What webcams use a CCD chip, rather than CMOS?

[Ben321]

I know that most webcams use a CMOS sensor, due to how cheap they are. However, CCDs have a feature that CMOS chips don't. They can accumulate light. CMOS chips output a voltage for a given light intensity, while CCD chips store charge for as long as light is falling on it. This means that with CMOS sensors they can only output a realtime signal, but CCD chips can be used like photographic film, for a true long exposure.

This is why I'm looking for a CCD webcam (already have a Toucam, but want another CCD webcam, in case my Toucam ends up getting dropped and broken). I know that few models of webcams were made that actually have CCD chips, and some of them are extremely rare (like the Philips Toucam, which is so rare that I probably am one of the few people who have one). I'm hoping that other companies than Philips may have made some CCD webcams, and I hope that there are some that might even be in production right now. I'm hoping that the Toucam's rareness is itself rare, such that most CCD webcams are easy to find. But I've not had much luck finding them. Are there any CURRENTLY manufactured (the older they are, the harder they will be to find, so currently manufactured is best) CCD webcams, made by any company? If so, is someone on this forum aware of who makes them, and what online-stores sell them?

[riklaunim]

CCD are pretty much dead, as Sony is not making them any more. CMOS are very capable of long exposure just like CCDs, and latest CMOS sensors offer much better performance than previous CCDs. There are no modern webcams with CCDs and there won't be any. To get long exposures with CMOS you need a camera that supports them - so no webcam. So in both CCD/CMOS cases you need a dedicated camera.

[Ben321]

9 minutes ago, riklaunim said:

CCD are pretty much dead, as Sony is not making them any more. CMOS are very capable of long exposure just like CCDs, and latest CMOS sensors offer much better performance than previous CCDs. There are no modern webcams with CCDs and there won't be any. To get long exposures with CMOS you need a camera that supports them - so no webcam. So in both CCD/CMOS cases you need a dedicated camera.

So how does CMOS do long exposure? CCD used charge wells. And I don't think CCD technology is dead yet. Professional astronomers like those at universities and observatories, and visible light cameras attached to multi-million-dollar deep-space observing satellites, they all use CCD cameras. Even most high-sensitivity security cameras (those so called "starlight cameras", that have sensitivity down to 0.00001 lux) still use CCD chips, not CMOS. Some cheap security cameras use IR LEDs and CMOS, but the more expensive ones use CCD chips and no IR LEDS, so they can see in almost perfect darkness (IR LEDs tend to have a dim red color, that can be seen in complete darkness, once the human eye has gotten dark adapted, and can notify a thief that they are being watched, when the whole point of a hidden security camera is to keep it hidden, so CCD cameras with no IR LEDs is preferred by security professionals).

And by the way, Sony still makes those ultra-sensitive EX-HAD CCD chips for use with "starlight" security cameras.

[riklaunim]

Consumer CCDs are dead. The high end CCDs still exist and will do for some time. If you really want to buy a CCD then there are still Sony and Kodak CCDs in DS cameras and will perform well. If you think that those CCDs are more sensitive than CMOS - then no, they are not / not always. Current Sony CMOS sensors have much lower read noise than their CCDs, and the QE is also as high or higher than CCD. Few years ago CMOS were way way behind, but not now. You can check astrobin, cloudynights and other sites for DS images taken with ASI178/174 or latest ASI1600. And if you want to see how a "professional" level  CMOS would look like then check this: http://qhyccd.com/QHY42.html .

Very low read noise also allows extensive lucky imaging at 1-few second exposures, for example http://www.astrokraai.nl/viewimages.php?t=y&category=7

Lux isn't comparable unit of sensitivity as it's measured with some sensor (variable pixel size), with some lens, and some light source. It also doesn't tell anything about noise. For pixel  check https://en.wikipedia.org/wiki/Active_pixel_sensor#Pixel both technologies do photon to electron conversion and accumulate charge but with different implementation. For sensor comparison: https://www.ptgrey.com/camera-sensor-review

 

 

15 minutes ago, Ben321 said:

 

 

[Ben321]

10 minutes ago, riklaunim said:

Consumer CCDs are dead. The high end CCDs still exist and will do for some time. If you really want to buy a CCD then there are still Sony and Kodak CCDs in DS cameras and will perform well. If you think that those CCDs are more sensitive than CMOS - then no, they are not / not always. Current Sony CMOS sensors have much lower read noise than their CCDs, and the QE is also as high or higher than CCD. Few years ago CMOS were way way behind, but not now. You can check astrobin, cloudynights and other sites for DS images taken with ASI178/174 or latest ASI1600. And if you want to see how a "professional" level  CMOS would look like then check this: http://qhyccd.com/QHY42.html .

Very low read noise also allows extensive lucky imaging at 1-few second exposures, for example http://www.astrokraai.nl/viewimages.php?t=y&category=7

Lux isn't comparable unit of sensitivity as it's measured with some sensor (variable pixel size), with some lens, and some light source. It also doesn't tell anything about noise. For pixel  check https://en.wikipedia.org/wiki/Active_pixel_sensor#Pixel both technologies do photon to electron conversion and accumulate charge but with different implementation. For sensor comparison: https://www.ptgrey.com/camera-sensor-review

 

 

 

Doesn't Sony still make the EX-VIEW HAD CCD chip? I think it's used in ultra-sensitive outdoors (not indoors) security cameras (like those near the front doors or parking lots at banks, military bases, and any place that requires top-notch security at night, using only starlight for illumination). And from my understanding, lux rating is all about signal/noise ratio.

When a camera is rated at 1 lux, it means that "with a typical lens, this camera can see light down to 1 lux before the signal is buried in the noise".

When a camera is rated at 0.0001 lux, it means "with a typical lens, this camera can see light down to 0.0001 lux, before the signal gets buried in the noise".

[riklaunim]

EXView HAD were top of the line and still are very solid DS cameras.  The difference between those and their current CMOS sensors is that ExViews have like 6-8e read noise where as their CMOS have 1-3e read noise (both for machine vision cameras. Astro cameras have bit less). IMX224 in ASI224 was even measured to have <1e read noise.

[Ben321]

39 minutes ago, riklaunim said:

EXView HAD were top of the line and still are very solid DS cameras.  The difference between those and their current CMOS sensors is that ExViews have like 6-8e read noise where as their CMOS have 1-3e read noise (both for machine vision cameras. Astro cameras have bit less). IMX224 in ASI224 was even measured to have <1e read noise.

What do you mean by WERE top of the line? Aren't they still? Do you mean that Sony no longer manufacturers EXView HAD CCD Chips?

[riklaunim]

Their CMOS sensors have less noise and are surpassing their CCDs in sensitivity. Plus they are faster and somewhat cheaper. In "low light level imaging" like night surveillance read noise is an important factor, so limiting it makes it possible to catch fainter signals (check the Point Grey charts and "absolute sensitivity threshold").

[Ben321]

9 hours ago, riklaunim said:

Their CMOS sensors have less noise and are surpassing their CCDs in sensitivity. Plus they are faster and somewhat cheaper. In "low light level imaging" like night surveillance read noise is an important factor, so limiting it makes it possible to catch fainter signals (check the Point Grey charts and "absolute sensitivity threshold").

Are you aware of any CMOS chips that can perform at least as well as the EXView HAD CCD chip? And I think Sony still makes the EXView HAD CCD, doesn't it?

[pete_l]

55 minutes ago, Ben321 said:

Are you aware of any CMOS chips that can perform at least as well as the EXView HAD CCD chip?

I don't know how you'd quantify it, but the sensor in the Sony A7S must come pretty close.

[riklaunim]

2 hours ago, Ben321 said:

Are you aware of any CMOS chips that can perform at least as well as the EXView HAD CCD chip? And I think Sony still makes the EXView HAD CCD, doesn't it?

IMX265, IMX250, IMX252, IMX178, IMX224, IMX185, IMX290. In general every latest Sony CMOS sensors with low read noise will surpass them in "short" exposures. Longer exposures depend on camera hardware.

Last year there was a boom for Neptune and Uranus infrared imaging. Before it was very hard to catch any detail as those planets are very faint in IR (but with high contrast of cloud bands) and CCDs weren't able to provide clear image "easily" (there was very little such images). But in ~2015  ASI224 showed up with < 1e read noise (6-8 times less than previous CCDs). It has good IR sensitivity but QE is bit lower than for example ICX618/ICX693/ICX692 CCD. The difference was read noise. Weaker signal got through noise and it become much easier to catch very faint Uranus/Neptune signal with such camera. Other IMX sensors followed that achievement.

[Dkblow]

Guys  Im wondering if I could use a night security camera? 

[atticphotography]

I'm in the same boat, trying to find a modern replacement/equivalent to long-obsolete webcams that were all the rage 15 years ago.  This may be of interest as I've just found (and purchased) a low lux camera from a Chinese supplier via ebay.  

 

It has a 1/3" Sony Super HAD II CCD sensor and is rated, according to the manufacturer, at 0.00001 Lux.  The lens unscrews allowing a 12mm to 1.25" adapter to be fitted. Having spoken to the supplier, they have had previous enquiries and had subsequent positive feedback, from numerous people wanting this for use with guide scopes.  I've also ordered the adapter with an integral IR cut filter.  It might be a while before I get the parts as they'll be coming by surface mail from China, but I'll try to post up some findings when I can

[michael.h.f.wilkinson]

I would just like to add that when moving from imaging with a Philips SPC-900NC (a.k.a. Toucam-III) to a CMOS-based planetary camera (which can indeed do longer integration times, as exemplified by practically any DSLR, or indeed digital camera), the step up in performance was HUGE.

SPC900NC

ASI120MC