• Ultraviolet Photography
  •  

Avro CF-100 In UV

22 replies to this topic

#1 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 26 June 2018 - 17:49

Avro CF-100 Mk5 "Canuck" RCAF Fighter (414 Electronic Warfare Squadron). This is a decommissioned plane, not a life-sized replica.

Note the dark canopy, the glass being totally transparent in the visible spectrum. If you look closely, there are 2 mannequin pilots in there. Nice to know that even in old planes, the pilots had plenty of UV protection up there.

Nikon D3200 full spectrum with Kolari Vision UV bandpass filter. 18-55mm kit lens at 24mm, f/4, 1/25 sec hand-held, ISO 800. Processed in Lightroom - played with the HSL a bit, added some clarity and dehaze.

Mounted on a concrete plinth, the plaque underneath reads: "The CF-100 all-weather fighter, powered by Orenda engines, was designed and manufactured in Malton. It was the first all-Canadian military aircraft. 692 CF-100s and over 2,000 engines to power them, were produced in Malton between 1951 and 1958. This historical memento was erected by Malton Branch 528 Royal Canadian Legion, in co-operation with the City of Mississauga. A.D. 1974."

Attached Image: DSC_0176a_RS.jpg

#2 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 26 June 2018 - 18:05

Yeah, you need a better lens for UV, the 18-55 VR is a poor UV lens.
The plane reminded me of this topic:
http://www.ultraviol...ispectral-plane

I will go shoot two UV comparison shots using 18-55mm VR and Kuribayashi 35mm.

Edited by Cadmium, 26 June 2018 - 18:26.


#3 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 26 June 2018 - 19:19

View PostCadmium, on 26 June 2018 - 18:05, said:

Yeah, you need a better lens for UV, the 18-55 VR is a poor UV lens.
Absolutely - that's why I spent this past weekend shopping on eBay :)

Still, not too bad, considering.

View PostCadmium, on 26 June 2018 - 18:05, said:

The plane reminded me of this topic:
http://www.ultraviol...ispectral-plane
That's a great set of comparison photos. I took some shots with my IR and with full spectrum, but since I forgot my tripod that day, there would be no way to play around with the channels.

View PostCadmium, on 26 June 2018 - 18:05, said:

I will go shoot two UV comparison shots using 18-55mm VR and Kuribayashi 35mm.

I'm thinking of returning with my 55-300mm at 80mm and my EL-Nikkor 80mm on a tripod, and doing some comparison shots of this plane. If I step back far enough, I should be able to get at least the nose cone and canopy in the shot, perhaps a little more. Should be a good indicator of the amount of focus shift and shutter speed between the 2 (same aperture, same ISO).

I'll also take some multi-spectral shots.

#4 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 26 June 2018 - 19:36

Well, this worked better than I had thought, but I am not shooting into the sky, and I am shooting at about 35mm to frame it like the Kuribayashi 35mm + M42 -to- Nikon Infinity focus adapter.
As I remember in the past shooting wider, and at the sky will give the color shift from center to edges like your plane shot shows.

First two pics share the same in camera white balance, no out of camera adjustments.
Nikon 18-55 VR @40mm + Baader U (f/8, ISO 400, 1.6s)
Attached Image: 18_55_VR_Baader_U_CAM_WB.jpg

Kuribayashi 35mm + M42 -to- Nikon Infinity focus adapter + Baader U (f/8, ISO 400, 1s)
Attached Image: Kuri_35mm_Baader_U_CAM_WB.jpg

Second two shots are white balanced from RAW using NX2 (marquee).
Nikon 18-55 VR @40mm + Baader U (f/8, ISO 400, 1.6s)
Attached Image: 18_55_VR_Baader_U__NX2_WB.jpg

Kuribayashi 35mm + M42 -to- Nikon Infinity focus adapter + Baader U (f/8, ISO 400, 1s)
Attached Image: Kuri_35mm_Baader_U_NX2_WB.jpg

Edited by Cadmium, 26 June 2018 - 19:37.


#5 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 26 June 2018 - 20:56

Good visual comparison.

I have to say, though, that I am a bit disappointed to see only a 0.6 second difference in shutter speeds. I suspect that gap will increase as each lens is opened up more.

Hard to tell from the photos, but I don't see much focal shift between the 2, either. There also seems to be haze near the top of the white-balanced photo out of the Kuri that doesn't seem to be so pronounced in the 18-55 image, although there's more tonal separation between the grass and trees with the Kuri.

Edited by Andy Broomé, 26 June 2018 - 21:00.


#6 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 26 June 2018 - 20:57

Anyway, here's a split-toned 950nm shot (I rarely go full monochrome in IR, even at 950nm):
Attached Image: DSC_0171-2_RS.jpg

Edited by Andy Broomé, 26 June 2018 - 21:06.


#7 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 27 June 2018 - 06:29

View PostAndy Broomé, on 26 June 2018 - 20:56, said:

Good visual comparison.

I have to say, though, that I am a bit disappointed to see only a 0.6 second difference in shutter speeds. I suspect that gap will increase as each lens is opened up more.

Hard to tell from the photos, but I don't see much focal shift between the 2, either. There also seems to be haze near the top of the white-balanced photo out of the Kuri that doesn't seem to be so pronounced in the 18-55 image, although there's more tonal separation between the grass and trees with the Kuri.

I don't see much difference between the RAW 'haze', you peak of, not in the white balance version either.
If anything, there is a color shift from the top center of the 18-55mm to the top corners (more blue), seen in the white balanced version, hard to see maybe, but this is one of the problems with the 18-55mm VR.

Attached Image: 18_55_VR_and_Kuri_35mm_Baader_U_CAM_WB.jpg

Attached Image: 18_55_VR_and_Kuri_35mm_Baader_U_NX2_WB.jpg

Edited by Cadmium, 27 June 2018 - 06:33.


#8 UlfW

    Ulf W

  • Members(+)
  • 660 posts
  • Location: Sweden, Malmö

Posted 27 June 2018 - 06:47

I see the haze too, both in the original image and the striped composition. Could it be that different monitors reveal more or less?
Ulf Wilhelmson
Curious and trying to see the invisible.

#9 Andy Perrin

    Member

  • Members(+)
  • 2,257 posts
  • Location: United States

Posted 27 June 2018 - 07:05

Not sure I follow this. All UV landscape pics have haze, just because shorter waves gets scattered more easily (as the inverse 4th power of the wavelength, in fact). That yellow-greenish haze in the WB version indicates the Kuri is letting the shorter waves through. If that's what you're talking about, it's a good thing!

Edited by Andy Perrin, 27 June 2018 - 07:11.


#10 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 27 June 2018 - 08:04

All I know is that a 18-55 VR will not compare to a Kuri 35mm. I could try a Sparticle test comparison if you want. :D

#11 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 27 June 2018 - 13:23

View PostAndy Perrin, on 27 June 2018 - 07:05, said:

That yellow-greenish haze in the WB version indicates the Kuri is letting the shorter waves through. If that's what you're talking about, it's a good thing!
Yes, you're correct. The wider tonal range makes the haze more apparent in the Kuri white-balanced version, and it's easily fixed in Lightroom/Photoshop with a gradient filter.

Still, the difference in shutter speed, given the same aperture and ISO, is not as good as I was expecting (assuming there was no significant change in lighting between the 2 shots). I hope my EL-Nikkor 80mm will make a bigger difference vs my kit lens.

#12 Andy Perrin

    Member

  • Members(+)
  • 2,257 posts
  • Location: United States

Posted 27 June 2018 - 15:09

I think your expectations are slightly out of sync here. The camera is less and less sensitive to UV the shorter the wavelength (see Jonathan's threads on measuring the camera sensitivity). In addition, there is dramatically less sunlight in the shorter waves too. While switching lenses will get you a small increase (going from 1.6sec to 1sec is 60%), but in terms of stops that's not even one stop. Using a de-Bayered sensor will get you a stop or two, but then you give up color. So there is no winning. A quartz lens even will only be a bit faster than the Kuri just because there isn't much UV at those short wavelengths. If you want a faster setup, optimizing your filter stack is probably the first consideration.

Edited by Andy Perrin, 27 June 2018 - 15:11.


#13 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 27 June 2018 - 19:23

Yes, I've already looked at swapping my S8612 3mm for a 2mm version. It'll be quite a while before I can consider buying a Baader U, so I'll have to stick with my B+W 403 and S8612 stack for now.

#14 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 27 June 2018 - 19:30

The Kuri shot has more vibration blur, and it also has a brighter exposure. I was not using a 'cable'. Close actual pixel inspection shows overall blur on edges of everything in the Kuri shot.
The exposure (dark/light) difference between the two shots is another reason you are surprised by the 1s vs 1.6s difference. The 18-55 VR shot should really be a little longer exposure time to equal the Kuri exposure shot.
These were shot AP which doesn't always yield exactly the same thing (center weighted).

The lens is the #1 bottleneck. I will do a Sparticle for it later. The deeper a lens goes, the stronger and wider the curve.
Transmission depth determines exposure time, as well as color, but it shows up more as exposure than color.
Here is a good example of filters and lenses, how the lens will define the useful range of the filter, how many UV filters work essentially the same with a fairly good UV lens,
but also how poorly a good UV filter will work with a poorly transmitting UV lens. Even though the 18mm lens transmits down to 360nm, the curve of that lens 'chops off' most of the transmission of any UV filter.
The Sparticle doesn't show the slope, just the cutoff depth.
By the way, the 18/4 lens is the best 18mm that I know of for UV, not good, but for 18mm the best I know of.
Transmission depth of the lens is important, no mater how you look at it.

Attached Image: Kuri_Baader_LUV_UVee_3.jpg

#15 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 27 June 2018 - 20:45

Oh good, I found an old test of the 18-55 VR using the Mini-Sparticle . This shows the difference in UV transmission depth.
Attached Image: UVA_lens_tests_sparticle_b.jpg

#16 Andy Broomé

    Invisible Light Enthusiast

  • Members
  • 148 posts
  • Location: Mississauga/Toronto, Ontario, Canada

Posted 28 June 2018 - 02:34

View PostCadmium, on 27 June 2018 - 20:45, said:

Oh good, I found an old test of the 18-55 VR using the Mini-Sparticle . This shows the difference in UV transmission depth.
Attachment UVA_lens_tests_sparticle_b.jpg
Sorry, what am I looking at here? The 18-55mm VR looks sharp, at least compared to the Kuri.

#17 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 28 June 2018 - 03:05

Andy, the Sparticle is a grouping of narrow bandpass filters. each are 10nm wide, 340BP10, 350BP10, 365BP10, 385BP10, and 400BP10.
In this case they are all in the UV-A range of the UV-only filter on the lenses.
If the transmission of the lens + the UV-only filter (Baader U on the lens in these shots) pass any of those bandpass ranges, then the filter shows light.
If the lens + UV-only filter don't transmit the range of the BP filter, then it doesn't light up.
So in these tests, the only lens that transmits below 360nm is the Kuribayashi 35mm, the other lenses are limited to 360nm and above.
I show the Nikon 18/4 in graph above.
Here is a Sparticle test showing the 18/4, see how it stops at 360nm.
However, keep in mind that the UV transmission curve and peak are very much more suppressed with most lenses that cutoff at 360/365nm, just like the graph illustrates only a 30% peak.
Attached Image: Nikon_18mm_vs_Bushnell_21mm_Sparticle_8.jpg

Edited by Cadmium, 28 June 2018 - 03:09.


#18 Andy Perrin

    Member

  • Members(+)
  • 2,257 posts
  • Location: United States

Posted 28 June 2018 - 03:59

Quote

The lens is the #1 bottleneck. I will do a Sparticle for it later. The deeper a lens goes, the stronger and wider the curve.
Transmission depth determines exposure time, as well as color, but it shows up more as exposure than color.
This is true, but what I was fighting against above was the idea that ANY lens is going to help the exposure time all that much. Even the best quartz lens and a Baader U still takes a very long exposure in UV compared to visible light because there is just not much light and the sensor is also weaker. If you pump up the ISO you can shoot hand-held, but UV photography is about dealing with severe light constraints no matter what.

#19 Cadmium

    Member

  • Members(+)
  • 2,602 posts

Posted 28 June 2018 - 05:09

I should say that if you have a UV-Nikkor (or the like, 'quartz' lens), then the bottle neck is not the lens.
With UV the bottle neck is always kinda of the sensor, but we can't change that (short of a monochrome conversions, and even then...).
So then we are left with two other bottle necks, the lens and UV filter, and short of the quartz lens, the lens then becomes the bottleneck, because the usual UV-only filter will transmit more than even the best non-quartz lenses.
Given that scenario, all you UV-Nikkor people can shuffle down the hall to conference room B, because the transmission depth of non-quartz lenses will vary widely and change the exposure time and color of UV photos.
Not only that, but there is uneven color (as seen with the 18-55 VR), and focal shift, and sharpness.
You can tell quite easily if a lens transmits deep or not just by comparing exposure time. It doesn't tell you how deep, but it is a sure indicator.
I did an exposure adjustment with the two images, to make them look the same brightness, which needed about 0.7 stop added to the 18-55mm VR, so given that it means the difference is about 1-1/3 stop.
But it is hard to say exactly, it would need some more testing, but the 18-55 VR shot is defiantly darker than the Kuri shot.

#20 dabateman

    Da Bateman

  • Members(+)
  • 885 posts
  • Location: Maryland

Posted 28 June 2018 - 10:42

Yes without a Quartz lens, I do think the lens is the bottle neck. Not the filters. Remember looking at spectrum data for filters a half drop is one stop. So 50% transmission vs 25% transmission is only one stop difference. Most of the uv filters we use are quite good. We may argue over 75%vs 60% transmission in our brains but this will not change the exposure much. Although spectral width is still important.
However comparing lenses. I tested my EL Nikkor 80mm f5.6 enlarger lens vs that Wollensak 25mm f1.5 lens the other day. At f5.6 on both lenses the Wollensak was 4 stops slower at gathering UV than the Nikkor! And this was using the best possible uv filter, the 390bp25 filter I have with 70% transmission peak at 392nm.
I think the Wollensak should be pulled from the sticky lenses. Really not great.

Edited by dabateman, 28 June 2018 - 10:45.