Ultraviolet photography at night

[kogakunippon]

Yeah, but you have to take the WB under sunshine or other wideband source. You're taking WB using the available light in the image, so it's not the same. But there are definitely reds in that picture, I agree.

 

I tryed it and waited long time......but Sunshine during the night is even rarer then red in UV pictures. :)

[Andy Perrin]

Haha, I mean you take a photo during the day, white balance off Teflon, then use a program (e.g. Photo Ninja or something similar) to transfer the WB to the night photo. In some cameras it is possible to set the WB in the camera during the day, and then just keep using it at night. That also works. But as I recently learned on another thread, not all cameras can do the UV WB in-camera.

[Cadmium]

If you want to find out where the red is coming from, test it.

Do a comparison test (night if you want) between the UV-only filter, and the UV-only filter stacked with various longpass filters to see if you can block out various colors from the shot.

The Baader U will start to leak IR somewhere in the 800-1000nm range, but only with long exposures that have UV blocked out by longpass filters.

The Baader will leak 'white' (monochrome) no color in that range of IR.

If you block the Baader U (or other UV-only filter) at a low enough range, say in the 400nm or 500nm range, then you will know if it is leaking red or white light with long exposures.

[kogakunippon]

Thats a very good idea, I will test this.

[kogakunippon]

You are right Steve, the Baader U-Filter leak in IR when you need very long exposure times. To find this out after coming back from my trip makes me very upset, because somebody from this forum was telling me before I started that its NOT necessary to stack filters if I am using my Baader U-Filter! This information was completly wrong. What a shame.... I did the wrong thing, because I was told here to stop using the Schott S-8612 filter. So much work and so much time wasted for nothing..... :(

 

Modified Nikon D7100

EL-Nikkor 105mm

Baader U-Filter

608 sec

F/5.6

ISO 800

 

http://gallery.photo.net/photo/18263482-lg.jpg

 

plus Schott S-8612 Filter 2mm thick:

 

http://gallery.photo.net/photo/18263481-lg.jpg

 

plus another KG-03 filter 3mm thick:

 

http://gallery.photo.net/photo/18263483-lg.jpg

[Andy Perrin]

The thing is, this is a long exposure, so SOME light leaks through no matter what, if you wait long enough. What matters is how much infrared there is relative to the UV. If the UV is drowning out the IR, as it WILL do during the day, then just the Baader by itself is fine. It's only under these night conditions that it becomes harder to tell what you are really seeing.

 

This has been a very instructive thread, particularly in terms of understanding the limits of the Baader. Lots of nice pictures too!

[nfoto]

Wolfgang: In my opinion, you are getting the wrong end of the stick as it were.

 

All UV bandpass filters "leak" IR. The question is more what are the ratio UV:IR in the light source(s) and what kind of attenuation of UV is provide by the lens (and camera sensor). If the incident UV is extremely low compared to IR, well that in effect means you are exposing through the side lobe comprising mainly IR to get any exposed capture at all.

 

To illustrate how efficient the Baader U really is, I routinely work in the studio with 2 or more studio flash units (800 Ws each, uncoated Xenon tube) and there is virtually no IR contamination present. Only if I take a shot straight into the flash head when it flashes, I can get some IR recorded.

[Andrea B.]

.......the Baader U-Filter leak in IR when you need very long exposure times. To find this out after coming back from my trip makes me very upset, because somebody from this forum was telling me before I started that its NOT necessary to stack filters if I am using my Baader U-Filter! This information was completly wrong. What a shame.... I did the wrong thing, because I was told here to stop using the Schott S-8612 filter. So much work and so much time wasted for nothing.

 

Wolfgang, slow down!!

Your night UV photos are valid because you were using UV-LED illumination and there was no additional source of IR - or only a very low incidental source of IR. None of us stack the BaaderU with S8612 or other BG type glass.

 

There were some night shots posted elsewhere on UVP for which we speculated that there might be some IR captured because they were photographs containing car headlights & taillights and bridge lights. It was simply a speculation.

 

As for the "red" in your photographs, I'll remind you and everyone else that when you convert UV files in ACR (CS/Photoshop), then you are not going to attain the typical UV white balance in your false colour palette that you see in most shots here. ACR simply cannot achieve that good kind of "click-white".

 

Also a reminder that if you click-white in UV photos made under narrowband UV-LED illumination, then you will get a very plain, low colour result in many cases. Sometimes an almost black & white look results. I'm not quite sure how click-white works for multiple narrowband UV-LED torches because I have never tried that. I would urge you to simply keep the attractive colours you have shown here in your night series and not stress out over this white balance thing.

 

Your work is good as it stands. Let it be. :D

 

If you want to see what one of your files looks like when white-balanced in an editor which can perform a good click-white, then you could upload a raw file to Dropbox and send me or Bjørn the link so that we can download the file and run it through a converter like NX2 or Photo Ninja, both of which can handle white balance of UV files.

[Cadmium]

Hi Wolfgang. You don't need to stack the Baader U with any filter. This was absolutely correct information.

The Baader U is a UV pass filter, and it is very efficient at blocking 'out of band' transmission, anything above 400nm, however if you 'push' that filter to the extreme, then you can 'squeeze' IR through it which will show up in the 800-1000nm range, as I said above, which is monochrome, below is an example of that, you will see no color (once white balanced).

Now, what I was meaning to say about stacking longpass filters is a test to isolate 'out of band' ranges of the filter, to see if for example at 200+ seconds there is some visual range color getting through.

Now with the Baader U, this would be extremely unlikely, and I have used and tested it so many ways that I seriously doubt any of the color you are seeing is from the visual or IR range.

Andy's filter may be a different situation, especially given that strong red, but I have not tested that filter.

 

Now, using S8612 stacked with the Baader U is of no use for the test I was suggesting, it doesn't cut out the UV and it doesn't cut out the visual, all it does is cuts out the IR, which is the opposite of the test I have in mind.

To test for visual range color leak, we need to suppress the UV passing through the UV-only filter (UV bandpass, as Bjorn calls it).

So to suppress the UV we are then forcing the UV-only filter (Baader U in this case) into an extreme situation to look for any possible leak, in this case a color leak.

If you have any questions about that, please let me know.

As I have said before, I think your red is a product of white balance yellow light from the UV range.

I would not expect you would find any red leak when doing a longpass filter test as I suggest with the Baader U.

The test might be something Andy could try.

 

Anyway, there is nothing wrong with your filter, it needs nothing stacked with it, unless you want to test some extreme situation.

 

OK, so here is a long exposure example of a Baader U stacked with a long pass filter, Schott RG610 which only passes red and IR above about 610nm, which removes the UV (and blue/green visual range).

Such a daytime shot might be a second or two long (depending on camera settings), but here because the UV is suppressed, it requires a much longer exposure time to see the 610nm+ light transmission,

which is actually 800nm+ IR. If there were any 600nm range light transmitted here we would see some red/brown, but there isn't any of that stuff, just high end monochromatic IR.

 

[Cadmium]

Wolfgang, partly what you are doing when you stack the Baader U with the S8612 is cutting off some of the Baader U UV amplitude and bandwidth. So that is probably why the second and third pics are slightly darker,

but also notice that the colors have not changed.

You don't need to stack your Baader U, other than a slight difference in exposure, I don't see much difference in those shots.

Your time was not at all wasted.

[Andy Perrin]

Andy's filter may be a different situation, especially given that strong red, but I have not tested that filter.

...

As I have said before, I think your red is a product of white balance yellow light from the UV range.

I would not expect you would find any red leak when doing a longpass filter test as I suggest with the Baader U.

The test might be something Andy could try.

I haven't got any red in mine, Cadmium. If you are referring to the wind turbine shot, that was full spectrum (no filter at all, just the lens). My speculations about infrared in mine were referring to the car headlights I saw, and the red OlDoinyo saw in HIS pictures with his Baader, which HE speculated might be IR at the time, in that old thread.

 

As for the long pass filter tests, I haven't got any long pass filters whose spectrum I'm confident in to stack with. I have the NEEWER filters whose actual spectrum I don't know, and I have some tiny Omega IR bandpass filters, but not long pass. I just didn't see the point in doing a test with the NEEWERs since I'm not sure where they even start to cut off, or worse, whether they might leak UV (which would make a test invalid). Do IR filters ever leak UV?

--

 

By the way, there is certainly infrared around at night. Here is a shot with the NEEWER720 filter, which I am assuming is showing IR here given the light-colored trees.

NEEWER720 Novoflex Noflexar F16 167" ISO100 (sharpened)

[Cadmium]

Oops, Sorry Andy, I missed the fact that your windmill shot was full spectrum.