IMACS – Updates

IMACS – Updates

  • f/4 detectors unstable behavior

As of 31 October 2023, the f/4 dewar has been returned to operations in the IMACS instrument with all detectors behaving properly following the resolution of an identified short in the 10V supply lines entering the dewar at the hermetic seal connector. Unfortunately, the dewar ion pump and vacuum began to exhibit unreliable behavior once the system was cooled down to operating temperatures. We are operating without the ion pump active for the time being until we can schedule another intervention to again remove the dewar to the clean room for a change of the ion pump and another thorough investigation of the dewar vacuum.

older new below…

The detector array at f/4 has intermittent unstable behavior in terms of elevated noise issues as well as individual detector readouts at zero counts. While recent interventions in the dewar as well as repaired clocking electronics boards had seemed to improve the situation, the detector array is not considered robust enough for continued scheduled usage until the underlying cause of the failures can be identified and resolved.

As of February 2023, the lower 4 chips of the f4 detector show intermittent behavior. The upper part of the array is operational. Thus, long-slit programs should not be affected by this. However, PIs with MOS programs, should have backup plans for their run, or they could choose to move their programs to f2.

  • Mask Blanks

New mask blanks are currently on site. The production costs for the blanks is substantially higher than earlier manufacturing costs with a per unit cost of USD$650. This is what the fee for masks will be as of 01 November 2023.

older news below…

We are currently reusing blanks of scarcely populated old masks to create new IMACS masks. This requires additional time, since the slits of old masks need to be taped over. Because of the substantial increase in effort, submission deadlines should be strictly followed. Along with submitting SMF files, please send an email prioritizing your masks with clear assignment to each night of your run. It is possible we will not be able to accommodate all requests. 

New mask blanks are in manufacturing but a delivery date has not been defined.   

Past Issues

  • U_Sloan filter damaged

On July 10th, 2022, during a filter change, the new U_Sloan filter fell down and was  irreparably damaged, thus it is no longer offered for science.

  • Flexure Control: 

As of October 2019, the IMACS GUI has been modified in order to keep flexure control OFF by default. It is possible to turn flexure control ON, but there are rotator angle ranges where the piezo-stages cannot control the flexure of the system properly, resulting to error messages. In the “control OFF” configuration, we expect an offset of <1.5pixels in the x direction and <1pixel in y for the f/2 camera, throughout the rotator range. 

If the observer notices larger motions, they should mention it in the relevant section of the night report. 

  • 400l/mm grism:

A new grism is available for use with the f/2 camera. Due to a manufacturing error, the ruling on the 400/mm grism is slightly rotated with respect to the prism. This results in a leftward shift of the dispersed images. Observers should be aware that targets more than approximately 12.3 arcmin left of the center (as viewed in the QL tool or in maskgen) will be shifted off of the detector mosaic.

  • IMACS readout noise modes:

As of July 25th the old “slow” readout mode in IMACS has been removed and the previously known “snail” readout mode has been renamed as “slow”. Currently both cameras have only two readout modes: Fast and Slow (which corresponds to previous “snail”). This is part of an effort to only offer configurations with stable noise levels. 

  • Minimum Exposure times:

In order to avoid shutter delay effects on flux counts, exposure times should be longer than 2sec for both cameras.

  • Dewar swap: 

As of Dec. 20, 2017 the Mosaic3 dewar is mounted on the f/2 and the Mosaic2 dewar is mounted on the f/4. Photometric zero points and grism/grating efficiencies after the swap can be found in the manual. Noise and gain values also updated for both cameras. 

  • f2 drop in efficiency – resolved

The f2 zeropoints, after cleaning the primary mirror, the adc and the work done on the f2 camera are back to nominal values. The original loss was of the order of 0.3mag and not 0.5mag as previously reported (the offset was caused by a numerical error in calculating the zeropoints). The original loss of 0.3mag, detected in February 2023, was due to various factors, external to the camera (status of the primary mirror and the adc and wildfires in the south of Chile that affected extinction).

Investigating the issue, the f2 camera has been inspected, various optical elements have been cleaned and the doublet and quartet had their oil flushed. The throughput measurements increased by 10%. On sky measurements of zeropoints also indicate a 0.1mag in all bands.

  • Chip#4 on f/2 camera: 

As of September 2019, we are experiencing intermittent loss of signal (0 counts) in chip#4 of the f/2 camera. We believe that this is due to problems inside the dewar. When this happens, currently the only options are to either change field (since by changing rotator angle, chip#4 might become responsive again), or keep obtaining data sacrificing the information on chip#4. 

  • Issues with the MACS collimator

UPDATE: as of April 15, 2022, the collimator has been serviced and the issue has been fixed.

Example of an IMACS image affected by the elongation artifact

On March 25th, 2022 we became aware that deep IMACS images were showing elongation of the stellar profile of saturated stars in most parts of the field. Based on pinhole images the effect also appears in not saturated images but it is more subtle. 

For the f/2 camera, the chips that are mostly affected are 1,2,3 and 4. The effect is worse in the edge of the mosaic, of the order of 10%, decreasing to 1% in the areas that are less affected. 

For the f/4 camera, all chips seem to be affected, with the intensity decreasing from right to left (chips 3,4,8,7). Here the ratio has a smaller gradient with an average of a 3% effect through the mosaic. Chips 1 and 6 seem to be the ones affected less. 

This artifact should have marginal effect on spectroscopic programs. 

The effect on imaging programs depends on the scientific goal. For example programs that rely heavily on PSF analysis and deconvolution might be unfeasible. 

The root cause of this issue has been traced back on the collimator optical elements and the instrument team is working on a solution. 

Table of Contents