UV Filter Issues

I am a real estate photographer. I use a D750 with the Nikon 20mm f1.8 lens. Recently I have noticed a considerable number of flare spots when shooting rooms with considerable sunlight. Certainly, I try to shoot angles that minimize the sun but its not always possible. I purchased a B+W UV Filter for the lens to see if it would help to reduce the flare and, in fact it has. However, I am noticing that the filter is giving the edges of the photos a greenish tint. Is this a normal occurrence with a UV filter? I have attached two photos. The first is with the filter and the other is without. Thanks!

This is the filter that I am currently using. - Hoya 77mm UV and IR Cut Filter

For a 20mm super wide angle, light hitting the edges of the frame would go through considerably more filter thickness. Perhaps that introduces the green tint??

 

BTW, I doubt that a UV or any filter will decrease flare. If anything, a filter might increase it.

 

A lens hood or some kind of shade can cut down the amount of flare, but obviously that is hard to apply with a super wide.

The Hoya UV+IR filter works on the interference principle. Multiple layers of coatings are spaced to block UV and IR while passing visible light. The transmission changes as the angle of incidence increases, which increases the effective thickness of the coating, hence the frequencies of light it blocks. A 20 mm lens is a very wide angle, which is probably responsible for the color shift at the edges of the frame. Increasing the path length would move the corner* value to longer wavelengths, impinging on the visible pass band.

 

I do not see the flare spots you describe. Flare spots usually occur when a light source, even outside the frame, reaches the front element of the lens. Particularly bright areas in the frame may result in veiling flare, which I do not see either. In either case, a filter can exacerbate flaring, because it is a flat surface parallel to the sensor, or simply another reflective surface in the light path.

 

You do have a lighting problem, and limited depth of field. For greater DOF, you need to stop down, and probably use a tripod. I find that a flash with a soft dome, aimed upwards, fills the room with light without shadows. The walls of the room diffuse the light like a big soft box. Another technique to try is a bracketed exposure, merged into an HDR image with tone mapping.

 

* Corner value means the wavelength where the transmission is reduced by about one stop. It's an electronic term with applications to optics or any other filter in the frequency domain (q.v., Fourier analysis).

Edited May 12, 2017 by Ed_Ingold

Normally, adding more layers of glass increases the likelihood of flare.

 

Differences in UV filters include the tendency for many of them to not be quite neutral. In a digital world, this is easily corrected.

 

I can't see much difference otherwise between your two pictures, and that is as it should be.

 

Over the years, many tests by the photo magazines tended to show that many so-called UV filters actually did practically nothing, but were just "protection" filters.

Thanks. I uploaded these photos just to show the comparison with the filter and without. There was no flare in either shot. My issue is with the greenish tint that is applied when using the filter. I do use a lens hood at all times. Thanks.

The point about a UV filter with high quality multicoating (e.g. B+W MRC or MRC Nano) is that flare is (in some situations) reduced relative to an uncoated filter, not relative to an unfiltered lens. If the filter is causing you other problems for whatever reason, take it off. Note also that even good manufacturers like B+W and Hoya make uncoated filters in addition to their (generally more expensive) coated ranges.

Take the filter off and through it in the trash. Modern cameras don't need a UV filter--there is one already built in over the sensor. Once upon a time I used a filter on my lenses but quickly discovered even the best coated ones were screwing up my shots as you describe. I've come to think they are totally useless.

 

 

Kent in SD

I suspect it's the IR cut "feature" that's the cause of the Cyan corners. It's difficult to get IR cut without some colouration creeping into the visible spectrum. Together with what Ed said about the oblique optical path at the edges of the image circle.

 

Ah well! A lesson to us all to avoid UV + IR filters on superwide lenses.

 

Incidentally the hoods provided with most lenses are pretty useless. Better than nothing - just. The way to prevent flares is to flag off the sun from hitting the lens with your hand or a piece of card. Obviously being careful not to get the flag in shot.

The Hoya UV+IR filter is not the same as a UV filter which simply absorbs UV, sometimes with a slight warming effect. Since the Hoya filter works on an interference principle, a high angle of incidence would move the UV and IR cuts to longer wavelengths. Consequently the less visible blue light is passed passed, and a greater amount of IR. Less blue means more green. More IR should be invisible, but in practice the sensor may render a green tint (false color) to shadows.. The effect of IR depends on the camera and sensor. It was pronounced in my D2H (green shadows) and probably the D200. Both cameras suffered from excessive IR sensitivity.

The Hoya UV+IR cut filter is not the same as a UV filter

A point some in this thread appear to have missed. I never had the issue the OP demonstrates using an UV filter and I use UV filter or clear filters on all my lenses.

The way to prevent flares is to flag off the sun from hitting the lens with your hand or a piece of card. Obviously being careful not to get the flag in shot.

+1, the OP just purchased the wrong tool for the job at hand.

Thanks all. Rodeo Joe is correct. I purchased the wrong filter. I had intended to purchase the B+W UV-HAZE which I use on my other lenses. I mistakenly order the IR cut. I have since ordered the correct filter. The UV-HAZE does a pretty good job of eliminating, or at least reducing flare on my interior shots. Thanks again!

Thanks all. Rodeo Joe is correct. I purchased the wrong filter. I had intended to purchase the B+W UV-HAZE which I use on my other lenses. I mistakenly order the IR cut. I have since ordered the correct filter. The UV-HAZE does a pretty good job of eliminating, or at least reducing flare on my interior shots. Thanks again!

 

 

Better than no filter at all? Flare is the main reason I avoid filters like a disease, with exception of a polarizer.

 

 

Kent in SD

Are you sure the UV+IR cut filter uses a dichroic principle Ed? All other Hoya filters are dyed-in-the-mass glass, and the only coatings applied are AR ones. In any case I'd see no need to use interference films for UV absorbtion. That could be left to the glass substrate.

 

The only thin-film camera filter I've ever come across is a Sony ND video filter. This works superbly because of its absolute neutrality and immunity to an oblique ray effect.

 

BTW, the following is from Hoya's description of the filter:

 

"PLEASE NOTE: The UV-IR CUT filter is intended for use with digital SLR cameras and cine cameras that do NOT have an IR filter installed in front of the sensor. Color balance effects can occur when this filter is used with cameras that have an IR filter installed in front of the sensor."

 

Nikon's DSLRs all come with an IR filter incorporated in the sensor AFAIK.

Edited May 14, 2017 by rodeo_joe|1

I'm not absolute sure, but I have an Hoya "Hot Mirror" filter for infrared, which has the same metallic appearance as the UV+IR filter illustrated in the B&H site. UV is pretty easy to remove. Glass alone cuts off at about 300 microns, but IR is best done with an interference filter.

 

Dichroic refers to a different refractive index for various planes of polarization. Hot mirror filters have an iridescent appearance similar to a true dichroic material like tourmaline.

Dichroic refers to a different refractive index for various planes of polarization.

 

You probably meant to say birefringent, but I don't see how it would be related to a "hot mirror" sort of filter.

You're right Ed.

I didn't read far enough through Hoya's description of the filter.

 

In fact, if you read to the bottom, there's a warning about using this filter with wideangle lenses. It seems it's really meant for astrophotography use.

 

FWIW, the word "dichroic" simply means two-coloured. This may refer to the interleaved layers of different refractive index material (or conductor/dielectric) used to create the filter. Alternatively, when used for strongly coloured filters, you can also see the complementary colour at a certain angle.

 

Wikipedia describes a dichroic filter as synonymous with a thin-film or interference filter.

We're getting into the weeds. Wikipedia gives a "second definition" which corresponds to my assertion. My physics books were written on parchment scrolls ;)

 

What matters is that the filter effect depends on the spacing of the reflective surfaces in the coating, which reflects light when the spacing is 1/2, 3/2, etc. the wavelength. As the angle of incidence (measured from perpendicular) increases, the effective spacing increases, shifting the reflectance to longer wavelengths. By Snelius' Law, if it isn't reflected, it is transmitted.

But if the cutoff wavelength is increased (further into the infrared), why then is there an effect in the visible region? With visible red light being reduced and a Cyan cast introduced.

 

I think we have to look at oblique planar wavefronts effectively "bunching up" for an explanation.

 

Whatever the physics of it. This type of filter just isn't really suitable for general use, and especially with wideangle lenses.

Sheesh! And I thought I was a nerd...;)

 

But I love it, guys. Never quit. I learn something new from you every day.:D

I have run into a condition like this and opted for Gelatin filters with the mount at a distance from the lens and large enough to cover beyond the angle of view of the lens. Also the use of rear lens element clips for Gelatin filters as long as their clean eliminates the similar issue.

But if the cutoff wavelength is increased (further into the infrared), why then is there an effect in the visible region? With visible red light being reduced and a Cyan cast introduced.

Less blue is more green. More IR often results in green shadows, due to false color.

 

It's amazing how sensitive digital cameras are to blue/green balance. Sample shots taken with an A9 (DPReview) rendered a blue ping pong table bright green, probably caused by metal arc lighting in the arena. A slight tweak while grading will restore the blue color (often at the expense of another color).

"Less blue is more green." - What?

The complement of blue is yellow, and the complement of red is cyan.

 

"More IR often results in green shadows" - Can't say I've noticed that effect. I've noticed a shift in blues toward purple due to IR sensitivity, and for many black materials to appear brownish. No green cast though.

 

Another consideration is that DSLRs already have IR cut filters over their sensors. So the fact that an additional IR cut filter passes a little more IR should have no visible effect.

 

No, the effect must be due to a peripheral/oblique absorption of visible red. It should be easy to verify that simply by looking through the filter at an angle.