ALMA Holography Troubleshooting
Last Update: JeffMangum - 22 Nov 2006
Table of Contents:
Stopping a Scheduling Block
The ability to stop an SB is a new feature with one important caveat: The existing array must be destroyed after the scheduling block is stopped before starting another one. Trying to execute another SB in the same array causes mount issues.
Use the "stop" button on the Scheduler to stop the SB and the "remove" button to remove the current array.
- 22 Nov 2006
antMountGUI Consumes Buttloads of CPU
The current antMountGUI is, well, a challenge to use. One "feature" is that it spews a ton of log messages while running. To help mitigate this from consuming too much CPU, simply make sure that either the terminal window that you started the antMountGUI from is covered or iconed or, if you are using tabs, not the current tab.
- 15 Feb 2007
Synthesizer Connected Light on Control GUI Does Not Turn Green After Power Failure
The current transmitter system uses a shitty NI PCMCIA 4-Port RS232 serial interface card. For unknown reasons the card may not initialize properly if the signal is interrupted, the card is removed, or the break-out cable (bit honking cable sticking out of the left side of the control laptop) is disconnected. To make matters worse, if the PCMCIA card goes berzerk and the Agilent sythesizer (located in the box at the base of the holography tower) reboots it will fail to do so due to spurious signals being sent to it from the fricking PCMCIA card.
The solution to this problem is:
- Exit the LabView transmitter control GUI.
- Remove the PCMCIA card from the laptop. You should probably do this with the usual hardware removal tools within Windows.
- Power cycle the Agilent synthesizer in the coffin at the base of the holography tower.
- Re-insert the PCMCIA card.
- Restart the LabView control GUI.
You should now see a green light on the synthesizer connection.
- 07 Nov 2006
Loss of Timing Event Due to Central Reference Generator Failure
The current implementation of the holography "mini rack" does not include ventillation. If the power fails and shuts down the HVAC system on the VertexRSI antenna the temperature in the receiver cabin can become warm. This seems to cause the CRG to stop delivering TEs. To fix this:
- Restart the HVAC system on the antenna (get help, as this is not trivial).
- Go to the receiver cabin.
- Open the back of the mini rack and check to see if the green "TE" light is flashing on the back of the ABM.
- If the green "TE" light is not flashing:
- Power-down the mini rack using the power switch on the top of the box.
- Open the mini rack box.
- Remove the front panel to the CRG box (the top-most bin in this rack).
- Let the electronics cool a bit (10-15 minutes).
- Power-up the mini-rack.
- You should see the "TE" green light on the back of the ABM flashing.
- Re-install the front panel on the CRG box.
- Close the doors of the mini rack.
- Stop wasting time on these stupid problems and continue taking data!
- If the green "TE" light is flashing:
- You are screwed...your problem is somewhere else.
- 07 Nov 2006
Receiver and DSP
- If you see erratic output from the DSP (undulating amplitude about a fixed value of either 0.0 or 0.5), the DSP has likely went into a fault condition. We have come to characterize these faults as follows:
- Symmetric Failure: In this failure mode the DSP produces:
- Garbage data
- An untouched most significant bit (i.e. it is unset)
- Total power amplitudes between -1.0 and 1.0
- The workaround for this problem is to do TWO DSP resets.
- Asymmetric Failure: In this failure mode the DSP produces:
- Garbage data
- Most significant bit set more or less randomly
- Total power amplitudes between 0.0 and 1.0
- The workaround for this problem is to power cycle the TE distributor (CRD?) in the receiver cabin with the smart plug strip.
- 05 Nov 2006