• Ultraviolet Photography
  •  

Mercury Hg lamp spectrum lens test - Sigma 30mm f2.8 EX DN MFT

18 replies to this topic

#1 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 13 April 2019 - 16:39

Gear used:
Olympus e-m5 body full spectrum converted myself.
Sigma 30mm f2.8 EX DN MFT

No filters
Attached Image: P4139249.jpg

#2 dabateman

    Da Bateman

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

Posted 14 April 2019 - 01:58

I think the two best Af lenses are the Sigma 30mm f2.8 lens and the Olympus 14-42mm f3.5/5.6 R2. Amazing the UV they allow.

#3 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 14 April 2019 - 08:08

My copy of Sigma 30mm needs repair :(

I hope to test the Olympus 30mm/3.5 macro.

There is potential for the older 4/3 Olympus 25mm/2.8 lens - it has the simplest design of 5 lenses in 4 groups.

#4 Cadmium

    Member

  • Members(+)
  • 2,280 posts

Posted 14 April 2019 - 08:24

Those are so pretty and everything, but if you don't put some numbers on those things they don't show me much.
If you are going to use gratings and lines, then don't leave it up to people to figure out what the lines mean, define it, delineate it.
I know you know what it all means, but for others, define it.
Don't just toss out an image that has no definition.
Without definition it is totally ridiculous.
Or get a spectrometer, which will do all of that for you.

These are colored lines, tell us what they are, put numbers on them.

Edited by Cadmium, 15 April 2019 - 03:06.


#5 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 14 April 2019 - 08:55

There is post in index sticky - lines
Left orange line is 365nm

#6 Cadmium

    Member

  • Members(+)
  • 2,280 posts

Posted 14 April 2019 - 09:21

You need to put numbers on ALL of these gratings to make it clear to EVERYONE what they mean.
Otherwise this is meaningless.
There are a few other people that have no clue, and no idea what the lines mean, or what the colors might mean,
and they are not going to go to your "lines" link to try to figure it out for each of your gratings.
If we are suppose to judge your gratings by color, then what are you white balancing on?
If you don't want us to judge your gratings by color, then number them.
If you number them then that is all you need, and it is self explanatory.
You have made about 5 new grating topics, and none of them are numbered.
If you want the general public to understand what they mean then I suggest you number all of them.

And you ignored Ninjin's question about white balance here:
http://www.ultraviol...dpost__p__26766

Edited by Cadmium, 14 April 2019 - 10:12.


#7 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 14 April 2019 - 09:37

You can just ignore whole Experimental Lens Spectra sub-forum or do better spectra than me. I do what I can, and what my equipment allows me to do. I use two common light sources with well-known lines of the spectrum, and they are explained in the index topic.

And about the white balance - I just did not recognize the question.



#8 dabateman

    Da Bateman

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

Posted 14 April 2019 - 11:30

Dmitry,
Your spectra do have value. But your not reporting the most critical information, that being the exposure you used to capture the spectra.
My 14-42mm f3.5/5.6 R2 lens can see with my 370bp15 filter. But more importantly its only 1 stop slower than the Sigma 30mm f2.8 lens. This make it useful for UV photography. I wouldn't have known that based on your noisey solar spectra of it.

Now that you have an Em5, you can capture images at 30 minutes. We don't know if your settings to get the spectra was 2 seconds or 2 minutes. Please update your images with the exposure settings.

Would be interesting to post side by side all at same exposure settings for the Sigma 30mm f2.8, 14-42mm R2 @30mm and the Olympus 30mm f3.5 macro. To get an idea.

Also it looks like you cut the above image off too much on the left. Can't tell if you saw any faint trace of the 335nm line.

#9 enricosavazzi

    Member

  • Members
  • 429 posts
  • Location: Stockholm, Sweden

Posted 14 April 2019 - 14:21

View PostDmitry, on 14 April 2019 - 08:08, said:

My copy of Sigma 30mm needs repair :(

I hope to test the Olympus 30mm/3.5 macro.

There is potential for the older 4/3 Olympus 25mm/2.8 lens - it has the simplest design of 5 lenses in 4 groups.
I tested the 30 mm f/2.8 Sigma and the Macro 30 mm f/3.5 Olympus just minutes ago, both with Baader U in sunlight, on full-spectrum converted Olympus E-PL6. Unfortunately small low clouds come in continuously today, so It was difficult to shoot with a constant source intensity, and my results are not reliable.

For what is worth, both lenses at f/3.5 gave me an exposure time of 1/10s at 1,000 ISO, and the image quality and false color are closely similar. I did not test the 30 mm macro in UV macrophotography, but it might perform as well as at infinity.

For shooting at infinity, the Sigma lens is probably preferable because it costs less, weighs less and its AF is much faster (both in UV and VIS).
-- Enrico Savazzi

#10 Ninjin

    Member

  • Members
  • 13 posts

Posted 14 April 2019 - 19:18

Cadmium,
Well, it's not really a question. I only said about settings.

Question, would be this: If the lens T-43 transmits to 336 (on the line), why not any 'green' band about 339, for the test zwb1 UV stack?

Such spectrometer, or the grating, or Sparticle tester, will show it. Just, if such settings as is, may be this result is not so obvious.

#11 Cadmium

    Member

  • Members(+)
  • 2,280 posts

Posted 14 April 2019 - 23:42

OK, not question. More of a white balance fix/adjustment.

I use my Sparticle, which is a set of bandpass filters. When white balanced with PTFE these bandpass filters have colors, for example green for 340BP10, and yellow for 360BP10, etc....
I usually always label the bandpass filters on shots to make it clear to anyone what narrow band range they represent.

I seldom use the grating method of defining transmission of lenses and/or filters. When I do I have matched the white balanced UV colors up with what I see with my Sparticle.
Thus, the depth of the UV transmission would shows in the width and the colors shown in the grating.
I have never labeled my grating shots, because mostly I have just compared them to the Sparticle.
If I were using lines to delineate gratings, I would certainly note the nm of those delineations along the bottom of the shot.
I understand that a full range grating such as UV+Blue+Green+IR is not going to white balance the same as a UV only shot. I have Sparticle shots showing UG5/U-330, with/without IR suppression,
and the white balance is different because of the wider range.
However, with UV only shots it would be helpful and more uniform to white balance those individually for that range, even if you also number the nm lines.
You can do it however you want, but for the general public to understand what they are looking a, it would be helpfull to show them nm numbers.

Edited by Cadmium, 14 April 2019 - 23:46.


#12 UlfW

    Ulf W

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

Posted 15 April 2019 - 08:21

Correctly identified lines in a grating shot is wavelength-wise far more accurate than any sparticle or colour-based method.
However the intensity of the lines is dependent on the intensity of the peaks in the lightsource and transmission and efficiency of the grating.
Ulf Wilhelmson
Curious and trying to see the invisible.

#13 dabateman

    Da Bateman

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

Posted 15 April 2019 - 10:34

Ulf,
And his exposure settings, which he doesn't indicate and should always. Did he use Iso 200, 2 seconds or Iso 800 and 5 minutes. We don't know and this is important.

#14 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 15 April 2019 - 10:35

Well, for now I just collected bare raw data for lenses, to determine what worth more detailed investigation. Testing conditions was not so stable, cloudy weather affect shutter speeds 10-100x times. I know, method is not perfect yet.

Recent spectra was taken at no more than 2 minutes exposure maximum at iso 200. I do like live composite on e-m5!

About cut images on left - nothing important, just lens flare or leaks from unstable test conditions.

I do consider to buy Olympus Macro 30mm f/3.5 instead of repairing my faulty Sigma 30mm.

Upcoming plans:
1. determine key parameters for testing
2. learn how to do stable results to compare lenses
3. test Viltrox EF-M2 speedbooster
4. more lenses in case I got something new

To be continued...

Edited by Dmitry, 15 April 2019 - 10:39.


#15 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 15 April 2019 - 10:54

Just in case admins prefer to split method discussion to separate thread.

First of all, equipment:
Nearly finalized choice is fullspectrum e-m5 without dustshaker. I have feeling it have more sensitivity to 334nm compared to e-pm1 with dustshaker. I do like many other handy features.
Spectrometer adapter - simple dumb DIY holder to attach spectrometer on filter threads on lens. Does not affect image in any way, maybe flares and light leaks sometimes.
Spectrometer itself - jewelry spectrometer containing grating and colimating lens. May limit UV reach, but does transmit 334nm

Light sources:
Sun :) unfortunately not stable and affected by weather
Mercury Hg lamp - very stable but have many gaps in spectra. Nothing between 334nm and 365nm for example.

Filters:
Cheap zwb and qb glass from China just to limit amount of light from unwanted spectra range. Always used same.

Lenses:
List to be discussed.

To be continued...

Edited by Dmitry, 15 April 2019 - 12:47.


#16 Dmitry

    Dmitry K.

  • Members(+)
  • 197 posts

Posted 15 April 2019 - 11:00

Image processing:
Most unclear for me. Because different angle of view and different magnification, same spectra on different lenses can have different width in pixels.

Do I need to crop just spectra and leave dark areas outside?
Do I need to scale it to same size? Which size?
Any other suggestions?


#17 UlfW

    Ulf W

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

Posted 15 April 2019 - 11:49

View Postdabateman, on 15 April 2019 - 10:34, said:

Ulf,
And his exposure settings, which he doesn't indicate and should always. Did he use Iso 200, 2 seconds or Iso 800 and 5 minutes. We don't know and this is important.
I just wanted to point out that there can be a very good wavelength accuracy using defined spectrum lines, compared to any filter method, but also some limitations.
Such lines are used to calibrate spectrometers and their wavelengths are known to small fractions of a nm.

A typical Sparcticle filter often has a peak wavelength tolerance of ±3nm, a ±50% width of 10nm or more and a much wider ±25% width than that.

There are limitations in all methods and by knowing them it is easier to interpret the results correctly.
I agree that it is important to include as much information about the setup as possible to make it possible to interpret the results.
Ulf Wilhelmson
Curious and trying to see the invisible.

#18 Cadmium

    Member

  • Members(+)
  • 2,280 posts

Posted 15 April 2019 - 15:52

View PostUlfW, on 15 April 2019 - 08:21, said:

Correctly identified lines in a grating shot is wavelength-wise far more accurate than any sparticle or colour-based method.
However the intensity of the lines is dependent on the intensity of the peaks in the lightsource and transmission and efficiency of the grating.

That is great, but without noting the nm of those lines on each test shot, you leave a lot of people out of understanding the test.
Put numbers on there... makes it a whole lot better.

Edited by Cadmium, 15 April 2019 - 15:55.


#19 Cadmium

    Member

  • Members(+)
  • 2,280 posts

Posted 15 April 2019 - 16:12

View Postdabateman, on 15 April 2019 - 10:34, said:

Ulf,
And his exposure settings, which he doesn't indicate and should always. Did he use Iso 200, 2 seconds or Iso 800 and 5 minutes. We don't know and this is important.

Not sure who you are referring to here, but usually these days when I do a Sparticle test, I shoot the first shot with the Kuribayashi 35mm, manual settings, using flash,
then shoot following lenses with the same setting. That gives me a direct real time Kuri reference to compare the other lens(s) to.

I compare everything to the Kuri, so far have not found anything better. Focotar II looks the same to me, but I prefer the built in focus of the Kuri for most things.

Edited by Cadmium, 15 April 2019 - 16:14.