What's the easiest and most affordable way for obtaining UV reflectance images?

Johannes Zander wrote:

I think to correct for removing the filter stack in front of the sensor, you need to bring the sensor nearer to the lens.

from: http://www.ir-photo.net/ir_mod.html

From the formula given on the above mentioned website you would bring the sensor round about 1,3mm forward to the lens. No shimming of the lens needed.

I have a Laowa 10-18mm that takes a flat rear filter. So, I just tested.

With the filter in, focus set to 3 feet is the same as focus to infinity with no glass.

Thus, the rear of the lens does need to be farther away with the filter in. So, I guess I did have it backward? Ooops. That means removing the cover glass can cost you infinity focus... which is very bad. How come I haven't heard everybody complaining about that?

Well, at least the rest of my post above is not wrong... right?

Enginel wrote:

ProfHankD wrote:

Yup, quartz lenses are the "right answer." That's why old EPROM chips used to be so expensive: to get enough UV to erase 'em, they needed a little quartz window and a ceramic package to mount it in.

Has someone crazy enough used them as an image sensor?

MOS DRAM would be a better choice I think. But you have to etch the top away.

https://en.wikipedia.org/wiki/Cromemco_Cyclops

Enginel wrote:

ProfHankD wrote:

Yup, quartz lenses are the "right answer." That's why old EPROM chips used to be so expensive: to get enough UV to erase 'em, they needed a little quartz window and a ceramic package to mount it in.

Has someone crazy enough used them as an image sensor?

Not that I'm aware of... which I wish you hadn't said. 

It takes A LOT of UV to erase a bit, so exposure time would probably be days if not weeks of sunlight, and without a quartz lens to focus the light, I'd probably try a pinhole, extending exposure time into months. However, I will admit that I kind of want to try it now. You should be able to get a huge DR by sampling over time to watch when each bit changes... except I don't think lighting really comes with a huge DR when averaging over a month....

BTW, an early digital camera, the Connectix QuickCam , actually used a RAM chip that had be specially packaged with a glass window. I still have my parallel-port-interfaced Color QuickCam somewhere.... In fact, in 1996, that was just about the only live view camera ("webcam" wasn't yet a thing then) I could hook-up to my 6400x4800-pixel cluster supercomputer-driven video wall, which was a little sad, because it delivered 320x240 images with somewhat wonky color.  Then again, it was the gross inadequacy of that which got me started doing computational photography... so I guess it was a very good thing in the long run?

Johannes Zander wrote:

Enginel wrote:

ProfHankD wrote:

Yup, quartz lenses are the "right answer." That's why old EPROM chips used to be so expensive: to get enough UV to erase 'em, they needed a little quartz window and a ceramic package to mount it in.

Has someone crazy enough used them as an image sensor?

MOS DRAM would be a better choice I think. But you have to etch the top away.

https://en.wikipedia.org/wiki/Cromemco_Cyclops

That's interesting. I did a lot with NorthStar Horizons back in the day, when they were Chevys and Cromemcos were Cadillacs, but I don't recall ever seeing a Cyclops. However, my first home machine was a HeathKit H8, so it's possible that I was just a tad too late to see those.

My first digital imager was actually a CDS cell strapped to a HIPL0T DMP-4to make a scanner. I actually got fairly decent resolution out of my little kludge, monochrome only. I think I still have it somewhere too.

So, are there any easily-windowed DRAM parts still in production?

I did find a straightforward (if unpleasant) way to remove a plastic case , but I'm not sure it leaves the die functional, and I'd need to bond wires to it and repackage it. Here 's a more official way.