Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.arcetri.astro.it/irlab/fasti/l3ccd/l3ccd.html
Дата изменения: Thu Jul 28 18:23:57 2005
Дата индексирования: Sat Dec 22 06:47:25 2007
Кодировка:

Поисковые слова: п п п п п п п п п п п п п п п п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п
Fasti Web Site




Fasti Use for L3CCD control.

Some background.

Some recent results (see) attracted the attentions of astronomers working on field which could benefits from photon counting techniques with high speed and spatial resolutions as adaptive optics. The key technology is a clever add to the standard CCD: an electron intensifier directly integrated into the silicon (details here, also, for a comparison of different techniques see here

Fasti use

Fasti has been designed to be fast and largerly configurable. So we can easily adapt it design to the L3CCD detector, with only minor modification in the digital part.

We plan to build a prototype of this system using the low cost video format detector available. Later on also the bigger scientific detectors could be used. To shorten developing time, we also plan to use the two analog boards which can interface the digital TTL electronic levels to the correct ones for this technology.

The waveform generation flexibility of Fasti can provide us with the ability to do also on-chip and off-chip co-adds. Also the availability of a substancial processing power can give us the possibility to test in software many different data pre-processing techniques before the integration of the right one in hardware.

Speed estimation

With a maximum output pixel rate of 11 MHz, the frame rate is around 66 Hz. To get higher speed, there are some possibilities:
- to use an internal (analogue) co-add in the column directions (2-4x gain)
- to use an internal (analogue) co-add in the row direction (no speed gain)
- to use only one half of the column, masking 1/2 of the detector (2x gain)
- to increase the clocking maximum speed to 20 MHz or more ( to be tested, maybe 2-3x)

With this techniques, we can gain a factor from 4 to 26 in frame speed, giving values ranging from 260 to 1500 frames/sec.

The higher clocking speed is only speculative, and has to be tested on the real device, being out of the official specifications. Moreover, for the possibility to introduce non linearity, charge losses, pixel contamination ecc., this procedure should only be used in photon counting mode.

The data rate attined in the best case (30 Mhz), give no impossible load to the digital acquisition in the photon counting mode, since data are 1 bit only, they can be parallelized, giving a maximum speed of about 940.000 32 bit word/sec.

Prototype configuration

The prototype we propose is a modification of the Fasti Infrared design. The main difference lies in the analog to digital translation. This prototype is targeted to test the photon counting mode, so the A/d is a comparator, which gives only the presence/absence of a photon in the pixel being measured.

The common parts are:
- the embedded computer, for the prototype it can be an ordinary desktop.
- the inner bus, consisting of ethernet and serial busses.
- the sequence generator, the SVB1, whose 8 channels are enough.
- the acquisition board, capable of more than 3.5 millions of 32 bit words/sec.

The custom parts are:
- the electrical level translation board (bought from Marconi Instr.)
- the A/D board, consisting of a comparator and a parallelizer.



Last Modified:
10 April 2001


Back one page Labir Home