Hellow from Poland (Minsk Mazowiecki)

[Doug A]

Fine shots. The new filter is a keeper. I also enjoy seeing UV cityscapes. Especially of places I've never visited.

 

I'm glad you are safe. Was worried about you with the turmoil nearby.

 

Thanks for sharing,

Doug A

[lukaszgryglicki]

UV ~ 340 nm: Nikon D600 full spectrum, mono, UV-Nikkor (f=4.5, 5.6 or 8), max ISO 1600, min shutter speed 1/15s, Hoya U-340 + BG39 2.3mm.

 

All hand-held - and because of 1/15 shutter I was usually doing a burst of shots and 1/5 was acceptable.

 

Note that this is a sunny day, almost no clouds - and looks very haze, also take a look at some cars paint - it looks at least strange. Head lights and windows are usually black, paint shows texture.

 

BTW: anybody see any IR leaks? I'm hoping to have clean <= 350 nm images...

[colinbm]

Looking good.
Some cars paintwork show differences in UV if they have had repairs & re-sprays, as can be seen in some of these pictures.

[lukaszgryglicki]

So it can be used to detect repainting - interesting...

 

[ulf]

IR-leaks are normally most visible in objects with strong IR reflectance that are UV-dark. That is typical for some flowers with UV-signatures.

Sensitivity for IR leakage in filters increase by more UV-marginal lenses and camera sensors. I do not think you have any problem here.

[Andy Perrin]

The plants would show any IR leak, but they are very dark here. The grass is darker than the dirt even. It seems quite unlikely there is a leak.

[lukaszgryglicki]

OK then, so maybe I don't need S8612 in 52mm and BG39 2.3mm does the job... Hoya U-340 has quite a significant IR leak, that seems to be surpassed (but Houa U-340 is 4mm thick).

 

[lukaszgryglicki]

Various trichromes:

Nikon D600 mono + UV-Nikkor, f=8, ISO=100, exposure (as long as needed - UV needed around 4s, sunny day).

- For UV: Hoya U-340 52x4 mm + BG39 52x2.3 mm = ~ <= 350 nm.

- For Vis: Houa IR/UV cut 52 mm = ~ 400-700 nm.

- For IR: IR850 52 mm = ~ >= 850 nm.

Assigned to different channels on different photos.

I've found it impossible to align them perfectly, because even if I do this manually on some very distinguishible pixels - the farther I'm from that point - the bigger differences are - it seems like "geometry" is slightly different at different focal points and wavelengths...

 

[colinbm]

Amazing.
That car at the bottom middle is showing some good paint mis-matching, even though not in visible light.

[lukaszgryglicki]

Yes, I can detect those pains changes in UV, next time I go for walking, I'll be specially looking for this.

U-340 + BG39 are good in detecting those flaws.

 

[Doug A]

Impressive color images. My preference is number two. Interesting that geometry changes with wave length.

The UV Nikkor is looking mighty fine. 

 

Thanks for sharing,

Doug A

 

[lukaszgryglicki]

Yep, just wanted to re-do this with wide angle - just hoping that when I bump time to 30s or so then I'll be able to do the same (tried with Nikkor 20/3.5 Ai).

But ... no!

Not even possible to focus using live-view with U340 4mm and BG39 2.3mm = almost completely black LV wide open and if I focus in visible there is a focus shift (there is none in UV-Nikkor btw).

So it *may* be possible with 20/3.5 but focus will be by guess-trial-error and exposure 30s might be not enough (Nikon doesn allow any more than 30s, Fuji allows...) - UV-Nikkor needs around 4s at ISO=100 and f=8. Seems like this filter stack is quite deep in UV.

I'm speaking about ISO 100 - UV-Nikkor shots were at ISO 100.

 

[Andy Perrin]

I think the alignment issues are partly due to very small rotations not being accounted for, not geometry changes with wavelength. The difficulty you describe “the further I am from that point - the bigger the differences are” is what you expect from a rotation. 

[lukaszgryglicki]

You are probably right, but camera was not on ANY tripod but was put on the balcony ground - but yeah this is perfectly possible.

[lukaszgryglicki]

Hmm - I think not - there was not noticable rotatation, I also can't imagine scale... I was struggling with my program (which uses brurte force) to account for rotation and best is fi=0 degees no matter what I try - I didn't try scale yet, but how "scaling" would be possible from the same place (+/2 say 1cm and focus is in infinity or maybe samy 100+ meters) - so why image shows different "distortion" over the frame? Any thoughts?

I was on the weeding and then slept - so I only tested extra rotation - is there any major error I'm doing here?

 

[lukaszgryglicki]

Now shot indoor - UV,Vis,IR -> R,G,B:

 

[Doug A]

Like the effect and colors. Surprised there was enough UV light indoors.

Thanks for sharing,

Doug A

[lukaszgryglicki]

Barely, UV is from Hoya U-340 4mm thick + BG39 2.3mm thick, it was 30s with ISO 400s if I remember correctly.

 

[lukaszgryglicki]

Hansnes, Norway, almost 70N (few kilometers short):

UV, Vis, IR -> RBG.

[lukaszgryglicki]

I often feel like I should work as a scientist (software developer is still quite close - researching & solving problems every day), whenever I have any free time I'm trying to play with that UV stuff, I also have thermal IR, many chemical elements (even radioactive one) - feel like this is some kind of addiction, not exectly doing anything useful, but just like spending time trying all this stuff with mixed results.

If I sometimes spam too much about my various/strange ideas please forgive me - I just like scientific discussions - basicaly not limited to any particular topic.

 

 

[lukaszgryglicki]

Got my Invisible 308nm filter. 

Today is too dark, when I try to do a photo with it using Nikon D600 mono - I'm getting blackness, even with 30s exposure and 1000 ISO.

When I get something I'll report - need a sunny day an dprobably longer exposure with maybe even 3200 ISO (but that would have a lot of noise) so right now waiting.

It is possible that my "mono" Nikon has quite UV-blocking glass - it is supposed to pass UV (and it easy does around 350nm maybe even less - U-340 makes really good photos) but it may start blocking in UV-B range (below 320 nm)...

 

[Stefano]

With my Canon EOS M (standard coverglass, color sensor) I can get noisy images at ~f/2.5, ISO 25600 and 8-15 seconds. I use two Chinese 310 nm bandpass filters rated at about 50% transmission, so the overall transmission should be about 25%. Under the same or equivalent conditions you should easily record images, even if they are noisy.

[Stefano]

Here Jonathan used f/8, ISO 800 and 4 seconds at 313 nm, to give you an idea.

[lukaszgryglicki]

I had very little time to test, will see over the weekend.

I used slow lens (UV-Nikkor at f/4.5) - live view shows blackness, I just did a few test shots, even without tripod.

On the weekend (sun permitting) I'll do it for 30s, f=5.6, ISO as low as possible to get anything.

If that is not working I'll do bulb/time exposures of several minutes.

 

[lukaszgryglicki]

LOOOOOL

So stupid mistake!!

 

Now I'm able to make a photo with it, using f=5.6, ISO=100 and t=5s (D600 mono + UV-Nikkor)

Guess what?

 

My WINDOWS are blocking 99% or even a lot more of UV-B. Yes - I'm stupid, I did test photo indorr through the window!!

Now when I open it and make a photo from balcony, all looks just fine.

Live View works when UV-Nikor is even at f=8 (but quite dark) and with f=5.6 and f=4.5 it is OK - able to magnfy in Live view and shoot!

 

LOL again.

 

When I use life view indoor to see through open balcony door I see everything but when I look at the window just next to it - it is pitch black!