Recent Changes - Search:
MtPleasant26m /

Fixing Problems Such As Recovering From Power Failures

Formatter requires power cycle after power failures

It's important that the formatter is the last device in the rack to power up. So, once power has been restored go around the back of rack 10 and pull out the formatter power plug. It has a sticky label attached to the cord and is located about 40 cm up from the bottom right corner of the rack. Wait a few seconds and replug it. You'll need to then run fmset on hobart, choosing "." to set the clock to the field system time, and then "s" to sync the formatter. Afterwards, press escape to close fmset.

DC power supply in rack 3

Check the power supply rails by rotating the switch in the bottom of the rack and observing that each supply rail, as measured on the panel meter, is present. If any rail is not at the right voltage then switch off the power supply and switch it back on again. If you do this then you will have to reset the station clock next and then restart drive PC, Rakbus, sam26m. This is because those PCs still obtain the time from the station clock.

Station Clock not working

Check UT matches GPS + 10 hours (or 11 during daylight savings) and check tick phase is a small near zero ms number. Setting the station clock is menu procedure being careful to use AEST and not summer time or UT. Start the clock during the second prior to the time you entered in AEST. Do not hurry, get the station clock right first.

Mk4 Formatter/Decoder clocks

On a terminal window on Hobart, ( not the field system console), type "fmset" to get access to the Mk5 formatter clock. The available commands are shown in the window. Use "+" or "-" to correct errors of a few seconds.

On at least one occasion we had to cycle the power to the Mk5 rack (rack 10) to get the decoder clock to sync with the formatter clock. This is done by pulling the four black 110 Volt power leads on the bottom right hand side from the back of the rack.

Mk5 has rebooted, requires login

In the absence of an UPS, the Mk5 computer (rack 9) will have rebooted. You need to log in again as oper. Then you need to run the commands:
setterm -blank 0
Mark5A -m0

Counter 2 not working

Counter 2 is the one above the DAS PC. It can be used to measure frequency but should normally be set to read maser PPS to GPS offset for the maser monitoring system. If it is not doing this then do 'recall, 1, enter'. The display will then show maser to GPS PPS offset.

Cable Counter not working

The cable counter is the one on the left below Hobart. It measures the cable length for the geodesy receiver. It should normally display 0.0159XXXX or 0.016XXXXXX. If it is not doing this then do 'recall, 2, enter'. The display will then show the cable length reading.
If this does not fix it, and the counter is reading 0.024,xxxxxx then you might need to power cycle the VLBI Delay Calibrator in the Mark4 rack. It is on the left hand side below the VCs. What you do is connect a BNC plug to give you some extra "pull". Then undo the 4 thumb screws securing the front panel. Then pull the unit out a few inches. Then slide it back in. If this worked the the counter will read 0.016,0XXXXX instead of 0.024,0xxxxxx. Do not forget to do up the screws again. Leave all the front panel switches where they were.

GPIB not working

Sometimes you notice messages in the PCFS log screen on Hobart that indicate "ibcon timeout", usually following clkoff or cablen commands. This means that either Hobart's GPIB card has locked up or that the GPIB extender in the focus cabin. The latter is the most likely. To fix this, you need to power cycle the GPIB extender. This can be carried out at anytime for an IVS S/X experiment but should only be done between scans for LBA experiments as the Agilent will also be turned off.

To power cycle the GPIB extender, use the middle toggle switch on the panel below the focus position display (labelled "Agilent and GPIB extender"). Flip it up and leave it off for a few seconds before restarting. Again, wait a few moments and retry the clkoff and cablen command in the fs window. If the problem persists, then it is Hobart that is the problem. To fix this you need to cycle Hobart's power by doing the following:

Terminate FS by entering "terminate" in operator input

Exit FVWM by left click on back ground and going to bottom of menu

Change to another window by pressing "ctrl Alt F1"

Then press "ctrl Alt Del"

When it starts its memory test, turn off the power to Hobart for 15 seconds then turn it back on.

The Mark4 rack alarm will sound because rubbish comes out of the PC serial port at start up. you will fix this soon, in the mean time you can silence the alarm by pressing the alarm mute button on the drive panel and silence the racks alarm by turning off the alert on the tty distributor in the mark4 rack.

Login as "oper" when prompted

To re start FS type "fs" in the beige coloured PCFS log window and watch for errors.

In the operator input you should load an experiment procedure file by entering "proc=experho"

You should then execute the setup procedure to load the frequencies and attenuator settings etc. Usually the snap command is "setupsx".

This should switch off the red lights in the rack. You may need to do setupsx again. you may need to do "if3=alarm" and "vc15=200,2" and "valarm".

If all the red lights are off in the rack, switch the alert back on on tty distributor.

Check the gpib is working by typing the snap command "cable" or "clkoff" the response should be sensible cable measurement or fmout-gps.

DAS levels incorrect

This section is not critical during an IVS S/X experiment.
If the DAS levels are wrong or the DAS alarm is sounding after a power failure then you need to check that another receiver has not been selected on the Multifeed Receiver Computer. If it is wrong, select the correct receiver for the current observation. You should check that everything else is correct, such as quad-in or -out and attenuator settings.

Remotely rebooting PSOS machines

To remotely reboot any of the old PSOS machines (sys26m or sam26m), use the command rem_reboot -r rakbus sam26m on newsmerd. rakbus needs to be alive to use this but it can be power cycled by the Remote Rebooting link in the sidebar.

/usr/bin/xterm -name monit2 -e /usr2/fs/bin/monit2 &

/usr/bin/xterm -name monit3 -e /usr2/fs/bin/monit3 &

Backing up Baldor drive settings

You can do this by logging into as observer@hobaldor. If the baldor serial logger (test with ps -ef | grep baldor_serial_logger)is running, then you'll need to kill it. If it's not running, then make sure that the serial port is configured correctly by running

stty -F /dev/ttyS1 1:4:1cb2:8a30:3:1c:7f:15:4:1:0:0:11:13:1a:0:12:f:17:16:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0

Once that's right, you can save the drive parameters with the following commands.

echo -e "A1\r\n" > /dev/ttyS1 && echo -e "P\r\n" > /dev/ttyS1 && cat /dev/ttyS1 > X2.DriveP.txt

echo -e "A2\r\n" > /dev/ttyS1 && echo -e "P\r\n" > /dev/ttyS1 && cat /dev/ttyS1 > X2.DriveP.txt

echo -e "A3\r\n" > /dev/ttyS1 && echo -e "P\r\n" > /dev/ttyS1 && cat /dev/ttyS1 > Y1.DriveP.txt

echo -e "A4\r\n" > /dev/ttyS1 && echo -e "P\r\n" > /dev/ttyS1 && cat /dev/ttyS1 > Y2.DriveP.txt

Avoid these LOs

Setting the second LO (SMLs) to these frequencies will cause problems with birdies and reduced sensitivities. This applies to both Ceduna and Hobart

560 - 592
373 - 395
280 - 296
224 - 236
186 - 198