Observing system

Since the introduction of the Sequencer we have had occasions where BIAS apparently ignores exposure commands from the Sequencer. After a certain amount of time the sequencer gives up and continues with its next command. This causes any dither script to skip an exposure (or more) in a sequence. The phenomenon is not limited to scripts, but also when giving exposure commands after each others on the command line. Extensive tests in September showed that, although the problem is intermittent, it seems to have become worse over time, and there is a strong correlation with the exposure time. A modification inside the data acquisition software was eventually made to do an automatic ``re-try'' of the exposure command if it is not executed. This works well for the most typical exposure times, but fails for longer exposure times. Until the underlying cause of the problem is found, this work-around at least makes sure that the exposure eventually is made, though at the cost of some extra overhead. The new controller should in principle not have this problem.