Detector

The new CCD controller for ALFOSC always gave an intermittent noise which was identified to be at $\sim$13.8KHz. The source of this noise was believed to be from the internal switch mode power supply (PSU). The occurrence of this noise seemed to becoming more persistent and replacement PSUs were requested from Copenhagen. Three new PSUs were brought down from Denmark, two identical versions and one other, but all three different to what we had. Installing one of the two identical PSUs resulted in a lot of random low level noise. The other variant of the PSU was tried and it was found to give no additional noise, and we now achieve the specified noise performance for the ALFOSC CCD.

The new CCD controller for ALFOSC came with a temporary external temperature controller which plugs to the side of the controller. This connector lacked a proper fixing device and came loose a couple of times. The connector has now been fitted with a proper cover and clamping system.

It has been noticed that the bias level with the new CCD controller changes from the typical 10000 ADU level to 12000 ADU or 14000 ADU as well as occasionally an apparent arbitrary level around 23456 ADU. It was noted that a specific sequence of low-level commands would restore the level to the nominal 10000 ADU and this was added to the sequencer command to reset the CCD to full frame and no binning.

During the Copenhagen summer school one frame was accidentally heavily overexposed and the subsequent frames had clearly visible excess flux in some parts of the detector. The excess flux disappeared after some time. The tests indicate that after heavy overexposing there is some excess flux in some parts of the detector. The shape of the pattern and the dependence on the readout speed indicates that the ``light source'' is inside the dewar. As the heavy overexposing during the night is rare, this should not be a big concern for a regular user.

Quality control of the ALFOSC detector indicates that with the current controller there is less readout noise when using Amplifier B than Amplifier A. The plan is to change to Amplifier B being default. However, there are some issues when windowing the CCD with Amplifier B that needs to be clarified before this change in default can be implemented.

A known feature with the ALFOSC CCD camera is that there is a delay (of 0.085 sec) in the shutter which causes the exposure time to be shorter then requested. The detector controller software does not allow for such an error, and this needs to be changed (i.e., the effective exposure time should be equal to the requested exposure time, while the image header should provide the proper headers as for the different timing parameters. Now the header only reflects what is requested which is different from the effective exposure time). A separate issue is that due to the design of the shutter an exposure will take at least 0.3 sec. I.e., if you request a shorter exposure time you still get a 0.3 sec exposure. The exposure command will be changed to warn for these features.

Thomas Augusteijn 2013-05-10