VLBI Software Documentation
Field System
Experiment Operations
N. R. Vandenberg and W. E. Himwich
NVI, Inc./GSFC
Operations Manual
NASA/Goddard Space Flight Center Version 9.0
Space Geodesy Program October 1, 1995
This manual covers the standard operations used in Field System experiments. All standard
operations normally done at the station are described.
This manual is an operations manual, written in "cookbook" format. Explicit instructions and
examples are given on how to perform the required operations. Reference is made throughout
this manual to other manuals for more information. In particular, the operator should be familiar
with the information in the following:
Overview manual
Operators' Reference manual
SNAP Command manual
File Name Conventions and File Formats manual
drudg: Experiment Preparation Drudge Work manual.
Sections of this manual are arranged roughly chronologically according to the activities related to
a VLBI experiment. The final sections of the manual contain basic checklists, which are
condensed lists of the operations described in previous sections.
The contents of this manual include:
Setup: Get schedules and notes; read experiment notes; prepare SNAP file, listings, and
labels.
Pre-experiment: Check equipment, procedures, pointing, tapes, send ready message.
Experiment: Start schedule; log equipment configuration, performance parameters, head
parameters, sky conditions, cable values; send start message.
Post-experiment: Send log to correlator, send stop message.
2.0 Setting Up for an Experiment
This section covers setting up for an experiment. Topics include:
procedures for transferring schedules to the disk area where the Field System expects to
find them;
preparation of experiment materials using the program drudg:
creating a SNAP file,
making listings with SNAP line numbers and tape changes indicated,
printing tape labels with bar codes,
printing text notes about the experiment.
2.1.1 Electronic Transfer
For geodetic experiments, schedule files are placed in the common VLBI disk area on CDDIS,
the computer at Goddard, for electronic transfer to the stations. Electronic transfer can be done
via a direct network connection or via a modem. Refer to the Electronic File Transfer manual
for descriptions of detailed procedures for getting the schedules electronically. This manual also
contains information on CDDIS and the standard disk areas. Most stations are now connected via
network, and so the most common method for accessing the schedule files is described next.
Below is a sample FTP dialog that would retrieve a schedule file. The dialog assumes that you
are logged in to the Field System computer which is on the network and you will want to
generate a SNAP file from a schedule file. You should be logged in as user oper.
/usr2/oper> cd /usr2/sched
/usr2/sched> ftp cddisa.gsfc.nasa.gov
Name: vlbigeo
Password: password
ftp> cd sep97
ftp> get ca018.skd
ftp> get ca018.txt
ftp> quit
The CA018 schedule file now resides on the Field System computer in the /usr2/sched area.
You also have a copy of the experiment notes in the same area.
2.1.2 Copy from Floppy
If you receive a package with the schedule on a floppy disk, you will need to copy the file(s) to
the /usr2/sched directory. Use these commands to copy the schedule and notes files:
/usr2/oper> cd /usr2/sched
/usr2/sched> mcopy -t a:polrs1.skd polrs1.skd
/usr2/sched> mcopy -t a:polrs1.txt polrs1.txt
The POLAR-S1 schedule file and the notes now reside in the appropriate disk area.
Read the notes that accompany the schedules. If you received a printed copy of the Notes, read
that. If you transferred the schedule files electronically, print and read the .txt file using the
following command:
/usr2/sched> lpr polrs1.txt
The experiment notes describe the purpose of the experiment, the type of schedule that was used,
any special operational procedures, and the correlator plan. The notes document the goals of the
experiment and the planned conduct of the experiment.
It is important to read the notes so that you are familiar with any special required procedures
described there, such as more frequent weather monitoring, whether tape prepass is required,
non-standard frequencies, etc.
The SNAP file is the file that the Field System will follow to conduct the experiment at one
station. The SNAP file is normally generated at the station by running drudg on the schedule
file. For certain special experiments the SNAP file may be provided by the experimenter or
scheduling center.
You create the SNAP file for your station by running drudg with the schedule file (.skd or
.drg file) as input and selecting your station ID. drudg automatically names the SNAP file
with a .snp extension. Refer to the drudg manual for more information about what the
program does. You should be logged in as user oper.
/usr2/oper> cd /usr2/sched
/usr2/sched> drudg
... drudg starts, prompts for schedule file name, station ID ...
... select option 3 to create the SNAP file ...
... exit from drudg ...
Now the SNAP file is in the schedules directory where the Field System expects to find it.
Operators need to have a hard copy listing of the schedule for easy reference. In drudg, there are
three list options. Select the option that best fits the needs at your station, or do all of them so
they are available when needed during the experiment. If you have a SNAP file (.snp file) but
you do not have a schedule file (.skd or .drg file) you can still make tape labels and print a
summary of the SNAP file with these drudg options.
1) List the SNAP file in compressed format, i.e. several commands per output line (option 4). Be
sure to select the width of the output to fit your printer.
2) List the schedule with start and stop times and full information on source position, az, el,
slewing, and configuration (option 1). A blank line for operator written comments is provided in
this listing.
3) List a summary of the SNAP file (option 5). The SNAP file is required to be present for this
option. The listing makes one line per observation giving line numbers for the source
commands, tape footages, and tape changes.
Use the tape label option in drudg (option 6) to print bar coded labels. Refer to the drudg
manual for label sizes and the recommended way to feed them into the printer.
At the end of this manual is a copy of a memo about tape handling procedures, which also
contains information on the different types of labels you will find attached to tapes.
3.0 Pre-experiment Checkout
Before the experiment starts, the equipment must be checked for proper operation and proper
software control must be assured. The following subsections very briefly list the appropriate
checks that are critical to the experiment. Refer to the Basic Checklists manual for a thorough
discussion of each check.
3.1 Check Phase Cal
Check that there is phase cal present in all channels.
3.2 Check Pointing
Check the pointing on a few sources to make sure that you're getting a signal and to verify that
the offsets are small. Verify that the SEFDs for the observing bands are in the nominal range.
Refer to the Antenna Performance and Antenna Pointing manuals.
3.3 Pre-check Tapes
Run the tapes to be used for the experiment through the pre-check procedures, described below.
If the pre-check can be done so that the tapes are kept in the same environmental condition as the
tape drive, then you do not need to prepass the tapes during the experiment.
Tapes are pre-checked by mounting the tape and running it all the way forward, thoroughly
cleaning the drive, and finally rewinding. This allows the tape to thermally equilibrate to the
drive and also scrapes most of the debris off the tape surface. Be sure to give the drive 15
minutes to warm-up to operating temperature before running the pre-check. After the tape has
been rewound, clean the drive again. If there is a significant amount of dirt on the drive, set this
tape aside and put a label on it that says "DIRTIES DRIVE". If you have enough extra tapes,
send it back to the correlator where it will be retired, otherwise hold on to it and use it only in
case you run out of good tapes. If the tape did not dirty the drive, it can be remounted for
recording. All of the tapes for an experiment can be pre-checked prior to the start of the
experiment as long as the tapes remain in the same environment as the recorder, with no changes,
until they are recorded.
At the end of this manual is a copy of the recommended pre-check procedures for recorders.
There is also a guide for recommended tape usage. Tapes are listed by types for both narrow
track and wide track systems. Operators should adhere to these recommendations.
Send an electronic mail message as you begin the experiment operations. In the message, note
any special circumstances at your station. These messages are used by the correlator as
confirmation of station status. In particular you should include the clock offset (sign and value)
between GPS and the formatter, the weather conditions, pointing offsets, SEFDs, a general
statement about your station's condition, and that you are ready and/or have started the schedule.
You could also include the source and time of the first observation. This is especially desirable if
the first observation is not the first one in the schedule file.
An alternative is to send both a ready and a start message. Typically you would send the ready
message at least one hour before the experiment is to begin to confirm that you are ready. In this
case send all of the information above except the clock offset. Send the clock offset in the start
message.
When reporting the SEFDs and pointing offsets, please use the following format:
xxxxx/ssssss b.bbbb c.cccc source @ aaa/ee weather
where:
xxxxx is the X-band SEFD (rounded to 1%)
sssss is the S-band SEFD (rounded to 1%)
b.bbbb is the offset on the first axis: Az., R.A., or X (rounded to 1% of beamwidth)
c.cccc is the offset on the second axis: El., Dec., or Y (rounded to 1% of beamwidth)
source is the name of the source
aaa is the azimuth of the source
ee is the elevation of the source
weather is a brief comment about the weather, e.g. "cloudy", "rainy", "clear", etc.
If possible make this report for more than one source.
4.1 oprin Commands
The program oprin handles operator input to the Field System programs. The program runs in a
window labelled operator input and is always waiting to receive input; its prompt is a right
arrow character >.
Refer to the SNAP Command manual for details on the commands used in this section.
When you are finished typing a command press Return and the command will be conveyed to
the Field System for processing.
4.1.1 Starting a Shift (op and status commands)
When you first start a shift, enter your name:
>op=your-name (enter up to 12 characters)
You should also learn the current observation and tape status using the STATUS command:
4.1.2 New Tape Ready (label command)
After you have mounted the new tape, tell the Field System it's ready to go by typing in the tape
label:
>label=number,check
Example:
>label=HS000123,3F28
If no check-label is written on the tape (an unusual situation), run program labck to generate
the check-label. It will prompt you three times for the label (just to be sure you got it right)
before generating the check label. Then get an oprin prompt and enter the label command. If
you make a mistake typing the command and an error message appears, just try the command
again.
If you are in a panic and do not have the time to get the label, you can just type
>cont
and put in the label sometime later while the tape is still mounted.
4.1.3 Enter a Comment ("---)
A comment can be entered into the log at any time. Type your message starting with a double
quote as the first character.
>"ALL IS WELL or some comment on the state of things"
Type the comment enclosed in double quote marks; the closing double quote is not required.
Liberal use of comments is encouraged during experiment operations. Let the folks back home
know what's happening!
Any SNAP command can be entered via oprin and will be passed on for processing. For
example, the tape drive can be started by using the proper SNAP command. Refer to the SNAP
Command manual.
4.1.4 Listing the Schedule (list command)
If you want to see what commands are coming up in the schedule next, and what command
you're currently on, use the list command:
>list
The display will show the next 10 lines of the schedule, including the one you're on now.
4.1.5 Terminating the Field System (terminate command)
When you are through running the Field System use the terminate command. Refer to the
Operators' Reference manual.
4.2 Starting a Schedule (schedule command)
These instructions apply when you are starting a schedule for the first time using a fresh tape. It
is assumed that the Field System is running properly and the equipment has been set-up correctly
for the observing mode (see the Setup and Operators' Reference manuals). A good way to
verify that everything is in a clean state would be to manually set-up the equipment using the
configuration procedure, e.g. sx2c1=, that will be used for the first observation. You might
also want to exercise the various monitoring commands wx, cable, sxcts, etc. to make sure
all devices are talking. To re-start a schedule, see section 5.0 in this manual.
You should allow plenty of time so that the steps below can be completed well in advance of the
actual first observation. The two most time critical steps are getting the antenna onsource and
positioning the tape. The order of steps given below is not the most efficient; it is just one that
works. If you are experienced with the Field System you will probably see several ways it can be
improved. If, for example, the required tape motion will take longer than the antenna motion, you
might prefer get the tape drive set-up first. Also, you don't need to dismount a tape if it is one
you are intending to record and you know the footage counter is correct.
For a successful schedule start, follow each instruction below, in numerical order. When an
instruction is marked as "nA", "nB", or "nC" do only one of the "A", "B", or "C" options.
Step 1. Determine with which observation you will begin the schedule and note its starting line
number. The starting line of an observation is one with the source= command. Some of the
drudg listings include this information for each observation. Start the schedule with the
command:
>schedule=xx,#n,1
Where xx is the name of the schedule and n is the line number where the observation starts. The
Field System will then process the source= command for this observation and halt the
schedule. This will start the antenna moving to the source.
If your antenna is an azimuth-elevation type you will need to consider which "wrap" of the
azimuth position the observation will start on. There are three possibilities: counterclockwise,
neutral, and clockwise. You can determine which wrap the antenna is supposed to be on from
either the schedule listing or the last parameter of the source= command. If the antenna is
moving to the wrong section of the wrap, do whatever is necessary to force it to the correct
position.
Step 2. If a tape is mounted, dismount it. See section 4.4.2 for the details. Do this even if the tape
that is mounted is the one you intend to record. This step will assure that the tape footage counter
is correct.
Step 3. Mount the tape you intend to record. See section 4.3.1 for the details.
Step 4. Examine the schedule using either the SNAP listing or using the list= command. There
are three cases that must be distinguished:
Step 4A. The first observation is the first observation in the schedule file. In this case the
schedule will take care of positioning the tape and will prompt you for the tape label. You should
just release the schedule from its halted condition by typing:
>cont
When prompted for the label, respond with:
>label=number,check
where number is the tape number and check is the check-label.
Step 4B. This isn't the first observation in the schedule file, but it is the first observation on pass
1 of a new tape. In this case, there will be an unlod procedure and possibly a parity check
procedure (e.g. check2c2) and/or a tape positioning procedure (fastr= or sfastr=),
between the source= command and the set-up procedure (e.g. sx2c1=). In this case, skip
over the schedule lines before the set-up procedure by issuing the command:
>schedule=xx,#n
where xx is the name of the schedule and n is the line number of the set-up procedure. The Field
System will continue executing the schedule from this point. When prompted for the label,
respond with:
>label=number,check
where number is the tape number and check is the check-label.
Step 4C. If the your first observation isn't the first observation in the schedule file or on pass 1 of
a new tape, then it is just some arbitrary observation in the middle of the schedule. There will
possibly be a parity check procedure (e.g. check2c2) and/or a tape positioning procedure
(fastr=, sfastr=, fastr=, or sfastr=) between the source= command
and the set-up procedure (e.g. sx2c1=). In this case you will need to position the tape and
enter the label manually before allowing the schedule to continue. Determine the tape footage for
the observation from a drudg listing (options 1 or 5) and position the tape using:
>tapepos=feet (for a Mark III, IIIa or IV tape drive)
or
>rec=feet (for a VLBA recorder)
Enter the tape label:
>label=number,check (number is the tape number, check is the check-label)
>label (this command logs the label as a response for the correlator's use)
Finally, start the schedule running by entering:
>schedule=xx,#n
Where xx is the name of the schedule and n is the line number of the set-up procedure. The Field
System will continue executing the schedule from this point.
The Field System tries to run a schedule as automatically as possible, but there are certain
conditions about which it doesn't yet have enough information and so the operator must make
some decisions. Eventually all of this will be automated after the appropriate operating steps
have been worked out. Your comments or suggestions on these procedures will be greatly
appreciated.
4.3.1 Mounting a New Tape
To mount a Mark III tape on an empty tape transport, follow these steps:
1. Open the vacuum column door by grasping its left edge and pulling towards you.
2. Using a Q-tip and tape head cleaner (alcohol) carefully clean the record head, the playback
head, the vacuum chamber (pay special attention to the four windows in this area) and the rest of
the tape path. Remember, when cleaning the drive there is no such thing as a tape drive that is
too clean. In general, dirt on any surface the tape comes in contact with will offset the tape
position relative to the tape heads and cause a jitter as the tape is drawn through the recorder. In
addition to the tape heads and the capstan, it is important that the fixed post and the vertical sides
of the vacuum columns, both on the deck and the door be cleaned.
3. Place the tape which is to be recorded on the upper spindle. Leave the sticky tape on while
you put the tape on the spindle.
4. While pressing the reel back firmly, press the latch on the spindle in until it clicks into place.
5. Remove the sticky tape from the magnetic tape.
6. Unlock the spindle brakes by pressing the footage reset button (just under the footage counter)
(Mark III) or issue the command command rec=release (VLBA).
7. Unwind about 3 feet of tape from the top reel.
8. Thread the tape.
9. Loop the tape around the takeup reel and hold the end in your left hand.
10. Spin the takeup reel several times clockwise to build a static charge.
11. Release the end of the tape and it should be attracted to the takeup reel. Be sure the end of
the tape has not folded on itself.
12. Turn the takeup reel 5 or 6 turns in a clockwise direction so that the tape is being "pulled."
13. Close the vacuum column door.
14. Close the cabinet door.
15. Mark III: Press the STOP button FIRMLY and HOLD it in until the tape stops squealing.
This will start the vacuum system and load the tape. The STOP button and the footage reset are
the only button on the tape drive that you should ever have to push. Leave the other buttons
alone. VLBA: There are not buttons to push, just issue the command rec=load.
16. If a schedule is running, enter the new tape label:
>label=number,check
17. Put the sticky label for this tape on the Field Label area of the tape, and put a red dot on the
tape.
4.3.2 Dismounting a Tape
These directions cover how to dismount a tape assuming it is on the drive and the vacuum has
been established.
1. BEFORE YOU DO ANYTHING ELSE. Be sure that the identifying schedule tape label has
been put on the face of the reel and that there is a red dot to show the tape has been recorded.
2. If the tape is not at the load point, rewind it by entering:
Mark III VLBA
>tape=low >rec=unload
>srw
3. Mark III: When the tape reaches the load point, have it unload automatically by typing:
>unlod
VLBA: The previous step should have automatically unloaded the tape.
4. Mark III: Release the spindle brake by pressing the footage reset button.
VLBA: Release the brakes by typing in the command rec=release.
5. Turn the supply reel counterclockwise until all of the tape is wound on it.
6. Place the sticky tape over the magnetic tape.
7. Open the spindle latch and remove the tape.
8. Place the tape with the other already recorded tapes.
4.3.3 Prepass Procedures
If you have pre-checked all of the tapes for the experiment (section 3.3), you can skip this step.
Approximately one-half hour before the start of an experiment, the first tape to be used should be
mounted and run all the way forward, the drive should be cleaned, and then the tape should be
rewound, an "accommodation prepass." This allows the tape to thermally equilibrate to the drive
and also scrapes most of the debris off the tape surface. Be sure to give the drive 15 minutes to
warm up to operating temperature before running the prepass. The tape heads and tape bearing
surfaces in the transport should be cleaned. The tape should then be remounted for recording.
During the experiment, when a new tape is called for there may be time allowed in the schedule
to "prepass" it. If the prepass procedure is called from the schedule, the new tape will be run
forward at high speed and then the schedule will be halted while you clean the tape drive of
material that may have been scraped off the tape. Finally the tape will be rewound. You will
prompted to take the appropriate action during this procedure. Finally, you will receive the
normal newtape prompt for the tape label.
All of the tapes for an experiment may be prepassed up to 24 hours before the start of the
experiment as long as the tapes will be in the same environmental conditions as the tape recorder
until their use. Use the prepass schedule to prepass each tape.
The frequency and exact manner in which data quality is to be monitored during the experiment
will be indicated by the shift supervisor.
4.4.1 Monitor Display
To display the system status, use the program monit. To activate it,
type Ctrl-Shift-2 or select Monit 2 on operator's menu
The time-like quantities in the display will update every second. To cancel the display, close the
window.
4.4.2 Tape Error Rates (parity command)
It is possible to check parity and sync errors by playing data back after recording. Procedures may
be set up at the stations that record error rates using parity command. Standard procedures
such as check2c1 and check2c2 are invoked in the schedule to verify that good data was
written on the tape during the previous observation. If excessive parity errors occurred, you
should try to clean the heads between observation, when there is sufficient time. Usually you
need a period where the schedule is waiting at least 2-3 minutes. Follow these steps:
1. Mark III: Open the vacuum door to release the tape. DO NOT PUSH THE FOOTAGE
RESET/BRAKE RELEASE BUTTON! VLBA: Issue the command rec=release to release
the vacuum.
2. Un-loop the tape from around the rollers so you can access the heads.
3. Clean the heads.
4. Re-loop the tape around the rollers and pull taught by moving the supply reel. The brakes will
make the reel hard to move, but not impossible. Don't release the brake switch or you'll reset the
footage counter!
5. Mark III: Press the STOP button and hold it down until the tape loads into the vacuum
columns. VLBA: Issue the command rec=load to re-load the tape into the vacuum columns.
4.4.3 Real-Time Phase Cal Monitor
In addition to monitoring the phase-cal at the output of the Video Converters or Base Band
Converters, it is possible to monitor the phase-cal using the reproduce system in the tape drive.
The presence of the phase cal signal is a pre-requisite for proper geodetic data acquisition. The
phase-cal amplitude should be between 5 and 10% (voltage) and fairly uniform across the
channels.
For Mark III racks, you can use the pcalr program to monitor the phase-cal. This program
should not run during a schedule unless there is a gap of several minutes and the program is
stopped before the schedule begins executing again. You can start pcalr by using the pcal
procedure, which may need to be customized for your site. Use the command:
>pcal
pcalr can be stopped by typing:
>sy=brk pcalr
For Mark III racks it is also possible to monitor the decoder output to verify the presence of
phase-cal in reproduced data during parity checks.
For a VLBA rack and recorder you can use the DQA module to monitor the phase-cal for any
tracks that are enabled for recording. In particular this means that you can monitor the phase-cal
while the data is be recorded. Use the procedures:
>dqaodd (for Mode C forward passes or Mode A)
and/or
>dqaeven (for Mode C reverse passes or Mode A)
Refer to the Phase Calibration manual for detailed information on checking the phase
calibration signals.
4.4.4 Status Checking
During normal operation, the chekr program will be running to check the status of every
module which was set up in this experiment.
NOTE: Any errors reported by chekr are probably important. If you cannot figure out what is
wrong or cannot fix it, call the shift supervisor.
If for some reason you wish chekr to ignore a certain module, use the check command. Here
are some examples:
>check=*,-v1 (ignore all errors from video converter 1)
>check= (stop all checking temporarily)
>check=all (resume checking of all set-up modules)
monit will display a message reminding you about the modules that are being ignored.
Note: As of Field System version 8.2, only some checking is implemented for the VLBA
modules. Operators should monitor the status of the modules via SNAP commands and
procedures.
4.4.5 Decoder Display (Mark III)
A good eyeball check that all is running well can be made by setting the decoder to the TIME
display and then simply glancing at it now and then. You should also check that the current time
is correct by comparing it to the formatter. Normally the time code is being decoded in bypass
mode.
If the time code does not show up in the decoder, make sure the following are set up properly:
1. Tape reproduce tracks must be selected and the correct bandwidth and equalizer set up. Check
the setup with the repro and enable commands.
2. Formatter mode and sample rate must be consistent with tape speed and video converter
bandwidth. This means (normally), mode A or C, 4 or 8 Mbits sample rate, 2 or 4 MHz
bandwidth, 120 or 240 ips. Note that some Mark III and VLBA drives do not reproduce 8 Mbit
data well even in bypass.
3. Tape drive must not be confused about what state it is in, so you may have to cycle it through a
few st and et commands.
4.5 Handling Problems with the Schedule
A list of typical problems which may be encountered during an experiment is given below, along
with suggested ways out of the difficulty. In all cases, type a comment into the log explaining the
problem.
4.5.1 Re-starting the Field System
If you need to re-start the Field System software for some reason, e.g. the computer is hung up
and will not follow the schedule, or if you need to cycle power on the hardware, do these steps
with reference to the indicated sections as needed. These are general instructions that should
work in all cases, however they will not be the most efficient in all cases. Experienced operators
will know quicker ways to get the system running again.
Case 1. The computer has crashed or is not responding, but the Mark III hardware is still
functioning, i.e., there are no alarm lights on.
1. Reboot - See Operators' Reference section 2.0.
2. Start fs - Operators' Reference section 5.1.
3. Re-start the schedule - Experiment Operations (this manual) section 4.6.2.
Case 2. The Field System software and the computer seem to be OK, but the Mark III or VLBA
hardware glitched and the alarm buzzer is going off.
1. Mark III: Reset formatter time - Mark III Hardware Setup section 4.0
2. Mark III: Reset MATs - Mark III Hardware Setup section 5.1
3. VLBA: Reinitialize the MCB modules with the vlbainit procedure.
4. VLBA: Set the formatter clock - VLBA Hardware Setup section 5.0
5. Reset the Field System time - Mark III Hardware Setup section 5.2 or VLBA Hardware Setup section 4.2
6. If the tape footage count got reset by the glitch, reset the footage counter. Mark III: A straight-forward but possibly inefficient solution is to do the following:
>tape=low (set the low tape sensor)
>srw (rewind as quickly as possible)
wait for the tape to reach the load point
>tape=low,reset (reset the footage counter)
Even though this is inefficient it is a reasonable approach as long as there aren't very many
power failures or other problems that cause the footage counter to be reset. If there are (or even if
there aren't) you may prefer to simply note the approximate footage and attempt to position the
tape manually. The tapepos command will not work the way it should since the footage
counter has the wrong value. You can skip positioning the tape at all when you restart the
schedule, but then you should attempt to catch the tape up to the correct position as soon
possible, probably no sooner than the next change of direction. You can reset the footage counter
the next time the tape reaches the load point before a forward pass. Be sure to enter a comment
with an explanation of what you did so that someone at the correlator can figure out what the
tape footage values mean.
For VLBA, run the tape forward a few seconds and set the counter to its current value:
>st=for,120,off (start tape moving forward no recording)
wait a few seconds after the drive reaches speed
>et (stop the drive)
wait for drive to stop
>rec=feet (set counter)
7. Re-start the schedule - Experiment Operations (this manual) section 4.6.2.
Case 3. Both the computer and the Mark III or VLBA hardware require re-setting. For Mark III
modules this situation occurs also if the power on the computer has to be cycled on and off since
this seems to implicitly perform a hard reset on the MATs.
1. Mark III: Reset formatter time - Mark III Hardware Setup section 4.0
2. Reboot - See Operators' Reference section 2.0.
3. Start fs - Operators' Reference section 5.1.
4. Mark III: Reset MATs - Mark III Hardware Setup section 5.1
5. VLBA: Reinitialize the MCB modules with the vlbainit procedure.
6. VLBA: Set the formatter clock - VLBA Hardware Setup section 5.0
7. Reset the Field System time - Mark III Hardware Setup section 5.2 or VLBA Hardware Setup section 4.2
8. If the tape footage count got reset by the glitch, reset the footage counter. Mark III: A straight-forward but possibly inefficient solution is to do the following:
>tape=low (set the low tape sensor)
>srw (rewind as quickly as possible)
wait for the tape to reach the load point
>tape=low,reset (reset the footage counter)
Even though this is inefficient it is a reasonable approach as long as there aren't very many
power failures or other problems that cause the footage counter to be reset. If there are (or even if
there aren't) you may prefer to simply note the approximate footage and attempt to position the
tape manually. The tapepos command will not work the way it should since the footage
counter has the wrong value. You can skip positioning the tape at all when you restart the
schedule, but then you should attempt to catch the tape up to the correct position as soon
possible, probably no sooner than the next change of direction. You can reset the footage counter
the next time the tape reaches the load point before a forward pass. Be sure to enter a comment
with an explanation of what you did so that someone at the correlator can figure out what the
tape footage values mean.
For VLBA, run the tape forward a few seconds and set the counter to its current value:
>st=for,120,off (start tape moving forward no recording)
wait a few seconds after drive reaches speed
>et (stop the drive)
wait for drive to stop
>rec=feet (set counter)
9. Re-start the schedule - Experiment Operations (this manual) section 4.6.2.
4.6.2 Re-Starting the Schedule
Restarting a schedule is very similar to starting it. You must first have the system set-up and
ready to observe (see section 4.6.1). You then need to pick an observation that is far enough in
the future that you can get the antenna and the tape drive correctly positioned before the
observation begins. Be sure to allow yourself some extra time since nothing is as easy as it seems
especially when you are hurrying to get data on the tape again. Remember to allow time to
change the tape if that will be necessary. It is helpful if you can practice restarting schedule
before you actually need the skill. The procedure for restarting a schedule is the same as for
starting (section 4.3) and so is written with reference to it. It needs to be stressed that these
instructions are written as generally as possible and may not be the most efficient approach in all
cases. An experienced operator may well find ways to handle things more efficiently. There are
two cases:
Case 1. You missed some runs on the tape and the next run you can make is still on the same
tape. In this case you should do only steps 1 and 4C in the procedure for starting a schedule
(section 4.3).
Case 2. You missed some runs and one (or more) tapes changes. In this case follow the procedure
for starting a schedule (section 4.3) but in step 4, only options 4B or 4C may apply to your
situation.
In either of these case if you are using a Mark III tape drive, you may prefer to start the schedule
at the next convenient change of tape direction in the schedule. This is particularly helpful if the
tape footage has not been reset since positioning the tape to either end, using srw or sff is
relatively easy.
After you have restarted the schedule, type in a comment saying why you had to restart the
schedule and giving an idea of the runs you missed.
4.6.3 The Telescope Was Late On Source
Enter a comment in the log describing what happened and give the time at which you did acquire
the source.
4.6.4 The Tape Did Not Start Moving
The tape drive was probably not loaded into the vacuum columns or in not in REMOTE (Mark
III). Check that the REMOTE button is depressed. Try pressing the STOP button and holding it
in or issue the command rec=load (VLBA), then type
>st=for,speed
or
>st=rev,speed
depending on which direction the tape is supposed to move. You might have to reload the tape,
then re-type the st command. Type a comment into the log.
4.6.5 The Tape Stopped Spontaneously During a Run
Reload the tape into the vacuum chambers with the STOP button (Mark III) or with the command
rec=load (VLBA). Re-start the tape in the appropriate direction:
>st=for,speed
or
>st=rev,speed
Then wait until the tape stops at the end of the current observation and advance the tape using the
fastf or fastr procedure for the amount of time lost or position the tape with the rec
(VLBA) or tapepos (Mark III) command. Enter a comment into the log.
4.6.6 Tape Runs off the Reel at the Load Point
Follow these steps:
1. Enter the halt command to stop the schedule and prevent more confusion from complicating
your task.
2. Mount the tape, section 4.4.1.
3. If you haven't missed any observations, restart the schedule by typing cont. Otherwise follow
the instructions for restarting the schedule, section 4.6.2. You may need to jump ahead one or
more observations.
4.6.7 Tape Runs off the Reel at Far End
1. Enter the halt command to stop the schedule and prevent more confusion from complicating
your task.
2. Mount the tape. This is more difficult than a regular tape mount since you are attempting to
thread the tape on the supply reel rather than the take-up reel. If you have a Mark III drive, you
can avoid reseting the footage counter if you can thread the tape without releasing the break, but
this is almost impossible, so you probably should not try. Instead you will need reset the footage
counter after remounting the tape.
3. If you have a Mark III tape drive, you will need to reset the footage counter. A straight-forward
but possibly inefficient solution is to do the following:
>tape=low (set the low tape sensor)
>srw (rewind as quickly as possible)
wait for the tape to reach the load point
>tape=low,reset (reset the footage counter)
Even though this is inefficient it is a reasonable approach as long as there aren't very many
power failures or other problems that cause the footage counter to be reset. If there are (or even if
there aren't) you may prefer to simply note the approximate footage and attempt to position the
tape manually. The tapepos command will not work the way it should since the footage
counter has the wrong value. You can skip positioning the tape at all when you restart the
schedule, but then you should attempt to catch the tape up to the correct position as soon
possible, probably no sooner than the next change of direction. You can reset the footage counter
the next time the tape reaches the load point before a forward pass. Be sure to enter a comment
with an explanation of what you did so that the correlator can figure what you tape footage values
mean.
4. If you haven't missed any observations and the tape is correctly positioned, restart the schedule
by typing cont. Otherwise follow the instructions for restarting the schedule, section 4.6.2. You
may need to jump ahead one or more observations.
5.0 Post experiment
After the experiment has finished, post experiment checks are made to verify proper equipment
operation. Refer to the Basic Checklists manual and any station specific checklists you may
have.
Transfer the full log file to make it available to the processing center. Either deposit the log file
onto CDDIS via electronic transfer or copy the log file to a DOS floppy disk.
5.1.1 Electronic Transfer
A sample dialog for transferring log files to CDDIS is as follows:
/usr2/oper> cd /usr2/log
/usr2/log> ftp cddisa.gsfc.nasa.gov
Name: vlbigeo
Password: password
ftp> cd sep97
ftp> put ca018.log
ftp> quit
The log file now resides on CDDIS and can be accessed by the correlator.
5.1.2 Floppy Disks
If you send logs via floppy, you will have to copy the file to a DOS floppy. The sample dialog
looks like:
/usr2/oper> cd /usr2/log
/usr2/log> mcopy -t polrs1s.log a:polrs1s.log
The floppy disk in the A: drive will now contain a DOS format file that can be mailed to the
correlator.
Send an electronic mail message at the end of the experiment. In the message, note any missed
scans, unusual weather conditions during the experiment, and any problems encountered.
Confirm that you have copied your log file to CDDIS.
Include in the message your formatter clock offset (value and sign) from UTC. The correlator can set up for processing most efficiently if stations provide this information in a timely manner.