Page History

Notes:

• Before an exposure can be taken we need to
  • Perform a clear on the CCDs
  • "Warm" the shutter motor (actually pretension the motor so the brake can be released). We assume that the shutter can only remain in this state for a short time (T_shutter-hot), so T_max-T_min < T_shutter-hot ~ 0.1 seconds (TBC)
• In order to avoid the overhead of these items in the time for the takeImages command, we would like to request that the timing of the initImage command relative to the takeImages command be specified with a tight tolerance.
• Proposed values:
  • T_min = 0.95 seconds
  • T_max =1.05 seconds
• If the takeImages command occurs without a preceding initImage, or outside the T_min, T_max window the command will still be acted on, however there will be an additional delay before the first exposure (alternately we could just define the command as being rejected in this circumstance).
• It is also necessary to "warm" the shutter before the close operation, but the timing for this is handled internally by the shutter subsystem. Similarly if any warming is needed between images in a multi-exposure visit this can also be handled by the CCS in combination with the shutter subsystem.
• It is not expected that any clear other than the initial clear will be required in multi-exposure visits.

We should have diagrams which represent:

• The normal case of a two exposure visit with all timing occurring as expected
• What happens if initImage is too late?
• What happens if initImage is too early? (or perhaps equivalently takeImages never comes)
• What happens if takeImages occurs without a preceeding initImage, or an initImage which timedout
  
  • (One option would be to simply reject the takeImages command)

Some parts of these diagrams already occur inside LSE-71 – but they are specifically called out as non-normative.

Gliffy DiagramnameTakeSingleImage