The status of the cameras is displayed on the monitor. The display is
updated every 10 seconds. The displayed options are ``Not Available''
(i.e., you did not select this controller port with CAMERAS), ``In
Use'' which means wiping, selecting binning pattern, etc., ``Exposing''
(i.e., shutter open and exposure timing going), ``Reading'', or
``Closed'', which means shutter is closed and camera is waiting to
read out. ``Stopped'' means that the exposure has been halted by the
word STOPALL. During the actual exposures, the exposure time for each
camera in use and its status are displayed on the VIURAM monitor.
This display is updated every 10 seconds. The temperature display is
updated every 3 minutes. (The status display does not function
correctly if you are logged in as SYSTEM.)
No exposure can be initiated during the readout of a camera (the software will not let you do it.) The exposure words check for projected conflicts in required access to the controller at the end of the exposure for closing the shutter and reading out the frame. If a conflict is projected to occur, the shutter will close for the first camera to end its exposure, there will be a delay while the second exposure ends and the shutter is closed, then the first camera will be read out, etc.
A brief summary log file called DAILY.LOG exists in the scratch disk.
After each exposure is terminated, a new entry is made into this log
file. The file can be read during exposures, but should be printed
out only when no exposure is going to terminate before the end of the
print job (i.e. ideally at the end of the night). If the file is
deleted, a new file is started as appropriate.
LASTLOG will print out the last n lines of the DAILY.LOG file, where n is normally set to 20. (If it prints out a different number, type LASTLOG L 20, and thereafter it will print out the last 20 lines.) I find this useful so that I can see the parameters of the most recent exposures.
The expose words divide themselves into normal exposures and drift
scan exposures. All expose words which save data check the available
disk space before initiating an exposure. If there is inadequate disk
space, a message is typed on the terminal and no exposure actually
starts. At the end of each expose word that does not go into batch
mode, the name of the disk file(s) generated is typed on the terminal.
Also, the terminal bell is sounded whenever an exposure terminates in
either the interactive or background mode. All the expose words
prompt you for the object name.
All short exposures (i.e. exposures that do not become subprocesses)
end with the frame being displayed directly from the Microvax memory,
as this is significantly faster than storing the frame on disk, then
using the standard display word IMAGE (described in 
XI) to read it
back from disk and display it. For each camera, the usual questions
regarding the display format must be answered (you can use
``
'' if you are happy with the default values). Then once
you are happy with the display itself, you are asked if you want the
curser to be enabled. If not, the display stage will proceed to data
from the next camera in use, or if only one camera is in use, there
will be a short pause while the data files are properly closed, then
you will be back in DCL, ready to initiate another command. If
multiple cameras are in use, readout of the nth camera does not occur
until the display stage of the n-1 th camera is finished. If you
choose to use the cursor, remember to type q to exit that section. Do
not type CRTL+C during the display section at the end of short
exposures; the output data files will not be closed properly.
EXPOSE will ask for the object name and exposure time. All cameras in
use will be wiped, then an exposure of the indicated time will be
made, the chips will be read out, the data stored on disk and written
to tape (if automatic taping is enabled). Note that all the cameras
start together and end together. (They are, however, read out in the
order of 0, then 1, then 2, then 3, if all 4 are in use.) Also note
that the calls to the telescope control system 11/73 occur for all
exposure words when the shutter is closed. A job will be spawned if
the exposure exceeds one minute in length. If you get confused about
what happened within any batch job, you can type out the log (TYPE
*.LOG, where * is the name of the batch job typed out on the terminal
at its completion, for example EXPOSEB0). To see what spawned jobs
are running in the background, type SHOW PROCESS/SUB or SHOW SYS.
SNAP will initiate an exposure of only one camera. You are prompted
for both the exposure time and the camera (0-3).
All the words described above produce frames whose names are S123F.dst.
To get an erase frame, do a zero second exposure. The shutter will
not be opened at all, and you will in fact end up with an erase frame.
To take a dark exposure, use the word DARK. This is essentially the
same as EXPOSE, but the shutter is never opened. The resulting frames
are called DARKn.dst, where n is the port number of the camera. Thus
if you take multiple dark exposures on a given night or run, you will
need to rename the various frames, otherwise you will end up with
multiple version numbers of the relevant file.
TESTALL and TEST are the same as EXPOSE and SNAP, but the data is not
written to disk nor to tape. It is, however, in memory and is
displayed (one chip at a time). The display is adjusted by answering
the questions. The data can be displayed at various black/white
levels and over various parts of the image. With TESTALL, the data
from any of the chips can be displayed in any order.
STOPALL will stop an exposure that has been spawned into the
background (i.e., with exposure time exceeding one minute). All that
will happen is that the shutter will be closed. There is no readout
and no wipe. (To stop a shorter exposure, use CRTL+C, or if that
fails try CTRL+Y. Note that exposures stopped in the latter manner
cannot be restarted.)
REGO will restart any of the above.
KEEP is a word for saving exposures that have been stopped by STOPALL,
among other things. It works on all cameras that have been stopped by
STOPALL. It first closes the shutter of each of them, then reads each
of the cameras, stores the data, wipes the chips. Taping will occur
if enabled. The names of frames saved with KEEP are of the form
K123F.dst.
MODEXP allows one to modify the exposure time of an exposure after it
has been initiated while the exposure is still going on. It functions
for exposures initiated with SNAP or EXPOSE only. The time which you
are prompted to enter is the total desired time for the exposure.
Various situations, for example trying to modify the time on an
exposure which is too close to finishing, modifying the exposure time
to a total exposure shorter than what has already occurred, etc. lead
to error messages on the terminal and no change in the exposure time.
MODEXP can only be used on long exposures that are spawned as
subprocesses. If the exposure to be modified uses multiple cameras
with EXPOSE, the time is modified for all the cameras.
There is a log file produced when MODEXP is used which contains
information useful for debugging purposes. If you have any problems
with this word, see XVI for a procedure which will cure the one
failure found in testing (and which could not be duplicated, so we
cannot be sure it was fixed) and please print out the log file
(MODEXP.LOG).
There are two words that are useful for focusing or other such tests.
These words produce disk files but data is not written to tape. (If
you want to save the data, you can do so -- see section on taping.)
MULTIPLE will do several sequential exposures (i.e., open shutter --
wait -- close shutter) on the same frame with a wipe before the
beginning of the sequence, and a readout of all the cameras that are
enabled at the end. A keystroke is required to initiate each segment
of a multiple exposure. (MULTIPLE is equivalent to the normal
focusing -- i.e., move telescope, change focus, expose, repeat, etc.,
then read out final image with several focus exposures. I also use
MULTIPLE to get many images of bright standard stars on a single CCD
frame, moving the telescope between segments of the exposure.) These
frames have names of the form M123F.dst. The sequential exposures in
MULTIPLE need not be all of the same length. The exposure time stored
in the header is the time of the first exposure segment.
FOCUS is a sequence of short exposures each of the same length. The exposure time stored in the header is the length of the individual segments. To initiate each exposure a keystroke must be detected, allowing you to change the focus or whatever between exposure segments. Between each exposure, the telescope is move 8 arc-sec north (with a spacing of 16 arc-sec north between the final pair of exposures). (This assumes that the telescope control program is version 3.0 or higher.) After the final exposure, the full CCD in each camera in use is read out. Again disk files are produced, but the data is not taped. This enables one to do a sequence of focus exposures. These frames have names of the form F123F.dst.
Note that for direct imaging at the P60, the focus encoder reads in units of 2/10,000 of an inch (about 5 microns/unit). As the focus encoder reading increases, the distance from the secondary to the primary is increased.
CONSEC will take several sequential exposures of the same object, each
of which is the same length, and will save only an average of the n
exposures. CONSEC asks which cameras are to be used (i.e., all in use
or only one of them). This is useful for flat fields. Frames
produced with CONSEC are saved on tape and on disk. The names of the
frames produced by CONSEC are of the form C123F.DST. (CONSEC will run
interactively even for exposures longer than 60 seconds.)
If you decide you do not want to save an image, you can delete the
file with the VMS utility routine DELETE (abbreviated DEL) (DEL
A123F.DST). This should only be done if you are having problems with
disk space. All accounts are set up to require confirmation of any
file deletions.
SCAN will take exposures in the drift scan mode (also called the
transit mode or time-delayed integration). At the present time SCAN
will only work for one camera, and we assume 1x1 readout is desired.
SCAN will ask for the camera to be used, the number of lines per scan
and the rate (milliseconds/line). We do not yet know the smallest
possible rate consistent with no loss of data or timing. At the
present time, a scan with a total saved image consisting of more than
about 4,000,000 pixels may fail (it won't start correctly, so you'll
know about it right away). This is because the data is not stored to
disk at all until the scan is finished. After prompting for the
necessary parameters and the object name, the CCD is wiped, and the
ramp up begins. No data is saved until the ramp up is finished. The
scan then continues until the required number of lines has been saved
for each camera. During SCAN (including the ramp-up phase) the output
of the camera is displayed. Since the display is only 512 wide, you
must specify where the display should begin in x, since in many
cases the CCD is wider than the display can show at full spatial
resolution. While SCAN is running, three terminal keys are
recognized, two of which adjust the black/white level (B or W) of the
display. Just type the key (without a carriage return). You will
then be asked for the new black or white level, type the value and hit
return. The letter P will halt the scan, it cannot be restarted.
(This letter was chosen as being far on the keyboard from B or W so
you wouldn't hit it by mistake.) You should type P without the
carriage return.
If you stop SCAN by typing P, you will be asked if you want to save the data or not. The display will scroll up (admittedly somewhat jerkily) if more than 512 lines are read out. Also, the observer must make sure to input to the telescope control program the correct rates for the scan. During SCAN, the exposure VIURAM will show the number of lines saved so far (or the number of lines of the rampup). Since SCAN is a purely interactive word, never goes into batch, and requires the full attention of the Microvax, you cannot use the Microvax to do anything else during SCAN, and in addition the temperature display does not function.
The names of frames produced by SCAN begin with the letter ``T'', i.e., T123F.dst.
TRANSIT is a modification of SCAN which allows you to continuously
SCAN a single CCD camera, averaging several lines of readout into a
single saved line, which is then stored on disk. It is intended for
use in situations where the scan goes on for many hours continuously.
The initial questions asked when TRANSIT starts up are the number of
lines to be averaged, the CCD readout rate in milliseconds/line, and
the name of the file to which the data should be written. It is
assumed that the desired readout format is 1x1, and no binning code is
sent to the camera. The data is displayed on the monitor in the same
way as for SCAN and the same keys control the black/white level and
stop the scan. The number of averaged lines saved is displayed in the
status display. The temperature display does not function during
TRANSIT.
The telescope control computer is asked for the UT time before the start of TRANSIT and at the end. Stored with each line of data is the associated system time in 64-bit format. There is another word, RTRANSIT to read and display the data. See the Fortran program OBSERVESRC:RTRANSIT.FOR for an example of how to access each line of data saved.
It is not yet clear how fast TRANSIT can run with full functionality.
TRANSIT can run for hours, only limited by filling up the available scratch disk space. (If you should manage to completely fill up the scratch disk, the program stops and the existing data is still on disk.) Thus the length of the output file cannot be specified in advance. Therefore the output of TRANSIT is NOT a FIGARO file so none of the FIGARO or OBSERVING commands will work on the output file. The resulting output file can be saved on tape using the VMS utility BACKUP.
It should be noted that this version of TRANSIT is equivalent in functionality to the version of TRANSIT created by J. Gunn for use at the P200 coude. It is not equivalent in functionality to TRANSIT as used by the 4-shooter itself, where data from 4 cameras goes directly to tape and the data rate is significantly higher.
TRANSIT2 is the same as TRANSIT, but it allows for binning of the CCD.
The details of the sequences sent to the CCD controller for these
words can be found in Appendix B of this manual.