Could a new type of Image Sensor be made available to Cameras soon. Pt2

[SniffTheGlove]

http://www.gizmag.com/graphene-imaging-sensor/27718/

Scientists at Nanyang Technological University (Singapore) have developed a new image sensor claimed to be 1000 better than normal ccd/cmos sensors at the moment. It is based on Graphene and it also uses less energy.

[rowan46]

that looks exciting

[michael.h.f.wilkinson]

Modern CCDs hit 50% to 85% quantum efficiency, which is the percentage of photons hitting the CCD that are actually detected. You cannot improve upon that by a factor of 1000. What could be done is pushing noise down. A factor of 1000 would be really impressive, but given the low noise levels of many current devices, I doubt that will have the impact claimed.

[rowan46]

if it can push the noise levels on a dslr down by a factor of 1000 it may be a whole new ball game for affordable astro photography

[long_arms]

Incredible stuff!

Likelihood of any camera manufacturer producing a camera with one of these sensors in the coming years? My guess is pretty low

What do you think the manufacturers think of the tech? Is it good from their perspective? Genuine questions, I'm interested to know.

Dan

[michael.h.f.wilkinson]

if it can push the noise levels on a dslr down by a factor of 1000 it may be a whole new ball game for affordable astro photography

I do not think the number of electrons of "dark current" (outdated term, strictly speaking) per pixel in a CMOS sensor is in the order of thousands (per second), but closer to 100 per second or below (I have seen figures of 250/s for bad pixels). If that could be reduced to 0.1 per second that would be interesting (and then there is readout noise as well). Ultimately we could approach the photon noise limit. When claiming 1000 fold increase in sensitivity, it would be best to explain what you mean by sensitivity. If my exposure time is in the order of tens of milliseconds (as in planetary work) the dark current per frame is very low as it is, and I would worry most about both readout noise and quantum efficiency. With very long exposures, QE and dark current start to dominate (assuming low LP, of course)

[Carbon Brush]

Sorry to pour clouds and rain over imager's expections.

I'm very sceptical of anything much getting into commercial production anytime this decade.

I keep seeing reports that Prof This at the clever university of That has found a new application for graphene.

Replacing transistors with super duper performance ones. Making super eficient solar panels as thin as a wire.

Revolutions in computing, etc, etc.

One day I'm sure we will see graphene in commercial device. But I won't hold off buying anything to use in the next few years.

By way of example. The transistor was invented in 1948. It took until the 60's for any serious inroads into consumer kit. Even then it tended to be low power and low frequency devices.

Electronic computers have a similar age and it took until the late 70s to see something that would play hangman on your desk - without the desk collapsing under the weitght of electronics.

Is my opinion valid? I drive a car that is old enough to vote. I still have a video 8 camcorder. What do I know about technology?

[JamesF]

Modern CCDs hit 50% to 85% quantum efficiency, which is the percentage of photons hitting the CCD that are actually detected. You cannot improve upon that by a factor of 1000.

Probably depends on how creative you get with the numbers. If you said the average QE was 50% then that means 50% of the photons are "lost". If this new sensor has a QE of 99.95% then that would mean 0.05% of the photons are lost. I'm sure any salesman or marketing type would tell you with a straight face that this represents a 1000-fold improvement because 50 / 0.05 = 1000.

James

[SniffTheGlove]

I just hope it makes it into manufactures lines quickly enough to help us.

[ronin]

As it is still at reaserch stage and although people are talking abiout graphene there is not exactly a lot of graphene products around. So I think the use of "soon" is somewhat optimistic.

Even when ccd's were available it took 20 years for them to become common/available in the public domain, and these graphene cameras are not even being produced yet.

[dragonlord]

Ok, apparently the 1000x increase is compared to previous graphene CMOS/ccd sensors, not compared to current tech.