Infrared Fluorescence

[Andrea B.]

Argh. I give up. I can't write anything correctly today. I don't know what's wrong with me. Too much snow shoveling maybe.

I'll go correct my latest blooper. And then just look at the pictures for awhile.

 

****************

 

Damon your #6 was my bloopered #1. Your #6 is just a reflected UV test with maybe a bit of blue/violet. You did this test in photo #3.

 

*****************

 

Damon's test photos are quite interesting!! Thanks again Damon for entering the ice cold UV shack yet once again!!

 

EDITS: 18 Feb 2015. After seeing transmission chart of Blak-Ray, the interpretation of the photos has been clarified.

 

#1 Blak-Ray with Hoya R72, 30 second exposure:

Shows either some high red or some IR that has been recorded. But 30 seconds is a very long exposure, so the Blak-Ray red/IR output must be very small.

 

#2 Blak-Ray with Baader UV/IR cut, .4 seconds: Shows a Visible exposure. How much is Visible emission and how much might be from any Visible output of the Blak-Rays causing reflection, I don't know. I always assume glowy stuff is the fluor and the rest is side effects. Any comments on that assumption welcomed. "-)

 

#3 Blak-Ray with Baader-U, .6 seconds: Shows an Ultraviolet exposure. Bullseye!

 

#4 Blak-Ray with unfiltered lens, 1/5 second: Wow, this is very interesting. You've got a faint UV bullseye and some Vis fluor and maybe other stuff? I surely do like the glowy tips of the sunflower in this photo and in the first photo. Cool. Given that a unmodified Canon 1DM4 is in use which can record UV, it looks to me like that has happened with the unfiltered lens.

 

#5 Blak-Ray with 093 IR-pass, 30 second exposure: Nothing. So potential IR output, if any, is likely confined between 700-800nm.

 

*****

 

Now, imho, I think we can conclude that these Blak-Rays with Woods glass don't seem to output enough IR or Vis to "contaminate" the fluorescence photos. So you can say that you are producing UV-induced Visible Fluor photos as long as you have filtered your taking lens (regardless of which camera you are using).

 

John, I'll trust your eyes and untired brain to review Damon's newest photos also when you get a chance. Thanks for any input here.

 

Col, you've been following this. Help check for errors too !! And tell us what you think Damon's photos are showing. Thanks for any input here.

 

I think I'll stay offline for a couple of days to reboot. :)

[Damon]

Andrea--welcome to my world. You are one of the sharpest tacks on this site so don't worry about it. My stupid crazy setup with the Blak-Rays is causing all the trouble.

 

Having said that: :)

 

Please help with the conclusions when you are ready, or John, or Col, or anyone willing to look into my rabbit hole.

 

1. IR at least 720nm or greater gets through to my sensor at this particular long shutter/high iso

 

2. Clearly some visible light is getting to my sensor because I can also see this same light with my eyes. Most likely UV Induced Visible Fluorescence and not UV induced IR Fluorescence due to the UV/IR cut filter on the lens. There might be small amount of IR getting past the UV/IR cut filter

 

3. There is UV light getting to my sensor. This conclusion is obtained by seeing the UV bulls-eye signature of the sunflower

 

4. Mishmash. Some UV--kinda see the bulls-eye, IR bleeding it out maybe. The same green tips and blue center as number 2. ?

 

5. No recordable IR below 830nm getting to my sensor at this crazy long shutter speed

 

How can #1 & #5 be right at the same time?

 

-D

[Andrea B.]

In your #5 conclusion you wrote "No recordable IR below 830nm". I think you meant "No recordable IR above 830nm".

 

It's certainly true by default that you get no recordable IR below 830nm with a longpass filter which doesn't start passing IR at all until 830nm. But the point here is that you not getting any recordable IR above 830nm - because the Blak-Ray doesn't output any IR which is above 830nm.

[Andrea B.]

Review Col's IR fluor photo on this Flickr page to see what I meant above by "glowy". Compare Col's IR-reflection Allamanda flower to his IR-fluor Allamanda.

[nfoto]

The designation "830 nm" for a filter means it transmits 50% at that wavelength. Thus, IR830 will pass *a lot* at wavelengths in the 700-800 region. one has to look at the transmission curve to see how steep the cut-off below the midpoint really is. Some filters, for example the IR1000, have a very gentle slope, whilst other cut much more sharply.

 

Do keep in mind most filters aren't anything like the ideal 'brick-wall' filter.

[Andrea B.]

Here's the chart for the B+W 093 which Damon used. It has pretty good cut-in around 800nm. At least we can be sure with this filter that there's no Vis leakage. And like Bjørn said it is about 50% at 830nm.

 

[JCDowdy]

IMHO the conclusion that the Blak-Ray does not emit sufficient blue and/or deep red/NIR to contaminate a UVIVF photograph is not supported by the data thus far. Forgive me for reposting but I want to show these two plots together. I also keep telling myself we are discussing an unmodified camera.

(Sorry for the interrupted post - I clicked on the wrong button)

 

A

 

B

 

The Red/NIR portion of the woodsglass transmittance (A) overlaps more of the native Canon IR filter / Baader UV-IR cut filter than the 720nm filter which just barely overlaps either or both of these filters.

 

Consequently the 720 filter should be expected to exclude a significant portion of the Blak-Ray Red/NIR emission that the unmodified Canon should be more sensitive to.

 

The fact that anything at all is recorded with the 720 filter suggests that the Red/NIR from multiples of this lamp is likely considerable.

 

A shorter wavelength Red/IR pass filter, perhaps one similar to a Schott RG 645 or 665 filter (C below) , which will not attenuate the Red/NIR portion of the woodsglass transmittance (A) would be required for a definitive test.

 

C

[colinbm]

I have got out of bed to post this........

Search Andrea's post, on "Chapter 8: Ultraviolet Photography from Applied

Photography, 1971", for "wood's" & it will be highlighted in yellow,

5x.....

http://www.ultravioletphotography.com/content/index.php?app=core&module=search&do=search&fromMainBar=1

Wood's Glass is mentioned 5 x in the text & is referred to in the figures &

tables.

Applied Photography 1971, is the 'Bible' of photography & I quote from

section 8.4.1,

"8.4.1 UV transmitting filters.

 

Most UV sources emit some visible light, so that a UV transmission camera

filter (e.g. Wratten 18A or 18B) is used if the photographic record is to be

confined to the ultraviolet. For UV-excitation of fluorescent specimens a

filter must be fitted to the lamp house. The filters have a secondary

transmission in the far red and infrared.

 

Gelatin filters are not available, although a solution of cobalt chloride

can be used for some purposes. The generic term 'Wood's glass' is used for

filters having spectral transmission of the form shown in Fig. 8.3a. q.v."

 

Do be aware this was written for film cameras of the day (1971) &

particularly refers to B&W film that was used to record UV images too, but

the film had no sensitivity to IR, so a IR blocking filter was not needed.

There is reference to Blak-Ray lamps too that predate UVProducts.

 

I hope I sleep soundly now :)

Col

 

PS, I have been reliably informed that I am to desist from using the term, 'Wood's Glass', no pain of a horrible punishment, but WG doesn't refer to anything reliably identifiable in the scientific community & shouldn't be quoted as a filter type of unknown quantity. We should endeavour to try & establish the true specs of the Blak-Ray lamps filter.

Col

[Andrea B.]

OK, thank you John. We certainly appreciate your tracking thru all this and keeping us on firm footing. :)

 

30 seconds is a pretty long time, though. The line has to be drawn somewhere. If there is little-to-no contamination at 1 or 5 or 10 seconds, could we then claim UVIVF?

 

What would you suggest as a label for this kind of recorded fluorescence? Mostly-UV-Induced Visible Fluorescence? I tried using that phrasing above - it is a little awkward. But until we can nail all this down, we need to label what we are seeing. I suppose that Woods Glass "Blak-Ray Induced Visible Fluorescence" is also a possibility.

 

In your charts the Baader UVIR Cut comes down pretty sharply at 700nm. I'm not sure how it can be passing much IR 730nm 720nm which is the Woods peak. Aren't we more likely to be seeing some high red contamination instead of IR contamination?

 

*********

 

And Col, thank you for these references about "Woods glass filters". Although for now, please do excuse my continuation of the terminology. I'll try to go back and edit eventually.

 

I certainly do agree with you about establishing the true specs of the Blak-Ray and its cover glass.

(Didn't we have a site link somewhere? I'll go find something.)

[Shane]

MHO the conclusion that the Blak-Ray does not emit sufficient blue and/or deep red/NIR to contaminate a UVIVF photograph is not supported by the data thus far.

 

I used Blak Ray lamps for 10+ years for fluorescence photography of gem materials and I concur with that. I measured the Blak Ray output (spectrum is somewhere? and I believe posted on Nikongear). It all comes down to how strong the fluorescence is, the stronger the fluorescence the shorter the exposure, the less likely the contamination plays a role. If you are trying to capture weak blue or red fluorescence then the Blak Ray leakage is an issue.

 

Again, for comparison leakage tests you should be using a consistent reference target that has no chance of producing fluorescence of any wavelength - ball bearing.

 

It is also important to remember that Woods Glass has become a generic term for UV pass/ visible IR block and that there are now many types of UV glass that are lumped into this "descriptor", hence the need to actually measure the "Woods Glass" being used.

[JCDowdy]

Where to draw the line depends on what the image is used for in the end. Until one knows how much of what VIS or NIR spectrum is recorded there is no way to really decide.

 

I don't know what one would formally call these images. They likely do include some reflected blue and red as well as UVIVF and possibly Vis-IVF. I would just call them what they are - Blak-Ray photographs.

 

Stimulating discussion, sorry if I have been to much a "Captain Obvious"!

 

Damon, dang yer hide! I am going to have to get one of these darn Blak-Rays now! :)

[Andrea B.]

Found this in the user instructions for a Blak-Ray UV Intensity Meter:

 

For instance, the unfiltered Model B - 100A 100 watt mercury vapor BLAK-RAY Lamp has an average infrared intensity of 10% of the total reading. The value decreases to an average infrared intensity of 2.6% of the total reading when the filtered longwave emission is viewed.

 

****************

 

Found this at UV Innovations website:

 

Mercury vapor spectra filtered with Kodak 18a (Woods). The Blak-Ray is a mercury vapor lamp, so this might be a useful chart. The red line is the mercury vapor output. The orange is the Kodak 18a transmission. Unfiltered the mercVap outputs great gobs of Vis, but filtered hardly any - maybe just the tad bit of violet. Really just the IR we should worry about.

[Shane]

Point in case - the Woods glass shown on Johns posting is nothing like the UV glass used on the Blak Ray which cuts off UV below 300nm.

 

Basically today, the term Woods glass as it is currently used really just infers a UV pass/visibleIR block filter.

[Andrea B.]

Blak-Ray will supply transmission data upon request, so let us all go request in our various capacities as owners of websites & businesses and so forth.

[Shane]

Andrea the peak IR emission in the Blak Ray mercury lamp is 730/732nm and red at 668nm.

[Shane]

18 FEB 2015 UPDATE: This chart shows 2nd order artifacts at 668nm and at 730nm which may be ignored. See Post #75 for an updated chart.

 

Blak Ray B100AP measured 2001.

[Shane]

Note the blue leakage at 405nm, red at 668nm and IR at 730/732nm.

 

The early Baader UVIR (version 2459210 380-680nm with an additional leaky region below cut-on) leaked both blue and red

 

whereas

 

the late Baader UVIR (version 2459120A 420-680nm) only leaks the red.

[Andrea B.]

Shane, thank you, thank you, thank you !!!!!!

 

That red & blue leakage is so very small. We should probably call that 405nm Blak-Ray output a violet instead of a blue leakage? (See below.)

 

What does the left side of your graph represent?

 

********

 

From the NASA website:

 

Violet ~ 400nm

Indigo ~ 445nm

Blue ~ 475nm

Green ~ 510nm

Yellow ~ 570nm

Orange ~ 590nm

Red ~ 650nm

[Shane]

People (general public) normally consider visible as defined blue through red and to be honest, colour definitions mean nothing anyway, just try visiting different parts of the world, not to mention that we each may see colours slightly differently. So although I somewhat agree with you, the only correct scientific definition that is accurate is defined by wavelength. After all how does NASA describe 430nm, violetish indigo? :)

[Shane]

What does the left side of your graph represent?

Sorry - it means nothing. The spectrum was captured on an emission spectrometer so the peak intensities are just relative to one another.

[Andrea B.]

So, we are now able to say the following, if you will pardon my approximation phrasing. Also remember that there is some amount of bandwidth around the specifically mentioned wavelengths.

 

Output Illumination with Blak-Ray B100AP:

• lots of UV-365nm and a bit of UV-334nm
  
• very tiny bit of Violet-405nm,
  
• very tiny bit of Red-668nm, and
  
• some IR-730nm (small relative to the UV output).
  

Recorded under Baader-UVIR Block Filter (newer version, in the dark):

• any Visible emissions mostly between 420-680nm (the fluorescence),
  
• very tiny bit of Violet-405nm and Red-668nm reflection - depending on the subject, and
  
• a fairly small amount of IR-730nm reflection - depending on overlap between Baader-UVIR Block and Blak-Ray output.
  
• no IR-730nm reflection because there is no overlap with the Baader-UVIR Block filter
  

I'll read this over to make sure I'm not making bleary-eyed mistakes like last night! But don't hesitate to immediately post any corrections or errors.

[Shane]

It is 730/732nm not 720nm.

 

In combination with the "Woods Glass" filter the Blak Ray 334nm output is insignificant compared to the 365nm output (perhaps only around 1/20th of the output).

 

Since the Baader UVIR version A drops to less than 1% at 700nm and the Blak Ray IR emission is a sharp emission peak at 730/732nm I don't see an overlap occurring.

[Andrea B.]

Shane, I agree that colour names are somewhat arbitrary. I'm enjoying calling 405nm Violet because I can now see Violet (and Blue) much better after the left-eye cataract surgery. :)

 

*******************

 

I think we are very close to being able to call Damon's Blak-Ray photos an example of UVIVF, aren't we? If he were to stack a BG39 over the Baader-UVIR-Cut that would omit any pesky IR from being recorded. And might also suppress the Red-668nm, too. I have to go look at the charts to be sure where that BG39 cuts off. The BG40 might also work?

 

*******************

 

[Made wavelength corrections as needed above.]

[Shane]

If he were to stack a BG39 over the Baader-UVIR-Cut that would omit the pesky IR from being recorded.

 

May we cross posted but there is no IR leakage when using the Blak Ray with the later version of the Baader UVIR cur filter (in fact you even thought that yourself earlier in this post)..

[Shane]

The 3mm BG39 barely squeezes by as a red 668nm filter as it reaches ~0% ~670nm