Orthogonal transfer CCDs

We have seen that a typical CCD has three gates per pixel and that the readout operation is via clocking the pixel charge in a three-phase fashion. The charge within a given pixel can only be shifted in the vertical direction, that is, from row to row, until it reaches the output register. This is due to channel stops running vertically, which are biased to keep electrons in a given column. One sees a dramatic example of electron trapping along columns in Figure 2.6. A new type of CCD, the orthogonal transfer CCD (OTCCD), has been developed that allows each pixel's charge to be shifted both vertically and horizontally on the array (Burke et al., 1994).

The ability to move the charge in both directions within the OTCCD comes about through the use of a complex, four-phase mode of operation in which the channel stops become an addition gate. Four gates are used in each pixel, two of which are triangular and make up the central part and two of which are rectangular but smaller in area and are split into pairs surrounding the pixel center. The larger number and size of the gates lead to a lower overall QE for the early OTCCDs and more detailed intra-pixel QE effects are to be expected (Jorden, Deltorn, & Oates, 1993; Jorden, Deltorn, & Oates, 1994).

The first application of an OTCCD was to compensate for image motion during an integration, similar to a low order adaptive optics tip-tilt correction (Tonry, Burke, & Schechter, 1997). A 1024 x 512 OTCCD device was built to allow one half of the OTCCD to image, quickly readout, and centroid on a bright star, while the other half was used to integrate on a scene of interest. As the bright star center wandered during the integration, the object frame half of the OTCCD was electronically shifted (both vertically and horizontally) many thousand times per 100 seconds in order to move the image slightly (~ 0.5 arcsec) and follow the bright star centroid position. The final result from the OTCCD was an image with much improved seeing and little loss of information.

The complex gate structure of the original OT design caused small dead spots within each pixel as well as allowing charge to be trapped in numerous pockets within a pixel during charge shifting. These two effects amount to about 3% flux losses, which were probably compensated for in the improved final image. Advances in gate structure design and impurity control in the device composition have eliminated most of these losses (Tonry, Burke, & Schechter, 1997, Burke et al., 2004). The OTCCD promises to be extremely useful for certain applications in astronomical imaging.

A next generation OTCCD camera, dubbed OPTIC, the Orthogonal Parallel Transfer Imaging Camera (Tonry et al., 2004) contains two 2K by 4K OTCCDs. Two end regions on each OTCCD (of size ~2048 by 512 pixels) are used for guide stars. The ideal operation of OPTIC uses four guide stars, one in each quadrant of the large format CCDs to provide real time, "no moving parts" tip-tilt corrections to the science regions. All the problems with pockets and dead cells in the OTCCDs have been eliminated. Howell et al. (2003) present a number of observational extensions to the original purpose of the OTCCD, using it for moving object charge tracking, high-speed, high-precision photometry, and actual shaping of the point-spread function during integration. More on the photometric results available from OTCCDs will be presented later.

OPTIC has performed so well that two large imaging projects (the WIYN observatory one-degree imager and the Pan-STARRS cameras) are planning to use a new generation of OTCCD in their focal plane. Figure 2.7 shows the schematic design for the new style of OTCCD. The individual OTCCDs (approximately 512 x 512, 10-12 micron pixels) are arranged in a single monolithic 8 x 8 checkerboard pattern called an orthogonal transfer array

lines

Pixel Independently OTA:

structure Addressable Cell 8 x 8 Array of Cells

Fig. 2.7. Cartoon of the pixel and CCD layout of the new generation of orthogonal transfer CCDs. In this schematic, we see how the OT pixels are arranged in small (~ 512 x 512) devices (to increase yield and provide local guide stars) and these individual OTCCDs are placed into a square array of 64 independent devices called an orthogonal transfer array (OTA).

lines

Pixel Independently OTA:

structure Addressable Cell 8 x 8 Array of Cells

Fig. 2.7. Cartoon of the pixel and CCD layout of the new generation of orthogonal transfer CCDs. In this schematic, we see how the OT pixels are arranged in small (~ 512 x 512) devices (to increase yield and provide local guide stars) and these individual OTCCDs are placed into a square array of 64 independent devices called an orthogonal transfer array (OTA).

(OTA). Each OTCCD in the OTA is independently controllable and can be used for science imaging, guide stars, fast readout, or simply turned off (Burke et al., 2004).

Figure 2.8 shows one of the first production silicon wafers made by STA/Dalsa. The wafer contains three OTAs, a few other CCDs, and some test devices. This wafer represents one of the first OTA wafers to be processed for the new WIYN observatory imagers.

Was this article helpful?

0 0
Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment