CCD Digitization Info

[dknguyen]

Can anyone point me to some resources to learn how CCD digitization works? Like the ADC, and how the electronics are arranged and how the charge from a single CCD is selected and transported to the ADC?

Maybe even how to build one.

THanks

 

[ClydeCrashKop]

I found some old literature for Micron Technology Inc. IS32 OpticRAM chip. They said it was different than the earlier generation of image sensing chip such as the CCD. It is a RAM chip with a window on top. See Steve Ciarcia’s Cercuit Cellar in BYTE magazine , Sept 1983 or maybe his web site. As I recall from 28 years ago, CCD (charge coupled device) and this volatile ram were both laid out like a rectangular memory array and made of capacitors. That is why ram needs to be refreshed before the caps loose their charge. It seems that caps loose their charge faster if exposed to light so in this application they charged the whole array and focused the picture on it. Then instead of refreshing it, they repeatedly scanned the array to see which pixels had gone to zero. The lightest pixels lost charge first and the darkest last. They used this info to create a gray scale picture.
This is old info and probably obsolete. This is all from memory so if any of it is wrong, I plead occasional insanity.

 

[hjames]

Texas Instruments used to make some smaller CCDs - they still have some documentation floating around:

Here's a pretty common one
**broken link removed**

**broken link removed**
**broken link removed**
**broken link removed**

These sensors were/are popular with the amateur astronomy crowd, you should be able to find piles of sites with some (obsolete) ADC circuits. (They'd strap a CCD onto a TEC (peltier device) and have minute+ long integration times, something that wouldn't work out too well with the new CMOS sensors).

For more recent info, look at the CCD digitization chips made by Analog Devices/National Semi/TI. (for example the AD9826)

If you're looking for more keywords, look for "correlated double sampling"

 

[dknguyen]

Yeah....I type CCD or CMOS imaging and I end up with commercial products rather than technology documents. Thanks.

I want to try and get something like a 16x16 array (or maybe a bit more resolution, limited by data throughput) with 4-bits brightness values per pixel so I can mess around with it in a uC...see if I can't make some sense of the data coming through.

 

[hjames]

Thing is that the sweet-spot for imaging sensors is in the 10000's of pixels and 10's of megapixels/sec. The smallest commercial 2D parts (not counting optical mouse sensors) are still at least 100'sx100's - with the cheapest ones at NTSC/PAL resolution. You can always go to linear sensors, but that might not be useful for you(www.taosinc.com / www.mouser.com have 100 pixel linear sensors pretty cheap).

In anycase, if you look at the TC211 doc, and a couple of those other links, you can see how CCD sensors work - I've heard it described as a charge "bucket-brigade" approach. There's no way of emulating that using discrete silicon. The CMOS sensor approach works like DRAM in that there is a "row select" and a big multiplexer that reads out a particular pixel. That one you can build using discrete photodiodes.

 

[dknguyen]

Yeah, I found a bit more on how an ADC "picks" the CMOS cell to digitize, but for CCD, all I found was that there is only one exit point for the signal from the CCD, and that the charge from a single cell is transported through surroundings cells to the exit point to be digitized by the ADC...didn't get much more than that.

I'm still not sure how it's "picked" unless it's almost automatic. I have read that with CMOS you can extract only a portion of the image (presumably you cannot with CCD unless it's specially designed to do so). Is it like moving boxes out of a narrow hallway? You take the first box out, which leaves a nice clean path so the next box can travel to you without disruption? So does the charged CCD cell travel along the path of uncharged CCDs (that have already been digitized) to get to the ADC?

 

[hjames]

It's old-school stuff. There used to be these things called "bucket brigade delay lines" which was just a linear array of capacitors with switches. Essentially you'd shift some charge into the first capacitor, hit the clock, and the charge would get shifted into the next capacitor over. Big analog shift register.

How it works is described in the TC211 data sheet I linked in the previous post - You've got charge "wells" in the silicon created by doping which act as isolated capacitors. Then you've got some gates on top which 1) pull the charge from the well into a temporary holding area, and 2) push the charge into the next well in the second phase of the clock, when the gate is pulled "high".

The rows are shifted into the readout array using the same idea, but from the other direction. Both CMOS and CCD's will have some similar limitations - usually it's a lot easier to chose particular rows, and harder to choose columns in a row.