The blue and red detectors each have a low and a high gain setting. The default setup for each detector is the high gain setting, as most observers are looking at faint objects. Low gain mode is useful for objects that need more dynamic range, such as extremely bright objects. These settings can be changed with the Detector Configuration Submenu (Camera button) of the Configuration Manager.
The same information but in table form:
| Detector Side | Gain Mode | Bias Level (cts) | Saturation Level (cts) | Nonlinearity Level (cts) |
| Blue | Low (setting=5) | 2k | 65k | ∼62k |
| Blue | High (setting=10; default) | 10k | 65k | ∼60k |
| Red | Low (setting=0) | 160 | 45k | ∼42k |
| Red | High (setting=1; default) | 300 | 63k | ∼58-60k |
Unfortunately, ANY change in binning on the blue side causes increased read noise. This means more noise after either 1x1-> 2x2 or 2x2 -> 1x1. The effect is large for the first 20 minutes, medium for the first hour, and smaller for hours after that. Please see the ATTACHED plots and document.
If you plan to start the night with 1x1 binning (the Small slicer), please let your SA know so that they can take some flushing frames to minimize the effect on your calibrations. This is critical because we run the daily afternoon focus in 2x2 binning (Medium slicer), so switching to 1x1 binning will always trigger increased blue read noise for your afternoon calibrations unless we prepare ahead of time.
Additionally, when first half observers decide to use 1x1 binning (Small slicer) for their science and second half observers are using 2x2 binning (Medium or Large slicer), we will have the first half observers stop using the blue detector 15 minutes before switchover so that the SA can switch the binning and start flushing the detector to mitigate the increased read noise effect on the 2nd half observers. The first half observers can keep exposing with the red detector for those 20 minutes, just not the blue. Second half observers who choose to switch to 1x1 binning will take flush frames during their own time.
Binning changes on the red side do NOT change the read noise.
The red detector is 500 microns thick, thicker than most normal red-sensitive detectors
(e.g. LRIS's red detector is 300 microns thick) to increase sensitivity in longer wavelengths, 800-1080 nm.
This presents a much large interaction area for cosmic rays, so we catch way more of them than LRIS's red detector.
The red detector has 0.8-1.0% pixels contaminated by cosmic rays in a 300s exposure using 2x2 binning.
The effect scales linearly with time, so if the exposure is longer, the more area of the detector is contaminated with cosmic rays.
We get cosmic rays of every shape: point sources, streaks, and curves.
Even more challenging, many of the cosmic rays have count levels similar to observers' target emission lines and continuum sources, making them hard to remove using normal methods.
We strongly recommend that observers plan to take <=300 sec exposures on the red detector in 2x2 binning and <=600 sec exposures in 1x1 binning.
To allow for cosmic rays to be removed, we encourage observers to take a MINIMUM of 3 exposures per position.
This will allow for medium filtering to help create a mask and remove cosmic rays as images are combined.
Most observers will take 1 blue exposure while take 3-4 red exposures.
To match the overhead of 2x2 binning exposures, one 1320 sec blue exposure will end at the same time as four 300 sec red exposures.
Similarly, one 990 sec blue exposure will end at the same time as three 300 sec red exposures.
Please see the Estimated Overhead Page and take into account blue and red readout times, as well as red reset times.
