• Ultraviolet Photography
  •  

AR coated full spectrum conversions

15 replies to this topic

#1 Cadmium

    Member

  • Members(+)
  • 893 posts

Posted 10 July 2016 - 07:05

There is a debate between converted camera sellers on eBay about AR coating the internal clear glass of full spectrum conversions.
The question of AR coated internal clear glass is something I have not seen discussed here, and I have been asked about this question from people shopping for these eBay cameras.
I think this questions needs to be discussed and delineated for people who are tossed to and fro by the claims of sellers.

Here are two eBay links, you can compare what each says about AR coating the internal clear glass:

1) AR good:
http://www.ebay.com/...R4AAOSw8-tWZRiw

2) AR bad:
http://www.ebay.com/...8kAAOSwepJXcKPF

#2 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 10 July 2016 - 15:11

[EDIT:
Apologies to any readers confused by my many edits.
I wanted to be very careful about what I said.
Last update: 10 July 2016 13:23 ESDT.]

There are several varieties of anti-reflective (AR) coatings used on photographic filters or filter glass.

Edmund Optics provides good information about how AR coatings work here: Anti-reflection (AR) Coatings

The point to be made is that it is very important to know exactly which AR coating is used on a particular filter or filter glass before purchasing a product making use of AR coatings. For what wavelengths does the AR-coating enhance transmission and what, if any, wavelengths does the AR-coating reflect?

Not all AR coatings reflect UV. There are AR-coatings which enhance transmission of UV.

I'm not sure why one vendor mentions magnesium fluoride coatings because MgF2 is used an an AR-coating to enhance visible transmission?

Either WG280 or Spectrosil would make a good UV-transmission window. The Spectrosil should be of a non-UV-fluorescent grade.

Summary:
(1) Type and properties of any AR coating used on filter glass should be fully disclosed by the vendor.
(2) Everyone should be aware that AR coatings can sometimes cause side effects depending on how the coating is applied and on how the coated glass is used.
(3) Type and properties of Spectrosil (or any other fused silica materials) should be fully disclosed by the vendor.
(4) Until the Ebay vendors linked above provide complete information supporting their claims and until the Ebay vendors curtail their hyperbolic, misleading advertising, my advice is to stick with the established conversion vendors such as LifePixel, Kolari Vision or MaxMax.


[DISCLAIMER: UltravioletPhotography.com is not associated with any vendor of photographic products. We derive no income from any vendor of photographic products.]


****

Addendum: We need to find out exactly what kinds of glass are used by the various conversion vendors.
Andrea G. Blum
Often found hanging out with flowers & bees.

#3 enricosavazzi

    Member

  • Members
  • 374 posts
  • Location: Stockholm, Sweden

Posted 10 July 2016 - 20:07

In qualitative terms, one can generalize by saying that an AR multicoating can be designed for a specific purpose. In cases relevant to our interests, we can summarize as follows:
  • Camera lenses designed for recording VIS have AR coatings optimized to transmit most or all the VIS spectrum, i.e., they have a relatively flat transmission curve across the VIS range, which tapers off at either end of the range (especially the UV).
  • Coatings optimized for narrower ranges (e.g., UV only) can be designed with a higher transmission (or efficiency, i.e. lower reflection/absorption) in a narrower wavelength range.
  • Coatings for multispectral lenses must have a transmission range broader than the VIS range. Their overall transmission/efficiency is lower than those optimized only for the VIS range (or narrower ranges).This is the case, for example, of the Coastalopt 60 mm Apo. The lower efficiency of its coatings, compared to a VIS-only lens, is discussed on the Coastalopt site, if I remember correctly. I am quoting for memory and can be wrong, but I think that that discussion cites 99% transmission for the AR coatings of a typical VIS-only lens, and 90% for the CoastalOpt 60 mm Apo.

-- Enrico Savazzi

#4 Cadmium

    Member

  • Members(+)
  • 893 posts

Posted 10 July 2016 - 21:43

Thank you Andrea and Enrico, very well thought out.
A typical AR coating might be one 'centered' at 550nm, to be used for the usual visual range, however AR coatings can be 'centered' at other wavelengths, such as perhaps 350nm (or other) in the UV range.
I don't know if there is some kind of AR coating that is non-discriminating about what wavelength it transmits or not.
The idea is that the coating is anti reflective, and more so at the 'center' wavelength. I don't know if that would mean it blocks (or reflect) anything outside that center range any more than the uncoated glass would alone.
So, would an AR coating 'centered' in the UV range reduce transmission of the visual and IR range compared to the uncoated glass.
If the internal glass is AR in the UV range, and this somehow reduces the visual and IR range transmission, then this might be a concern.

#5 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 10 July 2016 - 23:31

In theory, anti-reflectivity coatings enhance transmission of the wavelength they are designed for by reducing reflectivity of that wavelength. The coatings may or may not affect other wavelengths outside the design parameters. It depends on the coating type and how it works together with the underlying glass substrate in the filter. [This is mentioned in the above Edmund Optics link.]

The puzzling aspect of using AR coatings for a clear glass sensor cover in a full spectrum camera is that in such a camera we would naturally want equal transmission of all wavelengths within say 300 - 1000 nm, approximately. But the vendor is offering two types of AR coatings: UV-Vis and Vis-IR.

Then to confuse things further, as Enrico has noted, the AR coatings on the CO60 lens reduce transmission to promote better transmission-without-reflectivity. That is a good thing for a complex lens where we do not want reflections playing through the various elements. But I cannot see why AR at the cost of such reduced transmission is a good thing for a full spectrum sensor cover?

It is difficult to comment further because the vendor does not disclose the type of AR coatings in use or the manufacturer of the AR coatings except to say that the buyer has a aforementioned choice of UV-Vis or Vis-IR anti-reflectivity coatings for their full spectrum conversion.

According to what I have seen so far, there is no "universal" AR coating useful for all wavelengths. Again see the Edmund Optics link above for a list of AR coatings in common use. However, the implication from Coastal Optics is that such a thing is possible. Perhaps it is permissible to combine different types of coatings to cover all wavelengths?

*****

I've managed to become even more confused by pondering a coating which is designed to enhance transmission by reducing reflections and in the process also reducing transmission.
Andrea G. Blum
Often found hanging out with flowers & bees.

#6 Cadmium

    Member

  • Members(+)
  • 893 posts

Posted 11 July 2016 - 00:57

Andrea, I am unaware of any AR coating that is full spectrum, except for what Enrico describes on Coastal Optics 60mm, for example.
If I understand those percentages correctly, then both the 99% and the 90% are higher than the transmission % the glass has with no coating at all.
but i am not sure that thinking is correct. Perhaps Enrico can explain that more to me.
In my thinking, the AR coating can have two reasons for use, not only higher transmission (at least at some desired range), but also as a protective coating for the glass surface.
I have never heard of AR coating the full spectrum glass. I don't know what such sellers as LifePixel and MaxMax do.

AR coating a UV-only filter (or conversion) at some designated UV peak wavelength of that filter stack would make sense (yet more expensive), but I don't see this as being good idea for clear glass in a full spectrum conversion, nor for non dedicated filters which might be used for different ranges of the spectrum in varied stacks.
I am no expert on AR coatings, and I would like more information on this question.

This question is confusing people out there.

Edited by Cadmium, 11 July 2016 - 02:20.


#7 OlDoinyo

    Member

  • Members(+)
  • 454 posts
  • Location: North Carolina

Posted 11 July 2016 - 02:35

Another issue: With some older lenses that have small, far-back rear elements, the angle of departure to the sensor corner deviates far enough from perpendicular that significant dichroic angle degradation can happen, resulting in unpleasant ring artifacts which are very difficult to remove from the image, just as hotspots are.

#8 Cadmium

    Member

  • Members(+)
  • 893 posts

Posted 11 July 2016 - 02:57

Clark, Interesting thought.

Original internal UV/IR BG-glass type cut filters sometimes have various coatings, but I don't know what different coatings they have, and of course those would be specific to the visual range.

Edited by Cadmium, 11 July 2016 - 02:57.


#9 Reed F. Curry

    Member

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

Posted 11 July 2016 - 13:31

Such a complex subject!
If I was using a camera for strictly UV, then I might think it best to have MgF2 sputtered on the camera's fused silica or quartz cover glass. Using a particle size of 87.5nm, the MgF2 would give me optimum transmission at 350nm. However, I presently have some glass so treated for UV-bandpass filters. The coating increases the 350nm transmission by 2%. But, upon reflection (no pun), I have determined that weighting the transmission in favor of any particular wavelength within the UV might give a less than accurate rendering of the subject. The typical UV bandpass glass already has a Bell curve of transmission. Optimally, we want a perfect 100% transmission starting at, say, 300nm, with an abrupt cut-off at 400nm. We can dream. So, for UV alone, is there any benefit to coating? IMO, no amount of creative AR coating is likely to do more than add complexity to an already complex representation of air. For this is what we want, isn't it, to have our filter, lens, and cover glass as transparent as air for the wavelengths we wish to capture?

The lenses I prefer for UV are all pre-1971 -- the start of multi-coating. The early coating of lenses allowed more glass/air transitions without increased flare. However, I still like to keep the glass/air interfaces to six - three groups - or eight - Tessar, four groups. I digress.

Personally, I see no benefit in coating of the cover glass. Keeping the shaker makes even less sense for UV shooting, as it simply adds two more air/glass interfaces, with perhaps 8% more loss.
Best regards,
Reed
http://www.uvroptics.com

#10 Cadmium

    Member

  • Members(+)
  • 893 posts

Posted 11 July 2016 - 18:56

Reed, Thanks for that insightful information.
'Cover Glass', good term.
My feeling is that the cover glass should not be coated, would that be the general consensus? I wonder why the cover glass is being coated?

#11 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 11 July 2016 - 22:01

I think I would be leaning towards no AR coatings on a sensor cover glass. But - unless we could perform an experiment using a camera first equipped with an uncoated WG280 full spectrum filter (or other WB glass or Spectosil) and then with a coated version of the same, we cannot make a definitive statement. Maybe I will think about performing this experiment with my old D300? It is fairly easy to open. I've done it multiple times. And that D300 is not currently in use, so if I borked it no prob. And I'd have to figure out where to get the glass.
Andrea G. Blum
Often found hanging out with flowers & bees.

#12 Reed F. Curry

    Member

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

Posted 12 July 2016 - 00:01

Another argument against coating the cover glass is the issue we have with any thin-film interference (dichroic) filter -- the wavelength shift when the incoming light is not at 0AOI. It would mean that wide-angle lenses would adversely affect the image quality. See http://uvroptics.com...ex.php?dichroic
Best regards,
Reed
http://www.uvroptics.com

#13 enricosavazzi

    Member

  • Members
  • 374 posts
  • Location: Stockholm, Sweden

Posted 12 July 2016 - 12:31

View PostCadmium, on 11 July 2016 - 00:57, said:

...
If I understand those percentages correctly, then both the 99% and the 90% are higher than the transmission % the glass has with no coating at all.
but i am not sure that thinking is correct. Perhaps Enrico can explain that more to me.
In my thinking, the AR coating can have two reasons for use, not only higher transmission (at least at some desired range), but also as a protective coating for the glass surface.
...
Yes, in general terms, within the optimal wavelength range, transmission is always higher (i.e. reflection is lower) with coatings than without. Perhaps an uncoated camera lens might transmit 50-70% of incident light, depending on the number of elements.

It is also correct that modern AR coatings are very hard, significantly harder than the optical glass they cover. So they can also be seen as a protection of the front and rear optical surfaces. However, the thickness of the outermost layer of a modern coating is on the scale of 100-150 nm (lambda/4), so even minute changes in thickness caused by wear (I am thinking about the kind of wear caused by habitual wiping, which affects a significant surface, not about isolated micro-scratches) are bound to affect somewhat its optical properties.

Really old AR coatings were obtained by chemically altering the glass surface, and this produced irregular and non-repeatable results (more like the variegated iridescence of oil on water than the evenness of modern coatings). Some of these coatings were also very soft, and easily removed by wiping the lens. Still, they were better than no coatings at all. The earliest multilayer AR coatings were uneven, and often shifted in color from purple to blue from one side of the lens to the other because of uneven thickness of the layers.
-- Enrico Savazzi

#14 Shane

    Member

  • Members
  • 255 posts

Posted 12 July 2016 - 15:38

Quote

Another issue: With some older lenses that have small, far-back rear elements, the angle of departure to the sensor corner deviates far enough from perpendicular that significant dichroic angle degradation can happen, resulting in unpleasant ring artifacts which are very difficult to remove from the image, just as hotspots are.

Try the free download of CornerFix.

#15 D.Lafon

    Member

  • Members
  • 10 posts

Posted 23 September 2017 - 06:12

View PostAndrea B., on 11 July 2016 - 22:01, said:

I think I would be leaning towards no AR coatings on a sensor cover glass. But - unless we could perform an experiment using a camera first equipped with an uncoated WG280 full spectrum filter (or other WB glass or Spectosil) and then with a coated version of the same, we cannot make a definitive statement. Maybe I will think about performing this experiment with my old D300? It is fairly easy to open. I've done it multiple times. And that D300 is not currently in use, so if I borked it no prob. And I'd have to figure out where to get the glass.

Andrea,
I happen to have a piece of AR coated Spectrosil that was purchased from "Vendor 1" referenced in the first post of this topic. You, or anyone else on this forum, are welcome to borrow it for experiments if you'd like. I don't recall the dimensions of the piece off hand, but I had ordered it in a custom size to replace the sensor cover glass in Fuji X cameras.
Cheers,
Daniel

#16 D.Lafon

    Member

  • Members
  • 10 posts

Posted 23 September 2017 - 07:23

View PostReed F. Curry, on 11 July 2016 - 13:31, said:

... Personally, I see no benefit in coating of the cover glass. Keeping the shaker makes even less sense for UV shooting, as it simply adds two more air/glass interfaces, with perhaps 8% more loss.

Hello Reed,

In the Fuji X series cameras that I am most familiar with, the dust shaker glass and the sensor cover glass are two separate windows, just as you indicate. The original sensor cover glass could be removed all together, and the replacement window chosen for the conversion can be cut to size and fitted in place of the dust shaker window instead. This way no extra air/glass interfaces are added, and as I found was the case in the latest Fuji X camera that I converted, the shaker glass is held snug against a full perimeter foam dust seal by a metal frame that is secured with screws, so the sensor surface is still sealed from the environment to help prevent contaminants from getting between the sensor and the newly installed window. I am curious if this is the method that the vendors who retain the dust shakers use as it does seem to have some practical and economic advantages.

Also, I hope folks do not mind me resurrecting this older topic. :)

Thanks again,
Daniel