 mm2. The detector located at the focal plane is an Atmel (Thom-
son) TH7888A frame-transfer CCD with a 1,024 ? 1,024 format and 14-?m pitch detector
elements that provide a 179-?rad pixel (instantaneous) field-of-view (IFOV). A 12-position
filter wheel (FW) provides color imaging over the spectral range of the CCD detector. Eleven
spectral filters spanning the range 395?1,040 nm are defined to cover wavelengths diagnostic
of different potential surface materials. The twelfth position is a broadband filter for optical
navigation. The filters are arranged on the filter wheel in such a way as to provide comple-
mentary passbands (e.g., for three-color imaging, four-color imaging) in adjacent positions.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 269
The NAC is an off-axis reflective telescope with a 550-mm focal length and a collecting
area of 462 mm2. The NAC focal plane is identical to the WAC?s, providing a 25-?rad IFOV.
The NAC has a single medium-band filter (100 nm wide), centered at 750 nm to match to
the corresponding WAC filter for monochrome imaging.
One of several impacts of the thermal environment on calibration accuracy is the rela-
tive response of the CCDs at wavelengths longer than about 700 nm (Sect. 4.6.2). Response
at longer wavelengths increases strongly with temperature (Janesick 2001). If data are ac-
quired over a large temperature range, inaccuracies in correction for temperature-dependent
response will introduce systematic errors in spectral properties at 850?1,000 nm that ul-
timately could lead to false mineralogic interpretations. To protect the MDIS CCDs from
wide temperature swings, incoming thermal infrared (IR) radiation is rejected in the optics
by heat-rejection filters on the first optic of each camera. In the WAC, this rejection is ac-
complished using a short-pass filter as the outer optic; for the NAC the bandpass filter has a
specially designed heat-rejection coating on its first surface.
3.1.1 Wide-Angle Camera
Lens Design. The wide-angle camera (WAC) consists of a refractive telescope, dictated by
the required wide FOV and short focal length. The design approach was to select the sim-
plest lens design that gives acceptable image quality over the field; however, an important
constraint on the design is the limited selection of glasses because of the radiation environ-
ment. The MESSENGER mission is expected to be subjected to a total dose of <15 krad
(Si) (R. H. Maurer, personal communication, 2002). The design requirements were achieved
by starting from a simple Cooke triplet and splitting the central negative element. The re-
sulting design, a Dogmar (Fig. 10), gave good image quality although not over the full
395?1,040 nm spectral range. The uncorrected axial chromatic aberration was reduced by
varying the thickness of each filter, a technique used in the NEAR Multi-Spectral Imager
(MSI) (Hawkins et al. 1997).
The WAC is focused at infinity with a 78-mm focal length to spread the 10.5? FOV across
the 14.3-mm detector. The 14-?m square pixels of the CCD provide an IFOV of 179 ?rad.
The selected aperture of f/10 means that the lenses, manufactured by Optimax, are quite
small. Radiation-resistant glass was required, but only about 16 types were available. Fewer
than half of these have adequate transmission below 400 nm. Of these, a combination of
K5G20 and LF5G15 gave the best design solution. The optical transmissions of the coated
samples of the two glasses used are given in Fig. 11.
Because of the mapping requirement for the WAC, the optical system must have very low
distortion. The field curvature of the WAC produces a small focus variation across the field,
but it has a negligible effect on the final modulation transfer function (MTF). The distortion
has a maximum value of 0.06% at the corner of the field, which is less than a pixel. Figure 12
shows a representative MTF for the WAC. The two curves labeled ?On-axis+diffraction? and
?7?+diffraction? combine diffraction and the optical aberrations at these two field locations.
Here, the wavelength used for diffraction is 700 nm. Because the design is diffraction limited
Fig. 10 Optical schematic of the
WAC
270 S.E. Hawkins et al.
Fig. 11 Nominal passbands and
lens transmissivities for the WAC
along with the quantum
efficiency (QE) of the CCD
Fig. 12 Theoretical modulation transfer function (MTF) of the WAC. The various modeled contributors to
the MTF are shown, with the product of these labeled ?Total? for an on-axis ray and 7?-off-axis ray
over the entire field there is very little difference between the two lines. The figure shows that
the MTF degradation is dominated by the pixel size at high frequencies (note the Nyquist
sampling frequency of the detector). The surface degradation is dominant at low frequencies.
The actual surface wavefront degradation here assumes a root mean square (rms) wavefront
error of 0.1 waves through the system, including refractive index variations of the filter and
lenses. Pointing has a small effect if the movement is limited to less than one-half pixel
during an exposure.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 271
The on-axis MTF value of 0.1 occurs at 50 line pairs per mm (lp/mm), or 3.9 lp/mrad,
and is comparable to the Rayleigh criterion at which two features can just be distinguished.
The resolution defined in this manner changes across the FOV from 0.26 mrad on-axis to
0.30 mrad in the corner.
Spectral Filter Design. A 12-position multispectral filter wheel provides color imaging
over the spectral range of the CCD detector (395?1,040 nm). Eleven spectral filters are
defined to cover wavelengths diagnostic of common crustal silicate materials and have full-
width half maximum (FWHM) bandwidths of 5?40 nm (Table 6). A broadband clear filter
was included for optical navigation imaging of stars. Because the optical signal at Mercury
will be too high through the clear filter, the quality of the image through this filter was a
secondary requirement.
Each filter consists of two or three pieces of glass using a radiation-resistant substrate
of BK7G18 glass in combination with a long-pass filter glass. These colored glasses trans-
mit efficiently over a specified wavelength and have a sharp cutoff at shorter wavelengths.
The long-pass filter glasses are needed to block short wavelengths in the narrow bandpass
filters of the WAC. Two filters required an additional layer of S8612 to achieve the desired
passband. The designation G** added to the glass type identifies it as being radiation resis-
tant. The number after the G gives the percent cerium times 10 used as the dopant. Standard
glasses typically darken when exposed to radiation. Because of the uncertainty of the trans-
mission degradation by radiation effects on the long-pass filter glasses used in the WAC
filters, it was necessary to test the radiation effects on sample colored glasses provided by
the filter manufacturer, Andover Corporation. The samples were made of high-quality Schott
Glass that matched those used in the flight filters. The experimental setup and the results of
the radiation tests are provided in Appendix 1.
The variation in filter thickness used to remove residual chromatic aberration results in a
small variation in the focal length of the camera between filters. The extreme filters give a
focal length of about 78 mm at 480 nm and about 78.5 mm at 1020 nm, respectively. Table 6
Table 6 Detailed specifications for the WAC filters and effective focal lengths
Filter System System Peak Total Focal Scale
number wavelength bandwidth transmission thickness length change
(measured (measured (mm) (mm) (%)
at ?26?C) at ?26?C)
(nm) (nm)
6 430 18 0.694 6.00 78.075 ?0.216
3 480.4 8.9 0.875 6.30 77.987 ?0.329
4 559.2 4.6 0.810 6.30 78.023 ?0.283
5 628.7 4.4 0.898 6.20 78.109 0.173
1 698.8 4.4 0.892 6.00 78.218 ?0.104
2 700 600 ? 6.00 78.163 ?0.104
7 749.0 4.5 0.896 5.90 78.218 ?0.033
12 828.6 4.1 0.921 5.60 78.308 0.082
10 898.1 4.3 0.898 5.35 78.390 0.186
8 948.0 4.9 0.942 5.20 78.449 0.262
9 996.8 12.0 0.952 5.00 78.510 0.340
11 1010 20 0.964 4.93 78.535 0.372
272 S.E. Hawkins et al.
lists the filters in order of increasing wavelength and identifies the number assigned to each
filter, and the effective focal lengths for each one. This difference results in a variation in the
image scale of 0.7%.
By positioning the filter in front of the detector (Fig. 10), the size of the filters is mini-
mized. However, the incident angle ? of the beam on the filter varies with the FOV from 0?
at the center to 14.9? at the corners of the 10.5?-square FOV. This angle will cause a shift in
the spectral passband wavelength of the interference filters across the field according to the
theoretical expression
? = ?o
?
1 ?
(
no
n?
sin ?
)2
, (1)
where no is the refractive index of the external medium (no = 1 in vacuum), and n* is the
refractive index of the filter substrate. The effect of the incident angle is much more serious
than the spatial variations across the surface of the filter. With a maximum angle of 7.6? at
the corner of the field, spectral shifts of ?3 nm are expected. The problem with this shift
is that it is a variation across the FOV and not a constant offset; these small variations in
passband, however, are not expected to limit mineralogic identification or the mapping of
surface abundance variations.
Thermal Effects on Optical Design. The change of refractive index with temperature of
glass K5G20 varies between ?0.9 and +1.0 ? 10?6/K, depending on the temperature and
the wavelength. For LF5G15 the range is ?0.9 to 2.0 ? 10?6/K. Thus, for a temperature
change of 40 K the refractive index will change by a maximum of 8 ? 10?5. This is a
negligible change, as the index changes by this amount with a wavelength change of only a
few nanometers.
The performance of the WAC lens is almost constant with field angle over the design
range, so it is not necessary to show the variation with field angle. It does, however, vary
with wavelength, so compensation is added by varying the thickness of the filter between 1.7
and 2.9 mm. The results shown in Fig. 13 are for the spectral range 0.45?0.6 ?m, over which
there is very little change in the image quality with a single filter thickness. The lowest
curve is for a nominal temperature of 20?C and vacuum operation. At atmospheric pressure
the refractive index of air has a small effect as shown by the highest curve. The effect of
temperature is negligible as the curves show the change in the spot sizes at temperatures
of ?20?C and +20?C. The rms spot radii in this figure are only 0 to 4 ?m, which is very
small compared with the radius of the diffraction disk, 8.5 ?m at a wavelength of 0.7 ?m.
These results indicate that the WAC optical performance is insensitive to the operational
temperatures of the instrument.
Stray Light Analysis. The optical design of the WAC took into account various sources of
stray light including scatter from the optical surfaces, intrascene scatter, spurious reflections
between optical surfaces, and scatter from the CCD detector. Scatter from external surfaces
such as the optical housings, WAC light-shade, and surface contamination were also con-
sidered. However, at Mercury, MDIS will be exposed to a radiance source of large angular
extent. On orbit, the planet itself will be the dominant source of stray light.
The light-shade of the WAC was constrained in size and complexity because of mass
limitations. In addition, because the pivot platform was required to rotate 180? to stow the
cameras for launch and other possible contamination events (e.g., large thruster firings and
orbital insertion), the overall length of the shade was constrained. A two-piece shade was
constructed in order to minimize the size of the hole in the beryllium radiator needed for
The Mercury Dual Imaging System on the MESSENGER Spacecraft 273
Fig. 13 Variation of WAC geometric spot radius with temperature as a function of wavelength. Residual
chromatic aberration over wavelength has been compensated by varying the filter thickness from 1.7 to
2.9 mm. The variations shown are much less than a pixel (14 ?m square)
the light-shade to pass through, yet be large enough not to constrain the FOV. Concentric
grooves were machined into the titanium light-shade to minimize direct scatter into the
system from the shade.
Extensive measurements were made during ground calibration of the WAC to character-
ize the effect of stray light in the camera. However, limitations of the experimental setup
made separating chamber-induced stray light from internal scatter in the optical system of
the WAC difficult. In either case, all of the ground-based observations to characterize stray
light in the WAC showed the contribution to be small, on the order of 0.1% of the response
of the small extended source (lamp filament ?7 pixels) viewed on axis. In-flight calibrations
will be made to characterize the stray light performance of the WAC further.
3.1.2 Narrow Angle Camera
The primary purpose of the NAC is for high-resolution imaging of Mercury. Because of
the very bright optical signal from the planet, a large collecting area is not required for
sensitivity; the need to reduce blurring from diffraction, however, dictated a large (24 mm)
aperture. The reflective design has a long focal length that required folding the optical path
in order to fit in the available volume. The all-aluminum mirrors and telescope housing
were assembled and tested by SSG Precision Optronics, Inc. The monochromatic design
has a single medium-band filter centered at 750 nm with a FWHM of 100 nm. The center
wavelength was chosen to match filter 1 of the WAC (cf. Table 6).
274 S.E. Hawkins et al.
Fig. 14 Optical schematic of the
NAC
Telescope Design. The NAC has a 1.5? FOV that is spread across the 14.3-mm detector,
requiring a focal length of 550 mm. To limit the diffraction blur, the NAC has a 24-mm
aperture, resulting in an f/22 system. An off-axis Ritchey-Chretien design was selected
over a simple Cassegrain in order to avoid the central obscuration of the secondary mirror.
The optical ray trace is shown in Fig. 14. The ellipsoidal primary mirror and hyperboloidal
secondary mirrors are gold-coated aluminum with a surface roughness of 0.1 nm and 0.2 nm,
respectively. In this design, the image plane is tilted at an angle of approximately 9? for
optimal image quality.
A bandpass filter is the first optical component of the assembly and defines the spectral
range of the instrument. A specially designed interference coating serves as a heat-rejection
filter. Mercury absorbs solar radiation and reradiates this energy back into space. Because
of its slow rotation, the dayside of the planet can be modeled as a 400?C blackbody (Hansen
1974). The radiation from this blackbody results in a significant amount of IR radiation
that would pass through the NAC interference filter and heat the CCD. Figure 15 shows the
transmission of the interference filter for the NAC, with the response of the CCD superposed
on it. In addition, the reflectance spectrum of the heat-rejection coating is plotted with a
normalized 400?C blackbody spectrum.
Figure 16 shows the spot diagrams corresponding to selected field angles for the NAC.
The location of the focal plane was selected to balance the image quality across the field.
The circle shows the Airy disk. Low distortion of the image was an important design spec-
ification for the NAC. At the higher orbital altitudes, the NAC will be used to complete
the nadir and off-nadir mosaics at high resolution. The theoretical distortion of the NAC is
0.25% at the corner of the field. At the edge of the FOV this amounts to 1.28 pixels. The
all-aluminum assembly of the NAC makes it insensitive to thermal distortions over the op-
erational temperature range of the instrument. Figure 17 shows a representative MTF for the
NAC in the same format described earlier for the WAC.
The aperture stop is located in front of the primary mirror. Figure 18 shows the geomet-
rical spot radius out to an angle of 0.75?, the edge of the FOV. Note that the corners of the
field are 1.06? off axis. The geometrical performance is excellent, and the upper line shows
the limit imposed by diffraction.
Stray Light Analysis. The narrow FOV makes the NAC less sensitive than the WAC to off-
axis sources, making the need for a long light-shade less critical. Mass and size limitations
further constrained the light-shade design to be simple, without any internal baffling of the
light-shade. The effects of this approach for the NAC light-shade were analyzed by consid-
ering a full hemispherical illumination of the NAC, representative of imaging at Mercury.
This calculation shows that the scattered radiance as a fraction of the surface radiance of the
out-of-field stray light is negligible for this system, even with no light-shade at all. The two
The Mercury Dual Imaging System on the MESSENGER Spacecraft 275
Fig. 15 The monochromatic NAC spectral passband (700?800 nm) is shown overlaid on the quantum ef-
ficiency response of the TH7888A CCD. Also shown is the heat-rejection coating specifically designed to
reflect the longer IR wavelengths radiated from the hot planet, assumed to be a 400?C blackbody (shown in
the figure in dimensionless units)
Fig. 16 Spot diagrams for
selected field angles (azimuth,
elevation, both in degrees) for the
NAC. The diffraction-limited
performance of the telescope is
apparent when compared with an
Airy diameter of about 38 ?m or
2.7 pixels (solid circle in each
diagram)
main reasons for including the light-shade were to provide a surface to project through the
thermal system and to provide protection from stray light for optical navigation.
The NAC filter is mounted in the NAC light-shade and is tilted at 1? from boresight to
prevent direct scatter back onto the detector. With the final light-shade design and filter in
place, the calculated bidirectional transmittance distribution function (BTDF) has a value of
0.01 at 1? with a slope of ?2; this value is typical for an optical surface. A combination of
analytical and empirical methods was employed to identify and minimize scattered light in
the NAC telescope. Off-axis light entering the imagers will hit the internal walls or baffles
and be mostly absorbed by the black paint. As the absorption is not perfect some residual
scattering will result. No direct path exists to the detector, and most scatter requires striking
a minimum of two surfaces.
Off-axis rejection in the NAC was modeled extensively using the software package
TracePro? and the detailed computer-aided design model of the optics. The calculated level
276 S.E. Hawkins et al.
Fig. 17 Theoretical modulation transfer function (MTF) of the NAC. The various modeled contributors to
the MTF are shown, with products of these labeled total for an on-axis ray and a 0.75?-off-axis ray
Fig. 18 Geometrical spot radius as a function of field angle for the NAC
of rejection in the NAC is below 0.01%. Measurements to characterize stray light in the
NAC were acquired during ground calibration. However, as with the WAC, difficulties in
separating out internal camera scatter from scatter induced by the test chamber made mea-
suring the scattered light performance difficult. The measured results still show scattered
light rejection below 0.1% of the response of the small extended source (lamp filament ?60
pixels) viewed on axis in the NAC.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 277
Thermal Effects on Optics. The all-aluminum structure of the NAC telescope and gold-
coated aluminum mirrors was selected to minimize any thermal distortions to the optical
system. The telescope mechanical interface to the Focal Plane Unit (FPU) is a highly pol-
ished surface, and the input end of the telescope is supported by a titanium flexure mount.
To minimize heat entering the system, a heat-rejection coating was applied to the NAC fil-
ter as discussed earlier. The filter housing is also the light-shade and is made of titanium
to reduce heat flow into the telescope. The light-shade is painted black, as are the internal
surfaces of the telescope and FPU. The optical performance was verified over temperature
by measuring the wavefront distortion of the NAC telescope.
3.2 Electronics
The electronics systems of MDIS are fundamental to all aspects of the MESSENGER pay-
load. Not only do the fully redundant Data Processing Units (DPUs) provide the interface
between MDIS and the spacecraft, but each DPU also provides the interface for all the other
payload instruments as well. Because of all the redundant systems built into the spacecraft,
cross-strapping of these systems proved to be a significant task. The main electronics sys-
tems of MDIS include the DPUs, the DPU Switching Interface Electronics (DISE) box, and
the FPU camera electronics.
3.2.1 Focal Plane Electronics
The detector electronics for both the WAC and NAC are identical. The top-level block dia-
gram of the FPU is shown in Fig. 19. However, each CCD is bonded to a camera-specific
mounting bracket (heat-sink) prior to assembly into the FPU electronics. The NAC heat-
sink is tilted 9? to match the optimal orientation of the NAC focal plane, whereas the WAC
heat-sink has no tilt. Because of this unique mounting configuration for each camera, once
the CCD-heat-sink assembly is integrated into a set of FPU electronics, the electronics are
no longer mechanically identical and are uniquely defined as a NAC or WAC FPU.
Fig. 19 Block diagram of the MDIS Focal Plane Units and the DPU Interface Switching Electronics (DISE)
box. The NAC and WAC FPU electronics are identical in design
278 S.E. Hawkins et al.
Performance Flowdown. Each modular FPU contains a 1-megapixel CCD that has 14-?m
square pixels and antiblooming control. Due to thermal, power, and operational constraints,
only one camera operates at a time. The frame rate of each FPU is fixed at 1 Hz; however, the
frame rate to the spacecraft is not fixed but cannot exceed 1 Hz. Manual and autoexposure
control from 1 ms to ?10 s permits imaging over a broad range of intensities.
The requirement for a 1,024 ? 1,024 format imager with electronic shutter and an-
tiblooming dictates the general design of CCD that must be used. On-chip binning is re-
quired in order to achieve the 1-Hz data throughput to the spacecraft solid-state recorder
(SSR) and for data compression. The sensitivity required is not very difficult to achieve,
although the relatively short exposures, typically >7 ms, require a commercial rather than
scientific type of CCD. The frame transfer time is 3.84 ms, limiting the minimum useful
exposure to about 8 ms, and this limit requires post processing to recover a clean image
(Murchie et al. 2002).
General Design Considerations. A CCD from Atmel (formerly Thomson-CSF), the
TH7888A, was chosen because it was a good match for the requirements and also en-
abled design experience gained with the Comet Nucleus Tour (CONTOUR) Remote Im-
ager/Spectrograph (CRISP) instruments (Darlington and Grey 2001) to be used. The nom-
inally 1,024 ? 2,048 pixel array of the TH7888A has two on-chip output amplifiers (only
one amplifier is used in the MDIS design) and uses frame transfer to obtain electronic ex-
posure control. The active image forms an array of 1,024 ? 1,024 in which optical energy
is accumulated. At the end of the exposure period, the accumulated charge in each pixel is
quickly transferred to the masked-off 1,024 ? 1,024 memory zone. A tantalum radiation
shield not only protects the CCD from ionizing radiation but also blocks illumination to the
memory zone as shown in Fig. 20.
The CCD is operated in two modes: binned and full-frame. In the full-frame mode for
either the WAC or NAC, the first four columns of each image are taken from a region of
the CCD that is never exposed to light and, thus, represents a dark level that is purely a
function of bias and dark current. The dark columns are separated from the image section
by five isolation columns to avoid diffusion of signal from the active area. When the image is
read out, these four columns are mapped into the first four imaging columns, so the resulting
image is a square 1,024?1,024 pixels, with the first four columns replaced with the sampled
dark columns.
Fig. 20 The Atmel (Thomson)
TH7888A CCD with its tantalum
radiation shield in place
protecting the memory storage
zone of the frame-transfer-type
device
The Mercury Dual Imaging System on the MESSENGER Spacecraft 279
Fig. 21 Focal Plane Unit electronics, comprised of multiple rigid flex boards that fold into a compact design
Because of the severe mass limitations imposed on all aspects of the MESSENGER mis-
sion, the camera electronics were required to be smaller than the CONTOUR/CRISP FPU
electronics on which the design was based (Darlington and Grey 2001). A hybrid incorpo-
rating considerable parts of the clock driver circuits was developed by The Johns Hopkins
University Applied Physics Laboratory (APL). A type of construction of the printed circuit
boards known as rigid-flex enabled the electronics to be folded into a small space (Fig. 21).
The clock drivers use an integrated circuit (IC) designed to drive the gates of power metal
oxide semiconductor field effect transistors (MOSFETs) and are almost ideal to drive the
capacitive gates of a CCD. The Atmel CCD requires drive levels, speeds, and currents that
are compatible with the drivers. The TH7888A has a relatively high internal gain of 6 ?V/e.
This value is large enough to drive the correlated double sampler analog-to-digital con-
verter (ADC) integrated circuit, made by Analog Devices, directly while meeting the system
noise requirements. The output is a 12-bit digital word for each CCD pixel value. A field-
programmable gate array (FPGA) is used to provide clocks for the CCD and ADC derived
from an input clock from the DPU. The DPU also provides a coded exposure time signal
for the camera, as the autoexposure algorithm is in the DPU. Figure 22 shows the detailed
timing for short exposures (<1 s) in the top panel and long exposures (1 < exposure < 10 s)
in the bottom panel. The FPGA also formats the data for transfer back to the DPU. The in-
terface between cameras and DPU uses low-voltage differential signaling for low noise and
to tolerate differences in grounding between the units.
During normal operation the CCD is read out once per second whether or not images are
being saved. This action sweeps out accumulated dark current in the detector. The only time
this sequence is interrupted is when exposures of longer than one second are being made.
The CCD also has an antiblooming gate which removes accumulated dark current from the
image area before the start of an integration by effectively setting the full well size to zero.
Hybrid and Power Conditioning. The hybrid incorporates two driver ICs and a linear reg-
ulator with various decoupling and bias components into a package a little larger than a
single-driver IC. Figure 23 shows the MDIS hybrid with its lid removed and the two driver
ICs and linear regulator magnified. One hybrid can provide four CCD clocks, and a total
of three hybrids is needed by each camera. The frame transfer process, which takes place
in 3.84 ms, repeatedly drives high capacitance lines to transfer an exposed image from the
image area to the storage area of the CCD. The result is a large current requirement every
280 S.E. Hawkins et al.
Fig. 22 Detailed frame timing for the FPUs. All clocking, control, and readout of the CCDs is generated
within the FPGA in the FPU. The actual frame transfer (FT) time (xfer) is 3.84 ms
Fig. 23 The development of a
hybrid driver component
significantly reduced the overall
size of the MDIS FPU electronics
over earlier designs. (A) Custom
hybrid component with lid
removed showing external leads
and relative size.
(B) Enlargement of the two
MOSFET-driver integrated
circuits and linear regulator.
(C) Detailed view of one
MOSFET driver.
(D) Enlargement of the LM117
linear regulator
frame, occurring for a few milliseconds each second. This current would be enough to over-
load the converter driving the camera if it appeared at the power input. To avoid overload,
simple passive filtering is incorporated in the camera (see Fig. 19) to reduce the current
surge to an acceptable level.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 281
Table 7 Requirement flowdown
for the MDIS camera detector
electronics
Description Requirement
Size of image plane 1,024 ? 1,024 pixels, 14 ? 14 ?m
Other formats 2 ? 2 binned (512 ? 512)
Spectral coverage 400?1,100 nm
Electrical dynamic range Up to 4,000 : 1
Noise level 50 electrons
Exposure range 1 ms to 10 s
Linearity <1 %
Operating temperature, CCD ?40?C to ?10?
Operating temperature, housing ?40?C to +30?C
Power consumption <2 W
Test and Performance. The completed camera boards underwent a variety of electrical
tests before the CCDs were installed to ensure that they had the expected electrical perfor-
mance. After CCD installation, the camera design was tested with optical input to verify the
performance requirements.
In general the cameras met the requirements shown in Table 7 without difficulty. The dark
current for long exposures causes a small shift in dark level. The CCD provides specially
shielded columns to provide an accurate measure of the dark level. Four of these dark pixels
are sampled per line and substituted during readout, for four of the 1,024 active pixels in a
full-resolution image (1,024 ? 1,024). The four reference columns provide a measure of the
dark background. In the case of 2 ? 2 binning, no valid dark reference pixels are available.
The antiblooming feature of the CCD causes a scene-dependent nonlinearity. This non-
linearity is an artifact of the earlier design on which this one was based. The extent of the
nonlinearity was measured to be below the required upper bound. When large areas of the fo-
cal plane array are illuminated, there is no measurable nonlinearity. However, there is some
evidence that when small regions of the array are brightly illuminated, the linear response
could be outside the specification.
3.2.2 DISE
The MDIS instrument contains six major electrical subsystems: two FPUs, a filter wheel
motor, a filter wheel motor resolver, a platform pivot motor, and a platform pivot motor
resolver. The pivot motor contains redundant windings with one set going to each DPU.
The remaining systems must be controllable by either of the two redundant DPUs. Selection
between these units is provided by a DISE box. The DISE box is identified in Fig. 1, and the
overall signal flow is shown in Fig. 24.
The DISE box switches between the DPU and MDIS electronics while minimizing the
connections through the rotary twist-capsule feedthrough. There are two switching func-
tions: one selects which DPU is master and the other selects which camera will be active.
For the DPU selection, latching relays are used. When a DPU powers up, it asserts itself as
master by pulsing primary voltage on a control line to the DISE board. This pulse switches
relays that control the power and signal interfaces to the cameras. This pulse also switches
relays that pass DPU commands to the filter wheel, filter wheel resolver, and pivot motor
resolver. There is one relay for each control line needed for these interfaces. After the DPU
asserts itself as master, it then applies power to the camera of interest. Each DPU provides
separate power lines for each camera. The DISE board uses these power connections to
282 S.E. Hawkins et al.
Fig. 24 Detailed interface showing power and signals from redundant DPUs through the DISE box. Me-
chanical relays in the DISE box are switched depending on which DPU provides power to the DISE box. The
DISE box also selects which camera communicates with the DPU. The LVDS protocol used for all control
signals and data largely eliminated any length constraints for the harness. However, cable runs between the
DPUs and DISE were kept to a minimum for mass considerations
power its internal circuitry that selects the correct camera as well as powering the camera
of interest. This selection is done in a master/slave configuration with the NAC having pri-
ority. If power is supplied to both cameras at the same time, signals from the NAC will be
transmitted. This arrangement ensures that the master DPU communicates with the intended
camera.
In addition to delivering each camera with the 15 V provided by the master DPU, the
DISE box linearly regulates a 6-V supply. This regulation is done in the DISE box to limit
power dissipation and reduce mass in the cameras. An FPU draws significant current at 6 V;
the linear regulators therefore dissipate the majority of the 1.7 W of the DISE box. To ensure
good thermal conduction of these components so that they can safely dissipate their heat,
they are mounted directly to the lower corners of the DISE box frame. Since the DISE box
physically mounts onto a thermally nonconducting composite stand of the MDIS bracket
assembly, a copper thermal strap conductively ties the DISE box to the spacecraft deck.
A Nusil thermal gasket was used to ensure excellent thermal contact between the copper
heat-sink and the deck of the spacecraft.
The DISE box switches control lines using the low-voltage differential signaling (LVDS)
protocol to and from each camera. Each camera has a single LVDS interface, and each DPU
has a single LVDS interface. Switching between them is done with cold-sparing LVDS re-
ceivers and cold-sparing complementary metal oxide semiconductor (CMOS) multiplexers.
The power and enable signals for the LVDS and multiplexer chips, shown schematically in
Fig. 25, are derived from the active 15-V DPU camera supply power.
3.2.3 Data Processing Unit
The DPU electronics provide two distinct services: they act as a communications router and
interface between all the instruments and the spacecraft, and they provide complete support
The Mercury Dual Imaging System on the MESSENGER Spacecraft 283
Fig. 25 Low-voltage differential signaling (LVDS) interface between the redundant DPUs and MDIS.
Cross-strapping takes place in the DISE box, which selects either a WAC or a NAC through the enable
(ENB) of a tri-state bus transceiver when commanded from the active DPU
for the MDIS camera electronics. Two separate sets of electronics boards are physically
stacked and packaged together. Figure 26 shows a block diagram of the DPU. The first
set of boards, comprised of the LVPS, EPU, and UART hub board, provides basic power
conditioning, data processing, and communications functions. The second set supports only
MDIS and is comprised of two motor controllers, WAC and NAC power supply boards, and
a single imager interface board. The EPU and LVPS boards are common designs that are
used in all instruments (except MLA).
Event Processing Unit (EPU) Board. The EPU board, one of the two boards of the com-
mon payload design, provides data processing capability and creates the telemetry data pack-
ets passed (through the DPU) to the spacecraft. The design was based heavily on versions
that were flown on earlier instruments built at APL, including the Imager for Magnetopause-
to-Aurora Global Exploration (IMAGE) High Energy Neutral Atom (HENA) and CON-
TOUR CRISP and Contour Forward Imager (CFI) instruments. All these designs are based
on the Intersil (formerly Harris) RTX2010 16-bit processor. This processor, used in many
APL instruments and spacecraft subsystems for the past 15 years, directly executes Forth
language instructions. On MESSENGER the processor is operated at 6 MHz and can pro-
vide approximately 6 million instructions per second of processing power.
In addition to the RTX2010, the EPU board contains 64 KiB of fuse-link program-
mable read-only memory (PROM) for boot code, 256 KiB of electrically erasable PROM
(EEPROM), and 256 KiB of SRAM. An Actel RT54SX32S FPGA provides all the logic
284 S.E. Hawkins et al.
Fig. 26 Block diagram of the MESSENGER DPU
necessary to interconnect the processor with the stacking connector data bus, as well as
implement the interface UART (running at 38.4 kbaud) to the DPU, the test port UART
(running at 19.2 kbaud), and an I2C bus to control off-board serial peripherals. The EPU
also receives (via the DPU) a once per second timing pulse from the spacecraft. This timing
?SYNC? signal originates at the spacecraft IEM oven-controlled crystal-oscillator (OCXO
or coarse clock) and is used as an interrupt to the processor to maintain timing synchroniza-
tion throughout the payload. This synchronization pulse is a key part of the overall time-
keeping design to maintain <1 ms end-to-end timing error between the mission operations
center and the payload.
Low-Voltage Power Supply. The LVPS board converts spacecraft primary power, ranging
from 20 to 36 V, to isolated secondary power. The common payload LVPS design was based
heavily on versions that were flown on the CONTOUR CRISP and CFI instruments.
Most of the instruments operate from only four output voltages: ?5 V and ?12 V. The
EPPS instrument (Andrews et al. 2007) uses ?15 V instead of ?12 V. In addition to provid-
ing four secondary outputs, the converters also provide power switches for auxiliary primary
power loads and some analog housekeeping capability.
A design tradeoff was made between the use of a customized converter design and one
based on commercially available hybrid converters. Although a customized approach would
have provided better overall power efficiency, the cost and schedule limitations had a signif-
icant impact on the decision to use the VPT, Inc., family of converters, based primarily on
their small size and low noise output.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 285
Fig. 27 Block diagram of the low-voltage power supply board, one of the two boards common to all the
instruments (except MLA)
The functional block diagram for the LVPS board is shown in Fig. 27. The LVPS board
uses an in-rush limiter, a VPT electromagnetic interference (EMI) MIL-STD-461 filter, and
two VPT hybrid dual-output converters (one for ?5 V and the other for ?12 V). These
converters provide better than 1% load and line regulation for nominal operation and good
step response. Two radiation-hardened opto-field effect transistor (FET) relays were used
to switch up to 1 A of primary power to off-board loads. An analog monitoring circuit,
comprised of a 14-bit ADC, instrumentation amplifiers, a 16-1 multiplexer, and an Actel
FPGA, provided 16 input analog channels, which were used to monitor four voltages and
six currents on the board and five user-provided analog voltages from off the board; the
remaining channel measured board temperature.
286 S.E. Hawkins et al.
Table 8 MESSENGER payload requirements on the common low-voltage power supply board design
Instrument Load on Load on Total Estimated Total
?5 V ?12 V power efficiency efficiency
out (mW) on ?5 V
XRS 2,550 480 3,030 65% 55%
GRNS 3,250 1,440 4,690 67% 65%
EPPS 2,950 1,440 4,390 67% 65%
MAG 1,250 720 1,970 55% 48%
MDIS 1,000 480 1,480 50% 41%
DPU 3,250 0 3,250 67% 67%
MASCS 0 480 480 67% 30%
A major challenge in using a single-power-supply design for multiple instruments was
the ability to handle varying load requirements. Table 8 shows the range of output loads
between the payload elements that used the common LVPS board. A drawback of these
dual-output hybrid converters is the need to provide a minimum load on each output to
ensure that both outputs remained in regulation; however, this constraint did not result in
any significant power losses.
UART Hub Board. Communication between each DPU and the instruments (other than
MDIS) takes place through a Universal Asynchronous Receiver/Transmitter hub. This board
uses an Actel RT54SX32S FPGA, 1 MiB of static read-only memory (SRAM), and a number
of RS-422 receiver and transmitter circuits to implement eight 38.4-kbaud, RS-422, full
duplex UART channels. Each channel has a separate first circuit interface for maximum
fault tolerance. Independent SRAM-based receive buffers indexed by FPGA-based registers
are used to offload byte-by-byte operations from the DPU software. Data may be received
simultaneously from all eight channels. A single memory-mapped transmitter is steered to
any of the eight output channels, such that only one channel is actively being commanded at
a time. However, all output channels can be selected simultaneously for broadcast messages,
such as time distribution.
The local EPU board for each instrument (other than MLA) formats the instrument
telemetry packets. The telemetry is copied, byte by byte, into the SRAM buffers in the DPU
as the data are received from the payload. Data may be transmitted continuously or in bursts
and are instrument specific. The SRAM buffer is operated in a ping-pong fashion, where
one set of buffers is receiving data, while the other is made available to the DPU expansion
bus for processing by the DPU. Once per second, the DPU switches receive buffers, thus
introducing a 1-s latency between when instruments send data and when the DPU forwards
them to the spacecraft. The instruments are not allowed to send telemetry during a 300-?s
period once per second to accommodate this buffer swap operation. The DPU also sends
each instrument a differential one pulse per second (PPS) SYNC signal (also RS-422) to
help define this telemetry ?quiet time,? as well as provide a very good time reference for the
instruments to synchronize with the spacecraft 1-PPS clock.
1553 Interface Board. Communication between each DPU and the spacecraft is accom-
plished through MIL-STD-1553 busses. The 1553 Interface Board is an exact copy of a de-
sign used on the CONTOUR CRISP and CFI instruments. The design is based on the UTMC
SUMMIT UT69151DXE protocol/transceiver chip and also includes an Actel RT14100A
The Mercury Dual Imaging System on the MESSENGER Spacecraft 287
FPGA and 64 KiB of SRAM. Each DPU is a remote terminal on the spacecraft 1553 bus.
All payload commands and telemetry, except MDIS image data, are communicated via the
1553 bus.
The DPU receives a 1553 bus message from the spacecraft once per second containing
the time corresponding to the next hardware 1-PPS signal used to synchronize systems on
the spacecraft. This message is received at a fixed time before the 1-PPS signal. The DPU
transmits this time message to all instruments prior to the arrival of the 1-PPS signal via
the UART data links. This approach eliminates the software latency and jitter of processor
interrupt response and avoids the latency of UART transmission.
Imager Interface Board. All image acquisition control and hardware image processing oc-
cur in the Imager Interface Board. The detailed signal interfaces for this board are shown in
Fig. 28. This DPU slice is a modified version of a design used on the CONTOUR CRISP/CFI
instruments. It uses two Actel RT54SX72S FPGAs, 2 MiB of SRAM, and a number of
LVDS receiver and transmitter circuits to communicate both with the FPUs and the space-
craft high-speed IEM interface card. The Imager Interface Board can control only a sin-
gle FPU at a time. Camera selection and signal multiplexing occur in the DISE box (see
Sect. 3.4.2).
The Imager Interface Board contains two FPGA designs: an imager-interface FPGA and
an RTX-bus FPGA. Image data are continuously received from the active FPU and buffered
Fig. 28 Block diagram showing detailed signal interfaces for the imager interface board
288 S.E. Hawkins et al.
in SRAM. If the DPU receives a command to record an image, the image data are transferred
to either of the two redundant IEMs. The main function of the imager-interface FPGA is to
read images from the FPU and send them to the IEM. There are two possible image sizes
transmitted from the FPU: 512 ? 512 (CCD on-chip) binned images and 1,024 ? 1,024
full-frame images. Both formats use 12-bit pixels. A lossy compression method available
through eight selectable look-up-tables (Fig. 4) reduces the 12-bit image data to 8 bits.
A lossless algorithm, referred to as ?Fast? compression, is a second hardware compression
mode selectable by the imager-interface FPGA. With the two hardware compression options,
a total of eight possible image formats may be sent to the IEM.
The image data read from the focal plane are received from a serial LVDS interface at
16 MHz. A full 1,024 ? 1,024 image requires approximately 850 ms to be received in the
DPU. The data are sent at 4 MHz to the IEM, limiting the image frame rate. Binned images
may be sent to the IEM at 1 Hz; however, full-frame images can be sent only at 0.25 Hz. This
interface limitation to the IEM requires the imager-interface FPGA to buffer the images in
SRAM. The design consists of two main processes: one reads from the FPU and one writes
to the IEM. The FPU read and IEM write processes are completely separate, and each has
a handshaking signal to indicate that the respective process has completed. A third SRAM
control process determines which SRAM is being accessed by which process by monitoring
the 1-PPS and handshaking signals from the other processes.
Fast compression takes advantage of the fact that it is not usually necessary to represent
a pixel value by the full 12 bits. This technique can be further optimized by recognizing that
adjacent pixels often tend to be correlated. By differencing adjacent pixels, with a technique
known as differential pulse code modulation (DPCM), fewer bits are required to represent
the original pixel value. This technique, implemented in the imager-interface FPGA, permits
the maximum number of images to be stored on the SSR without paying the time penalty
for a more rigorous compression method. Each Fast-compressed image will be of variable
length, and its overall size will be scene dependent. Typically, a higher scene entropy will
result in a lower compression ratio.
The main functions of the RTX-bus FPGA are to store a 1,024-bit image header and the
FPU command words generated by software running on the RTX processor in the DPU.
This FPGA is also capable of generating a 32-bin histogram created from the pixel values
produced by the FPU. This histogram is made available to the RTX processor and is integral
to the MDIS autoexposure algorithm. The FPGA allows the RTX processor to control the
flow of the 1-PPS signal, and it makes the FPU status data available to the RTX processor.
A dedicated memory address on the RTX-bus backplane provides the RTX processor ac-
cess to the RTX-bus FPGA. Through this interface, the RTX processor can write the 1,024-
bit header and FPU control words, read the histogram data, and write the 1-PPS source and
output registers. There are three valid sources of the 1-PPS signal. The first two are from
each of the IEMs through the LVDS channels. The third is a register in the RTX-bus FPGA
to which the RTX processor can write. Another register to which the RTX processor writes
determines which 1-PPS source is sent to the FPU and down the DPU backplane.
Imager Power Board. The MDIS Imager Power Board provides switched +15 V to the
NAC and WAC electronics. The board is a modified version of the common LVPS board al-
ready described and shares the same in-rush current limiter, EMI and secondary filter, FPGA,
and telemetry monitoring design. The supply has only one VPT DVSA2815S DC/DC con-
verter model, which produces a regulated 15-V output. A separate MOSFET-switched output
is used for each set of NAC and WAC FPU electronics.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 289
Table 9 Characteristics of the actuators used on MDIS
Characteristic Pivot actuator Filter wheel actuator
Manufacturer Starsys CDA InterCorp
Primary drive 24 to 35 V 24 to 35 V
Two-phase stepper yes yes
Reference frequency 400 Hz 2.5 kHz
Maximum step rate 1.1?/s 75?/s
Electrical resolver 8 speed 1 speed
Motor input step size 5? 30?
Gear system Hybrid In-line planetary
Gear reduction 580 : 1 50:1
Output step size <0.01? 0.6?
Backlash <0.01? <0.6?
Mass (actuator + resolver) 916 g 139 g
Redundant windings yes no
Range of motion 240? 360?
Motor Controller Board. Each actuator in the MDIS design has a dedicated motor con-
troller board. Although the characteristics of the pivot and filter wheel actuators are differ-
ent (see Table 9), the board designs are effectively identical. Because most of the low-level
motor stepping functions are directly controlled by the DPU processor, the boards are fairly
simple. Each board uses an EMI filter, an Actel A1020 FPGA, and a bipolar UDS2998 mo-
tor driver to generate the appropriately phased motor drive waveforms. The angular position
of each actuator is measured via an AD2S80A motor resolver readout hybrid. The DPU
processor generates the motor phase drive timing signals and monitors the shaft position via
the DPU I2C bus. Opto-isolators are used to keep the primary power motor drive and control
signal grounds separate.
3.3 Software
3.3.1 Common Software
The APL-developed MESSENGER instruments share common software, an instrument do-
main library. The library was created in 1994 to serve the NEAR instruments. It is currently
used in MESSENGER as well as the Compact Reconnaissance Imaging Spectrometers for
Mars (CRISM) instrument on the Mars Reconnaissance Orbiter and the LOng Range Recon-
naissance Imager (LORRI) and Pluto Energetic Particle Spectrometer Science Investigation
(PEPSSI) instruments on New Horizons.
The common software runs as a layer below the application-specific software. The first
layer provides standard services, for example, command handling and telemetry packet
queuing. Another layer below the service layer is a host abstraction layer. This layer ab-
stracts the input/output (I/O) interfaces used by the instrument to communicate with the
spacecraft. The host layer must be customized for each mission given that spacecraft in-
terfaces are not standardized. For the MESSENGER instruments, two different host layers
were needed. The MDIS instrument is connected directly to the spacecraft via a 1553 bus.
The other instruments use RS422 serial lines to the UART hub board in the MDIS DPU.
Software in the DPU acts as an intermediary between the spacecraft and the instrument.
290 S.E. Hawkins et al.
The common software includes a boot program. Its most frequent use is to boot applica-
tion code by copying it from EEPROM to RAM and then starting the application. However,
the boot program also has commands for loading memory from the ground, dumping mem-
ory back to the ground, and copying memory to and from EEPROM. These features can be
used to upload new instrument programs. The boot program is stored in nonvolatile ROM.
In the event that an instrument?s program in EEPROM gets corrupted, the boot program will
allow the instrument software to be reloaded.
The common software provides command handling and telemetry queuing services.
Application-specific software registers command handlers with the common software. Reg-
istration provides the command?s operation code (opcode), the expected number of argu-
ments, and some code to handle the command. When the common software receives a com-
mand from the spacecraft that is properly formatted and has the correct opcode and suffi-
cient arguments, the provided handler code is executed. The common command software
can store and run sequences of commands, called macros. Macros can loop and call other
macros. Up to 64 macros can be executing at any time. The common software?s teleme-
try service provides formatting and queuing of telemetry packets. The software gathers the
packet?s data and constructs a Consultative Committee for Space Data Systems (CCSDS)
header with the appropriate flags, sequence number, and other information. The packet is
queued for transmission to the spacecraft. A 64-KiB buffer allows instrument software to
queue large bursts of telemetry that the common software then trickles out to the spacecraft.
The common software also includes status/housekeeping packet generation, time manage-
ment, voltage and current monitoring, and I2C bus management services.
3.3.2 MDIS-Specific Software
The MDIS-specific software controls the imager hardware, exposure time, and mechanisms.
It also serves as a communications hub for the other MESSENGER instruments. Time mes-
sages and timing pulses are broadcast to the other instruments. Commands are accepted from
the spacecraft and forwarded to the appropriate instrument. Telemetry is collected from each
instrument and forwarded to the spacecraft.
Imager Control. Only one of the MDIS cameras, either the NAC or WAC, can be used
at any given time. The MDIS software selects the camera to use based on user command.
Images are collected in three stages: expose the image in the FPU, read the image from the
FPU to the DPU, and finally transmit it from the DPU to the spacecraft. Different processing
steps are applied to the image during each of these stages. The steps include exposure, 2 ? 2
binning, lossy 12 : 8-bit compression, and lossless Fast compression. All of these steps are
done in hardware; in fact, the software cannot access the image at all. The software does
control the processing, for example, by enabling or disabling the image compression at
the correct time based on user commands. The software also manages the three stages of
exposure, read, and transmit as a pipeline. In the simplest pipeline, each stage takes 1 s and
therefore one image can be taken every second. More complex pipelines result when image
binning is disabled (unbinned images take 4 s to be transferred to the spacecraft) or when
multisecond exposures are commanded.
Exposure Control. The exposure time of images can be set manually by command or au-
tomatically by the software. In manual mode, the range of available exposure times is nearly
10 s. In automatic mode, the exposure time of the next image is computed by the software.
This computation has two distinct steps. The first step computes a new exposure time based
The Mercury Dual Imaging System on the MESSENGER Spacecraft 291
Fig. 29 Autoexposure algorithm decision tree. A 64-bin histogram is computed in hardware for each im-
age. If an image is determined to be underexposed, the actual exposure is computed as Actual = minimum
brightness such that the sum of the pixels above this brightness < saturation threshold
on the brightness of a previous image. The second step anticipates filter wheel motions and
adjusts the computed exposure time accordingly.
During the read stage of the image pipeline, the hardware generates a histogram of the
image. The histogram is analyzed by the software to determine if the image is overexposed
or underexposed. First, the histogram is scaled by a factor of four if it comes from a 2 ? 2
binned image. If the brightest histogram value exceeds a specified saturation threshold, the
image is considered overexposed and the exposure time is scaled back by a specified fallback
factor. Otherwise the image is considered underexposed. Histogram values are accumulated
starting from the brightest bin down towards the dimmest bin, until the specified saturation
threshold is exceeded. The brightness value that causes the sum to exceed the threshold
is defined to be the actual image brightness. The exposure time is scaled by the ratio of
the commanded target brightness, to the actual brightness, after a background brightness
is removed. The algorithm is characterized by uploadable parameters for the commanded
target brightness saturation threshold (the number of allowed saturated pixels), overexposure
fallback, and background brightness. The algorithm is depicted in Fig. 29.
The algorithm described so far compensates for changes in scene brightness and filter
wheel changes. The next step adjusts the exposure time further if the imager, binning mode,
or filter selected for the next exposure does not match what was used in the test exposure.
The exposure time is scaled by the ratio of the transmissivity of the old setup to the trans-
missivity of the new setup. An uploadable table of transmissivities for the WAC filters and
for the NAC in either binning mode are used. Finally, the computed exposure time is forced
to fall within an uploadable range but is always less than 1 s.
Mechanism Control. The MDIS pivot platform and filter wheel are controlled by stepper
motors. The motor controller boards are on an I2C bus, and each motor step must be com-
manded over this bus. There are also power supply boards on the I2C bus; power switches
must be commanded and analog housekeeping read from these boards. The common soft-
ware provides a mechanism for building and executing an I2C bus schedule. The MDIS
software uses this code to construct a four-slot schedule that repeats every 8 ms, i.e., 2 ms
per slot. One slot is used to step the pivot platform, another slot is used to step the filter
wheel, and the other slots are used to control the power supply boards.
Given the pivot motor gear ratio and the step rate above, the platform can move about
1.1?/s. Similarly, the filter wheel can move 75?/s. Given 12 filters, an adjacent filter can be
292 S.E. Hawkins et al.
reached in 0.4 s and any filter within 2.4 s. Note that the I2C scheduling method allows the
pivot platform and filter wheel to move simultaneously.
3.3.3 Software Metrics
The MDIS software, excluding the boot program, has about 8,000 lines of source code
(5,000 lines if only noncommand and nonblank lines are counted). Approximately 60% of
this total is shared/reused code in the common software. The compiled application is about
63 KiB.
3.4 Electro-Mechanical Design
The mechanical assembly of MDIS consists of two main components, a bracket assembly
that interfaces to the spacecraft and a pivoting platform on which both cameras and the
thermal control system are located. One leg of the bracket assembly supports the DISE box
and a rotary interface through which all electrical connections to the pivot platform pass.
The opposite leg of the bracket supports the precision pivot motor.
3.4.1 Bracket Assembly
The bracket assembly, made of the same advanced composite material (graphite/cyanate es-
ter fabric prepreg) as the MESSENGER spacecraft structure (Leary et al. 2007), was chosen
to minimize mechanical stresses due to mismatches in coefficients of thermal expansion.
The composite bracket assembly, being a very poor thermal conductor, isolates the pivot
platform and thereby the cameras from the payload instrument deck. Eight titanium feet
form the mechanical interface to the spacecraft. Four feet were bonded to each stand using
a precision fixture, then bolted to the stand. A trim-cut machined from a top layer of epoxy
on each composite stand ensured that the pivot axis of rotation was parallel to the spacecraft
deck.
Pivot Actuator. The pivot motor assembly is located on one of the composite stands and
is comprised of a stepper motor with a hybrid gear-train and electrical resolver. The stepper
motor contains redundant windings so that the two fully redundant DPUs can independently
control the motion of the pivot, enhancing the overall reliability of the pivot mechanism.
The total range of motion of MDIS is about 240?, limited by hard mechanical stops in
the pivot motor. Pointing knowledge is determined by first ?homing? the instrument, which
is accomplished by driving the actuator into one of the mechanical hard stops for a period of
time sufficient to ensure the orientation of the instrument if it had been previously stopped
at the opposite extreme of travel. The rotational speed of the pivot platform is ?1.1?/s, so
homing requires about four minutes. Once the location of the pivot actuator is known, the
flight software retains this knowledge, and subsequent pointing commands are achieved by
counting pulses (steps) to the motor.
The MDIS pivot actuator (Fig. 30) is capable of accurately stepping in intervals of 0.01?
(?150 ?rad) per step, achieved through a hybrid gear train within the actuator. The 5? stepper
motor is coupled to a 5 : 1 in-line planetary gear train, used to drive the wave generator of
a harmonic drive. Harmonic drives permit very precise pointing and high torque ratios and
weigh less than comparable conventional gear systems. These devices use a radial motion to
engage the teeth of a flex spine and a circular-toothed spline, rather than the rotating motion
of other gear systems. The harmonic drive in the pivot actuator provides an additional factor
The Mercury Dual Imaging System on the MESSENGER Spacecraft 293
Fig. 30 Schematic view of the pivot actuator, with callouts identifying the hybrid transmission system and
duplex pair radial bearings with outer races back-to-back (DB). An in-line planetary gear system drives a
harmonic drive, achieving the accurate small-step pointing of MDIS. Individual motor windings are routed to
each DPU, making the pivot actuator fully redundant
of 120 : 1 mechanical advantage, resulting in a total gear ratio of about 580 : 1. The hybrid
stepper provides for very high gear ratios and near zero-backlash performance resulting in
pointing knowledge <75 ?rad. The pivot actuator assembly also includes an angular position
sensor (electrical resolver). However, because of mechanical distortions in the drive shaft
and electrical noise in the device itself, the output resolution of the electrical resolver is
worse than a step size.
Protective covers for optical components are very desirable during ground-based testing,
launch, and trajectory maneuvers requiring large thruster burns. However, due to mass limi-
tations, conventional protective covers for MDIS were not practical. Instead, the critical first
optic of each telescope is protected by rotating the platform 180? from the spacecraft?s +Z-
body axis such that both cameras look downward into the deck. This innovative approach
acts like a reclosable cover and ensures a circuitous path for any particulate or molecular
contamination.
Drive Train and Pointing. A challenging requirement on the mechanical design was to
limit rotation-plane errors of the pivot platform to <100 ?rad. The pivot actuator mounts
on a titanium bracket, designed to isolate the platform assembly thermally from the com-
posite bracket stands and not introduce any pointing errors due to thermal gradients from
the stands to the platform. In order to minimize the error build up of the large number of
mating surfaces making up the drive train of the pivot axis, a significant effort was made to
hold very tight mechanical tolerances on all components affecting pointing of the platform.
These components include a bearing assembly, a hollow shaft, the pivot platform, the motor
shaft, and the pivot actuator. The tightest tolerances were placed on the mating surfaces of
294 S.E. Hawkins et al.
Fig. 31 Sequence of steps resulting in the precision drive train of the MDIS pivot assembly. (A) Duplex
pair of angular contact bearings pressed into bearing holder. (B) Hollow shaft press fitted into the diaphragm
and bearing assembly. Shaft then press fitted into platform assembly. (C) Motor shaft pressed into platform.
(D) Completed drive train of platform, prior to integrating onto composite stands. (E) Verification of rotational
trueness (runout)
these components, and interference fits were chosen to minimize error propagation in the
drive train. Once assembled, all mechanical interfaces were drilled and pinned to minimize
the chance of a change in the alignment due to thermal variations or vibration during launch.
The titanium bracket opposite the pivot actuator supports a bearing mount, flexible di-
aphragm, hollow shaft, and electrical feed-through. A duplex pair of angular contact bear-
ings (Fig. 31A) was chosen to support the hollow shaft of the pivot platform. The stainless
steel races and ceramic balls of these bearings ensure minimal molecular diffusion between
the two surfaces and reduce the risk of ?cold welds? that over time could decrease the torque
margin of the motor. The angular contact bearings use a dry lubricant, MoS2, and underwent
a run-in period prior to installation. After preliminary cleaning, the bearings were vacuum
soaked for 24 hours to eliminate moisture in the MoS2. The bearings were exercised in a
dry-N2 environment for 10,000 cycles (revolutions) clockwise, followed by 10,000 cycles
in the counterclockwise direction.
Installation of the angular contact bearings was accomplished by establishing a clear-
ance of ?12 ?m between the outer race of the bearings and the inner diameter of the bearing
holder. The bearings were pressed into the titanium bearing holder by heating the holder
while keeping the bearings at room temperature (Fig. 31A). The hollow shaft was fabricated
to have a line-to-line fit, i.e., zero clearance to within limits of mechanical measurement.
Assembly of the hollow shaft required cooling the shaft, then pressing it into the room-
temperature bearing and holder. A flexible titanium diaphragm is used to maintain the po-
sition of the hollow shaft in the radial direction; the diaphragm is secured to a bracket that
mounts to the composite stand. The diaphragm was designed to be sufficiently stiff to sup-
port the pivot assembly while remaining flexible in the axial direction, permitting limited
The Mercury Dual Imaging System on the MESSENGER Spacecraft 295
motion resulting from differences in thermal expansion coefficients between the composite
spacecraft deck and the magnesium/titanium platform drive train.
Perpendicularity of the mating surfaces on the hollow shaft, motor shaft, and pivot plat-
form were maintained to within machining limits (<25 ?m). The interference fits of the
shafts with the platform were achieved by heating the magnesium pivot platform before the
hollow shaft was pressed into place (Fig. 31B). The rotational orientation of the motor shaft
established the final alignment of the pivot platform relative to the spacecraft. Special keying
tools ensured proper orientation of the motor shaft prior to it being pressed into the platform
(Fig. 31C).
Once the rotating portion of the MDIS instrument was determined to satisfy the pointing
requirement in terms of trueness of rotation to within the allowed tolerance, it was necessary
to maintain the perpendicularity of the rotation axis to the mounting feet. A high-precision
fixture was fabricated to emulate the spacecraft instrument deck. Each composite stand was
mounted to the fixture, and the pivot platform assembly was loosely attached to the stands.
Using a Coordinate Measurement Machine (CMM), two parallel planes were established
relative to the mounting feet. The hardware attaching the motor drive train (bracket/pivot
assembly) to the composite stands was slowly tightened. An iterative process of measuring
the relative parallelism of each bracket to the mounting feet was performed. The diaphragm
bracket measured parallel to the motor stand feet to within 13 ?m. The assembly was bagged,
and the motor/diaphragm brackets were drilled and pinned to the composite stands. Runout
measured at the bearing end of the hollow shaft as shown in Fig. 31D was determined to
be <20 ?m (0.8 mils), corresponding to a rotational deviation of the pivot plane of motion
from normal of <85 ?rad.
Twist Capsule. A stainless steel rotary feed-through (twist capsule) mounts to the bearing
holder and passes through the hollow shaft. Set screws position the rotating shaft of the twist
capsule inside the hollow shaft. All electrical connections to the platform pass through the
110 conductors of the twist capsule. The angular extent of the twist capsule is +90? to ?220?
and exceeds that of the pivot motor. The combination of rigid-flex circuitry inside the twist
capsule and stainless steel bearings requires extremely low torque to rotate. The connectors
for each FPU were assembled after the twist capsule was installed. Most signal connections
through the twist capsule are redundant and spliced just before each potted connector. The
harness at the fixed end of the twist capsule largely connects to the DISE box with the few
exceptions used for spacecraft thermal control and monitoring.
3.4.2 Platform Assembly
Because of mass constraints, mechanical housings were fabricated from magnesium instead
of aluminum wherever possible. The pivot platform, the two FPUs, and the filter wheel
housing were fabricated from magnesium stock. To minimize the stress on optical compo-
nents due to thermal contraction, the WAC telescope, filter wheel, and the NAC and WAC
light baffles are titanium. The NAC telescope is aluminum with aluminum mirrors. The sup-
porting structure for the radiators and heat pipes is stainless steel, to match the coefficient
of thermal expansion (CTE) of the stainless steel heat pipes and to minimize parasitic heat
leaks from the radiators to the platform. The radiator panels are made of beryllium, selected
for its light weight and high heat capacity. Each beryllium panel is passivated with a thin
aluminum layer, which is then clear anodized to optimize its thermal properties.
296 S.E. Hawkins et al.
Fig. 32 Mechanical assembly of
WAC filter wheel. Unique design
of using an electrical resolver
with a spline shaft eliminated the
need for an additional bearing or
bushing opposite the motor. The
unusual shape of the filters
reduced the overall size of the
assembly
Filter Wheel Assembly. The 12 optical filters were bonded in place using a low-outgassing
epoxy. By optimizing the physical shapes of the filters, the overall size and mass of the
wheel assembly (Fig. 32) were reduced.
The filter wheel actuator for the WAC consists of a CDA InterCorp DS9-A stepper motor
with output step angle of 30?, and an inline gear train with 50 : 1 reduction resulting in an
output step size of 0.6?. The step rate is 75?/s, permitting 1-Hz imaging using adjacent filters,
provided the exposure is less than 490 ms. The output shaft of the motor has a precision
double-D interface to the wheel assembly and an internal spline to accept the shaft of the
CDA InterCorp electrical resolver (DT1-A) angular position sensor. The motor mounts to
the filter wheel housing, and the resolver mounts to the filter housing cover, eliminating the
need for a bearing and providing a simple robust design.
Focal Plane Unit. In addition to mechanically supporting the detector electronics, the me-
chanical assembly of the FPU was designed to maintain the position of the CCD at the focal
plane of the optics over the range of operating temperatures. The complex thermal design
(Sect. 3.5) required that the CCDs be thermally isolated from the electronics located in the
same housing.
The front panel of the FPU housing is a critical component in each camera, providing the
precision mechanical support for the CCD detector and the mechanical interface for each
camera?s telescope. The NAC front panel was fabricated from aluminum to provide the best
match to the physical characteristics of the telescope. The WAC front panel is magnesium.
Best focus in each camera was achieved using precision spacers providing a three-point
mount of the CCD to the front panel. Each focus spacer is pinned to the front panel to prevent
shifting of the CCD during environmental qualification and launch. The focus spacers also
thermally isolate the cold detector from the front panel of the FPU. Figures 33A and 33B
show the inside surface of the front panel and the three focus spacers supporting the CCD.
Epoxy-glass isolators increase the thermal resistance between the CCD heat sink and the
front panel. The CCD itself is bonded to an aluminum heat sink with a thermally conductive
epoxy. Panel C of Fig. 33 shows the assembled FPU with its electronics folded in place.
An internal multilayer insulation (MLI) blanket surrounding the CCD further minimizes
radiative coupling between the FPU electronics and the CCD.
Internal surfaces of the optical housing were painted black to minimize scatter. However,
because of the small size of the lenses and lens spacers required by the optical prescription
The Mercury Dual Imaging System on the MESSENGER Spacecraft 297
Fig. 33 Detailed integration of the NAC FPU housing. (A) Front panel, which supports the CCD on one
side and telescope on opposite side. (B) Three-point mount provides correct focal position and isolates cold
CCD from FPU front panel. Thermal blanket further isolates CCD from FPU electronics. (C) Assembled
FPU without top panel
in the WAC, painting of these small components was not practical. Instead, the titanium
lens spacers used in the WAC were anodized, resulting in a dark dull finish. Because these
spacers are not directly in the optical path, the trade between the titanium anodization versus
black paint was deemed acceptable.
3.4.3 DPU Assembly
The DPU design was optimized to keep overall mass low, accomplished primarily by us-
ing the 102 mm ? 102 mm (4?? ? 4??) slice architecture first developed on the CONTOUR
CRISP and CFI instruments. The slice architecture consists of a simple mechanical frame
that supports the printed circuit board. Both sides of the board are accessible for testing,
even after assembly into the mechanical frame. The generic slice design provides for mul-
tiple configurations of the DPU, which greatly facilitated the common payload elements of
MESSENGER.
3.5 Thermo-Mechanical Design
The MESSENGER spacecraft is protected from direct solar illumination by the spacecraft
sunshade. Strict rules govern the allowable attitude of the spacecraft in order to ensure that
Sun-keep-in limits are not violated. As a result, the thermal environment of the instruments
is largely benign with the exception of short intervals during the 12-hour orbital period
when Mercury?s illuminated surface subtends a large solid angle. The passive thermal design
of MDIS uses radiators to cool the instrument, but this radiative cooling system must be
disconnected during the short heat impulse that occurs near periapses at Mercury to prevent
the instrument from exceeding the CCD?s operational high-temperature limit (?10?C). This
thermal switching is accomplished by means of diode heat pipes. Survival heaters ensure
that cold operational limits are not violated.
The DPU is directly coupled to the spacecraft deck and blanketed with MLI. The ?40?C
to +50?C operating temperature of the DPU follows the deck temperature. The temperatures
of the imager interface board and the LVPS board are measured internal to the DPU and
reported in the instrument housekeeping data. A number of external spacecraft temperature
sensors report the deck temperature at the location of the DPUs and MDIS.
The MDIS temperature sensors are mounted at key locations on the instrument (Fig. 34)
and are reported in image headers. The CCD temperature is the most important temperature
for calibration (Sect. 4.2.1). The MDIS platform temperature and pivot actuator temperature
298 S.E. Hawkins et al.
Fig. 34 Location of the MDIS instrument temperature sensors. (A) Location of FPU housing temperature,
?CAM_T1,? for each unit. (B) Telescope temperature, ?CAM_T2,? for the NAC. (C) WAC filter wheel hous-
ing temperature, ?CAM_T2?. (D) Location of the CCD temperature sensor for each camera
are reported in the spacecraft housekeeping. These measurements are a spacecraft function
and are reported independently of the power state of the instrument.
3.5.1 Thermal Design of the MDIS Platform Assembly
Each camera has its own thermal control system, and each CCD is mounted to a heat sink
that bolts onto a bracket that is thermally isolated from the FPU. Precision titanium spacers
minimize heat conduction from the FPU to the CCD while maintaining focus over the oper-
ating temperatures of the instrument. A hole in the front panel of the housing allows light to
pass through the assembly and onto the CCD detector, and the surface of the FPU closest to
the CCD is unfinished to provide a lower emissivity surface to minimize radiative coupling
between the CCD and the front panel (Fig. 34).
The thermal link passes through a hole in the top of the housing and ties the CCD to
a phase-change material (PCM or wax pack) mounted to the top of the FPU housing. The
wax pack is also thermally isolated from the FPU housing using epoxy-glass standoffs and
MLI blankets. A thermal strap, secured to the top of the wax pack using a conductive epoxy,
thermally ties the wax pack to the evaporator portion of a stainless steel diode heat pipe.
Each camera has its own wax pack and heat pipe assembly. The two heat pipes mount inside
the radiator structure, and both pipes connect to each of the three radiator plates (Fig. 35A).
Each heat pipe is thermally connected to the inner surface of each of the three beryllium
radiator plates with copper saddles. A photograph and schematic of the thermal system are
shown in Figs. 35A and 35B, respectively.
Radiator Assembly. The three radiator panels are physically mounted to a stainless steel
structure, supported by thin titanium rods to form a spoke-like assembly that minimizes heat
The Mercury Dual Imaging System on the MESSENGER Spacecraft 299
Fig. 35 (A) Photograph of the diode heat pipes and radiator support structure. (B) Schematic diagram show-
ing key design elements that maintain the MDIS CCDs at nominal operating temperatures. Each CCD (NAC
on left, WAC on right) is thermally coupled to phase-change wax pack that buffers the detector temperature.
The wax packs are cooled by the radiator at times when the hot surface of Mercury does not occupy a large
part of the radiator FOV. The diode nature of the heat pipes acts as a switch and provides a thermal pathway
when the radiator FOV is cold and thermal isolation when it is hot. Hsg denotes housing
conduction from the radiators back to the platform (Fig. 35A). The structure also supports
the two diode heat pipes. Copper pads, or saddles, are soldered to the pipe at the location
of each radiator panel. The surface flatness of each saddle and radiator panel was controlled
to ensure optimum contact area. MLI blankets enclose the entire CCD-to-heat-pipe thermal
path, isolating the components from the rest of the instrument.
The three radiator panels are made of beryllium with a total area of about 340 cm2.
Beryllium was chosen for its large specific heat, and the panels were plated with aluminum
in order to obtain a finish (clear-anodization) with a low solar absorptivity (?) and high
emissivity. The low ? was required to protect against overheating of the instrument by direct
solar illumination as might occur in the event of a loss of attitude control of the spacecraft,
resulting in a tumble.
Diode Heat Pipe Description. The heat pipes, fabricated by Swales Aerospace, are made
from 9.5-mm-diameter stainless steel tubing filled with about 5 g of butane. During normal
operation, heat flowing from the CCD, and through the wax pack, will be absorbed at the
evaporator. Liquid butane is evaporated, and the gaseous butane flows down the center of
the tube to be condensed on the cold walls of the condenser. An internal stainless steel
wire mesh acts as a wick to transport the liquid from the condenser back to the evaporator,
continuing the fluid loop.
A separate pipe, the liquid trap, is attached to the evaporator section that allows the heat
pipe to act as a diode, i.e., allowing heat to flow only in one direction. A small tube con-
nects the liquid trap to the evaporator. The diode action occurs when the condenser/radiators
become warmer than the evaporator and the butane liquid in the condenser vaporizes. The
now gaseous butane flows down the pipe and condenses onto the walls of the relatively cool
(?10?C) evaporator and liquid trap. Since there is no wick in the small tube between the liq-
uid trap and the evaporator, the liquid cannot flow out of the trap, back to the evaporator, and
up to the condenser for reevaporation. As the radiator plates continue to heat up, more liquid
is vaporized and flows down to the liquid trap area. This process, referred to as ?burning out
the pipe,? continues until most of the liquid is located in the liquid trap and no more liquid
can flow back up the pipe to the condenser. In this state, the heat pipe is effectively shut
off and there is no longer any heat transfer between the evaporator and condenser. Once the
300 S.E. Hawkins et al.
radiators cool back down and the condenser again drops below the evaporator temperature,
the heat pipe will start back up and conduct heat like a regular heat pipe.
The heat pipe was made of stainless steel for several reasons. Stainless steel has a very
low thermal conductivity and a high melting temperature. When the heat pipes and radia-
tors/condensers are near their predicted maximum temperature (?230?C) due to the heat
impulse from the hot surface of the planet, the diode action of the pipe and the poor thermal
conductivity of stainless steel permits very little heat to be transferred back down the pipe
into the wax pack. Unlike an aluminum heat pipe that could soften at temperatures above
190?C, the mechanical properties of steel ensure that the pipe will retain its integrity above
300?C. Because the heat pipe is a pressure vessel, a leak in the pipe would cause it to cease
to function. A final essential element in the selection of materials for the heat pipe was that
the thermal interface of the mechanical connections between the beryllium radiator plates
and heat pipe condenser had to be maintained up to 230?C. The integrity of this interface
was maintained by soldering copper saddles to the stainless steel tube at both the condenser
and evaporator ends. The copper saddles bonded in this way were empirically determined to
tolerate the thermal stresses and maintain excellent thermal contact over the wide range of
temperatures (?77?C to +230?C).
PCM Wax Pack Description. The wax pack is an aluminum enclosure filled with paraffin,
which was selected for its melting point of about ?10?C. The camera CCDs are required
to be maintained between ?45?C and ?10?C throughout the mission. The bottom of the
wax pack has threaded inserts that provide a mounting interface to the CCD thermal link.
One side of a thermal strap connects to the top of the wax pack with a thermally conductive
epoxy and is overlaid with aluminum tape.
Thermal Strap and Link Description. The thermal link between the CCD and wax pack
was designed to carry 1 W of heat with a minimum drop of temperature. The thermal link
was made of multiple sheets of thin aluminum foil, to provide the required flexibility and
conductivity. Aluminum end blocks were swaged to each end of the foils to allow a mechan-
ical attachment to the CCD interface plate and to the bottom of the wax pack. The mating
surfaces on the thermal link end blocks, the CCD interface plate, and the wax pack were
polished to provide a good thermal connection.
The thermal strap between the wax pack top and the heat pipe evaporator was created
from two copper braids. One end of the braids was bonded into an aluminum end block, and
that block was then bonded to the evaporator of the heat pipe. The other end of the braids
was left free for permanent attachment to the top of the wax pack with thermally conductive
epoxy.
Functional Operation of Thermal Control System. In orbit about Mercury, when the space-
craft is away from the hot planet, the MDIS radiator plates do not see the planet and are
exposed only to deep space. In this condition, the radiator plates become very cold (?77?C)
and the diode heat pipes operate as normal heat pipes. The gas internal to the pipe moves
freely along the length of the pipe, transporting heat from one end to the other and radiating
that heat out to space. The heat from the CCD will flow from the CCD through the thermal
link, through the wax pack and thermal strap, to the evaporator of the heat pipe. As the liq-
uid in the pipe evaporates, heat flows along the heat pipe to the condensers and finally to the
radiator panels. When the paraffin in the wax pack is all frozen, the wax pack, thermal link,
and CCD will continue to cool below the freeze point of the paraffin. As the spacecraft starts
to approach the planet, the radiator plates begin to absorb heat from the hot planet. When
The Mercury Dual Imaging System on the MESSENGER Spacecraft 301
the temperature of the condenser rises above the temperature of the evaporator, most of the
heat-conducting gas in the pipe condenses in the liquid trap. In the burnt-out pipe condition,
very few gas molecules are present and the pipe is shut down, effectively disconnecting the
condensers and radiator plates from the rest of the thermal path.
With the heat pipe in its diode mode, namely, shut off, the heat that is still flowing from
the CCD will now be absorbed into the wax. The paraffin will warm up to its melting
point and remain at that temperature until all 240 g of paraffin have melted. The amount
of wax was selected so that all the wax would not completely melt during the hours sur-
rounding periapsis. The melting wax clamps the temperature of the CCD at its operating
warm-temperature limit (?10?C). As the spacecraft moves farther away from the planet, the
MDIS radiator and condenser temperatures will fall below the evaporator temperature and
allow the heat pipe to operate normally again. With the condenser temperature below the
evaporator temperature, heat will flow from wax pack to the radiators and begin refreezing
the paraffin. The partially melted wax pack will remain at its melting (freezing) point until
all the paraffin solidifies. At that point, the wax pack temperature will continue to drop down
to ?45?C where heaters and thermostats control the minimum temperature.
Other than the CCD-to-heat-pipe heat path, the majority of the camera components are
thermally connected to the main platform. Heaters on the platform prevent the camera optics
and FPU electronics from dropping below ?35?C. There are no radiators associated with
the platform. The cameras and platform are isolated from the environment by MLI blankets
and will not overheat due to the relatively small amount of heat dissipated in the electronics.
4 Instrument Calibration
Laboratory measurements described in the following sections were used to derive values
for the terms of the calibration equation as shown in (2) for both the WAC and NAC. Both
instruments measure relative light intensity in engineering units referred to as DNs. The
raw engineering units are converted to the physical units of radiance, L (W m?2 ?m?1 sr?1),
following the calibration equation:
L(x, y,f,T , ?, b)
=
[DN(x, y, f,T , ?, b,MET) ? Dk(x, y,T , ?, b,MET)]
Flat(x, y, f, b) ? Resp(f,T , b) ? ?
?
Sm(x, y, ?, b) ? Scat(x, y, f, ?, b)
Flat(x, y, f, b) ? Resp(f,T , b) ? ?
, (2)
where:
DN(x, y, f,T , ?, b,MET) is the raw DN measured by the pixel in column x, row y,
through filter f , at CCD temperature T and exposure time ? , for binning mode b and
Mission Elapsed Time MET ;
Dk(x, y,T , ?, b,MET) is the dark level in a given pixel, derived either from the dark strip
or estimated from exposure time and CCD temperature;
Sm(x, y, ?, b) is the scene-dependent frame transfer smear for the pixel;
Scat(x, y, f, ?, b) is the contribution of scattered light from elsewhere in the scene;
Flat(x, y, f, b) is the nonuniformity or ?flat-field? correction at this pixel location;
Resp(f,T , b) is the responsivity relative to the baseline operating temperature; and
? is the exposure time in milliseconds.
302 S.E. Hawkins et al.
The analysis used to derive each term of the calibration equation is described in the
following sections.
4.1 Ground Calibration Facility and Test Results
All calibration tests were performed in the Optical Calibration Facility (OCF) at APL. The
OCF consists of large, linked vacuum chambers with a host of support equipment. The
largest of these chambers, the instrument chamber, has an internal diameter of 1.3 m and a
length of 2 m and permits mounting the instrument on a three-axis motion stage. The motion
stage can rotate the instrument in azimuth and elevation through a range limited only by the
instrument harness and mounting hardware. Translation is possible over a limited range, but
it is sufficient to center the instrument in the collimated beam of the chamber. The interior
and ends of the instrument vacuum chamber are surrounded by cold walls, within which an
external refrigerator circulates a cooling fluid to reduce the internal temperature to approxi-
mately ?40?C. Most of the calibrations were performed at approximately ?30?C. However,
additional thermal configurations over the range ?34?C to +25?C were also performed, to
bracket the behavior of the instrument over the range of expected flight conditions. Fig-
ure 36 shows a schematic of the OCF test setup. Inside the instrument chamber, there are
two sources of calibrated illumination. In one configuration the instrument views a 508-mm
diameter integrating sphere through a quartz window. Dark cloth was draped around the
window and sphere opening in order to block unwanted external light from entering the in-
strument optical path. Sphere spectral radiance was calibrated by the manufacturer just prior
to the start of instrument calibration, for each configuration of the sphere?s four bulbs: two
45-W lamps and two 150-W lamps. One photometer was coupled through a fiber optic to
monitor the internal sphere brightness during all calibration runs. Two additional photome-
ters (not shown in Fig. 36) were mounted on the camera and allowed additional monitoring
of incident irradiance.
Rotating the azimuthal stage of the OCF 180? enables the instrument to look along a
beam tube into an off-axis parabolic collimating mirror, which has a focal length of 1.43 m,
operates at f/7, and is focused on a grating monochromator. The monochromator may be
illuminated by an incandescent lamp with quartz optics. At the input slit, a set of neutral
density filters may be used to attenuate the light by known amounts. At the exit slit, a set of
long-pass filters is used to remove higher orders (i.e., shorter wavelengths) from the grating.
The wavelength can be changed manually or can be set to scan under computer control. The
grating may be positioned to zeroeth order, which allows the full incandescent spectrum to
be passed.
A MgF2 window partitions the monochromator from the unit under test, allowing the
monochromator to be removed without breaking vacuum so that other sources may be placed
at the focus of the collimator. These sources include a point source (pinhole), pinhole array,
and test samples illuminated by an incandescent source and viewed off a silver fold mirror.
Four networked computers were used to acquire and monitor the calibration data. Each
computer had a specialized set of tasks to perform. The first was dedicated to running most
of the OCF instrumentation such as the motion stage, monochromator wavelength adjust-
ment, and filter wheel selection. The second computer served as a frame grabber that com-
municated directly to the instrument DPU. The operators ran the calibration scripts and
monitored the progress of the runs from the third computer. The fourth computer presented
real-time images and plots on its screen, allowing monitoring of calibration results during
each experiment.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 303
Fig. 36 Optical Calibration Facility hardware configuration used for MDIS calibration
304 S.E. Hawkins et al.
All data were saved locally on disk and were also automatically copied to the remotely
located MESSENGER Science Operations Center (SOC) within approximately 30 s of ac-
quisition. The SOC provides a Web interface to the data, organized by test, which includes
facilities for searching the database for images in particular configurations. In addition, the
SOC processed the image data, performing dark-correction (using the dark strips) and pro-
viding a variety of plotting and profiling options. These features were used in near-real time
to examine the data for instrumental issues and to validate test results.
4.2 Dark Level
The measured signal from the CCD, in the absence of incident photons, is the sum of three
major components: (1) dark current from thermal electrons, (2) an electronic offset, or bias,
of ?240 DN intentionally added to the readout to prevent occurrence of negative values
that would appear as zero DN words, and (3) noise picked up by instrument electronics.
Throughout calibration there has been no evidence of picked-up noise. In the following
discussion, the sum of dark current and bias is treated together as ?dark level.? The response
of pixel-dependent dark level to temperature and exposure time was investigated separately
for the NAC and WAC without on-chip binning (full-frame) and with on-chip binning.
4.2.1 Response of the Sensor System to Exposure Time and Temperature
Characterization of the WAC and NAC dark level was conducted over the CCD tempera-
ture range 25?C to ?40?C, as measured by a temperature sensor (AD590) bonded to the
CCD heat-sink mounting plate on each camera (Fig. 34D). This temperature sensor was
calibrated during instrument-level thermal-vacuum testing, where it was possible to install
multiple thermocouples throughout the MDIS instrument. In particular, a thermocouple lo-
cated in close proximity to the thermal link on the wax pack of each FPU was used as the best
measurement to the actual CCD temperature. A linear regression was used to fit the mea-
sured response of the CCD temperature sensor with the calibrated thermocouple response.
The other instrument temperature sensors, (also AD590s, see Fig. 34) located on each FPU
housing and the WAC filter wheel and NAC telescope, were calibrated in the same way. The
final expressions to convert the digital raw output to temperature are
T [degC] = A ? Raw[DN] + B. (3)
The coefficients A and B are listed in Table 10.
Figure 37 shows dark-image means (in DN) for the NAC and WAC full-frame (not
binned) and binned images as functions of exposure time and CCD temperature. In the
Table 10 Conversion coefficients from raw DN to ?C for internal MDIS temperatures
Description Name A B
WAC CCD temperature ccd_temp 0.2718 ?318.455
WAC FPU temperature cam1_temp 0.5022 ?262.258
WAC FW temperature cam2_temp 0.553 ?292.760
NAC CCD temperature ccd_temp 0.2737 ?323.367
NAC FPU temperature cam1_temp 0.5130 ?268.844
NAC FW temperature cam2_temp 0.4861 ?269.718
The Mercury Dual Imaging System on the MESSENGER Spacecraft 305
Fig. 37 Dark image mean as a function of temperature (DN) and exposure time for the WAC and NAC in
each binning state
NAC full-frame data, at short exposure times and low temperatures, the dark level is domi-
nated by the electronic bias and the image mean decreases with increasing temperature. This
relationship originates in the amplifier in the read-out electronics, which has a temperature-
dependent offset (not shown in this figure). After an image has been exposed, large current
pulses are sent from the power supply to the clocking signals controlling the movement of
charge in the CCD. These high-current clocking pulses increase the overall signal levels for a
very short time, as observed in the measured DNs. At longer exposure times (>200 ms) and
306 S.E. Hawkins et al.
higher CCD temperatures (>?30?C) the effects of dark current, which increases exponen-
tially with temperature, dominate. The NAC full-frame data show this expected behavior
of increasing dark level with increasing temperature and exposure time. Because on-chip
binning adds the outputs of four pixels prior to quantization, the impact of the dark cur-
rent is larger, and the onset of increasing dark level with temperature and exposure time
occurs at lower temperature and shorter exposure time. WAC dark levels are dominated by
the increase in dark current with higher temperature and longer exposure time.
4.2.2 Dark Models
For the purpose of modeling dark level, the WAC and NAC were treated as four different
instruments: NAC full-frame, NAC binned, WAC full-frame, and WAC binned. Two dark
models were derived from the OCF calibrations for each of the four modeled sensors: (1) a
forward model of the dark level (in DN) as a function of column number, row number,
exposure time, and raw CCD temperature; and (2) a backward model that uses the row-
dependent dark levels in the dark strips and extrapolates them across the FOV using the
same variables as the forward model. For both the forward and backward models, dark
level is treated as a linear function of column number, row number, and exposure time. The
exponential temperature dependence was approximated with a third-order polynomial for
ease of calculation. It is known that the CCD dark level temperature response will change
with time due to exposure to radiation in space, and this specific time dependence will be
determined in flight over the lifetime of the mission.
Forward Model. An overview of the forward dark model is illustrated in Fig. 38 depicting
the effect of dark level with increasing exposure and temperature. Consider the first of four
variables to the dark model: column position x, row position y, exposure time ? , and tem-
perature T . For the column-dependence at a particular row, exposure time, and temperature,
dark level is modeled linearly, where an offset ?(y, ?, T ) is the value at column x = 0 and a
slope ?(y, ?, T ) is the increase per column (per increment of x). Then, to find the dark level
at a given column, the model for this row, exposure time, and temperature is:
Dk(x, y, ?, T ) = ?(y, ?, T ) + ?(y, ?, T )x. (4)
However, the column-dependence will vary with row, so that in the second level of the
model, the offset ? and slope ? are each functions of row position y. Thus, linear expansions
Fig. 38 Schematic representation of dark model calculation decision tree. Each key parameter is identified,
and the arrow shows the direction of increase for that parameter. The gray scale suggests the effect of dark
level in the image as a function of these parameters
The Mercury Dual Imaging System on the MESSENGER Spacecraft 307
of ? and ? are
?(y, ?, T ) = A(?,T ) + B(?,T )y, (5)
and
?(y, ?, T ) = M(?,T ) + N(?,T )y. (6)
The offset ? to the column-dependent dark level has a value A(?,T ) in row y = 0, and in
each subsequent row y + 1 that value is incremented by slope B(?,T ). Similarly, the slope
to column-dependent dark level ? changes with each row by an offset M(?,T ) and slope
N(?,T ).
Note that at this level, the row-dependences of column-dependences are themselves de-
pendent on exposure time and temperature. A third level of linear regression can be added to
the model, such that A(?,T ), B(?,T ), M(?,T ), and N(?,T ) can each be expressed as its
own linear function of exposure time ? , using a temperature-dependent offset C(T ), O(T ),
E(T ), or Q(T ), respectively, and a temperature-dependent slope D(T ), P (T ), F(T ), or
S(T ), respectively:
A(?,T ) = C(T ) + D(T )?, (7)
M(?,T ) = O(T ) + P (T )?, (8)
B(?,T ) = E(T ) + F(T )?, (9)
and
N(?,T ) = Q(T ) + S(T )?. (10)
The expanded dark model at this level becomes
Dk(x, y, ?, T ) = C(T ) + D(T )? +
[
E(T ) + F(T )?
]
y
+
{
O(T ) + P (T )? +
[
Q(T ) + S(T )?
]
y
}
x. (11)
The fourth and final level of the model, C(T ), D(T ), E(T ), F(T ), O(T ), P (T ), Q(T ), and
S(T ), are all third-order functions of temperature, for example,
C(T ) = H0 + H1T + H2T
2
+ H3T
3. (12)
The forward dark level model was derived by first finding a linear fit of dark level as a
function of column number for each row in the dark calibration data (2). Second, the bias
and dark DN accumulation rate per column increment were fitted linearly as functions of
row (3?4). A, B , M , and N were fitted as functions of exposure time, from data binned
into 15 CCD temperature bins (each bin 10 DN wide) covering the range 1,000?1150 DN
(?40.8?C to ?6.8?C) using (5?8). Finally, the linear fit coefficients as functions of exposure
time were fitted across the temperature bins with a third-order polynomial (12).
The coefficients of these expansions are given in Table 11a for the four sensor models.
After launch these coefficients are serving as the bias for the time dependence of the dark
model, with the slopes to be calculated from post-launch observations.
To assess the accuracy of the model, the fitted model was differenced from the data used
to create it, and the rms errors between the data and model were measured. Figure 39 shows
that, for all four sensor models, over the expected range of temperatures and for exposure
times of 1,200 ms or less, the difference between the dark level models and the data used to
derive them is comparable to the read noise for each CCD (?1 DN for the WAC, ?2 DN
for the NAC).
308 S.E. Hawkins et al.
Table 11a NAC full-frame dark models
Coefficient H0 H1 H2 H3
Forward dark level model
C 4202.30 ?10.7314 0.00974273 ?2.94302 ? 10?06
D ?64.4884 0.179181 ?0.000165891 5.11765 ? 10?08
E ?2.58253 0.00754599 ?7.35219 ? 10?06 2.38873 ? 10?09
F ?0.000464774 1.31009 ? 10?06 ?1.23407 ? 10?09 3.88515 ? 10?13
O 0.0143372 ?4.21571 ? 10?05 4.12534 ? 10?08 ?1.34637 ? 10?11
P ?6.86389 ? 10?05 1.80933 ? 10?07 ?1.57409 ? 10?10 4.50925 ? 10?14
Q ?0.000169733 4.92790 ? 10?07 ?4.77059 ? 10?10 1.53993 ? 10?13
S 1.35800 ? 10?07 ?4.33913 ? 10?10 4.56621 ? 10?13 ?1.58643 ? 10?16
Backward dark level model
C 5451.46 ?14.2965 0.0131305 ?4.01470 ? 10?06
D ?66.9307 0.185911 ?0.000172068 5.30647 ? 10?08
E ?2.42407 0.00705565 ?6.84831 ? 10?06 2.21669 ? 10?09
F 0.00526574 ?1.44654 ? 10?05 1.32342?10?08 ?4.03222 ? 10?12
O ?1.81968 0.00519216 ?4.93259 ? 10?06 1.55995 ? 10?09
P 0.00351380 ?9.68964 ? 10?06 8.90189 ? 10?09 ?2.72453 ? 10?12
Q ?0.000402942 1.21434 ? 10?06 ?1.21845 ? 10?09 4.07108 ? 10?13
S ?8.26754 ? 10?06 2.6993 ? 10?08 ?2.07595 ? 10?11 6.32380 ? 10?15
Table 11b NAC binned dark models
Coefficient H0 H1 H2 H3
Forward dark level model
C ?5809.80 17.2831 ?0.0163855 5.17322 ? 10?06
D ?18.7770 0.0535211 ?5.08542 ? 10?05 1.61084 ? 10?08
E ?52.1256 0.148428 ?0.00014089 4.45864 ? 10?08
F ?0.00425778 1.22892 ? 10?05 ?1.18214 ? 10?08 3.78984 ? 10?12
O 0.676937 ?0.00190954 1.79397 ? 10?06 ?5.61987 ? 10?10
P 0.00180111 ?5.16803 ? 10?06 4.94001 ? 10?09 ?1.57305 ? 10?12
Q ?0.00688223 1.94574 ? 10?05 ?1.83419 ? 10?08 5.76568 ? 10?12
S 9.00182 ? 10?06 ?2.57628 ? 10?08 2.45929 ? 10?11 ?7.83098 ? 10?15
Backward dark level model
C ?8188.55 24.0208 ?0.0227352 7.16515 ? 10?06
D ?16.5122 0.0471223 ?4.48311 ? 10?05 1.42195 ? 10?08
E ?38.7242 0.110516 ?1.05155 ? 10?04 3.33600 ? 10?08
F ?0.0110371 3.13365 ? 10?05 ?2.96580 ? 10?08 9.35730 ? 10?12
O 7.69642 ?0.0217915 2.05305 ? 10?05 ?6.43961 ? 10?09
P ?0.00486634 1.36687 ? 10?05 ?1.27902 ? 10?08 3.98703 ? 10?12
Q ?0.0463603 0.000131138 ?1.23619 ? 10?07 3.88354 ? 10?11
S 2.89199 ? 10?05 ?8.17259 ? 10?08 7.69996 ? 10?11 ?2.41893 ? 10?14
The Mercury Dual Imaging System on the MESSENGER Spacecraft 309
Table 11c WAC full-frame dark models
Coefficient H0 H1 H2 H3
Forward dark level model
C 1238.24 ?2.76843 0.00256473 ?7.86953 ? 10?07
D ?3.48338 0.0101166 ?9.79576 ? 10?06 3.16249 ? 10?09
E ?2.42999 0.00714405 ?7.00585 ? 10?06 2.29185 ? 10?09
F ?0.00053432 1.49958 ? 10?06 ?1.40025 ? 10?09 4.34984 ? 10?13
O 0.0206338 ?6.29517 ? 10?05 6.39338 ? 10?08 ?2.16318 ? 10?11
P ?0.00033310 9.70986 ? 10?07 ?9.43818 ? 10?10 3.05943 ? 10?13
Q 0.000517513 ?1.51006 ? 10?06 1.46957 ? 10?09 ?4.77044 ? 10?13
S 1.17016 ? 10?07 ?3.34717 ? 10?10 3.19031 ? 10?13 ?1.01330 ? 10?16
Backward dark level model
C ?3182.94 9.81730 ?0.00936597 2.97916 ? 10?06
D 12.7966 ?0.0364940 3.46355 ? 1005 ?1.09384 ? 10?08
E 11.4352 ?0.0325653 3.08604 ? 10?05 ?9.73046 ? 10?09
F ?0.0755373 0.000216107 ?2.05838 ? 10?07 6.52700 ? 10?11
O 1.26711 ?0.00353596 3.28770 ? 10?06 ?1.01873 ? 10?09
P 0.00127217 ?3.57421 ? 10?06 3.33658 ? 10?09 ?1.03455 ? 10?12
Q 0.000637356 ?1.79417 ? 10?06 1.67841 ? 10?09 ?5.21619 ? 10?13
S ?2.32720 ? 10?06 6.63236 ? 10?09 ?6.28992 ? 10?12 1.98479 ? 10?15
Table 11d WAC binned dark models
Coefficient H0 H1 H2 H3
Forward dark level model
C ?484.568 2.11771 ?0.00206813 6.75547 ? 10?07
D ?10.2411 0.0299233 ?2.91567 ? 10?05 9.47476 ? 10?09
E ?27.9169 0.0813578 ?7.90653 ? 10?05 2.56248 ? 10?08
F 0.000646762 ?1.91510 ? 10?06 1.89122 ? 10?09 ?6.22884 ? 10?13
O ?0.550564 0.00152559 ?1.40630 ? 10?06 4.30355 ? 10?10
P ?0.00201059 5.92984 ? 10?06 ?5.83228 ? 10?09 1.91308 ? 10?12
Q 0.0127947 ?3.69327 ? 10?05 3.55415 ? 10?08 ?1.14037 ? 10?11
S ?1.54738 ? 10?06 4.29029 ? 10?09 ?3.95289 ? 10?12 1.20989 ? 10?15
Backward dark level model
C 876.088 ?1.72949 0.00155692 ?4.62411 ? 10?07
D ?8.67913 0.0253792 ?2.47483 ? 10?05 8.04835 ? 10?09
E ?24.2259 0.0705204 ?6.84466 ? 10?05 2.21526 ? 10?08
F ?0.00286560 8.40352 ? 10?06 ?8.21840 ? 10?09 2.68055 ? 10?12
O ?4.55575 0.0128502 ?1.20770 ? 10?05 3.78006 ? 10?09
P ?0.00660749 1.93028 ? 10?05 ?1.88063 ? 10?08 6.11108 ? 10?12
Q 0.00192523 ?5.01882 ? 10?06 4.27240 ? 10?09 ?1.17912 ? 10?12
S 8.80210 ? 10?06 ?2.61144 ? 10?08 2.58360 ? 10?11 ?8.52398 ? 10?15
310 S.E. Hawkins et al.
Fig. 39 RMS error of the dark models to image pixels as a function of dark image mean and exposure time
for the NAC and WAC, binned and not binned, from OCF calibration data
The Mercury Dual Imaging System on the MESSENGER Spacecraft 311
Backward Model. In the full-frame mode for either the WAC or NAC, the four dark
columns behave identically to the scene as a function of row, exposure time, and temper-
ature to within 0.26 DN. A second method to derive the dark level from the image data is
referred to as the Backward Model and uses the dark columns and a fitting procedure to
determine the dark level similar to that for the dark Forward Model.
In the binned mode for both cameras, true dark columns are unavailable due to an incor-
rect mapping of these columns in firmware. However, the second column of a binned image
provides a lower response than a column in the active image area. This lower-response col-
umn does show a temperature- and exposure-time response that can be modeled, making it
a functional ?dark column.? Therefore, the Backward Model simply uses the second column
of an image (binned or full-frame) to be representative of the dark strip properties.
The same fitting procedure used to derive the dark level for the Forward Model was used
for the Backward Model with the following exception. The bias for the fit to each row as
a function of column number was defined to be the value of the fit to the dark column at
that row number. Then, the regression error for a range of accumulation rates assuming this
fixed bias was calculated. The accumulation rate minimizing the regression error was used
as the accumulation rate of each row as a function of column number. From this set of co-
efficients (Table 11a), the exposure time dependence and the temperature bins were derived
and, as above, the rms error of fit regressions calculated. Since the bias derived from the dark
columns is not the same as the bias derived directly from the dark frames, it is expected that
rms errors will be slightly greater. For exposure times of less than 1,200 ms, the rms error
for the dark level derived from dark column Backward Model is, on average, larger than the
dark level derived from the Forward Model. Figure 39 compares the rms errors resulting
from these two models. The large disparity between the Forward and Backward models at
long exposures in the full-frame WAC is a consequence of a poor fit to the Backward Model
at long exposures.
Given the initial better fit of the dark level Forward Model to the data, as well as the
nature of the binned ?dark? columns, it is tempting to rely on the Forward Model exclusively.
However, the dark strips, even for binned data, serve as an indicator of the variations of each
CCD?s response to radiation and, as such, a means to calibrate the changes in the behavior
of the CCD with time. Thus, both dark models will be periodically reevaluated en route to
Mercury and during the orbital phase of the mission.
4.3 Frame Transfer Smear
Frame transfer smear corrections for MDIS follow the technique described by Murchie et
al. (1999) for the NEAR MSI imager. In brief, an image is exposed for a nominal integration
time and is then transferred to the memory zone of the CCD, from which the analog signal is
digitized line by line. Accumulation of signal continues during the finite duration of frame
transfer, inducing a streak or frame-transfer smear in the wake of an illuminated object in
the field of view, parallel to the direction of frame transfer, provided pointing remains stable
to the end of the frame transfer period. Quantitatively, the smear correction is:
Sm(x, y,T , ?, f, b)
=
y?1
?
1
t2
?
[DN(x, y, ?, b) ? Dk(x, y,T , ?, b,MET)] ? Sm(x, y, b, ? )
Flat(x, y, b, f )
, (13)
where Sm(x, y,T , ?, f, b) is the smear in column x and row y at exposure time ? and tem-
perature T in binning mode b and filter f . Dk(x, y,T , ?, b,MET) is the dark level in column
312 S.E. Hawkins et al.
Fig. 40 (A) Frame-transfer smear is apparent in this unbinned NAC short-exposure (1-ms) image of Earth
during the MESSENGER Earth flyby in August 2005. (B) Image after processing with desmear algorithm
x and row y at exposure time ? and temperature T in binning mode b at mission-elapsed
time MET . Flat(x, y, b, f ) is the flat-field correction in column x and row y in binning
mode b and filter f . ? is exposure time, and t2 is the time for frame transfer (3.84 ms) di-
vided by the number of lines in the image in the direction of frame transfer, i.e., 1,024 for
full-frame images or 512 for binned images. Empirically, it was found that this correction
removes frame transfer smear to the level of noise in typical, field-filling scenes for which
exposure time is >2 times the frame transfer time. For high-contrast scenes such as well-
exposed, non-field-filling extended sources imaged against a black background, artifacts are
at or below the noise when exposure time is >3 times the frame transfer time.
Figure 40 (left) shows an example of frame transfer smear for a full-frame NAC image
acquired during MESSENGER Earth flyby in August 2005. The image was exposed for
1 ms, less than the guideline, but taken to test the effectiveness of the desmear algorithm.
In the corrected image (Fig. 40, right), although the smear is slightly overcorrected, the
visual appearance of the corrected image is dramatic. The overcorrected artifact disappears
in typical scenes for exposure times ?8 ms. Thus, for analysis of MDIS filter passbands
(Sect. 4.5) and radiometry (Sect. 4.6), only images with exposure times ?8 ms were used.
4.4 Geometric Calibration
4.4.1 NAC and WAC FOV, IFOV, and Offset
To determine the fields-of-view and the relative alignment of the WAC and NAC, the output
slit of the OCF monochromator was imaged using white light at different positions of the
motion stage. The ends of the slit were measured from dark-corrected, desmeared images,
and the pixel positions and stage positions were used to determine the IFOV, FOV, and
relative pointing of each camera (Fig. 41). The angular difference per pixel position defines
the IFOV, and 1,024 times that value defines the FOV. The WAC FOV is 10.54? ? 0.02?,
and the IFOV is 179.6 ? 0.3 ?rad. For a 14-?m pixel pitch (specified by the manufacturer),
these figures imply a focal length of 77.96 ? 0.15 mm. The NAC FOV is 1.493? ? 0.001?,
and the IFOV is 25.44 ? 0.02 ?rad, implying a focal length of 550.3 ? 0.5 mm.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 313
Fig. 41 WAC and NAC fields of view as determined from the ground calibration. Note that the WAC and
NAC are rotated about 180? relative to one another. Scale, offset, and relative twist of NAC and WAC images
are given in the inset
This measurement also provides information on the relative orientations of the two FOVs.
The pixel positions measured at each stage position define a vector for each camera whose
angular separation defines the twist between the two FOVs. The NAC FOV is rotated relative
to the WAC?s by approximately 179.69? clockwise (CW), so that a scene appears ?right-side
up? in one camera and ?upside down? in the other. Allowing for this twist, solving for the
center pixel position in each camera in angular coordinates, and differencing the two results
yields the offset in boresight positions. The two boresights are offset by 0.21?.
4.4.2 Position within Pivot Plane
Through mechanical tolerances, deviation of the pivot plane from normal due to mechanical
error is less than 85 ?rad. Sampled measurements over the range of motion of the actuator
are consistent with the nominal stepping of 0.01?/step. We attempted to characterize fully
the single-step response of the pivot actuator but found systematic effects dominating the
measured response.
The flight software does not routinely command the pivot actuator to the operational hard
stop, but halts motion at a software stop located at +50?. To establish a second known posi-
tion within the ?240? range of motion of the pivot, detailed alignment measurements were
conducted during spacecraft-level tests. These measurements were repeated at the NASA
Goddard Space Flight Center facility after thermal-vacuum testing and just prior to the fi-
nal close-out of the spacecraft at Astrotech in Florida. The technique employed multiple
314 S.E. Hawkins et al.
Fig. 42 Final ground-based
coalignment measurement
between WAC, NAC, and
spacecraft. (A) Configuration of
spacecraft on test fixture,
showing theodolite in
foreground. (B) Close-up of
MDIS orientation with WAC on
left and NAC on right. (C) WAC
image and correct orientation
relative to geometry shown
in (A). Note cruciform projected
into theodolite gives a particular
pixel, mapped into the spacecraft
frame. (D) NAC image, acquired
by shining a bright white light on
subject located behind theodolite.
The reticule of the theodolite
provides the vector mapped into
the spacecraft frame
theodolites to reference the spacecraft master alignment cube and map the MDIS alignment
cube and boresights at a given pivot position into the spacecraft system. After commanding
the pivot to the 0 position of the actuator (apertures directed along spacecraft +Z-axis),
a reference flat was placed on the nonflight protective cover of the WAC. The theodolite was
autocollimated to this mirror, ensuring that the projected beam from the theodolite would
at least fall in the WAC FOV. The coordinates at this setting were recorded, and without
touching the spacecraft structure the red-tag cover was removed from the WAC. Several im-
ages were snapped by the WAC mapping a known field location in the WAC image into the
spacecraft coordinate system.
A similar procedure was done with the NAC, but because of the bandpass filter on the
NAC the green light of the theodolite could not be passed through the optical system of
the NAC. Instead, after autocollimating from a reference flat on the NAC red-tag cover and
recording the measurements in the spacecraft frame, a bright white light source was placed at
the eye-piece of the theodolite. After passing through the theodolite, the light was collimated
and successfully imaged in the NAC. Further, it was possible to identify the reticule on the
theodolite that was coaligned with the cruciform, which was autocollimated on the reference
mirror on the NAC before removing it. Figure 42 shows these two images. Note the WAC
image is rotated 180? relative to the NAC.
4.5 Wavelength Calibration
The NAC uses a single filter, while the WAC views the scene through a filter wheel outfit-
ted with 12 filters of varying widths. The WAC filters are labeled by their position counter-
clockwise around the wheel as viewed from the CCD. Using the OCF monochromator, NAC
and WAC filter transmissions were measured as functions of wavelength for CCD tempera-
tures ranging from ?35?C to ?26?C. The source appeared in the data as a bright rectangle
in the center of each image. After subtracting the dark model from the image values and
performing smear corrections, a mask was derived from each image based on bright and
The Mercury Dual Imaging System on the MESSENGER Spacecraft 315
dark pixel distributions. The values in the mask were set to zero for the dark regions and one
for the source rectangle. After multiplication, the pixel values for the resulting image were
summed to achieve maximum response with varying source wavelength. A Gaussian shape
was used to model most of the WAC and NAC filter centers and widths, except for WAC
filters 6 (430 nm) and 11 (1,010 nm), which were modeled with a combination of Gaussian
and polynomial fits. The center wavelength and FWHM passbands were calculated for fil-
ters 6 and 11 through a cumulative distribution technique, which is similar to a weighted
average. Specifically, after normalization, the data were summed as
C(?n) =
n
?
i=0
F(?i), (14)
where C(?n) is the cumulative distribution to ?n, F(?i) is the filter response at wave-
length ?i , and n varies between 0 and M , the number of monochromator wavelength settings
within the filter passband. The 50% point of the cumulative distribution was defined as the
center wavelength of those two filters, and the 75% and 25% points were differenced to find
the filter width.
Figure 43 shows fits calculated to WAC data acquired in the OCF at a CCD temperature
of ?26?C, which is about the median operating temperature expected in Mercury orbit.
These compare well with the manufacturer-measured passbands (Table 6). Filter 2 is a clear
fused silica filter spanning the entire passband of the CCD (395?1,040 nm) and is not shown
here. The NAC filter was measured to have a center frequency of 751 nm and a passband of
88 nm.
The center wavelengths shift systematically with temperature of the filter. However, with
the exception of the longest wavelength WAC filter (#11), in which the passband is affected
by a temperature dependence of the response of the CCD itself, they agree with the nominal
values to within 2 nm. Figure 44 shows the difference in filter nominal center wavelengths
between room temperature and a CCD temperature of ?26?C. The scatter in the wavelength
offsets at the lower temperature (?35?C) should be considered as a measure of the preci-
sion of determining the center wavelength (?0.5 nm). All filters show the expected shift to
longer wavelengths with increasing temperature. Filter 11 has a greater shift than the rest of
Fig. 43 WAC filter centers and
passbands at ?26?C
316 S.E. Hawkins et al.
Fig. 44 Offset in WAC filter
center relative to position at
?26?C, as a function of
temperature
Fig. 45 Shift in WAC filter 7
(750 nm) passband between
?31?C. Symbol labels give
temperature in DN and room
temperature
the filters, due to enhanced long-wavelength response of the CCD at higher temperatures.
Figure 45 shows fitted measurements for an example passband, the WAC 750-nm filter (#7),
and how the passband shifts with temperature.
4.6 Responsivity, Response Linearity, and Response Uniformity
Calibration of the signal accumulation rate per unit time per unit radiance at the sensor was
conducted in the OCF for the WAC and the NAC over a broad range of exposure times and
source light intensities. The exposure times to saturation were determined empirically in the
initial set-up. Using a variety of temperatures (?34?C to 25?C) and source intensities, im-
ages were acquired after varying the exposure times until saturation occurred. Source inten-
sities were stepped in a fixed decreasing-light pattern through the eight levels that could be
achieved with the integrating sphere?s two 150-W and two 45-W bulbs. The bulbs required
about 20 minutes of warm-up before asymptotically approaching constant light output. The
WAC calibration procedure began with all lamps on then subsequently turned lamps off in
The Mercury Dual Imaging System on the MESSENGER Spacecraft 317
sequence. The exposure sequence was repeated for each filter for the WAC, but only a single
exposure sequence was required for the single filter of the NAC.
With MDIS mounted in the OCF vacuum chamber with the integrating sphere set back
from the door looking into the window of the door, the sphere did not fill the FOV of the
WAC. It was not possible to acquire field-filling sphere measurements in vacuum and at cold
temperatures. However, room-temperature, ambient-pressure measurements were acquired
with the chamber door open and the sphere moved forward close to the instrument, filling
the FOV.
4.6.1 Response Linearity
The relationship between the raw DN output from the camera per unit exposure time and
radiance, as a function of variation in exposure time or radiance, is referred to as response
linearity. Using dark-corrected, desmeared DN image (DCDSI) values taken from the center
quarter of the images, response linearity was measured with both the NAC and WAC binned
and full-frame. Linearity was first examined separately at room temperature and cold for
(a) linearity with respect to exposure time at individual light levels, and (b) linearity with re-
spect to radiance at individual exposure times. To first order, the results are not significantly
different, so all data were merged to examine linearity with respect to measured photons (the
product of radiance and exposure time). However, it was found that each detector departs
from linearity at low signal levels. Nonlinearity only appears to be correlated with a specific
CCD (WAC or NAC), and not correlated with either binning state or detector temperature.
For each (WAC and NAC) filter, the responsivity was normalized to unity at a reference
dark-corrected, desmeared DN level of 1,500. Responsivity of each detector was found to
correlate with the background subtracted DN level in a linear-log relationship, as shown in
Figs. 46 and 47 for the WAC and NAC, respectively.
The measurements were taken using a calibrated integrating sphere, and a matrix of ex-
posure times and brightness levels. The clustering of the data in Figs. 46 and 47 shows that
there is no difference in system responsivity at a given signal level, whether that signal level
is achieved by differences in exposure times or by changes in scene brightness.
Fig. 46 WAC linearity response
shows no distinction between
short exposure and brightness
variation. These data were
acquired over all filters and a
variety of exposures and
brightnesses
318 S.E. Hawkins et al.
Fig. 47 NAC linearity response,
although different from WAC,
shows no systematic difference
between varying exposure,
brightness level, or temperature
4.6.2 Responsivity
Responsivity is the dark-corrected, desmeared DN per unit time per unit radiance at the
reference DN. Integrating-sphere radiances were calibrated by Labsphere using each of the
four lamps separately, over a wavelength range of 350?1,100 nm in 5-nm steps. To determine
radiances through each of the WAC and NAC filters, the band passes described in Sect. 4.5
were convolved with the sphere?s spectral radiance. For data taken at room temperatures, the
chamber door was open, but for lower-temperature measurements the sphere was viewed
through the quartz window in the OCF chamber door. Therefore, for the low-temperature
data an additional correction was applied to the sphere radiance for each filter to account
for the window. This correction was derived from room-temperature data, as the ratio of
corrected DNs per unit time with the door closed to that with the door open, for a single
configuration of sphere bulbs.
Figures 48, 49, and 50 show the responsivities measured for each filter for the full-frame
WAC at a range of radiances and exposures, for three different temperatures. Data points
are represented by crosses, which have widths showing 2-? errors of the dark model and
heights showing 2-? errors in responsivity. Each line connects the responsivities measured
at a single sphere radiance at different exposure times. For each filter, the data overlay to
within the errors in the dark model. Table 12 gives the actual responsivities for the NAC and
WAC. Figure 51 shows the temperature dependence of the full-frame WAC responsivities
for each filter, normalized to the CCD temperature of ?30.25?C. For these data, differences
between ?30?C and ?34?C at wavelengths <700 nm are assumed to be a measure of sys-
tematic errors. No significant temperature dependence of responsivity is expected in these
filters in that temperature range. Note that the plot shows the CCD temperature, and not the
temperature of the filters. The exact temperature of the filter is uncertain, but a characteristic
of these filters is a shift in wavelength toward the red with increasing temperature. Although
the temperature coefficients are small (0.015 nm/? at 400 nm and 0.025 nm/? at 1,000 nm),
the dispersions in the plots are largest near those filters at the edge of the quantum efficiency
curve and so could have a large effect on responsivity. This effect goes in the wrong direc-
tion, unless the filter temperature lagged in temperature from the CCD. A second factor that
may contribute to the large dispersion is that as the CCD temperature increases, the edges
of the band gap widen, which would tend to increase the responsivity, especially at longer
wavelengths. Figure 51 shows that around the operating temperature of MDIS, variation in
The Mercury Dual Imaging System on the MESSENGER Spacecraft 319
Fig. 48 Full-frame WAC responsivity at CCD temperature 1,267 DN (26?C). Radiance is in units
W m?2 nm?1 sr?1. The lines represent different light combinations; groupings of lines correspond to dif-
ferent filters, whose numbers as given in Table 6 are indicated. Note that the broadband filter (02) used for
optical navigation was not fully characterized and is not shown
Fig. 49 Full-frame WAC responsivity at CCD temperature 1,060 DN (?30?C). Radiance is in units
W m?2 nm?1 sr?1. The lines represent different light combinations; groupings of lines correspond to dif-
ferent filters, whose numbers as given in Table 6 are indicated. Note that the broadband filter (02) used for
optical navigation was not fully characterized and is not shown
320 S.E. Hawkins et al.
Fig. 50 Full-frame WAC responsivity at CCD temperature 1,046 DN (?34?C). Radiance is in units
W m?2 nm?1 sr?1. The lines represent different light combinations; groupings of lines correspond to dif-
ferent filters, whose numbers as given in Table 6 are indicated. Note that the broadband filter (02) used for
optical navigation was not fully characterized and is not shown
responsivity is a small effect. For the binned WAC, responsivity is greater than full-frame
WAC responsivity by a factor of 4.
Figures 52 and 53 show the responsivities measured for the NAC at cold temperatures.
Data points are represented by crosses, which have widths showing 2-? errors of the dark
model and heights showing 2-? errors in responsivity. Each line connects the responsivities
measured at a single sphere radiance through a single filter at different exposure times. Note
the highly exaggerated scale; the scatter in the determinations is only about 4%, which is a
measure of the absolute accuracy of the radiometric calibration.
Table 12 summarizes the responsivities of the NAC and WAC, both binned and full-
frame, at a reference CCD temperature of ?30.3?C. It is expected that the temperature
dependence of responsivity in both cameras over the CCD operating temperature range of
?45?C to ?10?C will be approximately quadratic. However, ground-based measurements
were acquired only at intermediate temperatures, and therefore the best correction avail-
able from these data is linear in form. A second-order correction will be determined from
targets of opportunity in flight (e.g., Canopus from Mercury orbit, at which time the CCD
temperature will vary over its operational range). The application of the linear correction is
Rf,T ,b = Rf,?30.3?C,b
[
correction_offsetf + TCCD[DN] ? correction_slopef
]
, (15)
where Rf,T ,b is responsivity in filter f at CCD temperature TCCD in units of DNs, and b is
the binning mode. Rf,?30.3?C,b is responsivity in filter f at CCD temperature of 1060 DN
(?30.3?C) as given in Table 12, and correction_offsetf and correction_slopef are camera-
and filter-dependent temperature correction offset and slope, also as given in Table 12. The
temperature correction defaults to unity at the reference CCD temperature of ?30.3?C.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 321
Table 12 WAC and NAC responsivities
Filter Responsivity Temperature Temperature
number (DN ms?1)/(W m?2 nm?1 sr?1) correction offset correction slope
WAC full frame
01 1.9359?105 1.249 ?2.3460?10?4
02 9.8165?104 0.956 4.1897?10?5
03 5.6441?104 1.831 ?7.8410?10?4
04 1.3643?105 1.569 ?5.3712?10?4
05 1.7488?105 1.533 ?5.0237?10?4
06 1.0788?104 2.303 ?1.2284?10?3
07 1.7803?105 1.091 ?8.6054?10?5
08 8.8803?104 ?0.353 1.2761?10?3
09 4.0546?104 ?1.284 2.1543?10?3
10 1.2662?105 0.240 7.1685?10?4
11 4.2273?104 ?1.855 2.6924?10?3
12 1.6698?105 0.746 2.3968?10?4
WAC binned
01 7.7437?105 1.249 ?2.3460?10?4
02 3.9266?105 0.956 4.1897?10?5
03 2.2576?105 1.831 ?7.8410?10?4
04 5.4572?105 1.569 ?5.3712?10?4
05 6.9952?105 1.533 ?5.0237?10?4
06 4.3152?104 2.303 ?1.2284?10?3
07 7.1210?105 1.091 ?8.6054?10?5
08 3.5521?105 ?0.353 1.2761?10?3
09 1.6219?105 ?1.284 2.1543?10?3
10 5.0648?105 0.240 7.1685?10?4
11 1.6909?105 ?1.855 2.6924?10?3
12 6.6792?105 0.746 2.3968?10?4
NAC full frame
3.5504?104 1.319 ?3.0184?10?4
NAC binned
1.3574?105 1.133 ?1.2592?10?4
Slope and offset are for correction to the average responsivity at a CCD temperature value of ?30.25?C.
Binned responsivity in each filter is four times the full-frame value, but the temperature correction offset and
slope are the same
4.6.3 Response Uniformity (Flat Field)
Response uniformity, or flat field, is a measure of pixel-to-pixel variations in responsivity.
Measurements of response uniformity of the WAC and NAC were conducted in the OCF, at
room temperature and while cold, by imaging the integrating sphere using two 45-W bulbs.
Figure 54 shows a 200-ms-exposure, binned WAC image acquired through the port hole
window in the OCF chamber door. As noted earlier, the port hole window is smaller than
322 S.E. Hawkins et al.
Fig. 51 Relative responsivity vs.
temperature for unbinned WAC
measurements, normalized to the
responsivity at ?30?C
Fig. 52 Full-frame NAC responsivities for CCD temperature 1,057 DN (?34?C). Radiance is in units
W m?2 nm?1 sr?1. The lines represent different light combinations
the WAC FOV. Four significant nonuniformities are evident in the data: (1) bright patches
around the rim of the sphere opening, nicknamed ?clouds,? (2) a slight fall-off in brightness
with field angle from near the center of the FOV, (3) slight horizontal striping, and (4) darker
spots scattered across the FOV. The clouds are believed to be reflections of the integrating
sphere off the radiator and thermal blankets that were mounted between the cold wall (on
the door) and the chamber door. Removing the thermal blankets in later tests significantly
reduced the clouds. The fall-off in brightness with angular distance from the center of the
FOV (Fig. 55) is the expected cos4 fall-off in response with field angle from the optic axis.
The horizontal striping is a function of the binning and is absent in the full-frame images.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 323
Fig. 53 Binned NAC responsivities for CCD temperature 1,053 DN (?35.2?C). Radiance is in units
W m?2 nm?1 sr?1
Fig. 54 Nonuniformities are
visible in integrating sphere
images acquired through the
quartz window in the OCF
chamber door of the calibration
facility
The darker spots scattered across WAC images are fixed with respect to the CCD re-
gardless of filter wheel setting, though their intensities do vary slightly with filter. The sizes
of the spots are consistent with shadows of 35-?m dust on the CCD window, and their
number density is consistent with the standards for a class-10,000 clean room in which the
camera was assembled. Also consistent with this hypothesis, following instrument vibration
during environmental testing, the locations of several spots changed. With the exception of
324 S.E. Hawkins et al.
Fig. 55 Profile cuts through an unbinned 699-nm filter WAC integrating sphere image, showing the cos4
falloff in response off the optic axis. Image is average of 100 images after dark-model and desmearing cor-
rection
a single particle (the black spot in the right center of Fig. 54) the dust spots do not signifi-
cantly affect the DN levels. Given this result, it is likely that the spots themselves will move
as the instrument is subjected to the vibrations of launch and flight. Images of the Venus
cloud tops acquired during the second Venus flyby will be used to redetermine the flat field
post-launch. Preliminary analysis of the flight images acquired thus far shows no significant
change in the number or location of spots.
To avoid contaminating flat-field measurements resulting from reflections in the OCF
facility, and to avoid having to construct the WAC flat field from tiling image data to form
image mosaics, it is desirable to determine the flat field data at room temperature with the
door open and the aperture of the camera in close proximity to the integrating sphere aper-
ture. To reduce noise in the derived flat field to approximately 10?3, ?100 images have
to be averaged together per filter, camera, and binning mode. Figures 56 and 57 compare
the central portions of binned, corrected, and averaged WAC integrating sphere images
acquired at room temperature and at cold temperature, respectively. When the images are
dark-subtracted, desmeared, averaged, and normalized to the image mean, the relative DN
levels are nearly identical on a pixel-by-pixel basis. Since the data are in good quantitative
agreement regardless of physical temperature, the flight flat-fields are derived from room-
temperature, door-open images.
Full-frame and binned flat fields are shown in Figs. 58 and 59, respectively. The values
are normalized to unity in the central part of the images used for responsivity determinations,
so that updates to responsivity from non-field-filling sources and to flat field from field-
filling sources can be decoupled. The NAC flat field is shown in the upper right panel in both
figures, and the 11 narrow-band WAC filters are shown in the remaining panels. Because the
spots in the NAC images were not observed to move during environmental qualification of
the instrument and they are immobile relative to the detector, redoing the NAC flat field
in-flight is not a requirement for the NAC as it is with the WAC.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 325
Fig. 56 Contrast-enhanced,
room-temperature, binned WAC
integrating sphere image with
chamber door open
Fig. 57 Contrast-enhanced,
cold, binned integrating sphere
image with chamber door closed
4.7 Point Spread Function
The point spread function (PSF) is a measure of the two-dimensional distribution of the
radiance measured at the detector emanating from a point source. For a conceptual but non-
physical imager, light emanating from a point would all fall into a single detector element.
In practice the PSF is broadened by diffraction, surface imperfections of optical elements,
and scatter centers on optical surfaces. The expected size of the Airy disk (approximately,
326 S.E. Hawkins et al.
Fig. 58 Contrast-enhanced, unbinned flat fields
the FWHM of the PSF including only effects of diffraction) is >2 pixels for the NAC and
?1 pixel for the WAC. In the particular case of the NAC, the PSF is significantly broader
than if the system were purely diffraction limited at the NAC focal length. For nadir imag-
The Mercury Dual Imaging System on the MESSENGER Spacecraft 327
Fig. 59 Contrast-enhanced, binned flat fields
ing at high southern latitudes, this degradation will limit effective sampling of the surface to
significantly worse than the desired 250 m/pixel spatial sampling.
The approach adopted for the NAC was to characterize carefully the PSF during ground
testing and use Fourier image restoration (optimal filter) techniques to improve the PSF dur-
328 S.E. Hawkins et al.
Fig. 60 Effect of image
restoration using the optimal
filter on an MDIS NAC image of
the Moon obtained during Earth
flyby. The left image is before
PSF correction
ing calibration. A similar strategy was applied to NEAR images to remediate PSF degrada-
tion due to contaminants on the imager?s optics (Li et al. 2002). The optimal filter deblurring
algorithm can be presented as
I = CF
|H |2
H(|H |2 + K)
, (16)
where I is the deblurred image, F is the original, calibrated image in units of radiance, H is
the PSF, |H |2 is the norm of complex number H (the square root of the sum of the squares
of the real and imaginary parts of H ), K is an empirically determined noise constant, and C
is an empirically determined constant that maintains radiometric accuracy. K is determined
as a value that provides the most sharpening of an image before ?ringing? is introduced
at abrupt bright-dark boundaries. The output of the correction is in arbitrary units that are
directly proportional to the input units of radiance, so C is the ratio of the summed pixel
values in the original and deblurred images.
For the deblurring procedure to work effectively, Cand K must be validated across a
variety of scenes, and the PSF must be both well determined and nearly constant across
an image. In the case of the NEAR imager, these constraints were all satisfied. There is
no requirement, however, that the PSF be radially symmetric about its central pixel. The
effectiveness of the procedure is shown in Fig. 60, a NAC image of the Moon acquired
during the MESSENGER Earth flyby.
For the NAC, the PSF was assembled from images of a subpixel pinhole imaged in white
light at the focus of the OCF collimator. Multiple exposure times were used, the lowest of
which is unsaturated, and the longer of which are progressively more saturated. At each
exposure time, multiple images are acquired. After correction for dark level, frame transfer
smear, and flat-field nonuniformity, the images are divided by exposure times to convert to
units of DN. The multiple images at each exposure time are averaged to improve statistics.
Starting with the longest exposure, saturated parts of the corrected, averaged image are ze-
roed out and replaced with unsaturated parts of the next longest exposure. The procedure
is continued until the central pixel is reached, and finally all values are normalized to that
of the central pixel. This procedure to assemble the PSF is repeated over a 3 ? 3 grid of
positions within the FOV (center, corners, and edges) to assess uniformity of the PSF across
the FOV.
Figure 61 shows the 3 ? 3 grid of NAC PSFs, displayed using a linear stretch between 0
and 1. The brightest part of the PSF exhibits its expected 2?3 pixel diameter, and the PSF is
closely similar across the FOV; there is no evidence for shapes of the PSF that are related
to position or distance relative to the optic axis. Figure 62 shows the PSFs similarly, except
displayed using a logarithmic stretch between 10?6 and 1. The shape of the PSF is well
The Mercury Dual Imaging System on the MESSENGER Spacecraft 329
Fig. 61 Two-dimensional intensity distribution of the NAC point spread function, at a 3?3 grid of positions
across the FOV. The ordinate and abscissa are labeled with pixel location. Pixel (0,0) corresponds to the
lower left. The stretch is from 0 to 1
determined out to a radius of 10 pixels. The shape is not exactly radially symmetric, but it
does appear nearly uniform across the FOV. Thus, the measured NAC PSF appears perfectly
suited to application of the optimal filter to improve effective spatial sampling of Mercury
in the NAC.
5 Mission Operations and Data Products
The MESSENGER Mission Operations Center (MOC) (Holdridge and Calloway 2007) will
conduct most spacecraft activities with reusable command sequences. These activities will
330 S.E. Hawkins et al.
Fig. 62 Two-dimensional intensity distribution of the NAC point spread function, at a 3?3 grid of positions
across the FOV. The ordinate and abscissa are labeled with pixel location. Pixel (0,0) corresponds to the
lower left. The stretch is logarithmic from 10?6 to 1
be translated into time-tagged commands that call macros prepared, uploaded, and veri-
fied by the MOC using the MESSENGER planning and scheduling system. Typically these
activities will include recorder operations, DSN track activities, and orbital and attitude
maneuvers and the integration of instrument operations. Instrument commands and house-
keeping commands will be merged at the MOC, and the sequences will be checked to ensure
that all activities are within operational limits and within available spacecraft resources (e.g.,
memory, power, thermal, recorder).
These reusable command sequences are built by the MOC with input from the subsystem
and instrument engineers. The sequences are built from a hierarchy of command groups. The
lowest level consists of fragments, which include one or more commands (with defined rel-
The Mercury Dual Imaging System on the MESSENGER Spacecraft 331
ative timing) required to implement simple actions on a single instrument or subsystem. The
next level consists of Canned Activity Sequences (CASs). A CAS includes one or more frag-
ments with defined relative timing required to perform an entire activity. This activity may
cross subsystem boundaries. CASs and fragments usually have input parameters to provide
flexibility in use, e.g., logic, timing, and command parameter values. An example of a CAS
would be an Optical Navigation Sequence, which first commands the Guidance and Con-
trol (G&C) subsystem to point the spacecraft at a star using G&C fragments. MDIS is then
commanded to take images using MDIS fragments. These CASs and fragments are tested
with hardware and software simulators and are configured on the planning and scheduling
system. The highest command level is the request, which calls one or more of these CASs
and associates a real spacecraft time and a set of input parameters defining each instantia-
tion of each CAS. Request files are created by the MOC for spacecraft commanding and the
science teams for instrument commanding. The concept of CASs and fragments is essential
to the philosophy of reusing pretested command blocks to save time in the preparation and
validation of spacecraft activities.
The MOC compiles spacecraft and instrument flight rules and operational constraints.
This rule set is translated into software run on the ground in the planning and scheduling
system. These rules assist the MOC and the science team in flagging command sequences
that may put the spacecraft or instruments in improper configurations, exceed onboard re-
sources (e.g., CPU memory, SSR space), or pass parameters to CASs and fragments that
are not permitted. These flagged errors are eliminated by adjusting timing between CASs,
altering CAS input parameters, or changing which CASs are used to make the command
sequence.
6 Data Acquisition Strategy
6.1 Flyby Imaging
The MESSENGER trajectory provides three flyby opportunities of Mercury: January 2008,
October 2008, and September 2009 (Table 2, Fig. 63). During the first flyby, approximately
half of the hemisphere not viewed by Mariner 10 will be illuminated (subsolar longitude
Fig. 63 Preliminary map, in simple cylindrical projection, of MDIS flyby imaging coverage at ?500 m/pixel
(at subspacecraft point) from a range of 20,000 km and emission angle ?70?
332 S.E. Hawkins et al.
190?E); the first Mercury data return from MESSENGER will thus observe new terrain,
including the previously unseen western half of the Caloris basin and its ejecta. During the
second flyby, illumination will be centered on the eastern edge of the Mariner 10 hemisphere
(subsolar longitude 4?E). The lighting geometry for the third encounter will be nearly iden-
tical to that of the second encounter with the subsolar point at the prime meridian (0?E);
the approach and departure phase angles will be less extreme, however, resulting in better
inbound imaging. During the second and third flybys, most of the remaining unseen portion
of Mercury will be imaged. Total coverage between Mariner 10 and the three flybys will ex-
clude only the poles and a small longitudinal gap ?6?-wide, centered at ?97?E longitude.
Due to the equatorial closest approach on all three flybys, resolution will diminish towards
the poles and limb.
Image mosaics from the various photometric geometries obtained during the flybys and
from orbit will require an accurate photometric model of the planet at the wavelengths of the
NAC and WAC filters. Therefore, MESSENGER will perform a sophisticated photometric
characterization of Mercury?s surface from data acquired during the flybys, as well as during
orbital operations, through observations of the same point on the ground acquired at various
emission, incidence, and phase angles. During each flyby one area on the surface will be
observed at 11 different phase angles in 10? increments. Because of Mercury?s slow rotation
these observations will have a fixed incidence angle?only the emission and phase angles
will change.
6.2 Orbital Imaging
Global Monochrome Basemap. One of the primary goals of MDIS is to acquire a global
monochrome base map at 250-m/pixel average spatial sampling, low emission angle, and
moderate incidence angle. For a given area coverage is obtained at the first opportunity when
local nadir is viewed at solar incidence angles of 55?75?. Selection of the combination of
emission and incidence angles dictates that the spacecraft will be nearly in a dawn?dusk
orbit, minimizing thermal disturbances to pointing. The choice of NAC or WAC is driven
by the necessity of maintaining both cross-track overlap and near uniform spatial resolution
(140 m/pixel, Fig. 64): the NAC will be used to image the southern hemisphere, whereas
the WAC will be used in the northern hemisphere (Fig. 64). For monochrome imaging, the
750-nm filter is used in the WAC to match the 750-nm filter of the NAC. The global nadir-
viewing basemap is planned for completion during the first Mercury solar day (i.e., during
the first half of the orbital mission).
Stereo Mapping. The off-nadir stereo-complement to the basemap will consist of images
taken at nearly the same local solar time. This situation occurs with the spacecraft in the
same nearly dawn?dusk orbit, again minimizing the thermal disturbance to pointing and the
propagation of pointing uncertainty into uncertainty in southern-hemisphere elevations. Off-
nadir pointing will be accomplished using the capability of the guidance and control system
to point up- or down-track by commandable offsets.
The downlink profile over the course of the orbital mission is heavily front-loaded due to
favorable Earth?Mercury distance, so the off-nadir complement will be time-phased to take
advantage of this high downlink rate. In the southern hemisphere, stereo mapping will be ac-
complished entirely during the first solar day. Imaging on one orbit will be at nadir (0? emis-
sion), then on the next orbit off-nadir (25? emission), and the sequence will be repeated so
that nadir and off-nadir mapping will be built up simultaneously. The northern hemisphere
will be imaged at nadir on the first solar day, then on the second solar day the same image
sequence will be repeated off-nadir, covering locations at the same local solar time as in the
nadir map. The vertical precision of the stereo map will typically be about 100 m.
The Mercury Dual Imaging System on the MESSENGER Spacecraft 333
Fig. 64 Preliminary map of spatial sampling in the MDIS global base map, in simple sinusoidal projection
Gap-filling the Flyby Color Map. Once in orbit, the remaining gaps in the flyby color maps
will be filled. Details of the mapping strategy are still under development, but acquisition
will mesh with the global monochrome and stereo mapping. The southern hemisphere will
likely be imaged as mosaics centered on given latitudes and longitudes, while the northern
hemisphere will probably be imaged as strips that are pixel-binned on-chip, in order to obtain
images in a greater number of filters with no gaps.
Polar Mapping. In order to identify permanently shadowed (and permanently illuminated)
areas, polar regions will be imaged repeatedly throughout the Mercury solar day. Over two
orbits (24 hours), the subsolar longitude will change by only ?2?. This strategy provides
coverage of all areas near their minimum solar incidence angle. Coverage will extend from
85? or lower latitude to the pole on the night side, and from 80? or lower latitude to the pole
on the dayside. At the south pole, the campaign will be divided between the two solar days.
On the first solar day, the WAC will be used while the spacecraft is at high altitude, providing
?1.5 km spatial sampling and extending equatorward to approximately 70? latitude on the
dayside. On the second solar day a more limited region will be covered at higher resolution,
using one 2 ? 3 NAC mosaic per day with its long axis aligned north?south. The latter
provides about 200 m/pixel spatial sampling.
For the northern polar region, the entire first solar day is required for the monochrome
base map, so the polar campaign will be delayed until the second solar day. The WAC will
be used because its wider FOV is required for latitude coverage. The imaging strategy will
parallel that used for the NAC in the southern polar region, using one 2 ? 3 mosaic per day
with its long axis aligned north?south.
334 S.E. Hawkins et al.
High-Resolution Imaging. Selected areas of the northern hemisphere, identified for the
most part in flyby imaging, will be imaged from orbit at resolutions of up to ?20 m/pixel.
For this purpose the NAC has a ?fast? 1-Hz mode, which will acquire images binned 2 ? 2
on-chip in continuous strips 512 pixels wide. At low dayside altitudes (280 km), spatial sam-
pling is 11 m/pixel but along-track motion smear is 18 m at the shortest exposure, consistent
with low artifacts from frame transfer smear. Because of MESSENGER?s highly eccentric
orbit, high-resolution imaging is possible only in the northern hemisphere.
Targeted Color. Selected regions of the planet will be targeted on the second solar day
for full-resolution color imaging with spatial sampling up to ?400 m/pixel using the ability
of the guidance and control system to track a commanded latitude and longitude on the
surface. The maximum spacecraft angular velocity dictates the minimum altitude and thus
best spatial sampling at which this imaging can be done. Initial targets will be identified
from Mariner 10 data and MESSENGER flyby results.
Color Photometry. During the orbital phase of the mission, photometric geometry com-
plementary to that measured during the flybys is possible: incidence angle variations will be
measured at nearly constant emission angle. Key features in low southern latitudes will be
imaged repeatedly while the spacecraft is high (10,000 km) above high southern latitudes as
the subsolar point moves from terminator to near noon-time. Several target areas will be se-
lected: one to represent average Mercury (heavily cratered highlands), another representing
intercrater plains, and a third representing smooth plains materials (probably Tolstoj plains).
Including the flyby and orbital observations, the photometric properties of Mercury will
be characterized over greater than 80? of emission angle, 100? of phase angle, and 80?
of incidence angle. Results from this experiment will not only characterize photometric
properties of the surface but also photometric normalization to a standard geometry (30?
solar incidence angle, 0? emission angle) for production of mosaicked products.
On-orbit Calibrations. In order to maintain accuracy of southern-hemisphere elevation
measurements, pointing calibrations will be interleaved with nadir and off-nadir global map-
ping. Once or twice per orbit, depending on orbital geometry, MDIS will be pointed off the
planet using the pivot at least at two different pivot angles, permitting star images to be
acquired using the WAC. Subframes centered on stars will be used to minimize the data
volume required. On-ground, these data will be used to solve for pivot plane orientation as
a function of temperature at the base of MDIS.
The MESSENGER project will archive all MDIS Experiment Data Records (EDRs) and
Reduced Data Records (RDR) with the Planetary Data System (PDS) in a timely fashion
(Table 13). EDRs consist of the raw image data (DNs) fully documented in terms of geo-
metric and radiometric variables. In addition to the raw images, the MESSENGER project
will also archive radiometrically corrected versions in units of I/F or radiance, as appropri-
ate. The information contained in this paper and prelaunch calibration files archived with
the PDS will allow a user to apply any advances to the state of calibration as needed to the
original raw EDRs.
Critical science observations will be obtained during the three Mercury flybys and are of
four basic types:
? quadrature monochrome NAC mosaics (200?600 m/pixel),
? medium-resolution NAC quadrature stereo mosaics,
The Mercury Dual Imaging System on the MESSENGER Spacecraft 335
Table 13 MESSENGER MDIS EDR delivery schedule to PDS
Mission data Product MESSENGER planned delivery date
Prelaunch calibration EDR 6 months after launch
Earth flyby EDR 6 months after 2nd Venus flyby encounter
Venus flyby 1 EDR 6 months after 2nd Venus flyby encounter
Venus flyby 2 EDR 6 months after 2nd Venus flyby encounter
Mercury flyby 1 EDR 6 months after 1st Mercury flyby encounter
Mercury flyby 2 EDR 6 months after 2nd Mercury flyby encounter
Mercury flyby 3 EDR 6 months after 3rd Mercury flyby encounter
Mercury orbit EDR every 6 months after Mercury orbit insertion
? synoptic 11-color mosaics (1?8 km/pixel), and
? approach and departure movies (4.4?18 km/pixel).
Data products from these observations will serve as the foundation for assessing the or-
bital mapping strategy and planning special targeted, high-resolution sequences. RDR prod-
ucts associated with the flyby observations will be reviewed and archived with the PDS on
a schedule similar to that for the EDRs.
During orbital operations systematic mapping with both the NAC and WAC will result
in:
? a global monochrome basemap with an average resolution of 250 m/pixel,
? merged flyby and orbital multispectral cubes and mosaics in up to 11 spectral filters at
?1 km/pixel,
? local stereo-based digital elevation models, and
? very-high-resolution NAC local strip mosaics in the northern hemisphere (best resolutions
?20 m/pixel).
7 Conclusion
The MDIS instrument and its associated DPU form an integral part of the MESSENGER
science payload. MDIS will provide critical measurements tracing Mercury?s origin and
evolution. The pivoting design of MDIS provides for an extra degree of freedom to map
the surface of Mercury and acquire critical optical navigation images without violating the
stringent Sun-keep-in rules to maintain the spacecraft sunshade in its proper orientation. The
innovative thermal design ensures that the MDIS detectors remain within their operating
temperature range even during periapsis at local noon.
On August 3, 2004, the MESSENGER spacecraft was launched from Cape Canaveral,
FL. At the time of this writing, all aspects of the instrument have been tested in flight and
verified to be working as designed. On May 11, 2005, when the spacecraft was 29.6 million
km from Earth, the spacecraft was commanded to look toward the Earth to test the pointing
of the MLA instrument. Figure 65 shows one of the six images snapped of the ?0.43 mrad
(18-pixel) Earth as observed by the MDIS NAC. In this same frame we observed another
bright object?the Moon. This serendipity bodes well for the discoveries yet to be made at
Mercury by MDIS and the entire MESSENGER science payload.
336 S.E. Hawkins et al.
Fig. 65 Taken May 11, 2005,
this processed image comes from
the NAC when MESSENGER
was about 29.6 million
kilometers from Earth. The Moon
is visible in this frame, but its
contrast has been enhanced.
Although the Earth only subtends
about 18 pixels, bands of clouds
between North and South
America are apparent on Earth?s
sunlit side
Acknowledgements The authors would like to gratefully acknowledge the very detailed and thorough
contributions of the referees in preparing this manuscript. Their comments significantly improved the quality
of the paper.
Appendix 1: Radiation Effects on Filters
We deemed it necessary to characterize the degradation in transmission in the long-pass
filter glasses used in the WAC by exposure to ionizing radiation. This appendix summarizes
the experimental setup and the results of those measurements.
The manufacturer-provided specifications for the Schott colored glass filters were closely
matched to the flight filters. The 25-mm diameter filters were 3.00 ? 0.25 mm thick. The
polished surfaces had a surface quality of 80?50 scratch and dig and parallelism to 2 ar-
cminutes. Only one filter, S8612, was 0.5 mm thick.
The equipment used in this experiment included a 0.75-m focal length triple-grating
imaging monochromator/spectrograph with a sample holding cell, a stabilized light source, a
silicon photodiode detector, and a computer-controlled six-slot filter wheel. All data collec-
tion was accomplished with a desktop computer to provide control of the monochromator.
A 60Co radiation source chamber was used to irradiate the samples.
The transmission of each test filter was measured prior to any exposure to radiation in
order to establish the baseline. The test filters were irradiated in a sealed chamber containing
a 60Co source and received radiation at a rate of approximately 0.18 krad per minute. The
error introduced in the actual dose due to the time taken to lower and raise the samples from
the chamber is estimated to be ?5%.
The experimental arrangement to measure the transmission of the test filters is shown in
Fig. 66. Light from the stabilized source passes through the monochromator order-sorting
filter wheel and enters the monochromator through a variable-width slit; it exits on the oppo-
site side through a similar slit to which the sample holding cell is attached. The sample cell
consists of a collimating lens, a sample holding base, and another lens to refocus the image
The Mercury Dual Imaging System on the MESSENGER Spacecraft 337
Fig. 66 Experimental test setup
for measuring effect of radiation
on MDIS filter glasses
Fig. 67 Inferred transmission
loss over passbands of WAC
filters as a function of radiation
dose
onto the silicon photodiode for measurement Every transmission test followed the same test
sequence under the same environmental conditions.
A transmission test sequence included acquiring dark measurements for background sub-
traction to ensure that only the light passing through the filter was considered in the test. The
lamp was turned on and allowed to warm up for ten minutes before the test sequence began.
Once the lamp warmed up its spectrum was measured with no filter present. The trans-
mission of the short-wavelength filters was recorded over the range 350?650 nm with no
order-sorting filter present in the monochromator beam. The longer-wavelength filters were
scanned over the range 650?1,100 nm with an order-sorting filter that attenuated all wave-
lengths below 550 nm in the beam. The spectrum of the lamp was recorded midway through
the measurements and at the end of the filter measurements to verify that the stability of the
lamp was below 0.2%. At the end of the filter measurements, the lamp was turned off and a
spectrum of the background was taken. Finally, the filter transmissions were calculated by
subtracting the dark signal from the values and dividing the filter signal by the lamp-only
signal.
The measured transmissions of the test filter glasses generally show a drop of transmis-
sion with radiation dose that is wavelength dependent. The maximum transmission of the
filters, roughly 90%, is limited by the reflection loss from the surfaces. The loss of transmis-
sion as a consequence of radiation is smaller at the longer wavelengths. This trend occurred
for each individual filter and continued from filter to filter.
338 S.E. Hawkins et al.
The transmission of the glass filters over the WAC filter passbands was inferred by aver-
aging the results for the test filter glasses over the spectral passband for which they are used
in each WAC filter. Figure 67 shows the transmission variation with radiation dose relative
to the transmission before irradiation. The transmission loss was observed to decrease with
increasing wavelength. The legend in the graph identifies the filter for each curve. Note that
these curves are for 2-mm thick glass samples, whereas the actual thicknesses may be as
high as 2.93 mm. At the estimated mission radiation dose of ?15 krad the transmission loss
will be <4% in the worst case and will average to <2% for all the filters. In-flight calibration
will be used to track this change if it is measurable.
References
G.B. Andrews et al., Space Sci. Rev. (2007, this issue). doi:10.1007/s11214-007-9272-5
D.T. Blewett, P.G. Lucey, B.R. Hawke, G.G. Ling, M.S. Robinson, Icarus 129, 217?231 (1997)
J.F. Cavanaugh et al., Space Sci. Rev. (2007, this issue). doi:10.1007/s11214-007-9273-4
B.M. Cordell, R.G. Strom, Phys. Earth Planet. Interiors 15, 146?155 (1977)
E.H. Darlington, M.P. Grey, Proc. SPIE 4498, 197?206 (2001)
R.E. Gold, R.L. McNutt Jr., S.C. Solomon, the MESSENGER Team, in Proceedings of the 5th International
Academy of Astronautics International Conference on Low-Cost Planetary Missions, ed. by R.A. Harris.
Special Publication SP-542 (European Space Agency, Noordwijk, 2003), pp. 399?405
O.L. Hansen, Astrophys. J. 190, 715?717 (1974)
J.K. Harmon, M.A. Slade, Science 258, 640?643 (1992)
J.K. Harmon, P.J. Perillat, M.A. Slade, Icarus 149, 1?15 (2001)
S.E. Hawkins, III et al., Space Sci. Rev. 82, 31?100 (1997)
M.E. Holdridge, A.B. Calloway, Space Sci. Rev. (2007, this issue). doi:10.1007/s11214-007-9261-8
J.R. Janesick, Scientific Charge-Coupled Devices. SPIE Press Monograph PM83 (SPIE, Bellingham, WA,
2001), 920 pp
W.S. Kiefer, B.C. Murray, Icarus 72, 477?491 (1987)
J.C. Leary et al., Space Sci. Rev. (2007, this issue). doi:10.1007/s11214-007-9269-0
J.S. Lewis, Earth Planet. Sci. Lett. 15, 286?290 (1972)
J.S. Lewis, Ann. Rev. Phys. Chem. 24, 339?351 (1974)
H. Li, M.S. Robinson, S. Murchie, Icarus 155, 244?252 (2002)
H.J. Melosh, D. Dzurisin, Icarus 35, 227?236 (1978)
H.J. Melosh, W.B. McKinnon, in Mercury, ed. by F. Vilas, C.R. Chapman, M.S. Matthews (University of
Arizona Press, Tucson, 1988), pp. 374?400
S. Murchie et al., Icarus 140, 66?91 (1999)
S. Murchie et al., Icarus 155, 229?243 (2002)
B.C. Murray, J. Geophys. Res. 80, 2342?2344 (1975)
B.C. Murray, R.G. Strom, N.J. Trask, D.E. Gault, J. Geophys. Res. 80, 2508?2514 (1975)
J.B. Pechmann, H.J. Melosh, Icarus 38, 243?250 (1979)
A. Potter, T.H. Morgan, Science 229, 651?653 (1985)
A. Potter, T.H. Morgan, Icarus 67, 336?340 (1986)
B. Rava, B. Hapke, Icarus 71, 397?429 (1987)
M.S. Robinson, P.G. Lucey, Science 275, 197?200 (1997)
M.S. Robinson, J.G. Taylor, Meteorit. Planet. Sci. 36, 841?847 (2001)
M.A. Slade, B.J. Butler, D.O. Muhleman, Science 258, 635?640 (1992)
S.C. Solomon et al., Planet. Space Sci. 49, 1445?1465 (2001)
P.D. Spudis, J.E. Guest, in Mercury, ed. by F. Vilas, C.R. Chapman, M.S. Matthews (University of Arizona
Press, Tucson, 1988), pp. 118?164
R.G. Strom, Phys. Earth Planet. Interiors 15, 156?172 (1977)
R.G. Strom, N.J. Trask, J.E. Guest, J. Geophys. Res. 80, 2478?2507 (1975)
N.J. Trask, J.E. Guest, J. Geophys. Res. 80, 2462?2477 (1975)
F. Vilas, in Mercury, ed. by F. Vilas, C.R. Chapman, M.S. Matthews (University of Arizona Press, Tucson,
1988), pp. 59?76.
T.R. Watters, M.S. Robinson, C.R. Bina, P.D. Spudis, Geophys. Res. Lett. 31, L04701 (2004)
G.W. Wetherill, Geochim. Cosmochim. Acta 58, 4513?4520 (1994)
D.E. Wilhelms, Icarus 28, 551?558 (1976)