60 replies to this topic

[Bernard Foot]

Hi, Cadmium.

The U340 leaks a bit of far red. And I'm sure the UV/IR cut filter in the camera doesn't cut off perfectly at the border between vis and IR - meaning that the camera will see a bit of IR that it shouldn't. So the U340 lets through a bit of far red and very near IR light that would seen by the camera. The S8612 would hopefully cut this out. The pictures of the Lily below (my first UVIF shots ever!) use a U340 over the flash: the first shot has no filter over the lens, the second a UV/IR cut filter (unknown spec.) over the lens, and the third an S8612 over the lens. This does seem to indicate red leakage which is blocked by the S8612.

Do I mean ZWB1 or ZWB2? By your graphs, I mean ZWB1. But the graphs posted by the seller on ebay (see below) show ZWB1 and ZWB2 reversed comapred to your graph, and that's why I said ZWB2. I'll need to get to the bottom of this before I buy.

[UlfW]

Bernhard,

Please do not trust these Chinese spectrograms close to 0%
I suspect that they can have a severe offset error.
The thickness of the filter is not stated here making the information very unreliable for higher transmission-levels too.
What are we seeing?
The graph of ZWB2 at 375nm is unrealistically steep too. Are the graphs hand-sketched?

[Bernard Foot]

Yes, the graph that I posted looks odd at 375nm. Also it agrees approximately with Cadmium's graph on the right-hand side, but seems to have ZWB1 and ZWB2 transposed on the left-hand side.

If I use Cadmium's graph, the choice is between ZWB1 with far-red leak or ZWB2 with far-blue leak. I guess the far-red leak is easier to cure using an S8612.

BTW - the ad. where I got the graph says the ZWB2 is 1.9mm thick.

Edited by Bernard Foot, 18 November 2019 - 22:04.

[Bernard Foot]

Here's a question for the experienced UVIF photographers: how can I confirm that I am capturing a UVIF image and that I'm not getting visible light contamination?

[dabateman]

Bernard Foot, on 18 November 2019 - 22:08, said:

Here's a question for the experienced UVIF photographers: how can I confirm that I am capturing a UVIF image and that I'm not getting visible light contamination?

When I take UV induced visible fluorescence images. I have mainly been using my 365nm led bulbs with Zwb1 filters in front.
But I started with Canon 199A flashes with Zwb1 and bg39 filters mounted in the front.
You want no lights on in a controlled room. I use my basement furnace room. Then a light source that is just UV for inducing fluorescence. You can also use a 405nm light or use a lee filter over white LED to isolate 488nm. But you want your light source to be monochromatic.
Then use a 2A, 2E or Baader UVIR cut filter. A BW 486 filter leaks too much into the UV. My Sigma SD15 filters work great as they cut from 405nm to 680nm.
Then for UV induced visible fluorescence you will need to cut the IR from the camera.
For UV induced IR fluorescence I use a 720nm long pass filter on the camera lens.

Some will also use a U340 filtered Convoy 365nm led flashlight to paint the subject.
Hope that helps.

[Andrea B.]

Here's a question for the experienced UVIF photographers: how can I confirm that I am capturing a UVIF image and that I'm not getting visible light contamination?

1) Shoot in the dark. All ambient visible (or IR or UV) light must be blocked.
In my old house I used the coat closet as a dark room. After I crammed myself in there with the camera, mini-pod, filters and torches, I would then roll up a kitchen towel to block light leaking under the door. The cat, annoyed at being left out of the closet and wanting to play, enjoyed poking her paw under the door to try to snag the towel and drag it out.

2) The light source must be properly filtered to ensure that the fluorescence inducing light is as pure as possible. And the taking lens must be properly filtered to ensure recording only the desired wavelengths. (No, I do not fully trust my cameras to block all IR. Some simply do not fully block IR.)

That is the "purist" method I just described. If you follow that, then you know what you have captured.

But understand that beautiful, artistic fluorescence results are produced without always using both sets of filtration, and we all enjoy that and support it also. Some of the prettiest work is a combo of fluorescence and ambient light. And then there are those fluorescent mushrooms in the photos made outdoors at night where some moonlight might have snuck into the record.

When displaying or posting your work, please be sure to list your filtration or lack of filtration and whether or not you made the photo in the dark.

This can all be reviewed in the Fluorescence Sticky. There is also a discussion of standardized white balance for fluorescent work in there for those who might like to pursue that.

[Andrea B.]

SHORT VERSION:
Filter the lens to block everything except visible light.
Filter the UV-source to pass only UV.
Shoot in complete darkness.

Test for leaks using the Shiny Metal Test described in the Sticky.


EVEN SHORTER VERSION: Filter, filter, dark.

The principles above hold also, with appropriate filtration, for UV-induced Infrared Fluorescence, Cyan-induced Fluorescence or any other induced Fluorescence. Filter, filter, dark.

[Andrea B.]

When working with a new flower, I sometimes ignore Filter/Filter/Dark and just test for fluorescence wherever I'm shooting in whatever ambient light is present. Here is a little test to determine whether there is any interesting fluor from the Crownbeard Daisy. And yes, there is. So I will put it on the list to make some fluor studies of these Daisies next time they are in bloom. (And after I've set up a darkroom somewhere. The closets in this new house have skylights.)

Exposure: f/16 for 1/5" @ ISO-800. There was quite a bit of ambient light to get 1/5" at f/16 even with ISO at 800.

---

I forgot what my point was. (It must be the altitude here!)
I think it was that you can photograph fluorescence in less than ideal conditions. So don't let the thoughts of ambient contamination or "impure" fluorescence or "leaky" filters stop you from trying to capture it. You can still get a beautiful photo. You can always pursue "purity" at a later time if the area continues to hold your interest.

---

[Cadmium]

Ha! The altitude!

Remember, ZWB2 is suppose to be like Hoya U-360 and Schott UG1, those all have some visual violet leak slightly above 400nm,
so those types are not good to use for UV only flash. Don't use those for ANY fluorescence inducing UV light source (use U-340 or UG11).
Personally I don't believe any of the Chinese graphs. That glass is not made or scanned by one source, and the data is often not inclusive, it looks manufactured to me,
and since there are so many versions of those graphs, I can't be sure of any of them being true.
Schott and Hoya glass is used in industry, military specification grade, which includes melt data per batch to define variances, and rejection limits.
Much better and more consistent data is provided.

The only thing worse than Chinese filter glass, in my opinion, are those filters that have no data at all, the X-name filters (so to speak),
such as the 'X-Y-Zee-700' (I made that fictitious name up, not a real filter as far as I know), which might be some 700nm longpass IR filter, made of unknown glass type or manufacturer...
Oh, and with no graph, no data... I never buy filters with no graph.

Only UV should get through to the target/subject. Any UV light 'reflected' off the subject should be blocked by a UV blocking filter, 420nm+ longpass type filter (GG, KV, etc.),
even really with a stock camera to be technically correct, because often stock camera internal UV/IR blocking filters don't cut off at or above 400nm,
they often cut off slightly below 400nm, which will let UV from the flash reflect off the target into the image.

In short, there are three things you want to block.
1) any transmission from the light source that is above 400nm (any Visual or IR).
2) any reflected UV below '400nm' (...420nm, 435nm depending on your preference or safety tolerance, keep in mind most such blocking filters have a sloped cutoff transmission).
3) any IR (which can be from UV induced IR which would then not be pure Visual Fluorescence, but more like visual + IR fluorescence,
depending on the actual cut off point of the internal stock blocking filter, or external visual only stack on a UV/IR converted camera).

Even shorter shorter version: 7,199 feet

Edited by Cadmium, 19 November 2019 - 04:20.

[dabateman]

Bernard,
Although you do have red leak in your first image. I do like the look of that one the best.
The other hard art in UVIVF is white balance. I will dial it in using the Kelvin values around 2700K to start. Maybe less or more to get what I visually see depending on what is fluorescencing.
I recently imaged a Kiwi berry, and the visual fluorescence was in the red. A S8612 on the camera lens cut out all the visible fluorescence. My SD15 filter was needed to isolate it. So you may need to play around.

[Andrea B.]

Remember that the camera colour space cannot capture all visible colours. I've had some green fluorescence that I was unable to properly record. It was such an intense green of a certain kind! But the photo just could not match it even when using the fluorescent grey standard to set the white balance.

Also a reminder that using a blue-green filter as an IR blocker can add its own colour cast to the outcome. It is better to use something like the Baader UVIR-cut which adds much less of a colour cast.

[Cadmium]

A BG filter (AKA blue green filter) does not look blue green to a full spectrum camera. BG filters do not add a "color cast" to a photo when used on a full spectrum camera,
BG filters remove "color cast", they return the camera to natural visual color.
BG38 or BG40 are the best ones to use for the truest visual color in most cases.
In fact, Baader UVIR-Cut filter looks too red, it has more of a "color cast" than do BG38 or BG40, on my Nikon cameras,
and most people with other brands of cameras say BG38 and BG40 look the most natural.

The problem with Baader UVIR-Cut is that it has a more red version of the visual range than does BG40, for example.
BG40 is more true to natural visible than Baader UVIR-Cut is, in my comparison tests.
Here is the blue to red visible scale:
More blue < S8612 - BG39 - BG38 - BG40 - Baader UVIR-Cut > more red
(S8612 is a BG type filter) (BG38 and BG40 look best to me).

Use what works best for you, but that is the tendency I have seen with that list of filters (blue to red).
In my tests, BG40 is maybe the closest out OOC natural visible filter for my UVIR cameras, compared to a stock camera.

With either the UVIR-Cut or the BG40 you will still need something stack with them the cut the visual at or above the 400/415/420/435nm point, because even the UVIR-Cut doesn't cut above 400nm,
it leaks some UV.

Some comparison tests I did:
https://www.ultravio...__fromsearch__1

In my opinion, the best combination is
U-340 2mm + S8612 2mm on the flash (or U-340 2mm alone on a Nichia 365nm LED torch),
BG40 2mm + KV418 on the lens for a UVIR camera.
You can replace the lens stack with a variety of other filters, some may work all alone as one filter only,
but that is the lens stack I use most of the time for UVIVF. You can use GG glass, like GG420, GG435, however if you get close to something you may get fluorescing GG, so be aware of that.
KV418 is not that easy to find and is expensive, and Ulf and Jonathan have tested some other candidates for that.
Ask them or search for that.
The same lens stack may be used on stock cameras.
If your flash is leaking the ~700nm~ Red/IR range, then you will have a mixture of true visual fluorescence + reflected visual red.
You may like that look, but don't be fooled about what it actually is, that red is not florescence, or at least not all of it, and you can't tell how much is reflected or fluorescence unless you remove the reflected red.

Edited by Cadmium, 19 November 2019 - 12:09.

[UlfW]

A possibly better alternative to filter a full spectrum camera and use it for fluorescence in visual can be to use a non-modified camera. Modern digital cameras often have quite good cancellation of UV and IR by their internal filters. Those filters are better optimised for the camera's sensor sensitivity than an external BG-filter.

[Bernard Foot]

Thanks for all the input. I should have realised this was not going to be as easy I was expecting ...

As usual, I want to get going without spending a lot of money until I know what areas I am interested in. So my first stage will be to see what I can do with what I have to hand. Second stage is perhaps to get some ZWB1 glass so I can have multiple flashguns on the go. Then maybe start investing in better filters ...

So for stage 1 I'm using a stock (vis only) camera, U340 over the flash, and S8612 + UVIR cut on the camera. To get multiple flashes for more illumnination, I'm using the camera's multiple exposure capability. So even if there is a small amount of background light, it's not going to be recorded with a shutter speed of 1/160.

I don't know, but I think I may be getting some visible light leakage. For example, I just tried a Poinsettia. Dust etc. fluoresced nicely, but I was surprised that the green leaves had a green colouration and the red leaves had a dark magenta tinge - see below. [White Balance was AWB.]This may be correct fluorescence, but it may be visible leakage.

What I'd like is a white visible target that does not fluoresce. Then if I photograph this, it should be black. If it has any colour, then I know I have leakage, and what colour it is.

Anyone know of a suitable, easily-accessible material I could use for this?

[dabateman]

Bernard,
A good reference is either smooth Aluminum foil or a big curved stainless steel spoon.
Then you know if you have leakage, as it will not fluorescence.
A crumpled ball of aluminum foil will reflect light due to the wavelengths. So you can add one to see that effect.
If you are cutting out all of the provided monochromatic light, than you shouldn't see anything in the spoon.

[Cadmium]

Craig, I probably missed something in all these two pages, but let me ask,
you are using a stock camera, Yongnuo yn 560 iii flash converted to full spectrum with the plastic Fresnel removed, with adapter/filter holder, and U-340 2mm + S8612 2mm filters.
Are you using any filter on the front of the camera lens for your shots you have posted?
Thanks!

[Craigk79]

Yes I am using a stock camera and I modded the flash head by removing the 2 plastic fresnel filters out then added the adapter with filters :) No filter on my lens in dark environment.

[Bernard Foot]

dabateman, on 19 November 2019 - 23:25, said:

A good reference is either smooth Aluminum foil or a big curved stainless steel spoon.

Thanks for that.

I had a go, and really don't know what to conclude.

The first 2 shots show a large steel spoon, a PTFE tile, and a wooden cooking spatula lying on a sheet of aluminium foil. In the first, the flashgun was right next to the lens, in the second off to the side (and around 1.5 times as far away). Exposure was ISO 4,000, f/1.8. Standard visible-only camera with UVIR Cut and S8612 filters. Flashgun had a U340 filter.

The first shot looks like there is a lot of visible leakage. In the second shot the spoon is pretty black, implying little leakage.

The 3rd shot is a photograph of the flash tube itself. Exposure was ISO 100, f/8. There's very little visible leakage, bearing in mind the extreme brightness of the tube. (You can see that there is a crack in the filter, but I doubt that this was leaking enough visible light to account for the dramatic difference in the first shot.)

So I really don't know what to think. Because of the third shot I'm tending to think there is little leakage. Any other opinions?

[Cadmium]

Thanks Craig, One more question comes to mind, the blue material used for the 3D printed adapter, is it completely light blocking?
If you fill the adapter with light blocking material, such as cardboard or such, does the blue material show any light when the flash is triggered?
Will the blue material transmit any UV, Visible, or IR light?

[Craigk79]

The holder is printed in Black PLA , there appears to be no light leaking through it. I will send you a holder so you can test for yourself.