Talk:Charge-coupled device

Untitled
This is the discussion/talk page for: Charge-coupled device.

Technicality
I've gotta second the technicality tag here. I got to this article from the main page, but i have absolutely no idea what it's saying, and i'm fairly well versed in electronics, for a non-professional. Archtemplar (talk) 17:03, 6 October 2009 (UTC)
 * can you elaborate on the particular section(s)? User A1 (talk) 23:16, 6 October 2009 (UTC)

Sure. Paragraphs 1 and 2 should be simple enough for us non-Nobel Prize winners to stretch our minds around. I don't mind doing some research to figure out later paragraphs, but the introduction could be a little more accessible. Archtemplar (talk) 01:50, 7 October 2009 (UTC)
 * OK, having read it with your comment in mind, I now see what you mean. I will fiddle with this in about 9 hrs. If you have any more sections that you think are inaccessible, just jot them down here. User A1 (talk)

the entire 'history' section is absolutely not wikipedia style at all.. definatly needs reworking. you'll understand what i mean after just merely skimming through. inspy

Thanks A1, i think i understand its purpose better now. It says, though, "for example conversion into a digital value"...it does other things too? Archtemplar (talk) 18:02, 8 October 2009 (UTC)
 * Yeah, Bubble memory for one. Here you are, instead of using an Analog to Digital Converter, you are using a simple 0-1 conversion. Theoretically it could be used to do anything where you need charges moved around -- you could even use it to do weird things, like store an audio signal in an analog manner for later output (without digital conversion). It would work, but it would be very odd AFAIK. User A1 (talk) 23:20, 8 October 2009 (UTC)

I googled "ccd bin" and "charge coupled device bin" with no relevant results. I think we need an article describing what the "bin" is and a link to that article. Eddietoran (talk) 10:59, 9 March 2012 (UTC)

A significant portion of the "history" section is plagiarized verbatim from a book called "Catchers of the Light" by Stefan Hughes with no credit to the author in the references section.

Remove Corporate References?
This articles seems to have references to amateur astronomy companies such as SBIG, Astrophysics and Losmandy even though they contribute nothing to the content of the article. In the case of SBIG, the claim is also very misleading because the A0-7 isn't even SBIG's only autoguider and other companies make autoguiders too. Additionally, the AO-7 is an adaptive optics system which uses a smal, fast mirror in addition to standard autoguiding methods to keep target stars centered and to compensate for errors which standard autoguiders cannot. Likewise, Astrophysics and Losmandy are not the only companies to make "study mounts" and debatably not even the best. Lastly, these references are only to amateur companies, not professional telescopes where some of the cutting edge work in CCD astro-imaging is going on (ex: The Sloan Digital Sky Survery).

I would make changes, but I thought I'd put it here for comment first.

I agree - take the corporate references out. SBIG is, IMO, misusing this page for commercial purposes.

- I don't think it's SBIG, just some user who likes SBIG stuff, because they keep describing the AO-7 incorrectly (or at least imprecisely) as an autoguider when it's really an adaptive optics systems. It's like describing a handheld computer as an electronic appointment book. I've changed it back (again). - In the text it is said that because of the quantum efficiency of the CCD detectors astronomers started using them. It worked the other way around. Astronomers were the ones developing CCD's for beter detection of astronomical sources. When telescopes reached a certain size, it was too expensive to build bigger ones. The only way to detect fainter sources was to build bigger telescopes at the time. This was the drive to design better instruments for detecting fainter sources. This lead to the CCD. Just as the photo sensitive emulsion in film was first developed by astromers and used in photographic plates. -J.P Kotze

Add circuit diagram?
Alison Chaiken 17:00, 26 September 2005 (UTC): A simple circuit diagram illustrating the topology of the array and the attachment to the D/A and power supply would help this article a lot. The image doesn't have to be fancy, just illustrate how things are connected. I'd volunteer to make a drawing myself if I knew the layout details for certain.

CCD's in Non-Tracking Telescopes
I was interested in your comment: "An interesting astronomical application is to use a CCD to make a fixed telescope behave like a tracking telescope and follow the motion of the sky." I've not heard of this application before, do you have any further details or links? -- The context I've heard about it is: take a bunch of frames of the moving target with relatively fast shutter speeds and align them based on some common bright objects in all of the frames and then use a stacking algorithm. Stacking is not as effective as long exposures though. --

Stacking has other advantages. By taking a large number of frames and selecting only the best to stack, one can avoid the image degradation of atmospheric seeing or other transient factors inherent in long exposures.Wschart 01:50, 17 June 2006 (UTC) -- I think what the comment actually refers to is not stacking, but drift-scanning. In this technique, the CCD is oriented so that the stars will appear to move (due to the earth's rotation) along a column of the CCD. The CCD is then "clocked out" at the same rate that the stars are moving. Thus, the accumulated charge travels along with the star's image. The main downside is that you have no control over the exposure time; the exposure time will be however long it takes the star to cross the CCD. The upside is that you're continuously reading out the CCD; you don't have to close the shutter and waste a couple of minutes doing a readout. Here is a web page about drift scanning with amateur equipment]; the technique is also used on the Palomar 48-inch Schmidt telescope and the Sloan Digital Sky Survey (they call it TDI, time delay and integrate). -- Coneslayer 19:33, 10 July 2007 (UTC)

Basis of 70%
PizzaPuzzle, I am a little puzzled by the statement "CCDs commonly respond to 70% of the light to which they are exposed", because I'm not sure what the 70% is a measurement of. Is it the percentage of the CCD surface that is covered by sensing material, as opposed to inert structures such as metal conductors? Or is it the percentage of the incident energy in the visible spectrum that falls on the sensors that is converted to electric charge, instead of being reflected or absorbed as heat? Or is it the product of both of these? I would be grateful for clarification. -- Heron


 * The latter, 70% of the photons react with the CCD - the rest are "wasted". As opposed to only 2% usage by a photographic plate. It would probably be useful to divide this page into astronomical CCD use, and more general/technical aspects. Pizza Puzzle

Thanks. I added your 2% figure to the article, and moved the section to "Applications" to avoid repetition. Perhaps only the general/technical aspects should be described here, with the applications (astronomical and otherwise) moved to digital photography. What do you think? -- Heron

Why does light cause an accumulation of charge? Omegatron

You have to know a minimal bit of quantum physics for that. When a photon strikes an atom, it can elevate an electron to a higher energy level. CCDs are commonly made of negatively-doped silicon with many electrons - when light strikes them, it frees electrons to move around and they accumulate in the capacitors (I believe these are also called "wells.") Those electrons are shifted along the CCD and "counted" by a capacitor, giving an effective black & white image of how much light has fallen on each individual pixel. Several things can affect whether a photon releases an electron: polysilicon circuits on the front-illuminated CCD surface can block light from entering, certain depths of the CCD may be penetrated by longer wavelengths without interaction with the atoms, certain screens may reflect off or absorb some wavelengths of the light, etc. (Thus BOTH of Heron's interpretations above are accurate.) Some wavelengths of light can free more or less than one electron per photon. Knowing how many of the photons which fall on the CCD will release an electron is an accurate measurement of the CCD's sensitivity. This figure is called "quantum efficiency." Some CCDs can be made to reach as high as 95% QE. Several Japanese companies have developed many methods of increasing the sensitivity of the CCDs. For example, illuminating the CCD from the side without circuits, increasing the area of the CCD which is receptive to light. Socrates's Universal Disclaimer: I don't know anything about anything. - Nocturnal [edit:] Correction: the doped layers have spare electrons, but if they were free from the start that would of course defeat the purpose of the CCD! Sorry, I'm not a physicist, just an amateur photographer.

Frigoris 12:06, 5 April 2006 (UTC)ithink we need more about *how* a CCD work in the article

Hey can anyone tell me avout how it works in taking X-ray images ?

Actually hitting a CCD with x-rays will destroy the CCD. All X-ray images taken with a CCD are imaging a phosphor/scintillator. —Preceding unsigned comment added by 128.135.176.118 (talk) 19:04, 10 March 2011 (UTC)

Achitecture
We certainly need a few diagrams here. The architecture is also not very clear from the text alone. Those who can, please do. JMK 19:20, 30 April 2006 (UTC)

-- A closely related device is the analog Bucket Brigade Device, which was (is?) used for audio delay lines. Some typical examples are the EG&G/Reticon SAD-1024 and the Panasonic MN3004. It may be useful to add this to the article.

Sensitivity and bandwidth
Can someone please say something about sensitivity (I read something that about 8 photons are necessary to stimulate a CCD while 70 are neccessary to "switch" a grain of photo-sensitive chemical in a film) as well as the bandwidth of sensitivity of a "normal" CCD (in nanometer wave length). THANKS -- Michael Janich 09:23, 6 June 2006 (UTC)

Camera specifications
Can someone explain the meaning of the measurements given for CCD size by camera manufacturers? For example, what is 1/2.5" vs. 1/1.7"?? Originally, I thought these were shorthand inversions of inches (ie 1/2.5" = 0.4 inch diagonal), but it doesn't seem to work out correctly. Thanks!  Slowmover 15:39, 6 June 2006 (UTC) EDIT: As far as I know the inch used for describing CCD-sizes isn't 25.4 mm, it's something around 16 mm. This came from the old vidicons. http://de.wikipedia.org/wiki/Vidicon

I can explain this. When imaging tubes were used in TV cameras, they were measured by their physical diameter. The last size for broadcast tubes, for instance was 17mm, or 2/3rds of an inch. The target size in these was necessarily smaller, to accommodate fro the thickness of the glass wall and the electrodes, so was 11mm diagonal. When broadcast CCD cameras arrived they kept this standard of 11 mm diagonal to be able to use the same lenses. THe same designation, 2.3rds, was used for the CCDs, although it did not make much sense for the. There fore a 2/3rds CCD is 8.8x6.6mm, which seemingly bears no relationship to the 2/3rds number. This was the size used in the first digital compacts, and was referred to as 2/3ds, but when smaller sensors arrived for compacts the manufacturers something illogically kept the same convention, calling them be a size which was about 50% bigger than the diagonal of the sensor. So it's illogical, but derives from TV nomenclature. —Preceding unsigned comment added by 94.171.20.117 (talk) 13:13, 11 September 2009 (UTC)

Technical Details
I added the technical template to the top of this page. I've had small electronics courses and I had trouble understanding the article--I can't imagine what it would be like for a layperson. Certainly more context, preferably less jargon, etc. I don't know enough to do it myself, though. Thanks to anyone who can help! 71.57.91.231 08:20, 16 November 2006 (UTC)

Cooling of CCDs
The article claims that astronomical CCDs are cooled because of their infrared sensitivity to the emission of blackbody radiation from room-temperature sources. This is not true. CCDs are cooled because of the "dark" current produced by stimulation of electrons into the conduction band by thermal excitation from the temperature of the CCD itself. Astronomical CCDs, and perhaps all CCDs, are not sensitive to the wavelengths where the blackbody peaks for room-temperature (around 10 microns or so). AmberRobot 19:08, 31 January 2007 (UTC)

Physics of operation
The "Confusing" template was added to the "Physics of operation" section in April 2007, but the comments about the confusing nature of the section were inserted into the article rather than being placed on the talk page, so I'm transplanting them here. These aren't my opinions necessarily, but just what was on the page, mildly edited for reference translation and grammar. --Drake Wilson 23:43, 27 May 2007 (UTC)


 * "The first two paragraphs of the section are extremely confusing. The paragraphs below, concerning the physics of operation, are confusing, but it seems they are necessarily so. See Janesick's book (2001) Scientific Charge Coupled Devices, for any serious study of astronomical imagers (i.e., imagers like the one described below)."


 * The problem with this section is that it contains very little physics. The bulk of it provides details of the manufacturing process, which is not physics.  I would like to see a discussion of what happens when a photon interacts with the CCD, the conduction band, etc.  For example, at visible light frequencies, I do not believe a photon has enough energy to create an electron-hole pair with the electron in the conduction band.  I imagine that at such frequences the photon will create phonons (lattice vibrations), and if enough photons interact, the cumulative energy will be enough to get some electrons in the conduction band.  This is just speculation on my part, but I feel that it illustrates the type of discussion that belongs here. Ty8inf (talk) 16:00, 1 October 2008 (UTC)


 * A simple calculation shows that an IR wavelength of 9800A corresponds to an energy of about 1.3eV. I chose 9800A because that is the wavelength of the infrared signal from an IR remote, which can be recorded using a video camera.  But I thought that CCDs have a band-gap of about 3-4eV, which means that a 1.3eV photon cannot directly excite an electron from the valence to the conduction band. Ty8inf (talk) 17:45, 1 October 2008 (UTC)

Physics of operation, first two paragraphs
I find these two paragraphs (cable and ring electrodes stuff) incredibly confusing and they don't make any sense to me at all, despite reading them ten times and having studied a very fair deal of semiconductor physics as well as electronics. Shouldn't they be deleted or rewritten? The guy who made the edits (Arnero) seems like a real contributing user but as i said this makes absolutely no sense.

Grapetonix 20:50, 4 July 2007 (UTC)

I totally agree with this assessment. These two paragraphs should be removed and replaced.Brews ohare 18:24, 6 November 2007 (UTC)

Interline transfer LESS sensitive to IR than frame transfer?
Hello, I was just look at "http://www.tpromo2.com/ssmag/cctv/cctv3.htm" where it says that frame transfer CCD's are much less sensitive to infrared light than are interline transfer CCD's. This article says the exact opposite??? 65.183.135.231 (talk) 14:33, 23 May 2008 (UTC)

4/3" sensor size wrong?
If you look at the the wikipedia article Four Thirds System it says the sensor size is 18×13.5 mm (22.5 mm diagonal), with an imaging area of 17.3×13.0 mm (21.6 mm diagonal), while the table here says it 22.500×18.000 mm (28.814 mm diagonal). What is the actual number? The one here seems to be wrong. —Preceding unsigned comment added by 74.64.99.12 (talk) 01:28, 19 June 2008 (UTC) The name "4/3rds" is derived from the broadcast TV "2/3rds" standard which is 11mm in diameter.The "4/3rds" is approximately double this, so the 22.5mm dimension is the correct one. "2/3rds" itself is ultimately derived from imaging tubes, where it referred to the overall diameter including the thickness of the glass walls.Wmck (talk) 17:28, 10 July 2012 (UTC)

Use of CCDs in scanners
Can someone please add a Scanners section, describing how CCDs are used in optical scanners?

In particular, please could someone who is qualified explain the difference between a "3-line CCD (R,G,B)" and an "Alternative 6 line CCD" that you see are used in various models of scanner? Which one produces better quality images?

81.141.52.169 (talk) 09:51, 20 July 2008 (UTC)

CCD blooming
I think that the article should mention CCD blooming. What do you think? --pabouk (talk) 12:51, 10 October 2008 (UTC)
 * Sure -- Be bold and add it, if you have the time! User A1 (talk) 08:47, 11 October 2008 (UTC)

It looks like the photo labeled as 'vertical smear' is in fact an image of blooming/pixel overflow —Preceding unsigned comment added by 128.135.176.118 (talk) 18:59, 10 March 2011 (UTC)

Thank you for the links, pabouk. I agree that this article should mention "blooming". So I added a new section on "blooming" and used your links as references. --DavidCary (talk) 04:09, 6 June 2012 (UTC)


 * I edited it a bit, as it didn't agree with what I know about anti-blooming structures. Dicklyon (talk) 04:30, 6 June 2012 (UTC)

Gain
There is no discussion of gain in the article. —Preceding unsigned comment added by Lzkelley (talk • contribs) 07:44, 13 November 2008 (UTC) The article could most certainly use some work -- you can add it! Then again I could have, but havent -- were all passing the buck :) User A1 (talk) 09:08, 13 November 2008 (UTC)

Provenance and the Nobel Prize
There is an issue brewing with regard to the completeness of credit to those behind the work for which Smith and Boyle are being honored with a Nobel Prize in Physics. According to the Institute of Electrical and Electronics Engineers (IEEE) Spectrum, as available on http://spectrum.ieee.org/podcast/semiconductors/devices/who-deserves-credit-for-the-nobelprizewinning-ccd, Michael F. Tompsett and Eugene Gordon, also then at Bell Labs, may have had a signficant hand in the work. There is a series of links provided, in chronological order, all within the IEEE site.

I lack sufficient experience in Wikipedia to know how best to proceed. Thanks! Dhugot (talk) 14:33, 15 October 2009 (UTC)

Aluminum vs Aluminium
Both are fine to use, there is no argument there. However, since the rest of the article uses American English not British English, (based on the spelling of color), the use of aluminium is inconsistent. Consequently, I am changing it to aluminum.The Lamb of God (talk) 02:45, 23 January 2010 (UTC)

Media?
Not that it actually matters, but isn't the Media Wikiproject a very long way from getting to *this* article? One might as well include "rechargeable batteries" in media, after all, journalists use rechargeable batteries...or "leather" because they wear shoes....or "tomatoes" because journalists often eat pizza. I've taken the liberty of removing that tag. --Wtshymanski (talk) 21:39, 13 September 2010 (UTC)

the photoactive layer
What the basics of operations section needs is a description of how the photoactive silicon layer works. Or a link to a description of how it works, but I couldn't find one. — Preceding unsigned comment added by 74.38.239.72 (talk) 14:53, 22 August 2011 (UTC)


 * See if the edit that I just made is helpful for that. Dicklyon (talk) 01:08, 23 August 2011 (UTC)
 * I have changed your edit, because CCDs use MOS capacitors to do the job done by p-n junctions in photodiodes. But the article still lacks a device physics perspective on the generation of electron charges. I'll give it a try based on Sze's Device physics. Mathieu Perrin (talk) 16:12, 26 November 2011 (UTC)

intensity vs. irradiance
I just reverted the change of "intensity" to "irradiance" in "causing each capacitor to accumulate an electric charge proportional to the light intensity at that location." I can agree that intensity of is not very specific, but "proportional" can hide all times of factors, like the integration time and spectral weighting. However, "irradiance" is probably too specific, as it suggests equal weight for all wavelengths, which would certainly not be accurate (not even to first order); it would be no more correct than "illuminance", which is wrong in the same way, even though it is commonly used, in the measurement "lux-seconds". Getting into more detail makes sense only if we have something that's correct and sourced. Dicklyon (talk) 20:38, 17 October 2011 (UTC)

Microelectronic
Can you imagine a non-IC CCD? Picture rows upon rows of 6SN7s glowing away...it would be like a vacuum-tube CD player or cell phone. --Wtshymanski (talk) 14:23, 9 March 2012 (UTC)


 * And the relevance of this piece of trite to improving the article is what exactly? DieSwartzPunkt (talk) 13:20, 6 December 2012 (UTC)

Sze
The 4th line below the heading "Design and manufacturing" contains the string "Sze". Is this a typo? If not, please explain it. I think this article is NOT confusing at all, though I, too, would like to see a little more physics. Tim Coahran (talk) 02:12, 21 March 2012 (UTC)


 * It's Dr. Sze's last name; not a typo. Dicklyon (talk) 02:44, 21 March 2012 (UTC)

CMOS VS CCD
"In applications with less exacting quality demands, such as consumer and professional digital cameras, active pixel sensors (CMOS) are generally used; the large quality advantage CCDs enjoyed early on has narrowed over time."

Um, no. In video at least, CCDs are still generally preferred to CMOS sensors, partly because CCDs use intra-frame compression, while CMOS chips use inter-frame. The practical consequence is that some CMOS camera moves can produce flutter that is easily visible onscreen.Jim Stinson (talk) 18:29, 22 May 2014 (UTC)


 * For scientific instruments (read satellite telescopes) CCD's are still the preferred camera. A lot of high end cameras have post processing to deal with stuff like flutter. CombatWombat42 (talk) 20:47, 22 May 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Charge-coupled device. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120727032200/http://learn.hamamatsu.com:80/articles/ccdsatandblooming.html to http://learn.hamamatsu.com/articles/ccdsatandblooming.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 18:17, 19 November 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Charge-coupled device. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20071228122423/http://www.nae.edu/NAE/awardscom.nsf/weblinks/CGOZ-6K9L6P?OpenDocument to http://www.nae.edu/NAE/awardscom.nsf/weblinks/CGOZ-6K9L6P?OpenDocument

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 00:38, 3 August 2017 (UTC)

EMCCD Formula Wrong
The giant formula given for EMCCD counts is wrong. In fact, it is nutty. If the gain is g=1, then P(n|m) = kronecker_delta(n,m). The formula fails the simplest possible sanity check.

2600:1700:C280:1CF0:69FB:6EAF:DFA5:32FB (talk) 21:31, 10 February 2021 (UTC)