this table...

CMA - EXOSAT CMA Images/Lightcurves



The results from two EXOSAT LE telescopes. Each telescope had a channel multiplier array, CMA, detector in the focal plane. The CMA/telescope combination covered the 0.05-2.0 keV energy range with a field of view of 2 degrees and an on-axis angular resolution of 20 arc sec (HEW). The detectors had no intrinsic spectral capabilities; however, different filters were used to make broad band measurements. The most commonly used were thin lexan (no 7), Aluminum-Parylene (no 6) and Boron (no 8).

An image was generated for each observation using a particular filter. A detection program was used to generate one database entry per source detected above a particular threshold. The CMA database contains a list of detected sources, details about those detections, plus the names of the files containing the associated image and lightcurves. Since many objects were observed many times, there are multiple entries per object.


De Korte P.A.J. et al., 1981, Sp. Sci. Rev., Vol. 30, p. 495.

White N.E. and Peacock A.P. 1988, in X-ray Astronomy with
  EXOSAT, [eds. N.E. White and R. Pallivicini], Memoria S. A. It, p. 59.

Data Products

Images and lightcurves are available for this database. The data products were created during the post-operational phase at the EXOSAT Observatory using a standard, automatic analysis of the data.


The LE auto-produced high-quality results in the vast majority of cases. In some instances, however, problems were encountered. Most of the resulting inaccuracies or errors have been removed during a systematic program of quality checking. Although great care has been taken with this quality checking, the monotonous nature of the work means that a few bad entries may be lurking in the database.

The most frequent problems experienced are:

 * Extended sources: Many extended sources will not have been detected and
   the count rates of those that are detected can be severely underestimated.
   This is due to the fact that the program used to detect sources was
   optimized for point-like sources. Users interested in studying extended
   sources should extract the relevant images (which are available online)
   and perform a more detailed analysis.

 * Confused sources: In a few cases sources were detected near a bright
   object. In these cases it is possible that the lightcurve of the fainter
   source is contaminated by photons from the brightest object. One such
   example is the quasar EXO1102.7+2523 which lies only 2.5 arc min from the
   bright AM Her-type object CW 1103+254.

 * Boron filter: A special Boron PSF was used to estimate the count rate of
   sources detected in this filter configuration. However, since the Boron PSF
   is a function of the source's spectral shape, it is impossible to apply the
   most appropriate PSF in an automatic process. This limitation leads to a
   possible underestimation of the count rate which can be as large as a
   factor of two for unabsorbed sources with very steep soft X-ray spectra.

 * Spurious sources: There are two kinds of spurious sources - those due
   to statistical fluctuation of the background and those due to the inability
   of the software to cope with unusual situations. In the latter case these
   are removed from the database as soon as they are recognized as such. These
   sources are found near bright objects in very short exposures ( <1000
   seconds), in images where a diamond filter was used, or when a bright,
   extended source is present in the field of view (e.g., Cas A).
   The detection threshold of a source is a complicated function of
   exposure time, image background and position in the field of view.
   This threshold is set so that at most one spurious detection is
   expected in every 5 images. More than 3000 images were analyzed and
   about 600 spurious sources are expected to be present in the LE
   database across the full field of view. The vast majority of these will
   be outside the central 15 arc min radius of the detector.
   The reality of a source can be determined on the basis of
   additional information such as significance of detection, exposure
   time, background level, distance from the center, etc.

Count Rate to Flux

To convert LE count rates to fluxes, do the following:
 * log in to the `` anonymous ftp area
 * go to `caldb/data/exosat/cma/cpf`
 * copy the `.rmf` file for the filter instrument combination you need
 * start XSPEC, then:

 model wabs po        (or whatever your model is)
 fakeit none
 10000                (or whatever your exposure is)
 flux 0.1 2.5
The `fakeit` and `flux` commands will give the flux and count rate for the model you have specified.

UV Leak

The CMA detectors were sensitive not only to X-rays, but also to ultraviolet photons, which could cause unwanted contamination when observing O or B stars. The different filters could be used to reduce, estimate, and eliminate UV contamination. The program 'leuv' is used to predict the UV contamination from a star for a given filter. Only the Boron filter is completely free of UV contamination, although the aluminium parylene was relatively immune (except for the very brightest and earliest stars).


The average source detection threshold for a typical 10,000-second exposure within the central 12 arc minute radius region of the detector, using the 3000 lexan filter, is 0.002 ct/s. For a power law spectrum with an energy index of 1.0 and an interstellar column density typical when viewing a high galactic latitude, an extragalactic source of 4x10**20 H cm**-2 gives a flux of about 7x10**-13 erg c**-2 s**-1. If the exposure time is increased by an order of magnitude (a typical maximum observation time) the detection threshold decreases by a factor of about 3.


Counting rate from the Al/P filter, if available from an observation of the same source made within one day. `AL count rate error` gives the error on the Al/P counting rate.

Al/Par Count Rate Error

Background Level

The galactic latitude of the source.

Counting rate from the boron filter, if available (see the `AL count rate` description). `BO count rate error` gives the error on boron counting rate.

Boron Count Rate Error

B-V Color

Source Classification. Type 'class' at the browse prompt for more info. Source not identified as the class filed set to 'uniden'.

The `count rate` and `count rate error` are the average counting rate (counts per sec) over the observation and its associated one sigma uncertainty.

Error on Count Rate

Detector Deadtime Correction Factor

The Dec is the Declination of the detected source.

The probability of the detection being due to random chance. Note, this is not scaled for the independent number of search attempts in an image.

Some observation were carried out with an OBC program that sets a 'diamond' filter. This filter electronically screened out the outer part of the field of view. The resulting image, which is shaped as a diamond, covers only a fraction of the 2 X 2 deg. If the diamond filter was in used than the DIAMOND_FILTER paramater is set to 1 otherwise its value is 0.

The 90-percent confidence uncertainty in arc seconds for the LE1 telescope.

Expected Variance in TA1

Expected Variance in TA2

The `exposure` is the total on-source observation time in seconds. This includes all dead time effects, interruptions in coverage, etc.

File Name of Background Light Curve

The file_image parameter contains the file name of the associated image.

File name of the source lightcurve.

The filter numbers are

   * 7 = 3000 (thin) lexan

   * 6 = al/p

   * 8 = boron

   * 2 = PPL

   * 3 = 4000 (thick) lexan

Source Order Number

The instrument paramater records which of the two telescope+detector combination, LE1+CMA1 or LE2+CMA2, were used in the observation to detect the source. The values are L1, for the combination LE1+CMA1, and L2, for the combination LE2+CMA2.

IPC Count Rate

IPC Count Rate Error

12-micron IRAS Flux

25-micron IRAS Flux

60-micron IRAS Flux

100-micron IRAS Flux

Date of Last Record Update

Rates Buffer Identifier (First Letter)

The galactic longitude of the source.

ME Count Rate in 1-3keV Band

ME Count Rate Error in 1-3keV Band

ME Count Rate in 3-6 keV Band

ME Count Rate Error in 3-6 keV Band

Counting rate from the ME detector during this observation (warning: this may not cover exactly the same time interval), it also covers the central field of view and so will include all sources seen by the LE. `ME count rate error` gives the error on the count rate.

ME Count Rate Error

The name of the EXOSAT target/field. There are many serendipitous sources in the EXOSAT LE images, and in some cases the name given here will reflect the original target name and not the actual source name.

21 cm NH thru the galaxy H cm**-2 from Starke et al.

Observed Variance in TA1

Observed Variance in TA2

Distance from xpix=15, ypix=15

Dead Time Correction

NH (10**19 cm-2) in the Galaxy

Pricipal Investigator Code

Point Source Flag

Date of Buffer Creation

Proposal Type (HLX etc.)

Quality flag for the LE. This is a number between 1 and 5, where 5 is excellent, and 1 means there may be a serious problem with the entry. A value of 3 means that the entry should be checked. For example, all boron filter entries have a qflag of 3, to warn of the problem with the psf for soft sources observed with this filter.

ME Quality Flag

Visual Quality of Detection

The RA is the Right Ascension of the detected source.

Radio Flux (mJy) at 5 GHz

Flag for Radio Emission

Redshift or HD Number of Counterpart

Gives the time resolution of the associated data product lightcurve in seconds. The time resolution was 2^n seconds, where n was optimized such that the average number of counts per bin is 0.3.

ROSAT ID (spare)

Sequence Number

Flag for Serendipitous Sources

Archive Tape # Where Files Are Stored

The `time` of the observation refers to the start time. This is listed as, where yr is the last two digits of the year and day is the day of the year e.g., day 300 of 1984 is 84.300. All times are stored internally as a short history file key (SHF), which is the number of seconds before or after 1 January 1980 at 00:00 hrs UT. Times are accurate to the nearest second.

U-B Color

V Magnitude

For imaging instruments, the X pixels coordinate of the source in the detector is given by the parameter `X pixels`. The X value ranges between -1024 and 1023.

For imaging instruments, the Y pixels coordinate of the source in the detector is given by the parameter `Y pixels`. The Y value ranges between -1024 and 1023.

Contact Person

Questions regarding the CMA database table can be addressed to the HEASARC User Hotline.

Page Author: Browse Software Development Team
Last Modified: 8-Nov-2004