Below are listed those faults for which 2 or more hours were lost. Also repetitive errors which loose little time but occur frequently (if present) are included here.
The safety system was activated and stopped the telescope. It was found that this was due to the sensor of the safety system being iced over.
The Stegmann absolute encoder for azimuth showed failing bits and after a while too many for the TCS software to accept it. It cannot be excluded that it was damaged by the (at that time) allowed power-off command while the building was rewinding cables in the morning when `parking' the building. It now is nolonger possible to power-off while parking the building.
There were intermittent problems with filter wheel no. 1 not reaching its position when moving it to a specific position or when initializing it, which resulted in various amounts of time lost on different nights. As it takes at least 10 days to warm-up and cool-down the instrument this could not be investigated in depth during the observing run.
When the instrument was later opened one of the bearing in the worm wheel drive for filter wheel #1 was found to be stiff, though functioned at room temperature. The bearing was replaced, and the instrument worked well during a subsequent observing run.
The building crashed various times consecutively. In the end it was found that the support arm of one of the IR sensors at the bottom of the main entrance stairs case was damaged. The support of the sensor was replaced after which no more crashes occurred.
The sensor support might have been damaged during the "open day" which took place one and a half days earlier.
There were intermittent problems with the building crashing. Several solutions were tried which every time seemed to work, but after a few hours of observing the system would crash again. In the end it was found that a power supply of the brakes of one of the two pairs of motors which rotate the building was faulty. This caused the brakes sometimes to be activated which made the telescope run into the building.
One problem encountered in trying to solve the problem was the lack of documentation, both of the original design as well as of modifications that had apparently been made later (see below).
In is not entirely clear if the cause of the fault reported above was different, but given that there were more than 3 weeks between these faults and there was a well defined cause for the first fault we believe this to be the case.
The cross-dispersion profile of star images observed with SOFIN were found to be very asymmetric. It was first thought that this was a problem with the telescope optics, but images with StanCam showed normal round stars at the nominal telescope focus and in the end it was concluded that the problem was with the new cross-disperser which will be send back to the manufacturer.