• Ultraviolet Photography

Lee Blue filter test to get the Aerochrome Red in camera

149 replies to this topic

#1 dabateman

    Da Bateman

  • Members+G
  • 1,976 posts
  • Location: Maryland

Posted 12 March 2019 - 17:02

I ordered a Lee Filter swatch book. Quite cheap. In around there area of Scuba blue is two other filters with slightly different curves. The swatch book is 39mm x 89 mm, perfect for microscope slides, which I intended to play with "optical colour".

Holding the Scuba 729 filter infront of my Sigma 30mm f2.8 lens on my EM1 full spectrum, I was able to get an in camera WB. I was hard now for some reason and I needed the White Color checker to get it. Not sure why my EM1 is acting up with WB settings now.

Here is BW486 regular photo:
Attached Image: BW486.jpg

Here is straight out of camera using Suba blue 729 with it for the WB setting:
Attached Image: ScubaBlue_729.jpg
Scuba blue in my book as peak max at 500nm at 40% with about 30% peak at 800nm

Here is a neighbor filter QFD Blue 727 using the Scuba blue WB setting:
Attached Image: QFD_Blue_727.jpg
This filter has a 495nm peak max at 40% with 25% peak at 800nm

Here is an other filter Medium Blue Green #116, with Scuba blue WB setting:
Attached Image: MediumBlueGreen_116.jpg
This fitler has a peak max at 500nm at 60% with 50% peak at 800nm.

I don't have a weak IR blocking filter to test. There also seems to be other filters in the area that might be interesting. So if you really want that in camera Aerochrome look on the cheap, I would recommend ordering a cheap Swatch book from Lee and find out the best filter for your camera.

Edited by dabateman, 12 March 2019 - 17:12.

#2 GaryR


  • Members+G
  • 270 posts
  • Location: London, Ontario, Canada

Posted 12 March 2019 - 19:29

This is a SOOC shot (FS-Sony a7R) with a Rosco E-Colour+ 729 Scuba Blue gel, with a maxmax Xnite CC2 filter, CWB on a Kodak gray card. The CC2 filter does make a difference in the reds, and less cyan sky. As for the aerochrome look, visible reds (car tail lights) remain red, instead of yellow/green.

Attached Image: _DSC0932.JPG

#3 Cadmium


  • Members+G
  • 3,691 posts

Posted 12 March 2019 - 21:33

So the second shot is Lee Scuba Blue #729 only?
And the lest shot above is Rosco E-Color+, + Lee Scuba Blue #729, + CC2 (Schott KG1)?
David Twede (who did this in the link you found this idea in) uses KG3 (which has a little more IR transmission than KG1.
Other than that, this looks close to what he shows, don't you think?
I am not familiar with the Rosco E-Color+, can you explain your reason for adding that?

You will not get yellow tail lights with any filter or stack without post processing/swapping/etc..
No filter with transpose red to yellow without post processing the image.

Edited by Cadmium, 12 March 2019 - 22:48.

#4 GaryR


  • Members+G
  • 270 posts
  • Location: London, Ontario, Canada

Posted 13 March 2019 - 01:00

My shot is a single Rosco E-Colour+ 729 Scuba Blue gel stacked with a CC2. The "+" symbol is part of the name. The Rosco version was easier to obtain here in Canada, and appears to be the same blue/green color as the Lee Scuba Blue. At any rate it's dirt cheap for a 20x24" sheet!

I originally got the XNite CC2 to see if it would work with the Nick Spiker aerochrome profile, instead of the recommended Kopp 7093 glass. The KG3 that David Twede uses, has the same 900nm cutoff, so I figured the CC2 would produce similar results.

Post processing is the only way I know of, to get a believable aerochrome look, but the Scuba Blue method also works for video.

#5 Cadmium


  • Members+G
  • 3,691 posts

Posted 13 March 2019 - 04:38

Gary, thank you for your examples!
OK! This is cheering me up a little then. Not sure how much difference there would be in the actual IR transmission of the Rosco version. Rosco is a different brand of the Lee #729, I take it then?
Yes, the + is part of the name.
So when you said Lee, you didn't mean Rosco + Lee, you are just using the Rosco version?
Yep, these Lee filters are ultra cheap.
Yes, the Kopp 7093 is the same as KG3, CC2 is the same as KG1.
Guess I have said that already 5 times probably.
Not sure if I have shown a graph that shows all 4 of those.
Yes, nothing wrong with a filter that does which does what your test does at all, just that there would never be any filter that actually transposes... we will have to wait for a transposable Bayer matrix for that... :)

KG#, CC2, and Kopp 7093 comparison graph (all at 2mm thick) actual transmittance, linear:
Attached Image: KG_All_CC2_Kopp7093.jpg

Edited by Cadmium, 17 March 2019 - 21:45.

#6 GaryR


  • Members+G
  • 270 posts
  • Location: London, Ontario, Canada

Posted 13 March 2019 - 17:18

Thanks for the informative graph Steve.
I kind of figured that the CC2 would be close enough, but nice to see a comparison.

Yes, I'm only using a single Rosco gel with the CC2.
Rosco is a different brand, but they have a Lee to Rosco Equivalents chart. Scuba Blue is apparently the same in both brands.

Yellows in particular are missing in the Scuba Blue technique, but will be interesting to compare with the Kolarivision IR Chrome filter.

#7 Cadmium


  • Members+G
  • 3,691 posts

Posted 13 March 2019 - 22:38

This is not the perfect test, simply because I don't know what KG I am using here. It is KG glass, but I don't know the type, and it is also 3mm thick.
I have more KG ordered, but this 'Mystery KG' will have to do for this test.
Not sure how much the results will change once I get some KG3 2mm in front of the lens instead, which is what I intend to use.
The Lee Scuba Blue #729 shots were done later in the day than the B-410 shots.

Lee Scuba Blue #729 alone:
Attached Image: Lee_729_1280.jpg

Lee Scuba Blue #729 + Mystery KG 3mm thick glass:
Attached Image: Lee_729_KG_3mm_1280.jpg

Hoya B-410 2mm alone:
Attached Image: B410_1280.jpg

Hoya B-410 2mm + Mystery KG 3mm thick glass:
Attached Image: B410_KG_3mm_1280.jpg

#8 dabateman

    Da Bateman

  • Members+G
  • 1,976 posts
  • Location: Maryland

Posted 14 March 2019 - 00:12

Interesting, that KG glass really has an effect.

#9 Cadmium


  • Members+G
  • 3,691 posts

Posted 14 March 2019 - 05:45

Yep, It adds definition, and saturates the red/orange colors.
I am guessing that is because it cuts some of the higher IR, which ads more IR to the Blue and Green channels, where as the camera sensor is more sensitive to lower range IR via the red channel (by way of the Bayer filters),
thus a combination of red Bayer filters transmitting more % of the lower IR, and the suppression of the visual red.
It is a little hard to wrap my mind around, but I think that is what is going on.
Still, my example isn't very strong compared to Gary's example above, or to David Twede's examples that you first found for us.
I have not found a good way to talk to him David Twede, only by way of publicly posting on his blog, which is how I asked about the KG3 thickness he uses.

Edited by Cadmium, 14 March 2019 - 05:48.

#10 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,000 posts
  • Location: USA

Posted 15 March 2019 - 17:28

good thread with nice examples.

I'm trying to decide what to label a tag for this kind of IR work. Would an "EIR" tag be sufficient to cover all the various types that EIR/Aerochrome experiments produce? "Aerochrome" is rather long for a search tag maybe?
Andrea G. Blum
Often found hanging out with flowers & bees.

#11 UlfW

    Ulf W

  • Members+G
  • 1,110 posts
  • Location: Sweden, Malmö

Posted 15 March 2019 - 17:55

VIS+IR might be a more accurate label.
A VIS+IR tag could then include more types of images, like red foliage experiments, goldie images, and mabe B-410 images.

I have learned from Master Steve and fully embraced, red foliage is not the only needed telltale for proper EIR simulation. You need Green Red => Yellow too! :)

Edited by UlfW, 21 March 2019 - 06:28.

Ulf Wilhelmson
Curious and trying to see the invisible.

#12 Andy Perrin


  • Members+G
  • 3,222 posts
  • Location: United States

Posted 15 March 2019 - 20:04

I agree, Vis+IR makes more sense to me.

#13 Cadmium


  • Members+G
  • 3,691 posts

Posted 15 March 2019 - 21:04

I leave all labeling up to the labeling committee. :)

Making definite progress here now.
This is The Real McCoy, no 3mm thick Mystery KG filter, this is using real KG3 2mm.
This looks much closer to Gary's example above, and to David Twede's examples on the link.

I present to you Lee Scuba Blue #729 + Schott KG3 2mm stacked.
And yes Virginia, Santa remains red, and so does the little red shed, and tail lights too, but that is to be expected with an EIR/Aerochrome simulation using only a filter.
Attached Image: Lee_729_KG3_2mm_Landscape_1280.jpg

Edited by Cadmium, 18 March 2019 - 01:13.

#14 GaryR


  • Members+G
  • 270 posts
  • Location: London, Ontario, Canada

Posted 15 March 2019 - 23:11

That last example looks quite good, with the KG3. I expect that the KV IR Chrome filter will likely be perfectly calibrated for this type of false color shot. We'll likely be seeing a lot of IR Chrome images when they start shipping.

#15 Cadmium


  • Members+G
  • 3,691 posts

Posted 16 March 2019 - 03:21

Other tests from today.

Visual (using Schott BG38 alone):
Attached Image: Visual_Back_Land_1280.jpg

Hoya B-410 alone:
Attached Image: B410_back_land_1280.jpg

Hoya B-410 2mm + Schott KG3 2mm stacked:
Attached Image: B410_2mm_x_KG3_2mm_stack_back_land.jpg

FLD filter alone:
Attached Image: FLD_back_land_1280.jpg

#16 Cadmium


  • Members+G
  • 3,691 posts

Posted 16 March 2019 - 23:10

Two more KG3 stacks for the road.

Wratten 38A + Schott KG3 2mm
Attached Image: Wratten_38A_x_KG3_2mm_1280.jpg

Schott BG25 2mm + Schott KG3 2mm
Attached Image: BG25_2mm_x_KG3_2mm_1280.jpg

#17 Cadmium


  • Members+G
  • 3,691 posts

Posted 17 March 2019 - 20:57

With these I used in camera Custom White Balance. WB target was white Komatex.
I have made no post color adjustments.
All three are shot using Lee Scuba Blue #729 + Schott KG3 2mm filters stacked.
All the same as before, except I found my usual 18-55mm VR lens, and in camera Custom WB.
D7000 UV/IR, Aperture Priority, f/11, ISO 200

Attached Image: Hill_729_x_KG3_2mm_1280.jpg

Attached Image: Hill_Close_729_x_KG3_2mm_1280.jpg

A quite rare visitation by Pandora the cat! I have been told she has a PhD in Phenomenality.
Attached Image: Pandora_729_x_KG3_2mm_1280.jpg

Three shot pan
Attached Image: 729_pan_1280.jpg

Edited by Cadmium, 17 March 2019 - 22:30.

#18 GaryR


  • Members+G
  • 270 posts
  • Location: London, Ontario, Canada

Posted 18 March 2019 - 14:52

Took the Rosco Scuba Blue + CC2, out for a run at a local lakeside village.

Attached Image: _DSC0973.JPG

Attached Image: _DSC0990.JPG

Attached Image: _DSC0991.JPG

Attached Image: _DSC0996.JPG

Attached Image: _DSC0999.JPG

#19 Cadmium


  • Members+G
  • 3,691 posts

Posted 18 March 2019 - 22:21

Gary, Thank you for posting these photos. :)
You are using in camera custom white balance? And what camera? Oh yeah FS-Sony a7R.
I use Nikon, which has always been completely impossible to do in camera WB with most beyond visible filters,
so I never use that method anymore, I get quick and easy and versatile results white balancing in Nikon's CNX2/NX-D or Photo Ninja,
but comparing to your results I decided to try white balancing in camera, which works well in the Nikon with visual range filters,
and I was surprised how much better the in camera white balanced looked for this filter stack compared to any result in software from RAW.
The Lee and Rosco 729 filters are acrylic, and can be laminated between glass using the appropriate optical adhesive similar to how the Schott KV filters were constructed.
Also, the KG filters are something I never gave much attention to, I didn't think they would be useful for photography, but this has changed my mind about that,
after testing a few other filters stacked with KG I can see there are possibilities that I had not realized before.
Thanks again for posting your pics! :)
And David, thanks again for starting all of this with your link to David Twede's examples and formula. :)

Edited by Cadmium, 18 March 2019 - 22:27.

#20 dabateman

    Da Bateman

  • Members+G
  • 1,976 posts
  • Location: Maryland

Posted 19 March 2019 - 04:06

Yes I do find all this interesting. Using this idea, I am looking for an in camera goldie combination. This is more of my favorite IR look. I still need to test, but I think the lee filters:
363 special medium blue filter peak at 45% at 450nm
713 winter blue 25% at 450nm
722 bray blue 45% at 460nm, slight peak at 700nm
May work. Testing the 722 bray blue on a normal, stock camera. The image looked almost black and white, as green had moved to gray and some color remained.

My thinking is that since 500nm, went red. That lower wavelengths may go yellow. But I might be wrong. I may also have to try the greens:
139 primary green
90 dark yellow green
736 Twickenham green

These all have distinctive peaks and lowish IR. I still have to play around. Good thing the lee test swatch was cheap and I made a holder out of a 37mm to 58mm stepup ring. So I can drop filters in. The 12-32mm and 14-42mm lenses take 37mm filters in the m43rds world.

I still don't have a KG filter.