Bluish image from old lenses?

[Roger Hicks]

I’m not really sure, I’ll reflect on it and do a doodle or too

Dear Stewart,

I'm not 100% sure, now you've raised doubts in my mind, but I still think I am, because I'm not sure that an interference coating can ever be anything other than dichroic. I am however open to being persuaded. I look forward to your reflections, even if some of them are accidental puns.

Cheers,

R.

 

[Armoured]

Best way to check would be to use slide film (of the right colour balance type, of course) for a roll and see. The photo with the flowers looks like a flare issue to me, and I have seen flare give a colour cast; but scanning is probably the more serious issue.

 

[Finder]

Color reflection in optical coats are cause by the fact that a coating cannot compensate for all wavelengths at once. So reflections from an optical surface (that is what coating eliminate) will have a color. both multi-coatings and single coating can give a color reflection, just the multi-coated optics will have less reflection. Since reflections make up such a small portion of the light transmitted through a lens, coatings have no effect on the color of the image.

RFaddict000, you are comparing a transparency film with a negative film--color can easily be corrected in the printing of the negative. The problem you are seeing is the flare in the optical system, not the film response.

 

[Papa Smurf]

Use a skylight filter

Use a skylight filter

I've certainly noticed it out of doors with uncoated lenses: veiling flare out of doors tends to be blue. Indoords, under tungsten, it's yellow, for the same reason. A lens hood (shade) will help quite a bit.

Cheers,

R.

I have often noticed the same affect with older lenses or cameras. Lens coatings are not my area of expertise, but I suspect it is due to the lack of coatings or ineffective early coatings applied to old lenses. The blue cast bothers me a lot so I prefer to use a Skylight 1A or B when shooting out of doors.

 

[RFaddict000]

I am going to do exactly that. I have just got a few rolls of Velvia100F and am going to test it out with a few filters.

Best way to check would be to use slide film (of the right colour balance type, of course) for a roll and see. The photo with the flowers looks like a flare issue to me, and I have seen flare give a colour cast; but scanning is probably the more serious issue.

 

[RFaddict000]

I think all three are negative film?

After reading all the responses, I am going to test out with Velvia100F and a few filters. That will probably settle the debate.

Also, the image with the flowers was taken in a mist. Think it is a factor?

RFaddict000, you are comparing a transparency film with a negative film--color can easily be corrected in the printing of the negative. The problem you are seeing is the flare in the optical system, not the film response.

 

[Finder]

I think all three are negative film?

After reading all the responses, I am going to test out with Velvia100F and a few filters. That will probably settle the debate.

Also, the image with the flowers was taken in a mist. Think it is a factor?

Sorry, my confusion with names.

 

[Sparrow]

OK, tried to measure the colour temp of light coming through an old Rolleiflex SC lens compared to a direct reading ... and failed

However looking straight through the lens my example looks gin-clear, but its reflections do have a definite blue cast.

All I can think of is this, a single coating can only cancel unwanted reflections at one point in the visible spectrum. So if it's say the blue end that needs correcting then a coating 1/4 of the wavelength of blue light would reflect a wave that cancelled out the wave reflected from the glass' surface.




... however moving towards the red end it would get progressively out of phase and start looking orangeish, this clearly would have no impact on light going through lens,

Personally I don"t think the fault is in the lens.

 

[RFaddict000]

OK. I edited the midtones in photoshop. I just moved red all the way to +100 and get this second image:
http://www.flickr.com/photos/chanhongsiu/4543346272

But still, need to do more real shooting to find out the cause.

 

[Sonnar Brian]

Pictures taken in mist will have a bluish cast.

The shot in Sunlight looked good to me.

 

[Finder]

OK, tried to measure the colour temp of light coming through an old Rolleiflex SC lens compared to a direct reading ... and failed

However looking straight through the lens my example looks gin-clear, but its reflections do have a definite blue cast.

All I can think of is this, a single coating can only cancel unwanted reflections at one point in the visible spectrum. So if it's say the blue end that needs correcting then a coating 1/4 of the wavelength of blue light would reflect a wave that cancelled out the wave reflected from the glass' surface.




... however moving towards the red end it would get progressively out of phase and start looking orangeish, this clearly would have no impact on light going through lens,

Personally I don"t think the fault is in the lens.

It may be worth noting: coatings don't filter transmitted light. They cause destructive interference to reduce reflected light.

In an optical diagram, entering light should come from the left, image plane to the right--your drawing needs to be flipped.

 

[umcelinho]

I agree that shooting a slide film would solve this doubt 🙂 easiest way to find it out, for sure.

 

[notraces]

This is a scanning issue -- period. I'm guessing you don't color manage your system? You might want to figure out how to manage your color -- from input (scanning) - to editing (Photoshop and your computer screen) to final output - whatever method you choose. I have no doubt that scanning the negs properly will remedy the situation. Why not shoot a roll of color film -- take it to a decent lab -- or send it off -- and have them scan AND print for you - let me know what you come up with.

 

[Roger Hicks]

It may be worth noting: coatings don't filter transmitted light. They cause destructive interference to reduce reflected light.

In an optical diagram, entering light should come from the left, image plane to the right--your drawing needs to be flipped.

Of a given wavelength.

Other wavelengths are part transmitted, part reflected. Hence the dichroic effect, of which UV/IR filters are an extreme example (after the IR has been reflected, the UV is then absorbed by the glass). Otherwise, after all, how do you account for dichroic filters which are, as far as I know, invariably produced by coating, and therefore do filter transmitted light? (I accept that I may be displaying my ignorance here).

As I said to Stewart, I'm not 100% sure, but I still think I'm right.

Incidentally, why 'should' entering light should come from the left? Surely this is a fairly empty convention?

Cheers,

R.

 

[Dave Wilkinson]

Incidentally, why 'should' entering light should come from the left? Surely this is a fairly empty convention?

Cheers,

R.

It all depends which hemisphere you live in.

 

[Roger Hicks]

It all depends which hemisphere you live in.

Dear Dave,

Seriously, Frances was once asked in Malta, by a sincere young American, 'Now we're in the other hemisphere, does the sun still rise in the east?'

Cheers,

R.

 

[Finder]

Of a given wavelength.

Other wavelengths are part transmitted, part reflected. Hence the dichroic effect, of which UV/IR filters are an extreme example (after the IR has been reflected, the UV is then absorbed by the glass). Otherwise, after all, how do you account for dichroic filters which are, as far as I know, invariably produced by coating, and therefore do filter transmitted light? (I accept that I may be displaying my ignorance here).

As I said to Stewart, I'm not 100% sure, but I still think I'm right.

Incidentally, why 'should' entering light should come from the left? Surely this is a fairly empty convention?

Cheers,

R.

You are speaking about two different problems. Coatings effect the reflected light, not the transmitted light. The light reflected off the lens and coating surface are out of phase and so eliminate each other. This has no impact on the unreflected (transmitted) light--coatings and glass do not reflect at a specific wavelength and so the balance of wavelengths transmitted are not changed.

Dichroic filters effect transmitted light.

I guess the convention on optical diagrams is silly. I feel the same about letter order in writing. ?rettam ti seod tahw ,naem I After all, mc2 = E.

 

[Sparrow]

You are speaking about two different problems. Coatings effect the reflected light, not the transmitted light. The light reflected off the lens and coating surface are out of phase and so eliminate each other. This has no impact on the unreflected (transmitted) light--coatings and glass does not reflect at a specific wavelength and so the balance of wavelengths transmitted are not changed.

Dichroic filters effect transmitted light.

I guess the convention on optical diagrams is silly. I feel the same about letter order in writing. ?rettam ti seod tahw ,naem I After all, mc2 = E.

I can model it in my head but it gets slippery when i try to type it

Like this, most of the light goes straight through, a small amount gets reflected from the surface of the glass and interferes with the following light-waves causing flare (that's what happens with un-coated lenses)

However if a thin coating is applied with the same reflective index as the glass (1/4 of whatever the nm of the lightwave is I think) then the reflection from the surface glass would be cancelled by an equal and opposite reflection from the surface of the coating at the interface of the coating and the air. In which case the following light-wave would be free of flare, yes?

But; frequencies in other than the one corrected some light is reflected (or refracted?) and that's the bit we see.

I always thought of diachronic effects as being structural like a grid or shutter thing, dependant on angles like polarisers.

Did I draw the light going the wrong way? I only just managed O level physics, sorry didn't realise there was a convention

PS I didn't draw the transmitted light in the diagram as it seemed irrelevant

 

[Roger Hicks]

You are speaking about two different problems. Coatings effect the reflected light, not the transmitted light. The light reflected off the lens and coating surface are out of phase and so eliminate each other. This has no impact on the unreflected (transmitted) light--coatings and glass do not reflect at a specific wavelength and so the balance of wavelengths transmitted are not changed.

Dichroic filters effect transmitted light.

I guess the convention on optical diagrams is silly. I feel the same about letter order in writing. ?rettam ti seod tahw ,naem I After all, mc2 = E.

As I say elsewhere, I am open to being convinced, and it's late now and I'm tired. But as far as I can see, you can't separate reflection and transmission. A quarter-wavelength coating can only be set up for the destructive interference of a specific wavelength, which is by definition suppressed. All other wavelengths are reflected or transmitted to some degree. The light passing through is therefore 'white light' minus the suppressed wavelength(s) - which is a colour shift.

Also, Leica makes specific reference to using coatings to regulate colour balance: Google it and you'll find it. I'll read it tomorrow when I'm wider awake.

Cheers,

R.

 

[Finder]

I can model it in my head but it gets slippery when i try to type it

Like this, most of the light goes straight through, a small amount gets reflected from the surface of the glass and interferes with the following light-waves causing flare (that's what happens with un-coated lenses)

Just to be on the same page and following your argument, let me go paragraph by paragraph.

The only flare coatings eliminate are reflections off optical surfaces. Coatings need a slightly different index of refraction to either air or glass or there will be no surface--no change in refractive index, no surface.

The entering light hits the first surface (coating), some of it is reflected (flare).

The transmitted light continues through the coating to the second surface, some of it is reflected (more flare).

BTW, this is true when passing through the other side of the lens.

However if a thin coating is applied with the same reflective index as the glass (1/4 of whatever the nm of the lightwave is I think) then the reflection from the surface glass would be cancelled by an equal and opposite reflection from the surface of the coating. In which case the following light-wave would be free of flare, yes?

Right.

Because the coating is a 1/4 of a wavelength thick, the light reflected off the second surface is 1/2 a wavelength behind the the light reflected from the first surface as it travels to the lens surface and back. The two reflected waves cancel each other out.

But; frequencies in other than the one corrected some light is reflected (or refracted?) and that's the bit we see.

Right, since a single coating can only be made for a specific wavelength, other reflected wavelengths will not be eliminated or eliminated as effectively. This is where multi-coating become ideal as a group of wavelengths can be dealt with.

I always thought of diachronic effects as being structural like a grid or shutter thing, dependant on angles like polarisers.

Dichroic filters, or interference filters, use destructive interference as well. But rather then eliminate the reflected light, they impact the transmitted light. Because the optical path length is critical with dichroic filters, the entering light has a very limited angle in which it can pass though the filter or the exiting light will not be at the desired wavelength. (If you put a dichroic/interference filter on the front of a wide-angle lens, you will see the center of an image a different color than the edges.) The structure of the filter is quite complex, but I don't believe there is a standard construction or material--it really depends what you are doing with it.

Did I draw the light going the wrong way? I only just managed O level physics, sorry didn't realise there was a convention

No problem, but I could not tell if you were showing the destructive interference of the light reflected off the surfaces or after they were transmitted.