IX/25 Einstein Observatory IPC Parameters (HEAO-2, 1978-1981)
Einstein Observatory Imaging Proportional Counter (IPC) Parameters
Giacconi et al.
<Einstein Observatory (HEAO-2), 1978-1981>
ADC_Keywords: X-ray sources ; Surveys
Mission_Name: Einstein
Keywords: instrumental parameters
Description:
This catalog gives information for each Einstein Observatory Imaging
Proportional Counter (IPC) observation used in the Einstein
Observatory Source Catalog (EOSCAT). It also contains information
relating to individual "HUTS" (HEAO Universal Time), time segments of
data which were summed to create IPC Sequence numbers. Each HUT is
designated by its initial major frame
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
ipcfield.dat 243 4295 Field Parameters for the IPC
ipcihut.dat 153 13265 Component Huts for IPC Sequences
--------------------------------------------------------------------------------
See also:
IX/22 : The CfA Einstein observatory extended deep X-ray survey
IX/23 : Einstein Observatory Source Catalog
IX/24 : The Einstein Observatory High Resolution Imager Source List
IX/26 : Einstein Observatory, Revision 1B. Processings
IX/27 : Einstein Observatory Catalog (OCAT or Yellow Book)
Description of table ipcfield.dat:
This table gives information for each Einstein Observatory Imaging
Proportional Counter (IPC) observation used in the Einstein
Observatory Source Catalog IPC observations will not be included here
if the observation time was less than 300 seconds or if there was a
filter used during the observation. A brief description of the
instrument is given under "Additional Information" section below.
Byte-by-byte Description of file: ipcfield.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
2- 6 I5 --- IPCseq IPC Sequence number (1)
8- 91 A84 --- ObsTitle Title of the observation (1)
93- 94 I2 h RAh Right Ascension (B1950) (hour)
96- 97 I2 min RAm Right Ascension (B1950) (min)
99-100 I2 s RAs Right Ascension (B1950) (sec)
102 A1 --- DE- Declination sign (B1950)
103-104 I2 deg DEd Declination (B1950) (deg)
106-107 I2 arcmin DEm Declination (B1950) (arcmin)
109-110 I2 arcsec DEs Declination (B1950) (arcsec)
112-113 I2 h RA2000h Right Ascension (J2000) (hour)
115-116 I2 min RA2000m Right Ascension (J2000) (min)
118-119 I2 s RA2000s Right Ascension (J2000) (sec)
121 A1 --- DE2000- Declination sign (J2000)
122-123 I2 deg DE2000d Declination (J2000) (deg)
125-126 I2 arcmin DE2000m Declination (J2000) (arcmin)
128-129 I2 arcsec DE2000s Declination (J2000) (arcsec)
131-137 F7.5 rad GLON Galactic longitude of field center
139-146 F8.5 rad GLAT Galactic latitude of field center
148-155 F8.2 s livetime Live time of observation (2)
157-160 I4 yr strt.Y Start date (year) of data (3)
162-164 I3 d strt.D Start date (day) (3)
166-169 I4 yr stop.Y Stop date (year) (3)
171-173 I3 d stop.D Stop date (day) (3)
175-183 E9.4 cm-2 nH Column Density (4)
185-192 F8.5 rad rollang Average roll angle of instrument (5)
194 I1 --- f_rollang [0 1] Flag for bad roll angle 0=good; 1=bad (5)
196-198 A3 --- refid Field identification (6)
200-202 A3 --- fldflag Concatenation of all field flags (7)
204-215 A12 --- fitsfile Name of fits file from EOSCAT (8)
217-243 A27 --- comment Reason field not included in EOSCAT if no
fitsfile (9)
--------------------------------------------------------------------------------
Note (1): These columns give the sequence number and the title of the
observation. The sequence number is a unique numeric identifier for
each observation which was allocated sequentially at the time of
proposal submission. It is an internal index used to key on all
references to an observation. Observation titles are, for the most
part, from the "Yellow Book" (Seward and Martenis 1986, Cat. IX/27).
Several observations have been given new titles in this catalog
because of misnaming or mispointing errors discovered since the last
revision of the "Yellow Book."
Note (2): The live time (given in seconds) is a measure of the effective
exposure time at the field center. It is the total number of seconds
during which the detectors gathered data. The live time has been
corrected for intervals lost because of Earth occultation, passage
through the South Atlantic Anomaly, unacceptable aspect solution, and
detector dead time.
Note (3): These columns give the start and stop date of the observation.
The year is given as a four digit number and the day is given as a
number between 1 and 365 (note that 1980 was a leap year). These two
dates bound the observation, but their difference is not the observing
time because of inter-observation gaps, Earth occultation, etc.
Note (4): This column gives the column density of neutral hydrogen in
atoms per cm2, integrated along the line of sight through our Galaxy
in the direction of the field center of the observation. Specified in
exponential notation, NH values are obtained from sky surveys of
neutral hydrogen using 21 cm radio observations. For declinations
north of -40 deg, they are from the Bell Lab survey of Stark et al.,
private communication. For fields south of -40 degrees they come from
Heiles and Cleary (1979AuJPA..47....1H 1979AuJPA..47....1H) and Cleary, Heiles, and Haslam
(1979A&AS...36...95C 1979A&AS...36...95C).
Note (5): The 'rollang' column gives the angle in radians which measures
the roll of the spacecraft about the axis pointing towards the field
center. It provides the angular orientation of the focal plane
detectors with respect to the celestial coordinate grid. A positive
value of the roll angle indicates a clockwise rotation of the
(detector's projected) field when viewing the celestial sphere. The
'f_rollang' column (int) is a flag which is either 0 or 1, depending
on whether the roll angle given is meaningful. A "0" indicates a good
roll angle; a "1" indicates there is no unique roll angle which can be
ascribed to the observation.
The algorithms in Rev 1B took the nominal roll angle of the first
observation interval ("HUT", which stands for "HEAO Universal Time",
and is used to name data segments which make up an observation) as the
nominal roll angle for the observation. For normal observations this
procedure was satisfactory, since the aspect solution compensated
subsequent HUTs for any differences, which were usually quite small.
However, there are occasions for which the first HUT had zero exposure
time, and/or subsequent HUTs were obtained with roll angles which were
significantly different from the nominal value for the first HUT. This
led to erroneous roll angle labels, even though the data were
correctly handled.
In order to correct this problem, we have calculated a time-averaged
roll angle for each field by using the nominal roll angle for each HUT
together with the small offset from that value derived from the aspect
solution. Occasionally small data segments had roll angles which were
grossly different from the average. If any HUT had a roll angle which
differed from the average for the observation by more than 4 degrees,
it was considered to be "discrepant". If the exposure time of
discrepant HUTs was less than 5% of the total exposure, these HUTs
were ignored in calculating the final average roll angle. The roll
angle given in this column is this new average, not the Rev 1B value.
However, if discrepant HUTs contributed more than 5% of the exposure
time, the concept of a unique roll angle for the observation loses its
significance, and the "bad aspect flag" is set to 1 because the given
roll angle is not an accurate description of the angle of the
instrument during the entire observation. The bad aspect flag will
also be set if HUTs comprising more than 5% of the exposure time had
pointing offsets which differed from the average by more than 2
arcmin.
Note (6): This column gives the identification for the field which has
been taken from published papers which discuss Einstein data. In order
to reference papers which contain results on a particular field, or on
field sources which were not detected by Rev 1B processing or did not
pass our selection criteria, we provide a reference and identification
flag for the field similar to that used for the sources (see
IX/23/srcids.dat). This facilitates identification of parts of large
angular diameter sources (e.g., the Cygnus Loop) when there are no
discrete sources detected, and also accommodates papers which give
source names but no positions.
Our list of papers is based on a bibliography of articles maintained
by F.D. Seward, Guest Observer Coordinator for the Einstein project,
rather than on a systematic literature search, and contains papers
published prior to January, 1988. We apologize to authors whose papers
may have been omitted. It should be noted that field identifications
have been transcribed from the references without evaluation; the user
should consult the references (see the tables IX/23/srcids.dat and
IX/23/reflist.dat) and use his or her own judgment as to their
accuracy.
The main difference between the Ref/ID flags for fields and those for
sources is the addition of a new, nonspecific flag, "!", and the
omission of the "+" flag, germane only for sources.
The "!" flag is used to alert users that there are entries in the
source identifications table (IX/23/srcids.dat) which refer to
sources in the field which are not in the source table
(IX/23/source.dat). These sources may be at the field center, but
more often are either large extended features such as parts of
supernova remnants or off-axis sources which were not detected with a
signal-to-noise ratio greater than 3.5. This type of entry pertains to
a source, but can only be tied to the field, since the source is not
"officially" in the catalog. When this happens, the entry in the
IX/23/srcids.dat table will have an "S" in the position code and may
be much more than 100 arcsec from the field center.
The Ref/ID flags for fields are listed below:
-------------------------------------------------------------------
Flag Meaning
-------------------------------------------------------------------
AGN Active galactic nucleus
BL BL Lac object
CLG Cluster (or group) of galaxies
CV Cataclysmic variable
G Galaxy
GLB Globular cluster
P Pulsar
Q Quasar
RS Radio source
S Star
SNR Supernova remnant
SY Seyfert galaxy
* Nonspecific indicator for separations < 100"
! Nonspecific indicator for a source within the
field, but not contained in the source table
-------------------------------------------------------------------
Note (7): This column gives the field flag, an indicator which denotes
observations which have additional information available, which have
been processed in a slightly nonstandard way, or which were found to
have a peculiarity. The format will be a string, of 1 or more
characters, which indicates where to find ancillary information. The
flags which are defined, and the number of occurrences is listed
below:
---------------------------------------------------------------------------
Flag Meaning # Description
---------------------------------------------------------------------------
B Background suspect 4 An extremely rare flag, indicating that
a poor Mdetect background map may have
affected the reliability of M detect
sources (and the values of sizcor given
in IX/23/source.dat). Not a reference to
further information.
C Combined field 451 The current observation is one of two or
more pointings that have been combined
to form a new sequence number. The
results of the analysis of the merged
data appear only in Appendix J of the
catalog, not in the source table.
D Deleted sources 139 Detected sources have occasionally been
deleted from the catalog because they
were judged to be spurious. For the most
part this occurred for detections within
extended emission such as supernova
remnants and clusters. This flag
indicates that one or more deletions
have been made from this field.
Approximately 470 sources were deleted
through this process. For those cases
where discrete sources were expected a
priori to be embedded in extended
emission (e.g. M31, Orion), deletions
were not made.
G Ghost image 3 Ghost Image from strong source just
outside the field of view. For details
see the description of the IPC instrument
under "Additional Information" section below.
L Ldetect only 965 Indicates that Mdetect was not run. No
reference to further information.
P Particle 1 Background contaminated by particle
contamination events.
S Sources missed 257 Known sources were missed by the
detection algorithms. This flag
indicates that there are one or
more entries in the missed source
list (Appendix G of the Catalog,
see IX/23/missedsrc.dat table)
which pertain to this field.
T Time correction 82 Rev 1B was redone to correct faulty
dead-time corrections. This condition
exists whenever the dead-time correction
used in Rev 1B was in error by 3% or
more.
---------------------------------------------------------------------------
Note (8): The 'fitsfile' column gives the name of the FITS file of this
field from the FITS version of the catalog. The FITS version is
available on-line at the Center for Astrophysics through einline, or
from cdroms and tapes distributed throughout the astronomical
community. For more information, contact An
example of a FITS filename is I0458N41.XIA, where I indicates IPC, the
next four digits indicate the right ascension, N41 describes the
declination (41 degrees north), XI indicates this is the X-ray
intensity version of the map, and the last digit is a sequential
lettering for pointings close enough that the right ascension and
declination would not distinguish them. For instance, if there were
two pointings at this right ascension and declination, the other FITS
file would be called I0458N41.XIB.
Note (9): The comment column provides the explanation as to why the field
was not included in the Catalog, when there is no FITS filename given.
This is usually because the exposure time was less than 300 seconds.
--------------------------------------------------------------------------------
Description of table ipcihut.dat:
This table contains information relating to individual "HUTS" (HEAO
Universal Time), time segments of data which were summed to create IPC
Sequence numbers. Each hut is designated by its initial major frame
number; each major frame is 40.96 seconds. A particular observation
normally will contain more than one hut, but these have been merged by
standard processing to construct an image. All background measurements
are made after the huts have been merged. For information on complete
sequence numbers, see the table ipc_field.
Byte-by-byte Description of file: ipcihut.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
2- 6 I5 --- IPCseq IPC Sequence Number (1)
8- 9 I2 --- fieldno IPC hut number relative to field (1)
11- 19 I9 --- MinFno Minor frame number (2)
21- 27 I7 --- MajFno Absolute hut number (minor frame / 128) (2)
29- 30 I2 h RAh Right Ascension (B1950) (hour)
32- 33 I2 min RAm Right Ascension (B1950) (min)
35- 36 I2 s RAs Right Ascension (B1950) (sec)
38 A1 --- DE- Declination sign (B1950)
39- 40 I2 deg DEd Declination (B1950) (deg)
42- 43 I2 arcmin DEm Declination (B1950) (arcmin)
45- 46 I2 arcsec DEs Declination (B1950) (arcsec)
48- 53 F6.3 rad hutroll Hut - nominal roll angle (3)
55- 62 F8.3 --- aspyoff Aspect Y offset in pixels (4)
64- 71 F8.3 --- aspzoff Aspect Z offset in pixels (4)
73- 83 E11.5 rad asprot Aspect rotation (4)
85- 92 F8.3 s livetime Livetime (5)
94- 98 F5.2 --- avgbal Time-averaged BAL (6)
100-112 E13.8 us ustartim Start time (7)
114-117 I4 --- astartday Absolute start day (offset from 1978.0) (7)
119-122 I4 yr start.Y Start year (8)
124-126 I3 d start.D Start day of start year (8)
128-135 I8 ms starttime Start time (8)
137-140 I4 yr stop.Y Stop year (8)
142-144 I3 d stop.D Stop day (8)
146-153 I8 ms stoptime Stop time (8)
--------------------------------------------------------------------------------
Note (1): The IPCseq column gives the sequence number of the observation,
a unique numeric identifier which was allocated sequentially at the
time of proposal submission. It is an internal index used to key on
all references to an observation. The 'fieldno' column gives the
sequential number of the HUT in this field. There will be one entry
for every hut in an observation.
Note (2): These columns list the minor frame number and the major frame
number of the beginning of the HUT. The major frame number is the
absolute HUT number (±) one, and is equal to the minor frame number
divided by 128. The major frame duration is 40.96 seconds; the major
frame counter was started somewhat prior to launch.
Note (3): The 'hutroll' column gives the angle in radians which measures
the roll of the spacecraft about the axis pointing towards the field
center for a given hut. It provides the angular orientation of the
focal plane detectors with respect to the celestial coordinate grid. A
positive value of the roll angle indicates a clockwise rotation of the
(detector's projected) field when viewing the celestial sphere.
Note (4): The first two columns give the aspect offset, (in pixels), which
is the average offset in Y and Z from the nominal values given in the
columns hutrarad and hutdecrad. The 'asprot' column gives the aspect
offset (in radians) of the roll angle from the nominal values given in
the column hutrollrad.
Note (5): This column gives the livetime, in seconds, a measure of the
effective exposure time at the field center for each hut. It is the
total number of seconds during which the detectors gathered data. The
live time has been corrected for intervals lost because of Earth
occultation, passage through the South Atlantic Anomaly, unacceptable
aspect solution, and detector dead time.
Note (6): This column contains the time-averaged BAL for the hut of the
given observation. The BAL is a measure of the gain of the IPC; it
gives the pulse height (out of 32) where the aluminum (1.49 keV)
calibration signal peaks. The centroid of the count distribution
across the pulse height channels from the on-board AL calibration
source (1.49 keV) is used to specify the IPC gain.
Note (7): The 'ustartim' (microstartime) column gives the start time (in
microseconds) of the observation. The 'astartday' column gives the
absolute day during which the observation started, offset from the
first day of 1978.
Note (8): These columns give the start (and stop) year, day, and time of
each observation. The year will be between 1978 and 1981; the day will
be the day number of that year (1 to 365) (note that 1980 was a leap
year); and the time will be the start and stop times (in milliseconds)
of that day.
--------------------------------------------------------------------------------
Additional Information:
The Imaging Proportional Counter (IPC) has been described by
Gorenstein et al., (1981), and by Giacconi et al. (1979). The
following summary is meant only as a convenient reference for features
of the IPC with which users should be familiar.
The Instrument:
The IPC is a gas-filled counter with two "switch-back" pattern wires
sandwiching the anode. An incoming photon produces a cascade which
induces a pulse in both wires. The pulse height depends on the
intensity of the cascade, which in turn depends on the energy of the
incident photon and the gain of the counter. The rise time of the
pulse lengthens as the pulse propagates along the wire in both
directions. The electronics are thus able to assign a (y,z) location
(by comparing rise times at both ends of each wire), a pulse height,
and a clock time for each event.
Spectral Resolution
The entrance window of the IPC has been coated with "Lexan" to absorb
the UV. The window itself absorbs X-rays in the range 0.28 to 0.5 keV,
and the mirror ceases to reflect X-rays above about 4.5 keV. The
discriminators which measure pulse height encode the information on
the basis of 32 pulse height channels (0-31) which, for most
subsequent purposes, have been reduced to 16 (0-15). That is 0 and 1
become 0, 2 and 3 become 1, etc., although channel zero is not used
because of excessive electronic noise. Throughout subsequent data
processing channel specification will be via a 5-digit octal number:
--------------------------------------------------------------------
(low) Energy (high)
--------------------------------------------------------------------
PH Channel 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Contributes 4 2 1 4 2 1 4 2 1 4 2 1 4 2 1
Octal Flag 7 7 7 7 7
--------------------------------------------------------------------
e.g. Channel 7 and 11 only would be 00420K.
To assign an energy to the incident photon, we need to know the
detector gain which is a function of high voltage and gas composition
(vary with time) and irregularities in the wire spacing, etc. (vary
with position). The temporal dependence is defined by "BAL", the PH
bin (0-31) at which the pulse height distribution for the on-board,
calibrator (radioactive aluminum) peaks. Only the central 4x4' of the
IPC has been accurately calibrated with a celestial source for the
positional variations of gain. A less accurate correction map derived
from pre-flight calibration data is used for the rest of the field. In
Rev 1B processing, each event has a "PI channel" defined as well as a
PH channel (PI standing for pulse independent, i.e. binned according
to energy). This PI binning incorporates both the temporal and spatial
gain corrections.
Spatial Resolution
The Point Response Function (PRF) of the IPC is a quasi-Gaussian
function with low level wings at large distance (due to mirror
scattering). The size of the core (Gaussian part) is determined by how
accurately the electronic processor can assign the correct location to
each event. As the pulse height diminishes (lower gain or softer
photons) the S/N worsens and photons are assigned locations over
larger and larger areas, even though they actually arrived at one
location. The broad wings of the PRF, however, are caused by small
scale imperfections on the mirror. These cause occasional but large
deflections, which preferentially affect higher energy (shorter
wavelength) photons.
Timing
Time resolution is 63 micro-sec; the telemetry limit is 125 cts/s
("primary science" channel), which causes a "Poisson" dead time of
approximately 4% for fields without very high count rates.
Ghost Images
During pre-flight calibration, it was determined that a strong source
outside the field of view but within 2 degrees of the optical axis
could produce (ring-like) features within the field. We reproduce here
a table of Effective Area of the mirror at 0.25 keV as a function of
distance from the field center (at a position angle of 45 deg in
detector coordinates):
Table 5.1
---------------------------------------------------------------
Off-Axis Angle Effective Area
(arc min) (cm2 at 1/4 keV)
---------------------------------------------------------------
0 104
40 21
42.4 = "Masked out" edge
45 17
50 6
53.7 = Physical field-of-view edge
55 1.7
60 1.3
65 0.8
70 0.6
75 0.7
80 0.5
85 0.2
100 0.06
>112 0
---------------------------------------------------------------
The use of IPC data outside a radius >16 arcmin is fraught with
difficulties:
1) The rib shadows (tic-tac-toe pattern from the detector window
support structure) are not centered on (Y=511,Z=511); the
geometric center of the ribs actually lies at (Y=515.3,Z=523.7)
in a non-ASPECT-corrected image file.
2) The effective "size" of the ribs is somewhat dependent upon
x-ray energy (because the cone angle of entering, focused x-rays
depends on energy).
3) The gain of the detector, which varies widely across the field,
is generally depressed in the region near the ribs (because of
electrical field distortions caused by the presence of the ribs).
4) The IMAGE files smear out the position of the ribs because
different detector orientations are combined by the ASPECT
solution in creating the files in (alpha, delta) space.
As a consequence of these effects, data near the ribs must be used
with caution.
History:
Prepared from the tables available at the "ADS Catalogue Service"
(CfA, Harvard-Smithsonian Center for Astrophysics, Cambrigde MA)
References:
Giacconi, R., et al. 1979ApJ...230..540G 1979ApJ...230..540G
Gorenstein, P., Harnden, F.R., Jr., and Fabricant, D.G. 1981,
Trans. IEEE, NS-28, 869.
Seward, F.D. and Martenis, P. 1986, Einstein Observatory Catalog of
Observations (5th ed.; Cambridge: Harvard-Smithsonian Center for
Astrophysics, Cat. IX/27).
Stark, Heiles, Bally, and Linke, private communication.
(End) Francois Ochsenbein [CDS] 23-Apr-1999