CCD Primer

Binning
Bracket Pulsing
CCD Grading
Cosmic Rays
Dark Current
Deep Depletion CCD
Detection Modes
Dual Capacity Mode
Dual Readout Mode
Dynamic Range
Etaloning in CCDs
eXcelon CCD-EMCCD
UV Extension
Fiber Optics
Flat Fielding
Full Well Capacity
Gain
Image Calibration
Imager Architectures
Image Intensifiers
ITO CCD
Kinetics Mode
Linearity
Matching Resolution
MPP Mode
Noise Sources
On-chip Multiplication Gain
Open Poly CCD
Optical Window
PVCAM
Quantum Efficiency
Readout vs Frame Rate
Reducing Dark Current
Saturation/ Blooming
Signal to Noise Ratio
Spurious Charge
XP Cooling

 

 Fiber Optics

in the majority of CCD applications, light reaches the CCD through a lens- or mirror-based optical system. However, in some situations it is advantageous to use an image-preserving fiber optic bundle in place of conventional imaging optics. Significant gains in the amount of light collected can be achieved by directly coupling the light source to the CCD using fiber optics. Depending on the amount of demagnification, the gain in light collected can exceed 10x that of a f/1.2 lens.

Imaging fiber optics are commonly used to couple light from x-ray or neutron scintillator screens, chemiluminescent markers, image intensifiers, or streak tubes. Fibers can be bonded to most front-illuminated CCDs as well as some back-thinned devices.

The Coherent Fiber Bundle
A coherent fiber bundle is a collection of single fiber optic strands assembled together so that the relative orientation of the individual fibers is maintained throughout the length of the bundle. The result is that any pattern of illumination incident at the input end of the bundle re-emerges from the output end with the image preserved. Imaging fiber bundles can be made in a variety of shapes and sizes, with the most common having a circular cross section. Magnification can be achieved by the use of tapered fibers in the bundle.

Proprietary Fiber Bonding Process
In order to successfully couple light from an imaging fiber bundle to the CCD, the CCD and fiber bundle must be in very close proximity. Light emerges from the individual fibers at large angles, and a gap between fiber and CCD will lead to a loss in resolution. Princeton Instruments uses a proprietary bonding process to minimize the distance without sacrificing CCD performance. This process directly bonds the fiber to the CCD without oil layers or the use of intermediate fiber stubs that introduce losses in spatial resolution and transmission efficiency. In addition, the bond is stable and will survive the repeated thermal cycling that occurs in HCCD camera systems. Princeton Instruments continuous innovation in fiber bonding has extended available fiber tapers to over 165mm in diameter, coupled fibers to the largest commercially available scientific sensors, and even mated fiber bundles to high efficiency back-illuminated sensors.

Efficiency vs. Magnification
Besides the transmission losses through a large piece of glass, fiber-optic bundles have a transmission loss due to changes in the fiber diameter as light traverses the bundle. When light travels down a tapered fiber, a decreasing reflectance angle results in some of the light paths exiting the fiber. This appears as a loss in "effective" numerical aperture (NA). The relative loss between fibers with different magnifications can be estimated as the ratio of their magnifications squared. The larger the fiber bundle's magnification, the greater the reduction in effective NA. Fiber bundles with a 1:1 magnification, known as "stubs," provide the highest throughput. Applications requiring the highest possible light collection efficiency benefit most by using large CCDs to reduce the amount of demagnification required.

Limitations of Imaging with Fiber Optics
A disadvantage of fiber imaging systems is that field of view is limited by the size of available fiber bundles. Currently, the largest available fiber optic tapered bundle is 165 mm in diameter at the large end. However, to enable imaging of even larger areas, Princeton Instruments can create a mosaic of fiber bundles which are connected to multiple CCDs. This assembly can either be packaged in a single camera head, or into multiple camera heads, depending upon the number of bundles in the mosaic and whether or not the bundles are tapered. A second limitation of fiber optics is the introduction of distortion and non-uniformity of response. These defects are introduced during the fibers manufacturing process. Because these defects are static, they can be corrected through image processing. For example, response non-uniformity can be handled in most cases by flat-field correction. Gross distortion can be corrected by appropriate scaling and warping of the image data. Shear distortion, sudden dislocation in the alignment of adjacent fibers, is more difficult to correct for due to its discontinuous nature. Photometrics fiber defect specifications are available for customers requiring detailed information.

Fiber Optic Options
Many of our cameras are available with imaging fiber optics. Fiber bundles range in magnification from 1:1 fiber stubs to large 6:1 fiber tapers, and in diameters up to 165mm. Supported CCDs vary from 512 x 512 pixels to 4096 x 4096 pixels. Fiber bundles are available with extramural absorption (EMA) fibers to improve contrast, and low-thorium glass to reduce background from radioisotopes. At a customer's request, Princeton Instruments will also attach scintillating fiber optic faceplates to the front of fiber optic tapers.