Down Time

The down time statistics are based on individual fault reports. In Table 1 I give the general down time statistics for period 52.

A total of 123 fault reports were submitted, with an average time lost of 6 min per fault, for a total down time of 0.8% (1.2% on scheduled observing nights). Of these, 91 reported no time lost, 27 reported $<$ 2 hrs lost, and 5 reported 2 or more hrs lost. The latter includes 2 fault reports which were caused by a power failure external to NOT, and are not included as technical downtime. The effect of including this downtime in the statistics is show in Table 1 in parenthesis.

This compares to a down time of 0.8% over all nights (0.4% on scheduled observing nights) in period 50, and 0.3% over all nights (0.2% on scheduled observing nights) in period 49. Of the 101 fault report in period 50, 79 reported no time lost, 20 reported $<$ 2 hrs lost, and 2 reported 2 or more hrs lost. Of the 68 fault report in period 49, 49 reported no time lost, 19 reported $<$ 2 hrs lost, and none reported 2 or more hrs lost.

Table 1: Technical down time statistic period 51: 2015-04-01 to 2015-10-01

Night included	Time lost	Nights	Percentage$^a$	Last semester	Last Summer
All nights	760(1290) min	183	0.8(1.3)%	0.8%	0.3%
Scheduled observing nights$^b$	665(1195) min	102.5	1.2(2.2)%	0.4%	0.2%
Technical nights	95 min	24	0.7%	3.2%	0.0%
Service nights	295 min	46.5	1.2%	0.8%	0.5%
Visitor instruments	0 min	0	0.0%	0.0%	0.0%
$^a$ Taking the average length of time within nautical twilight. Exact numbers for each night
are used when looking at ``All nights''
$^b$ Excluding technical nights and visitor instruments

Excluding the downtime caused by the external power failure, the downtime is still somewhat on the high side when compared over the last few years, but basically in line with the longer-term average (0.75%). At least partly, the height of the downtime was caused by the absence of our technician, as this would likely have limited the amount of time lost in one of the two `normal' major failures. Also, the effect of the external power failure would likely have been limited if he would have been able to complete one of the projects he was working on. Below this is discussed in a bit more detail.

As for the total number of fault reports, like in the last semester, there were relatively many fault reports due to some new software scripts having some teething problems, while the `regular' problems with the detector software is being reported more diligently.