• Ultraviolet Photography

Are longpass filters useful for transmission testing?

48 replies to this topic

#41 UlfW

    Ulf W

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

Posted 04 November 2018 - 11:21

View PostCadmium, on 04 November 2018 - 09:16, said:

OK, then how does this idea sound, would you all like me to classify the U-360 1mm stack as a UV only filter?

It all depends on what we/you reads into the word "only"
This is something that would be good to discuss further.

Even a Baader U or a U-360/S8612 2mm+2mm-stack transmit ca 1% at 400nm.

The U-360/S8612, 1mm+2mm-stack transmit ca 8% at 400nm.
At 390nm it transmit 43% and at 410nm ca 0.4%

If OD4 or OD5 should be used as a decision level for "UV-only", there would just be UG11-stacks left.

None of these filters transmit anything significant further away from 400nm, for the normal UV-photography we are doing, regardless of if it is floral macro- or landscape-photography.

Functional-wise I would like to say they all belong in the UV-only group as images are completely dominated by the UV-light.
My reasoning behind that the decaying filter transmission slope at 400nm is steep enough to make that happen.
Ulf Wilhelmson
Curious and trying to see the invisible.

#42 Andrea B.

    Desert Dancer

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

Posted 04 November 2018 - 15:37

I agree with Ulf's analysis. Functionally, the various filters discussed here are UV-only. Photos made with a U360/2+2, BaaderU, Seu2 all successfully capture UV signatures of whatever subject is being photographed.

I don't think any of these filters need classification as "UV-only". They have always been offered as "UV-pass" filters or dual-bandpass filters (when not IR-blocked). Buyers can look at transmission charts and make up their own minds about what OD they want.

I think I should try those 3 filters I just mentioned on the Monochrom now that I have learned it can shoot UV so well. I've ordered an adapter so that I can use the UV-Nikkor on the MM1. Perhaps seeing UV images made without the distraction of false colour would be useful to this discussion?? Jonathan also could try some experiments with his Mono cams to try to find actual evidence or not of viole.t
Andrea G. Blum
Often found hanging out with flowers & bees.

#43 Cadmium


  • Members+G
  • 3,691 posts

Posted 05 November 2018 - 02:19

As I said, "I leave it to others to define what a 'UV-only' filter is, should be, or what you want it to be".
However, for as long as I have been involved in this, a UV-only filter cuts off at 400nm.
The VV (visual violet) light may only change the blue color a little, but technically it is a mix of UV and visual light, not UV only.
So it is a little like saying UVIVF that is illuminated with unfiltered UV light is visual fluorescence only.
Of course usually any blue light emitted by the unfiltered torch only changes the results slightly, just like the UV+VV filters may only have slightly different color,
take a look at the difference in color between Ulf's comparisons here:

He shows three stacks:
UG2A 2mm + S8612 2mm
UG1 1mm + S8612 2mm
UG1 2mm + S8612 2mm
These have progressively less VV mixed in with the UV, and you see a difference in the blue colors.
This is not unlike filtered and unfiltered UV light used to excite fluorescence.
It is a very similar scenario. If you want to ignore the facts because they don't matter, then that is your prerogative, I am only pointing out the facts.
Artistically, they are all nice to look at, but which one has more visual light mixed in to the UV photo?
If we thought that way about UVIVF then none of us would be filtering out torches.
Science and art both matter. We try to be accurate, clear and precise, and present the facts for people here.

"I don't think any of these filters need classification as "UV-only". They have always been offered as "UV-pass" filters or dual-bandpass filters (when not IR-blocked). Buyers can look at transmission charts and make up their own minds about what OD they want."

I think we do, and always have differentiated UV-only filters from UV+X+X filters.
Dual band is a whole other thing, and not what we are talking about here.
Once upon a time most people didn't accept the idea of a UV+Blue+Green range filter, such as the UG5/U330 + S8612 stacks.
What do we call those? UV pass filters? No, they are UV+Blue+Green pass filters, they are not classified as UV-pass filters.
"Buyers" may not understand what they are buying.

Edited by Cadmium, 05 November 2018 - 03:12.

#44 Andrea B.

    Desert Dancer

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

Posted 05 November 2018 - 06:24

The problem with mutually agreed upon scientific definitions is that that that biology does not always recognize them. While it is very useful for science to define a standard by declaring that 400 nm somehow separates ultraviolet radiation from visible in a continuous spectrum, try telling that to the human eye which can often see down to 370 nm. Indeed, visible violet is most often defined as those wavelengths in the interval between 380 - 450 nm. That is quite a lot more visible violet than has been discussed so far in this topic where everyone is quibbling over a mere 5 nanometers in the no-man's-land between 400 - 405 nm. If the standard definition of violet were applied, then all our commonly used broadband filters or stacks would need labeling as UV+Violet.
Andrea G. Blum
Often found hanging out with flowers & bees.

#45 Cadmium


  • Members+G
  • 3,691 posts

Posted 05 November 2018 - 06:46

400nm to 405nm is not mere. It makes a big difference in exposure time and color balance, and it may be a little more than just 5nm.
This is not about violet or the definition of violet. The point is about 400nm, which is the defining line between UV and visual as those are defined.

Ultraviolet (UV) is electromagnetic radiation with a wavelength from 10nm to 400 nm, shorter than that of visible light but longer than X-rays
Ultraviolet A UVA 315–400
Near ultraviolet NUV 300–400

Violet is the color at the end of the visible spectrum of light between blue and the invisible ultraviolet. Violet color has a dominant wavelength of approximately 380-450 nanometers.

This is why I keep telling you that violet is a poor word to use in this context, because it crosses the line between UV and visual. You have insisted that I use that word,
so the only way to use that word in this context is to divide into two terms. UV Violet and Visible Violet.
I would just as soon call anything above 400nm 'blue', in the context of the BGR visible range.
The best term would just be 400nm+ or even just 'visual'.
The definition between UV-only and not is the 400nm line.

Enough said.

Edited by Cadmium, 05 November 2018 - 06:52.

#46 Andrea B.

    Desert Dancer

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

Posted 05 November 2018 - 15:25

I understand and respect your comments. (((Except for the part about wanting to call 'violet' as 'blue'. I get complaints about that. :lol: )))

I rather like the idea of letting violet be 400-450 nm and labeling the visible colour under 400nm as something like 'violet-uv', similar to a labeling such as 'yellow-green' or 'red-orange'.

But...... why is 400nm the divider? Some references give 390nm and some give 380nm.

It is interesting to look at a solar spectrum for UV and visual only because it spreads the chart out a bit so that more detail is available.

In the attached chart (from a reptile website) we can easily see the case for 400 nm as a divider because that is where the solar spectrum takes a giant leap for sol-kind. But that is a physics/astronomy approach. If we take a biological human-vision approach, then 380nm or 390nm makes more sense as a divider between UV and visual. If we were to take a hypothetical bee-centered approach, then there is no divider. :lol:

So any one of 380/390/400 as a defining divider is valid when taken in proper context.

I tend to place the UV/Vis divider lower at 380 or 390 nm because a significant number of the floral UV-signatures have something going on which shows up best around 350-360 nm.

I've been trying to find this chart without the lamp overlay.
Attached Image: spectrumExoTerraSolarGlo160W.gif

Andrea G. Blum
Often found hanging out with flowers & bees.

#47 Reed F. Curry


  • Members
  • 348 posts
  • Location: New Hampshire, USA

Posted 05 November 2018 - 18:49

I think we have strayed from any meaningful subject. We don't need or want, imo, marketing phrases like "UV-only" in order to make our purchasing decisions. The purpose of superlatives such as "pure" are intended to provide a cover for impurity. :) For example, if you are buying a goose down jacket, you probably assume that one that is labeled as "white goose down" is superior to a jacket that is sold as "90% goose down." You would be wrong. The clothing industry has received a deferment from truth on the words "goose down". By law "goose down" must only be 75% goose down, not 100% goose down, as the makers hope you will assume. The other 25% may be cheaper duck down, waterfowl feathers, landfowl feathers, etc.

Just so, the notion of "UV-only" is an attempt, not to aid the consumer, but to mislead the consumer. Why does the UV-photographer require such an ambiguous term when he/she has access to a transmission curve? What does "only" mean? By law, terms like "pure", "full", and "only" are seldom permitted in advertizing of critical products. This is because "pure", "full", and "only" are states that are difficult to attain and prove.

"Only" is an absolute condition, there are not degrees of "only". No commercial UV-bandpass filter available for less than $1000, has an instant drop at 401nm to OD5 and continues or exceeds OD5 to 1100nm. That, however, is what UV-only means to me and probably others. We don't require ambiguous marketing phrases to make our purchasing decisions if the filter dealer provides solid data from testing.

Just one opinion, YMMV
Best regards,

#48 JMC


  • Members+G
  • 1,041 posts
  • Location: London, UK

Posted 21 November 2018 - 14:11

As promised, I re-visited my filter transmission measurements once my calibration standard for my spectrometer had arrived. It has and I have now recalibrated my spectrometer. As expected, it was out, slightly, but not by as much as I thought. Here are my two scans for the SEU II filter (before and after calibration of the spectrometer). Firstly from 280nm to 480nm.
Attached Image: SEU II after cal.jpg

And from 395nm to 410nm, again before and after calibration.
Attached Image: SEU II after cal 395 to 410.jpg

Finally here are the values from the calibration lamp. Two columns, first column the theoretical Hg peak positions for the lines, and second column the values from my spectrometer.
Attached Image: Cal results.jpg

I'm within +/-0.1nm in the region I'm interested in (330nm to 405nm) which is probably as good as I can expect from the device.
Jonathan M. Crowther


#49 Cadmium


  • Members+G
  • 3,691 posts

Posted 21 November 2018 - 19:18

About 40% at 400nm...