ING Banner
Home > Astronomy > Detectors > Measuring Linearity

Measuring Linearity of the ING CCDs

The linearity was measured using an external LED driven from the CCD SDSU controller. The current to the LED was of the order of 100microamps so as to reduce self heating and subsequent dimming. The LED was pulsed on at the beginning of each linearity measurement for between 1ms and 2s. For the lower pulse times, some correction was necessary due to small timing errors in the SDSU code. Actual pulse width was confirmed using an oscilloscope.

After each measurement exposure the bias level was noted as was the mean signal in a small box positioned close to the illumination maximum. The signal-bias was then calculated and divided by the length of the LED pulse to yield the 'signal per unit illumination' (SUI) at each exposure level.

The SUI at each exposure level was then compared to the SUI at an exposure of approximately 1000ADU, the difference was expressed as a percentage. The graphs that appear in these web pages show this percentage plotted against the actual exposure level in ADU (i.e. the exposure level on an un-reduced image, where the bias has NOT been subtracted).

The choice of 1000ADU as the reference point is somewhat arbitrary. It seems fair to assume, however, that at small signal levels such as this , the electronics will be linear and that non-linearities will only begin to appear at larger signal levels.

Further information on the measurement technique can be obtained from Simon Tulloch at ING


Simon Tulloch
June 2001

Top | Back

Contact:  (Detectors)
Last modified: 21 November 2010