Jump to content
UltravioletPhotography

does anyone have a dead cmos dslr sensor to donate to science?


Mark Jones

Recommended Posts

Dupont is trying to help me determine what is the best photo chemical to use to strip the bayer layer off a dslr sensor so it is 6x more senstive to UV light

does anyone have any dead dslr sensors they can mail to dupont?

DuPont/EKC Technology

Attn: Chiu Tse

2520 Barrington Ct.

Hayward, CA 94545

if NOT I will have to buy a camera off ebay and take out the sensor.

Link to comment

Ebay has many scratch and damaged cameras up for sale. But you will want to find one similar to the one you will convert, as they may test a copper one and you have an easier Aluminum one or vise versa.

 

I have thought hard about a monochrome converted camera. The maximum increase in sensitivity is only 2 stops. This has been tested by many people. So is that enough? Is it worth the cost? Those are questions you need to ask yourself. So is going from 1 minute to 15 seconds, worth it. Or is it 1 hour to 15 minutes.

You give up the fun false colors that are part of the joy in UV and even IR.

Also better dangerous lighting can get you far with much less cost. A 302nm bulb can be assembled for under $50.

Also to really take advantage of a monochrome converted camera, you should have an expensive quartz lens.

Using a monochrome converted camera in just UVA range is really not worth it.

 

Link to comment

David, the increase in sensitivity for the monochrome conversion was quoted by Prutchi in his book as being about 600%, although his source data for that was not discussed. With my work using a Baader U filter in my spectral sensitivity work, I've seen about an 800% improvement (when the effects of the light intensity as a function of wavelength was normalised, and using a Rayfact UV lens). I've finally got my paper on spectral sensitivity work accepted for publication, and I can share this on here once it's out, as there is a figure in there comparing a monochrome and standard multispectral camera with the Baader U filter.

 

For my sunscreen work, I'll take any improvements in UV sensitivity I can get, so three stops is is certainly worth it. It really depends on what you want it for. Do I like the false colours with the Bayer filter in place - absolutely. However when imaging skin and sunscreens black and white is ideal.

 

Granted, if you're not using quartz and calcium fluoride lenses, the benefit is reduced, but it is still significant. Is it worth the large amounts of money involved in having it done? That's a different question. If I weren't using mine for work, I doubt I'd have had the conversion done.

 

It'd be interesting at some point to compare the two cameras with a lens which isn't quartz and calcium fluoride using the Baader U. I'll add it to the list.

Link to comment

Jonathan,

I look forward to reading your paper.

Don't get me wrong, for me 30 seconds to 8 seconds may be worth it. As 8 seconds is the limit for High resolution shot mode on the Em5mk2.

Also a practical limit of 20 to 30 seconds is best for flowers, as they do move and wilt over the exposure time. Less than 20 seconds always leads to sharper shots, unless they are completely dry.

 

I just wanted Mark to put the gains vs cost into perspective and if there is a cheaper way to get to the gain, then that should be done first.

 

Remember:

Best brightest lights

Best transmission filters

Best transmitting lens

Highest sensitivity sensor.

 

I also now wonder about the Fuki Xt3 camera. Green pixels are what is detecting the lower UV and the X-trans array has more green than typical Bayer. Also its BSI and the newest Sony sensor, matching in resolution with the just announced A7R4 camera. Also Kolari took it apart and there are no issues with IR shutter monitor. So maybe something to investigate.

 

Link to comment

Jonathan,

Also do you have the Nikkor 80mm EL f5.6 lens?

I was surprised that it reproducibly had a faster shutter speed than my UAT when using a 370bp filter. It falls off hard after that, but for UVA imaging it really is good. My copy in my hands was always 2/3rds stop faster than my UAT at f8, at same focus distance and with same lights.

I remembered you tested the Rayfact vs the UAT and found the Rayfact faster. But I can't afford that lens. For others the EL80 may be a really good UVA alternative.

So you may not need a quartz lens if you want the best in UVA.

Link to comment

Hi David, the paper was a long time coming, and has recently been accepted. When I have the final version I'll be able to share.

 

Funnily enough I just bought a 80mm f5.6 El Nikkor on ebay - I didn't have one until now. I hope it's the right version. I also hope to do some tests with it before too long, although I am up to my eyeballs with work at the moment (which is good because it means I can buy lenses, but I don't have the time to test them).

 

I think defining 'the best' is a hard thing to do. I'd expect it to have a huge dependence on the wavelengths being imaged, so would depend on the lighting and filter. What I can try and do, is a comparison between the UAT and 80mm el-nikkor, at the same f stop, on the same camera, with the same filter, and same lighting. I could image a diffuse reflectance standard and see which one is brighter and by how much.

Link to comment

BTW, FWIW.....we don't recommend the EL-Nikkor 80/5.6 (metal version) as a reference lens because it is prone to hotspotting. So far we've seen this in moderate to long landscapes. Can't recall at the moment about near work, but I think I've seen some hotspotting there too. It is a very diffuse kind of hotspot. Not particularly objectionable as these things go. It can be remediated with a little Photoshop (or other app) work as long as the luminance and chrominance layers are separated. Andy Perrin had some photos showing the 80/5.6 problem. Link?

 

There are so many factors which go into making a good UV-capable or UV-dedicated lens. Slight differences in exposure times is not particularly important.

Link to comment

BTW, FWIW.....we don't recommend the EL-Nikkor 80/5.6 (metal version) as a reference lens because it is prone to hotspotting. So far we've seen this in moderate to long landscapes. Can't recall at the moment about near work, but I think I've seen some hotspotting there too. It is a very diffuse kind of hotspot. Not particularly objectionable as these things go. It can be remediated with a little Photoshop (or other app) work as long as the luminance and chrominance layers are separated. Andy Perrin had some photos showing the 80/5.6 problem. Link?

 

There are so many factors which go into making a good UV-capable or UV-dedicated lens. Slight differences in exposure times is not particularly important.

 

The last two years I have used two different EL-Nikkor 80/5.6 (metal version) for several thousand exposures, without any sign of hot-spotting.

Almost all of them was in close-up, the magnification range the lens is designed for.

It has become my favourite lens as the images are quite sharp and distortion-free.

 

Could the hot-spotting problem be in some way related to the camera / sensor configuration?

My UV/IR-cut filter replacement window from astronomik.de is wide-band AR-coated. That might affect possible hot spots.

 

The importance of different factors which go into making a good UV-capable or UV-dedicated lens varies a lot, depending of the type of photography or purpose it is intended for.

IMHO the EL-Nikkor 80/5.6 (metal version) could work well as a reference for the single purpose of determining and comparing the transmission of more or less UV-capable lenses around 370nm.

 

As long as such a lens is optically sound, without haze or mould the optical characteristics is likely very close between individual items.

Link to comment
Andy Perrin

I didn’t have a hotspot problem with that lens. I had an issue with the first copy I received of the EL-Nikkor 80 having haze issues. Second copy was good.

 

I had a separate issue last year with shooting objects that were in line with the sun, but that was not really the fault of the lens.

Link to comment

I bought a broken canon t6, ($45) now I just have to get off the top plate glass.

if you convert the camera your self it's not that expensive. Only the 20D is easy to convert with scotch tape removal of bayer though. seems all other options risk destryoying the sensor too easily, which then means doubling the cost.

 

the max max is so epensive, it woudl be just as cheap to get a mono sensor camera like the leica used and remove the IR cut off filter internally which has much less risk.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...