Colorchecker Passport in Solar UV

[rfcurry]

John,

 

Our erstwhile Northern Bear has spent quite some time testing the CopperU. His feedback has been absolutely invaluable. Tomorrow I am sending him the latest, and probably definitive, version for testing.

 

Concerning the release of the CopperU: I expect to offer the CustomU in 52mm threaded mount for sale on my website on 3/14. I have been working on the CopperU, off and on, for four years. I think you will like this filter.

[Andrea B.]

Is copper sulphate toxic should the filter get broken?

[rfcurry]

Here is the MSDS for CuSO4 -- http://www.sciencelab.com/msds.php?msdsId=9923597

 

It is used to treat swimming pools against algae, as well as aquariums, so you have probably encountered it. It is even sold in children's chemistry sets. The lowest dose to have a toxic impact on humans is 11g/kg. Basically, it is safe, though exposure can irritate the skin.

 

The amount of CuSO4 in solution used in my filters is only about 6.50 ml (0.44 tablespoons). The filter is very robust, however, and breakage is not likely.

[Andrea B.]

Thanks, Reed. Good to know this stuff. :D

 

John, thanks also for the Fotozones link.

Which reminds me that I've wondered if there is any easy way to index topics & posts to provide links. I suppose if it had been done on an on-going basis it would be easy enough to keep such a link list going. But to wade through so many Fotozones posts now seems daunting.

 

I wanted to try again to shoot the CC Passport with my GH1. But the sun which was out this morning for about 5 minutes has once again gone away. The cam is all set up and ready to go when I get a chance.

[Andrea B.]

SOMETIMES I AM SO DIM-WITTED !!!!!!! :D

For cryin' out loud I was setting the white balance on my GH1 incorrectly because I was triggering the shutter instead of the WB button during the process. I should have immediately known I was doing something wrong when I didn't get the perfect blues in the shot that I was complaining about above in Post #22.

Arrgghhhh!!! I'm so sorry for being so brain-dead. Maybe I can blame it on an excessively long snowed-in winter? :D

 

Here is my quick GH1 test in Solar UV using the BaadU on a Kaligar 35/3.5 showing perfection in white balancing my GH1.

Now I can test other filters in Solar UV.

 

Visible No Edits

WB was made on the 99% white standard.

f/8 for 1/400" @ ISO-400 EV=-.3

The EV was accidental because the EV button is placed where I manage to bump it constantly. There is also some free cat hair in the foto. Sigh.

 

Visible with Photo Ninja Profile Made from CC Passport

Not much difference. A little more saturation.

 

 

UV No Edits

WB was made on the 99% white standard.

f/8 for 1" @ ISO-400 EV=-.3

Clearly not on tripod.

 

UV with Vis Photo Ninja Profile

And, at last, there are the Blues!

[rfcurry]

Andrea,

 

Thank you for the hard work.

I wish I knew why I get an array of colors, rather than a monochrome palette. Multiple colors will allow us to distinguish relative changes in wavelength reflection with greater ease. Given the narrow dynamic range of the Lumix sensors, it would seem to be better to spread that over various colors rather than only one. But how? Do I have a funky camera that took too much acid in the '60s?

[Andrea B.]

oh lol !! :D

 

Reed I have no idea. I'm sure it is explainable. I'm going to repeat the preceding test with my old PrecisionU. Then we'll try to figure it out.

 

All I can suggest right now it that you make sure you are pushing the WB button when you set in-camera white balance and don't make the same mistake I did. With Nikons we push the shutter to set in-cam WB. That's why I went haywire when trying to set the GH1 WB the first time forgetting that the GH1 was different.

 

Added: No hard work. I'm having fun!!

[rfcurry]

I tried setting the WB as a color value, as I did yesterday with 2500K. This is not one of the custom WBs. The same result as you worked with yesterday in P1110879.RW2. So, it is not my WB setting. I checked my Standard Film Mode. I had that set to +1 contrast, +2 sharpness, and +1 saturation. I changed those, but no appreciable change in WB.

[Andrea B.]

GH1, WB=2500K, BaadU

 

No Edits

 

 

Same Foto with WB on White Standards

Note to Self: Don't use GH1 at ISO1600 for UV.

[Andrea B.]

Nothing mysterious going on - it is the Filter.

 

OK, finally, with the early version of the PrecisionU, GH1, in-camera white balance made on 99% white reflective standard, ISO-1600 (sorry, had such low amount of sunlight !!).

Now I'm getting the yellows.

 

No Edits

 

 

 

Now a conversion in Photo Ninja with saturation bumped way up to show where the yellows & blues are. I also re-balanced on the white standard just for safety's sake.

Again, sorry about this ISO-1600 which is rather destructive, but we are not having a big UV index here today. No matter, the point is that the early version PrecsionU clearly produces approximately the same yellows & blues in the same places as the later version Precision U.

 

 

 

Preceding photo but with addition of GH1 visible colour profile to "restore" original GH1 colour curve.

 

 

**********

 

Now, if I recall correctly, the PrecisionU peaks at a lower nm than does the BaadU. So that probably explains why we get more into those false yellows but still keep some false blues. When/if I ever get the sunlight back, I will shoot again and try for a clearer, prettier example with the GH1 and the PrecU.

[rfcurry]

Yahooooo!

Andrea has done it!

Okay, so it is the filter(s) I am using. The PrecisionU gives me much the same colors as the CopperU, but the CopperU provides more saturation, faster and should go deeper into the UV.

 

Thanks, Andrea. I was beginning to think I was having flashbacks myself. :D Ah, a misspent youth.

 

Now, if only I had a CaF2 or fused silica triplet, to see how deep we can go with colors.

[JCDowdy]

WOW, what an eye-opener!

I have occasionally seen faint hints of some of these colors on my BaaderU CC shots.

I thought it was because I was doing something wrong in WB or that my filter coatings had been scuffed letting color bleed through. Now I am thinking it was maybe just a combination of strong southern sunlight and deeply transmitting quartz lens!

As we open up more of the shorter wavelength UV spectral range the colors should liven up a bit it seems.

[Andrea B.]

One of the problems here is that using a CC Passport to judge anything at all in UV is just really all wrong. :D

Someday when I get a gazillion dollars I'll build a nifty filter board with non-leaky narrowband UV-pass filters which show different false colours. If that makes any sense. And if the backlighting can be standardized in some fashion. (Maybe a couple of Blak-Rays?)

 

The most interesting experiment which everyone is missing here would be to shoot a flower (or other subject) with all these different filters and compare the results in Black & White to see what changes, if anything, about the UV signature of that flower (or other subject). There must be some double-humped UV signatures out there in the world which would show up differently with different filters having different bandpass widths or peaks, doncha think? Or some single humps which could be caught with one filter but not another.

 

The Siren Call of False Colours.

[JCDowdy]

The most interesting experiment which everyone is missing here would be to shoot a flower (or other subject) with all these different filters and compare the results in Black & White to see what changes, if anything, about the UV signature of that flower (or other subject). There must be some double-humped UV signatures out there in the world which would show up differently with different filters having different bandpass widths or peaks, doncha think? Or some single humps which could be caught with one filter but not another.

 

Oh, don't think it has been missed!!!

[JCDowdy]

All this exuberance aside, one thing clearly remains inconsistent.

 

If we attribute the difference in the false colors to shorter wavelengths then we should not be seeing them under window filtered sunlight. Windows filter the UV significantly as Bjørn correctly noted in post#2. The transmittance UV cut-off (50%) will vary with thickness and quality of glass.

 

The CopperU filter must be tested under full spectrum sun with a Schott WG320 to reduce the bandwidth to ~BaaderU and with a WG395 or GG400 to see how much of this color is true color rather than false color.

 

Apparently the PrecisionU should be checked with a GG400 also. A regular old UV-blocking filter would probably suffice,

[nfoto]

The unanswered question is what happens in the range immediately below 400 nm and say down to 380 nm, where the Baader U will not respond much while other filters might.

 

Agree that we probably don't observe effects of shorter wavelengths for these indoors captures.

[Andrea B.]

I'm not sure I agree with you all's observations about window glass. A lot of window glass up here in the Northeast US is pretty old stuff without coatings. Window glass can transmit down to 350nm.

[Andrea B.]

For example.........I just made this at 4PM in the afternoon. The sun is under a haze of clouds. No interior lights are on.

 

GH1-broadband, Kaligar 35/3.5, BaaderU, Window Light

f/3.5 for 1/1.3" @ ISO-1600

No Edits.

[JCDowdy]

Window glass transmission varies with thickness. Thinner single window panes can go down to 50% at 320nm but are more common ~335nm. Thicker commercial glass can cut off much longer.

 

I was speaking of the spectral region below the BaaderU which is specified as "HBW/320-380nm" on the box. I read this as 50% cutoff on the short side at 320nm. So if these unusual colors were due to CopperU transmitting more than BadderU below 320nm window glass would counfound that comparison.

 

Are the CopperU and PrecisionU colors due to 380-400nm, as Bjørn notes above? Try stacking a Hoya L-37 or or L38 on them and see.

 

added later.....

 

I am beginning to suspect something else could also be at play.

[rfcurry]

Darn, I forgot to filter out the Dark Matter! :D

[JCDowdy]

It is a dark matter, the question is how dark?

 

You description of the filter quoted from your website:

"

The CopperU has a mean peak at 339nm of 90.3% with a FWHM (Full-Width at Half Maximum) of 111nm, with midpoints located at 270nm and 381nm. Note the shape of the transmission curve in the graph above. The broad passband exhibits a steep drop to >OD3 at 402nm. From 402nm to 1100nm, the CopperU has a minimum out-of-band rejection of >OD3 and an average of >OD4."

 

An OD of 3 is 10 stops, that is a Lee Big Stopper neutral density filter.

 

Reliable sources report a difference of 8-10 stops from BaaderU UV to Vis with a UV-Nikkor.

 

My dark question then is "minimum out-of-band rejection of >OD3 and an average of >OD4" enough?

[rfcurry]

John,

 

Let us do some comparisons:

"The BPU2 filter is a ultraviolet bandpass filter made so that only light in the 320 to 390 nm portions of the spectrum are transmitted; it will effectively block all else spanning 200 nm to 1120 nm. This U-Filter has improved transmission peaking with 85% transmission at 350nm with a 70nm bandwidth, and with a 5 stop reduction in the IR portion of the spectrum." -- Baader U from Company Seven [Note the 85% peak transmission is impossible with a 2mm thick UG11 substrate.]

 

Some time ago, Bjørn R. did a comparison of the PrecisionU and the Baader U for NIR leakage and found them similar. He recommends either - http://www.fotozones.com/live/index.php/topic/49117-the-uv-season-has-commenced/?hl=precisionu

 

Pelham Research finds OD3 and OD4 to be adequate for their scientific customers -- http://www.pelhamresearchoptical.com/broadband.html

 

Edmund Optics touts their "High Performance OD4 Longpass Filters -- http://www.edmundoptics.com/optics/optical-filters/longpass-edge-filters/high-performance-od-4-longpass-filters/3044

 

Yes, IMO, >OD3 minimum and >OD4 average is dark enough, because that is what has been well-received thus far (Baader U, Andrea U, and PrecisionU). I can produce the CopperU with a much higher OD without sacrificing UV bandwidth or increasing the cost, but the EV would be affected. Through discussion, I concluded that most strictly-photographic users would not be interested in taking 30 second exposures. If some astronomical customers request greater out-of-band OD, it is no problem to deliver it.

[JCDowdy]

Reed,

 

I certainly appreciate that the BaaderU & PrecisionU are top picks and appropriately recommended as Bjørn did in your 2013 link.

 

As for what is acceptable for scientific, or any other, use depends on the application. The Edmund Optics OD4 filter you linked is intended to be combined with a short pass filter thus combining their out of band blocking.

 

More demanding applications such as Edmund's Fluorescence Bandpass Filters are specified at OD6. That 85% @ 377nm, FWHB 50nm, OD6, DAPI Excitation filter for example is also typically paired with another high OD emission filter. Not counting what the dichroic beam splitter is contributing the combined barrier OD for that set-up could be up to 10-12.

 

I am merely asking a studied question, with filters that provide only 3-4OD VIS blocking aren't we close to detection threshold for VIS? Going back to the earlier reference to 8-10 stops difference between UV(BaaderU) and Visible and considering the still unexplained intermittent appearance of unexpected false colors are we occasionally exceeding that threshold?

 

You have posted several images with your filters stacked with longpass IR filters for evaluation of IR leakage. I propose you likewise test for total VIS&NIR leakage by stacking your filter with a Schott GG400, GG420, GG435, Hoya L-42 or Kodak 2E or some equivalent UVIVF barrier filter.

 

Please don't think I am picking on you, I find you filter design an innovative adaptation of a classic design. I am proposing a stringent but objective test, not throwing down the gauntlet.

[rfcurry]

John,

 

You said

I am merely asking a studied question, with filters that provide only 3-4OD VIS blocking aren't we close to detection threshold for VIS?

 

No. Just look at the transmission (not internal transmission) spectra of Schott UG-11 (below), a popular UV-bandpass glass. I use somewhat similar glasses. The Vis is blocked quite effectively, not rising above OD3 until 664nm. The use of CuSO4 knocks this rise down dramatically.

 

So, Vis is not an issue. And CuSO4 is particularly effective in the NIR; just ask astronomers, they pay $700+ for a 1.25 inch CuSO4 filter.

 

I know that my filter exceeds the statements of min >OD3 and average >OD4. However, I am happy presenting conservative claims that are acceptable to the public, rather than publishing exact claims only to have someone with a spectrophotometer screaming that I was high by 0.00001% at 723nm. :(

 

Re: other Edmund filters. If you wish to use their 50mm 377nm CWL, 50nm Bandpass filter as your UV-bandpass filter, have at it. It only costs $630.00. :) Add in the cost of the additional filters you want and you are probably looking at $1500.00. But if it gets you to OD12, I'm sure it will be worth it. Of course, all those additional filters are going to degrade your 85% transmission and probably your FWHM as well.

[Andrea B.]

Well, if you don't do stack tests Reed, someone else will. If you don't provide a spectral transmission chart, then someone else will. You know that's how it always works out. :(

 

John's business is Precision Testing Labs so naturally he asks these kinds of questions. There is no need for you getting all defensive!! Most of us here are simply interested in generic UV photographic use. While our needs may not be as stringent, we are no less interested in the technical details. :)