The design of the CCD camera, and the choice of CCD type, has been carefully made to enable the scientific goals of EIS to be met. The scientific goals of EIS, and more details about the instrument as a whole, can be found at the solar EIS home page. Some of the key scientific drivers for the camera design are listed below:

It is important to be able to measure suitable wavelengths across a broad temperature range corresponding to the transition region, corona and solar flares. In addition, it is also important to be able accurately co-align images from EIS with those provided by SOT and XRT. Consequently, two wavelength ranges were selected, 170-210, and 250-290.

Thus, two CCDs are required, one for each wavelength range. In addition, to maximise the Quantum Efficiency (QE) of the detector the CCDs will be backilluminated. At the EUV above wavelengths, the QE of the detectors should be around 80%,
 

In addition, it is important the resolution of the CCD is not degraded for example, by charge spreading or broadening of the spectral line due to loss of charge transfer efficiency.

Charge spreading will occur as the design of a backthinned CCD leads to a small region which is known as the 'field free region'. Here, the electron charge cloud generated by incident EUV photons will have to diffuse (and hence spread out) through this region before being collected in a potential well. The effect of this diffusion is that some of the charge collected in a region corresponding to the 'illuminated pixel' will in fact be collected in adjacent pixels, degrading the effective resolution of the CCD. For EIS, the charge spreading is predicted to degrade the effective resolution of the CCD to around 16um, which is still sufficient to achieve the spectral resolution goals.

During the mission, the effect of radiation induced charge trapping in the CCD will be to reduce the overall Charge Transfer Efficiency (CTE). This means that as the CCD is being clocked out, charge in individual pixels will be trapped and then re-released into other pixels. One effect of this trapping may be to degrade the spectral resolution. Although the physics of CTE is well understood, the effect on actual images is much harder to predict. Such predictions for Solar-B EIS can be found here.

To minimise the potential effects of reduction in CTE, a trade off can be made between CCD operating temperature (which generally, should be as low as possible) and which will 'freeze out' the effects of the charge traps, and the amount of shielding used, which will minimise the initial radiation induced damage in the first place. Again, this subject is discussed in detail here.

In addition, it is important reduce the thermally generated noise to levels where it has very little effect on the overall noise. For the integration times expected for quite Sun regions, a temperature of below -40 degrees C should be sufficient to ensure that the main source of noise is the shot noise on the incident photons.
 
 

• selecting a fast readout time - the EIS CCD camera will digitise pixels at a rate of 500,000 pixels per second. Thus, it would take around one second to fully digitise the largest camera image (2048 by 512 pixels in size);
• adopting on-chip windowing so that CCD rows which do not contain information can be quickly 'dumped'.