PFIS Home > Documents > Code 3170: Operations > OCDD

PFIS supports 3 ways of using the CCD, characterized by the readout style (you can see a table of modes). The modes are shown as allowed or forbidden, and are labelled with their OCDD code names. OCDD modes with an asterisk (*) will be commissioned by the PFIS team.

Normal. In normal mode, the CCD is used to take a single image. The shutter opens to begin the exposure, and closes at the end. A full readout occurs after the shutter is closed. The CCD image is accurately exposed and time-stamped.

Science Modes:

Frame Transfer. The time-resolved mode uses the frame transfer capability of the CCD controller. In this mode, half the CCD is masked off from light; science photons are allowed to fall only on half (4 arc-minutes) the CCD. After some time, the illuminated half is quickly (0.2 s) transferred to the masked half, to be read out. The unmasked half starts fresh, and receives new photons during both the readout and any additional exposure time desired. The shutter is opened at the beginning, and not closed again until all the requested transfers and readouts are complete.

There are no polarimetric modes with frame transfer. The beamsplitter creates E&O fields, which prohibits the use of a masked store area on the chip.

Frame Transfer (Open Loop). Open loop means no internal PFIS activity while the shutter is open. In this mode, a frame count and exposure time are loaded into the CCD control computer, and PFIS just stares during the CCD's frame transfer and readout loop. This mode produces frames accurately spaced in time, and with precise exposure times. The timing is controlled entirely by the real-time CCD control computer.

Science Modes

Frame Transfer (Closed Loop). Closed loop means that PFIS must interact with external subsystems over the network while the shutter is open. In this mode, PFIS must coordinate an interaction between the frame transfer and one of 3 other subsystems: the telescope, the Fabry-Perot etalons, or the polarimetry waveplates. The PFIS is not a real-time system, and cannot execute reconfigurations anywhere near as fast as a frame transfer. Moreover, it must await completion of the reconfiguration (telescope nod, etalon step, waveplate motion) before signalling the CCD controller to transfer the frame, during which time the CCD is still being exposed. If this "expose while reconfiguring" is bad, then every other frame must be wasted in a loop that looks like this:

Science modes:

Charge Shuffle. In this mode, the shutter is opened, charge is moved up and down on the CCD, but without any readouts, and then finally the shutter is closed and a readout occurs. Unlike the time-resolved frame transfer mode, in which a time-tagged series of frames are read out, this mode produces a single frame, whose contents are integrated over time, in synchrony with some other operation, like a telescope nod, etalon step, or waveplate movement. Note that there are no useful open-loop options in shuffle mode.

Science Modes: