MIT/Lincoln Labs CCID20 W7C1 CCD Test Results

This is the fourth standard epi, thinned, backside MIT/Lincoln Labs 2Kx4K CCD to be characterized in the UCO/Lick detector lab. It is very similar to other standard epi CCID20s tested at Lick.

A summary report is available. Noteworthy points include:

Brick wall pattern. With this device we did some tests of the amplitude of the brick wall pattern as a function of temperature. As the report shows a dramatic reduction in the pattern is possible if the CCD can be operated at higher temperature. Further measurements on other devices will occur.
CTE. Near perfect CTE is achieved using the serial and parallel clocks which generate low spurious charge. This is typical of Lincoln CCDs.
Read Noise. Read noise is around 2 electrons. This is as good as or better than any thin CCD in the world.
Low Level Traps. Only four low level traps were found using charge pumping. This is a very small number for such a large device and is evidence of Lincoln's excellent processing capabilities.
There is a small "dimple" in the CCD which is about 2mm in diameter. This is probably caused either by a void in the epoxy which attaches the thinned CCD to its supporting silicon wafer or by a problem in the thinning itself. We have not measured the physical depth of the depression but it is not likely to be more than a few microns deep.

Original postscript files are available from our anonymous ftp server and these provide better resoultion and clarity than is usually possible on a web page. Check the INDEX file for a description of what the other files contain. Here are a few figures to illustrate device highlights:

QE curve: This is a pretty typical QE curve for the CCID20s. The QE is an average over a fairly large portion of the CCD, so this result is an average over the brickwall variations.
Surface plot: The CCD has a bow and a twist. The peak-to-peak variation is about 27 micrometers. This does not include the "dimple" which is probably caused by an epoxy void.
Brick wall versus Temperature: This plot shows the dramatic change in the brick wall pattern amplitude with temperature. Other tests show that the amplitude continues to decrease as the temperature is raised.
Dimple: The small depression in the CCD about 2mm in diameter is seen in images only at very long wavelengths where the interference pattern is changed.

This plot shows the QE measurement made at a device temperatures of -130°C. It seems a little lower at the peak than other devices we've tested recently (compare W20C1 or W6C1), but not by much. We've found that the QE can be somewhat temperature sensitive, especially in the less sensitive areas of the brick wall pattern. So this result will change a little with temperature, with the QE increasing with increasing temperature.

plot of QE.

The two plots shown here are the same data set. The second plot is an edge-on view to better show the full extent of the curvature and the twist. The CCD connector is along the right edge. The laser beam used to measure the surface shape was badly distorted by the small "dimple" in this CCD, so that region of the device was not measured and the dimple does not show in these plots.
Surface plot showing bow and twist. p-p=26 microns.

Edge-on view of same data set as above.
Same data show edge on

This is a series of line plots across the CCD showing the sensitivity variations known as the brick wall. The four curves show the same row in four separate images obtained when the CCD was at the indicated temperatures and they illustrate the change in sensitivity with temperature. All of these data were obtained at 6500Å. Other tests indicate that the amplitude of the sensitivity variations will continue to decrease with increasing temperature. Since these CCDs don't have MPP the dark current is certainly a problem at -100°C and above. We have not yet begun to investigate whether or not dynamic dark suppression will help us work at higher temperatures with low dark and low noise.

Plot showing that the variations known as the brick wall decrease with increasing temperature.

This 9000Å interference fringe image shows the small section of the CCD in the region of the apparent depression. The area is roughly 130 pixels or 2 mm in diamter.
9000Å fring image showing area of the dimple.

9000Å fring image showing area of the dimple.

Back to Lincoln Labs Top Page