Panic Detector

The new detector implemented in PANIC since 2022 is a monolithic HAWAII-4RG

Full details on detector performance and characterization can be found in this document.

Without any correction (left), the detector is linear within 1% from 3000 ADU to 27000 ADU. After applying the NLC (right), the residuals are less than 0.5 % all the way up 48000 ADU (99% of saturation).

For very high precision, the correction should be applied to the individual images (not coadded). However, if saving the data as "integrated all" (coadded) is prefered, the correction may still be applied without problems up to about half full well (~ 20000 ADU). Otherwise, please have in mind that precision is lost due to averaging, being worst towards saturation.

The PSF of the objects may not be uniform across the full field of view (FoV). We have observed a distorsion of the PSF in the corners of the FoV, which is typically more pronounced in the northeast corner (top left). This phenomenon is more noticeable under good seeing conditions. The irregularities form a diagonal pattern across the FoV. The graph displays the PSF FWHM in a sequence corresponding to the J filter. This pattern is not constant: the diagonal may be shifted.

Recommended Readout mode: cntrs (continuous sampling read).
The detector is read N times during the integration time, obtaining a total of N frames. The desired number of frames is adjusted so that always a multiple of the minimum integration time is used. The image is calculated by subtracting the first frame from the last frame, ignoring all the other frames on the ramp (the user is able to additionally save all the frames on the ramp, if desired). Advantage: recommended mode por PANIC. Fully characterized. Disadvantage: only multiples of minimum integration time allowed as integration time.

For the rest of the readout modes, please consult the user manual.