• Ultraviolet Photography
  •  

reflected UV light source question for ink differentiation

UV Lighting
15 replies to this topic

#1 LC2

    Member

  • Members
  • 13 posts

Posted 30 November 2018 - 19:08

I work as a forensic image analyst, and part of my duties includes researching and teaching forensic photographic techniques. A couple years ago, I was tasked with procuring a camera system capable of photographing both the UV and IR spectrums. I put together a good camera kit for our forensic unit: a Nikon D7100 converted to full spectrum, Jenoptik UV-VIS-IR 60mm APO macro lens, Baader U-Filter and a couple IR pass filters. However I've been having trouble with the UV spectrum (specifically, light sources and and ideal (forensic) subject matter for reflected UV). I am hoping you guys will be able to help me, as you all seem incredibly knowledgeable!

What's puzzling me is the different results I am getting with a LED light source (PowerTac NH375, rated at 365nm) versus a regular (UV filter not removed) speedlight (Nikon SB-910). My test subject was a piece of paper that I had written on with a variety of pen ink brands/types/colours. With the speedlight as my light source, some of the inks disappear. But with the LED torch, the inks did not change... I would expect the LED torch to produce better results than the speedlight (and when I say better, I just mean I would like to see visible changes amongst the different inks, like I did with the speedlight, but even more pronounced).

I plan to try out our Polilight 400 (forensic light source, 400 watt metal halide lamp - the UV output is 350nm, 80nm bandwidth), but I haven't had a chance yet...


Thank you in advance!
Lieah.

Edited by LC2, 30 November 2018 - 19:10.


#2 nfoto

    Former Fierce Bear of the North

  • Owner-Administrator
  • 2,349 posts
  • Location: Sørumsand, Norway

Posted 30 November 2018 - 23:19

As you describe the setup, using the LED will enhance UV, whilst the speedlight certainly delivers a huge amount of IR (besides visible light).

Reflective UV will pick up surface detail and IR can sometimes show non-visible inks or lettering. I used IR (830 nm filters) for an earlier documentation of Dead Sea scrolls.

#3 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,367 posts
  • Location: USA

Posted 01 December 2018 - 02:02

Hi Lieah -

That is a very nice reflected UV kit you have put together. And that Polilight 400 is amazing with all those different output peaks possible! (And my goodness, that thing is very expensive. Woo!)

What is not clear to me from your post is what kind of Speedlight UV filter you are referring to? On the typical photo website the phrase "UV filter" indicates a UV-blocking filter which cuts haze. Here a "UV filter" usually refers to a UV-pass filter used to capture reflected UV.

Usually speedlight flashes have UV-blocking or UV/IR-blocking filters. So if your Speedlight is blocking UV, then it is output-ing lots of Visible and perhaps some IR light. Thus I would not be surprised that the inks look much different under mixed Vis/IR than under UV illumination. But as mentioned, it's not clear to me which "UV filter" is in use. Perhaps you can clarify?

If you look under our search tag UV Lighting then you will find most of our posts about UV illumination. UV-Led torches are discussed a lot. We have found that many of them emit some visible violet light. So we filter them with a good UV-pass filter. We've also found that some UV-Led torches do not meet the stated 365nm (or other) peak because of poor quality LED chips. Nichia chipped torches are usually the best. But even within Nichia, the chip bin lots can vary. So check the type and quality of the LED chip in any torch you use for forensics work. UV-LED torches are somewhat narrowband. But I don't know what you need for forensics? Broadband UV or narrower band UV? The current favorite UV-Led torch is the Convoy. Here is just one post about it: http://www.ultraviol...-uv-flashlight/
The search tag will bring up others.

Most of us doing documentary work (such as UV botanical signatures) use modified Xenon tube flashes for a broader and stronger UV illumination than can be offered by a UV-Led. Such flashes have had any tube coatings removed and also have UV-pass filters in place to supress most of the Vis/IR output. A few folks use Xenon lamps like the Blak-Ray, but they are large and thus a bit difficult to filter. Our founding member, Bjørn Birna, uses an enormous Broncolor UV studio flash which is quite nice and probably has the best broadband UV output of anything I've seen so far. [Added: The output is peaked I think around 350?]

(Birna, please return and offer a few details about that Broncolor because I have totally forgotten what model it is and how strong its output is.)

Do you have any reflected UV standards that can be used to confirm that your gear is working properly? Well, that is not quite what I mean..... We know from your gear list that it is all going to work properly!!! But when setting up any documentary kind of photo shoot, one usually makes an initial photograph of some known UV-reflecting and UV-absorbing subjects as "proof" that the gear and the illumination is all OK at the time of the shoot. I've been using Spectralon standards such as those made by Labsphere. Targets can also be used. And being a botanical photographer, I always try to find a sunflower to shoot when testing gear under UV illumination. Sunflowers have a known UV-dark central bullseye and UV-bright petal tips. And you can usually find them even in winter at the florist or grocery stores.

[UVP has no affiliation with any vendor. Any vendor links are provided only for your information.]
Andrea G. Blum
Often found hanging out with flowers & bees.

#4 LC2

    Member

  • Members
  • 13 posts

Posted 03 December 2018 - 15:33

View Postnfoto, on 30 November 2018 - 23:19, said:

As you describe the setup, using the LED will enhance UV, whilst the speedlight certainly delivers a huge amount of IR (besides visible light).

Reflective UV will pick up surface detail and IR can sometimes show non-visible inks or lettering. I used IR (830 nm filters) for an earlier documentation of Dead Sea scrolls.
Thank you Birna, so maybe the changes in the inks I saw was due to IR light from the speedlight, leaking in through the Baader U filter on my lens?

View PostAndrea B., on 01 December 2018 - 02:02, said:

Hi Lieah -

That is a very nice reflected UV kit you have put together. And that Polilight 400 is amazing with all those different output peaks possible! (And my goodness, that thing is very expensive. Woo!)

What is not clear to me from your post is what kind of Speedlight UV filter you are referring to? On the typical photo website the phrase "UV filter" indicates a UV-blocking filter which cuts haze. Here a "UV filter" usually refers to a UV-pass filter used to capture reflected UV.

Usually speedlight flashes have UV-blocking or UV/IR-blocking filters. So if your Speedlight is blocking UV, then it is output-ing lots of Visible and perhaps some IR light. Thus I would not be surprised that the inks look much different under mixed Vis/IR than under UV illumination. But as mentioned, it's not clear to me which "UV filter" is in use. Perhaps you can clarify?

If you look under our search tag UV Lighting then you will find most of our posts about UV illumination. UV-Led torches are discussed a lot. We have found that many of them emit some visible violet light. So we filter them with a good UV-pass filter. We've also found that some UV-Led torches do not meet the stated 365nm (or other) peak because of poor quality LED chips. Nichia chipped torches are usually the best. But even within Nichia, the chip bin lots can vary. So check the type and quality of the LED chip in any torch you use for forensics work. UV-LED torches are somewhat narrowband. But I don't know what you need for forensics? Broadband UV or narrower band UV? The current favorite UV-Led torch is the Convoy. Here is just one post about it: http://www.ultraviol...-uv-flashlight/
The search tag will bring up others.

Most of us doing documentary work (such as UV botanical signatures) use modified Xenon tube flashes for a broader and stronger UV illumination than can be offered by a UV-Led. Such flashes have had any tube coatings removed and also have UV-pass filters in place to supress most of the Vis/IR output. A few folks use Xenon lamps like the Blak-Ray, but they are large and thus a bit difficult to filter. Our founding member, Bjørn Birna, uses an enormous Broncolor UV studio flash which is quite nice and probably has the best broadband UV output of anything I've seen so far. [Added: The output is peaked I think around 350?]

(Birna, please return and offer a few details about that Broncolor because I have totally forgotten what model it is and how strong its output is.)

Do you have any reflected UV standards that can be used to confirm that your gear is working properly? Well, that is not quite what I mean..... We know from your gear list that it is all going to work properly!!! But when setting up any documentary kind of photo shoot, one usually makes an initial photograph of some known UV-reflecting and UV-absorbing subjects as "proof" that the gear and the illumination is all OK at the time of the shoot. I've been using Spectralon standards such as those made by Labsphere. Targets can also be used. And being a botanical photographer, I always try to find a sunflower to shoot when testing gear under UV illumination. Sunflowers have a known UV-dark central bullseye and UV-bright petal tips. And you can usually find them even in winter at the florist or grocery stores.

[UVP has no affiliation with any vendor. Any vendor links are provided only for your information.]

Thank you so much Andrea for your response, it has really helped. And I agree, the Polilight is pretty great! We’re very lucky to have a healthy government budget to work with :)

Sorry for the confusion about the speedlight… what I mean is that the speedlight has not been modified (so I’m assuming the UV blocking filter that comes with it would still be on there). So I figured the speedlight must be outputting a lot of visible and IR light, but since I’m using a Baader U filter on the lens, I figured the IR and visible light was being blocked, and so what I was seeing was a very little UV output that must be escaping from the speedlight… but I guess it could also be that the Baader U is leaking IR/visible light? Or I’m just not understanding the process?

I did have a search through the UV lighting section, and I did see some very good UV LED torch discussions, but I was still unsure of why I was getting the results I got, so I decided to ask my specific question. But I will continue to read through the discussions - there’s definitely a lot of good info in there! And I did read that many LED torches emit violet light, but I figured I was ok on that front, because I’m using the Baader U on my lens. But maybe I have that wrong?

And I too considered the possibility that the LED torch I’m using is perhaps not actually 365mn, so I wrote to the manufacturer, and they said that they “produce their forensic lights specific to the nm”, and that this is achieved by the LED supplier. They went on to say they have “very little spectral bleed due the the high tolerances they meet.” So their concerns were that the LED could be malfunctioning, or that there might have been a mishap when the dealer provided the light, thinking it was a 365nm but actually wasn’t. So they suggested I contact the Canadian distributor, so they can work with them to resolve any issues. So before I got into that process, I decided to reach out to you knowledgeable folk to get your thoughts first… but I still plan to start this process if I can’t figure out what’s going on. But it’s sounding like the targets and standards you gave me links for will tell me whether the LED torch is working as it should… I will be getting some of those ASAP, thank you!

And I will definitely check out the Convoy, thank you for the link. And I really like the sounds of those light sources… Birna’s studio flash sounds awesome!

So I think I will start with getting standards / targets to test the LED torch, as well as try out some stronger light sources, such as modified Xenon tube flashes or lamps.

Thank you so much, this has been so helpful and informative.

#5 Andy Perrin

    Member

  • Members(+)
  • 2,449 posts
  • Location: United States

Posted 03 December 2018 - 15:47

Yeah, using a source with a lot of IR and almost no UV due to the blocker probably means you were getting IR bleeding through the Baader. I have seen this problem with halogens sometimes (I don’t own a flash).

Edited by Andy Perrin, 03 December 2018 - 15:47.


#6 LC2

    Member

  • Members
  • 13 posts

Posted 03 December 2018 - 20:33

View PostAndy Perrin, on 03 December 2018 - 15:47, said:

Yeah, using a source with a lot of IR and almost no UV due to the blocker probably means you were getting IR bleeding through the Baader. I have seen this problem with halogens sometimes (I don’t own a flash).
Ahhh ok. Thank you Andy. I think I need to start all over again here... But I'm learning lots, that's for sure!

#7 nfoto

    Former Fierce Bear of the North

  • Owner-Administrator
  • 2,349 posts
  • Location: Sørumsand, Norway

Posted 03 December 2018 - 21:52

I'm using Broncolor 1600 W/s studio flashes with uncoated Xenon tube and quartz protective glass. They are really powerful and emit a lot of UV. Despite a very healthy IR output concomitantly with the UV and visible light, I have no issues simply using a Baader U (latest generation) filter for my UV photography. The usual lenses would be the UV-Nikkor 105/4.5 or the Coastal (Jenoptik) 60/4. I never had any issue with IR leakage through my Baader filters.

I often have to turn down the output of these flashes as one easily gets say an indicated exposure of f/22 at ISO 100, which is troublesome on its own.The actual reflector type of the flash head might change the output to a significant degree. One needs some experimenting to get the setup optimised.

Do note there is an early generation of the Baader U that does leak IR. However, I don't think that problematic filter has been in sale for 10 years or so.

#8 Andy Perrin

    Member

  • Members(+)
  • 2,449 posts
  • Location: United States

Posted 03 December 2018 - 22:12

nfoto, the problem is not with the filter, it is with using a light source from which all uv has been removed. If you illuminate only with vis+IR, even the Baader will leak. LC2's speedlight has not been converted.

Edited by Andy Perrin, 03 December 2018 - 22:17.


#9 JMC

    Member

  • Members(+)
  • 786 posts
  • Location: London, UK

Posted 03 December 2018 - 22:50

As has been mentioned LC2, an unmodified flash will have no (or essentially no) UV. While not Nikon sb910 data, this is the link to some of my work on flash output from Canon flashes - standard, and UV modified;

http://www.ultraviol...dpost__p__16912

The standard flash does not emit UV, so even with a Baader u filter there is no UV there to image. The good thing is, a converted flash will be relatively cheap, especially compared to some of your other kit.

#10 nfoto

    Former Fierce Bear of the North

  • Owner-Administrator
  • 2,349 posts
  • Location: Sørumsand, Norway

Posted 04 December 2018 - 09:08

Any flash will emit a lot of IR -- problem is to have some decent UV output as well. An unmodified flash will not do, as demonstrated many times on UVP. The problem we face when sufficient UV is delivered is whether our filtration manages to keep unwanted IR out. The Baader U does this, even when using high-powered modified studio flashes (that deliver massive UV by the way, never use them without protective eyewear).

#11 LC2

    Member

  • Members
  • 13 posts

Posted 04 December 2018 - 14:28

Thanks so much everyone, i think I have a good idea now of where I need to start, and you have also explained why I got the results that I did. Thanks so much!

#12 Shane

    Member

  • Members
  • 373 posts

Posted 07 December 2018 - 17:12

The appearance and effect of IR leakage in an image captured with a Baader U filter is going to depend on several factors, the strength and ratio of UV to IR in the source, the UV and IR reflectivity of the subject, the wavelength of the reflected UV and the UV transmission properties of the lens. Experiments from the old Nikongear forum showed that a dandelion photographed with an El-Nikkor and a Baader U, under typical noonish sunny day conditions, required a 6 stop boost to reveal the presence of IR leakage. This would have little effect on the UV contrast for that particular subject under those conditions. In the case of unknown subject UVIR reflectivity, it might pay to experiment to determine if IR leakage is a problem.

Edited by Shane, 07 December 2018 - 17:13.


#13 Andy Perrin

    Member

  • Members(+)
  • 2,449 posts
  • Location: United States

Posted 07 December 2018 - 18:08

Shane, she is using an unmodified speedlight with essentially zero UV. That was the main issue.

Edited by Andy Perrin, 07 December 2018 - 21:09.


#14 Shane

    Member

  • Members
  • 373 posts

Posted 07 December 2018 - 20:41

I got that..... I was just laying out some things to consider for this and future troubleshooting for her.

Edited by Shane, 10 December 2018 - 16:42.


#15 LC2

    Member

  • Members
  • 13 posts

Posted 07 December 2018 - 20:58

Thanks Andy and Shane... now that I am using the 365nm LED torch (until I can get my hands on a converted flash), I'm getting much better results. I'm going to try to post some photos of a scar... I got some interesting results!
(And, I'm a 'she'... I should probably update my profile ;) )

#16 Andy Perrin

    Member

  • Members(+)
  • 2,449 posts
  • Location: United States

Posted 07 December 2018 - 21:10

Whoops, corrected above! Sorry for inadvertently misgendering.