RoboDIMM monitoring techniques
Seeing monitoring
Our RoboDIMM uses the standard theory of the differential image motion monitor (DIMM), which is by now a standard and widely spread method of measuring astronomical seeing. With the DIMM method, seeing (FWHM in arcseconds) can be measured on a real-time basis: once every 30 seconds. Using the DIMM technique, tracking errors, wind shake errors, small focus errors and telescope optical quality have virtually no influence on the measurements.
Two independent measurements are provided by the motion in the longitudinal and transverse direction (parallel and perpendicular to the aperture alignment), as given by the variance:
sigma2l/t = sigma2[1–kl/t(d / S)1/3]
with
kl = 0.541 and kt = 0.810 ,
where d is the size of the entrance pupils (Vernin & Muñoz-Tuñón, 1995), which is 60 mm in our RoboDIMM.
The FWHM of the long-exposure seeing-limited point-spread function (PSF) in large telescopes is computed with the standard formula:


The Fried parameter r0 is estimated from the variance of the differential image motion in two small apertures ('entrance pupils' in a mask in front of the telescope tube). Each entrance pupil contains one optical wedge, to be able to focus all spots. The entrance pupils in our RoboDIMM have a diameter of 60mm, and are 124mm separated from each other.
Both star images are projected onto a CCD detector in the plane of focus, which is in our case an SBIG ST5C. The ST5-camera has 10 micron square pixels in an array of 320 x 240 pixels (= 3.2 x 2.4 mm). The shortest exposure time is 10 ms, which is good enough to effectively freeze the rapidly changing star images.
The exposure time is dictated by the speed with which atmospheric changes take place. Depending on the circumstances, exposure times in the order of 5-20 ms are needed to freeze these variations.
The star spots are approx. 35 arcseconds separated from eachother. The CCD plate scale is determined by using a double star with well-known separation. Wavelength = 500nm.
Model ST-5C CCD Specifications | |
CCD | TI TC-255 |
Pixel Array | 320 x 240 pixels, 3.2 x 2.4 mm |
Total Pixels | 76,800 |
Pixel Size | 10 x 10 microns |
Full Well Capacity | 50,000e |
Dark Current | 5e¯/pixel/sec at 0° C |
Antiblooming | Variable Rate |
Readout Specifications | |
Shutter | 2 Position Wheel, Internal RGB Ready |
Exposure | 0.01 to 3600 seconds,10ms resolution |
Correlated Double Sampling | Yes |
A/D Converter | 16 bits |
A/D Gain | 2.0e¯/ADU |
Read Noise | 25e¯ RMS |
Binning Modes | 1 x 1, 2 x 2 |
Pixel Digitization Rate | 30 kHz |
Full Frame Acquisition | <3 seconds |
Optical specifications (f/10) | |
Field of View | 5.4 x 4.0 arcminutes |
Pixel Size | 1.0 x 1.0 arcseconds |
Limiting Magnitude | Magnitude 14 in 1 second |
(for 3 arcsec FWHM stars) | Magnitude 18 in 1 minute |
If the RoboDIMM is combined with the automatic weather station (AWS), the "Greenwood frequency" could be calculated also. However, currently this is not implemented (yet).
In the field RoboDIMM has shown sensitivity to all seeing conditions, registering averages as low as 0.35" and (during the passing of a warm front - La Palma) as high as 7".
Scintillation monitoring
Scintillation index in each aperture is computed as the variance of intensity normalized by the average intensity squared. In this way the scintillation index does not depend on the brightness of the star and reflects only the strength of atmospheric scintillation. Contribution of photon noise and CCD read-out noise is carefully subtracted in the calculation by using dark frames and flat fields.
Exctinction monitoring
Parallel to the seeing measurements, RoboDIMM will measure the first-order extinction coefficient (real-time). This is done by determining the star magnitude in several hundreds of exposures, with a fixed wavelength (500nm).
