RoboDIMM

RoboDIMM was ING's DIMM monitor for measuring the local seeing routinely. Below it's a description of its components:

The Monitor: a Meade 12" telescope with an SBIG ST-5C CCD detector at Cassegrain focus, controlled via serial link by a PC running custom software. Instrumental focus is also under PC control. The same camera head has been used since installation in August 2002. The controller was changed for an identical model after a lightning strike in 2007. Image scale is 0.72 arcseconds per pixel.

The telescope forms four separate images of the same star on the CCD thanks to a pupil mask with 4 subapertures covered by small-deviation glass prisms. This allows four independent differential motions to be measured, from which the monitor program derives four simultaneous estimates of seeing.

Tower: The telescope is mounted 6 meters above ground, thus escaping much of the ground level turbulence, on a tower that also provides a high degree of vibrational stability, thanks to its serrurier truss-type structure. The original Arcetri observatory tower has been modified to improve access and lightning protection.

Dome: The monitor and PC are housed inside a fibreglass clamshell dome (from Astrohaven, Canada) which is mounted on an access platform at approximately 5 meters above ground.

More about the hardware.

Accuracy of data

We stress that RoboDIMM was not designed as a site characterization tool. If data are used for the purpose of site characterization, we urge the user of the data to pay attention to calibration issues.

Data from before 6 Dec 2007 are publicly available here.

Since 2004 RoboDIMM measures good seeing less frequently than expected from the statistics of previous campaigns (see DIMM 1995-98)), but this can't be ascribed to worsening seeing at ORM. Rather it appears that RoboDIMM is progressively less sensitive as the image motion, hence seeing, decreases.

Shortly after it started operating in August 2002, several samples of a few hour's length were used to show that RoboDIMM seeing agreed with the the IAC's DIMM seeing monitor to within better than 10%. However, this result is valid in the range 0.63 to 1.1 arcsec and for one night only. RoboDIMM also shows a similar level of agreement with SloDAR and JKT-SCIDAR monitors in summer 2004 and January 2005. More detail of this comparison is available. The algorithm used by RoboDIMM to calculate seeing is made publicly available here .