Infrared Camera Description

Dewar

The chip is mounted in a vacuum Dewar with cooled re-imaging optics. The Mark II version of the camera uses a cooled Offner relay system to reimage the focal plane onto the detector. This system has an intermediate pupil (image of the telescope primary), that is cooled and so the field of view looking out of from the chip will be tightly controlled to only see the telescope mirrors and the sky beyond. All other rays will strike cooled surfaces that emit little thermal background radiation.

The filters are mounted in rotatable filter wheel and are cooled to reduce thermal background. A stepper motor rotates the wheel. An LED and photosensor establish the home position.

There are four AD590 temperature sensors mounted in the Dewar. One is on the 77K cooling station and measures the main cooling temperature. A second one is on the radiation shield of the optics. The third one is mounted on the filter wheel case. The fourth is on the chip mount. All temperatures are reported in Kelvin.

Detector

The camera uses the Rockwell Picnic Array with the following properties.

The detector is composed of two parts: a thin substrate of HgCdTe (pronounced Mer-kad Tel) which is aborbs the infrared light and generates electrons and a silicon multiplexer which contains the complex switching array which routes the charge from the pixels to one of four on board output amplifiers.

Controller

The camera is controlled by several commercially available cards in an IBM PC clone. A level shifter, interconnect box is mounted on the vacuum Dewar. A diagram of the system is shown in ircam001 drawing. The clock sequences needed by the camera are generated by a PCIP-AWFG arbitrary waveform generator card made by the Keithley Instruments. It has can generate 8 different TTL compatible clock levels with a maximum pattern length of 32K. They are converted to 0-5 volts clocks by a High Speed CMOS chip.

The four outputs of the detector pass through amplifiers with a gain of (X5.1) and a manually adjustable offset to a 16 bit analog to digital converter card model DAS1602/16 made by Computerboards. It has a 10 microsecond conversion time so the entire chip can be read in 0.65 second. The data from this card pass through a very large FIFO (Mega Fifo by Computerboards) so that they do not have to be read in real time.

Readout

The chip is readout by clocking the multiplexer so that each picture element (pixel) is sequentially connected to one of the four output amplifiers. When the pixel is accessed the charge is measured but not removed. This behavior is different from that used CCD (Charge Couple Device) detectors in which the process of reading the chip by its nature resets the pixels. Instead a separate command is needed to reset the pixel and clear the built up charge. Additionally the charge can be read many times in order to get a more accurate measurement (decrease the readout noise). Since the camera does not have a mechanical shutter, timed exposures are made by resetting the chip, waiting for the desired exposure time, then reading the chip again. While the chip is not being read out it is constantly clocked with a reset waveform to flush out the building up charge.

The waveform generation is discussed in more detail here. Our camera has several ways in which it can be readout out. Note:: in the following the term expose means wait the requested duration.

When a sequence involves two reads the difference of the second - first read is returned. This usually leads to a positive value of the signal. To select these we have added a subcommand to the ccd_status command can be used to select the sequencer by specifying its name in the seq field. For example:

tx ccd_status seq=Fowler

Cooler

The chip is cooled using a Model 21 closed cycle helium refrigerator made by CTI Corp.  Specific information for our refrigerator unit is found here.  The high pressure gas is supplied by a compressor mounted downstairs manufactured by Austin Scientific.

Thermal Design

Below is a work sheet for the redesigned IR camera so that the thermal backgrounds will be small in comparison to the sky the signal.

First we need to know how cool the inner box surrounding the detector.


[ IRCamera Index | Leuschner Index ]

Last updated on 2000 November 21