Diffraction question

[The_Traveler]

If diffraction is an important issue at small apertures, how are macro lenses made to minimize that?

[MFloyd]

It's all about optic physics; there is not much you can do about it.  But there is a way to go around the problem: focus stacking.

[Bjørn Rørslett]

Yes and no. Diffraction cannot be avoided. However, some lenses are diffraction-limited in their performance and might deliver good results even well into the diffraction danger zone. The Coastal Optics 60 mm f/4 APO lens is a prime example. It can be stopped down to f/45 and the results, although clearly softer, are still highly acceptable.

[rosko]

If diffraction is an important issue at small apertures, how are macro lenses made to minimize that?

Any lens having a diaphragm will show diffraction if you stop down, macro lenses included.

However, if I need to stop down, I do stop down, as a lack of DOF is more annoying than a hypothetical diffraction effect, that being of course my own opinion.

[Roland Vink]

Diffraction is an edge effect, so a circular aperture opening will minimize diffraction since it has the shortest edge length. The same f-stop with a non-circular opening will have a longer edge length, and proportionally more diffraction.

Most macro lenses these days have rounded apertures, although I suspect that's more to improve the bokeh at wider apertures, not to reduce diffraction. By the time you stop well down the opening usually become polygonal so less than ideal, but with 7 or more aperture blades the difference is minor.

[pluton]

If diffraction is an important issue at small apertures, how are macro lenses made to minimize that?

Short answer:  They aren't.
Long answer:   They aren't, but some lenses seem to produce slightly more diffraction induced softness than other similarly spec'd lenses, and used under similar conditions.

[richardHaw]

i am basically a bug shooter so diffraction is something that i have to deal with a lot of time.

here is my study of the 105mm VR (my main bug lens) and you can see for yourself how much this lens can handle before the images become real ugly (when pixel peeped).

http://richardhaw.com/2016/01/24/study-diffraction-on-the-105mm-vr/

apologies for the images. I wrote this when the blog was new so i didnt know how to embed stuff. not only that, the images were too big to be uploaded as-as but just read the commentaries and take my word for it.

macro lenses are just like any other lenses with the exception that it has a really long helicoid and an optical design to match. like what was mentioned, diffraction is connected to the iris and i am not aware of any macro lenses that have special irises.

another thing to consider is that as you get closer, the effects of diffraction increases because of the magnification (dont quote me on this yet, im investigating but this seems to be the case). a lens in a bellows unit set to full will exhibit that. the cause might be due to the iris being too far from the film plane as well. i am not an optical engineer or expert so i am curious as well.

dont worry too much about it unless you absolutely need to show a bug's compound eye or the like in clear detail.

one more thing that puzzles me is that FX cameras get at least 1 stop more before diffraction starts to be obvious. i tested this on a D7000 and a D800 a long time ago so the pixel pitch is roughly the same.

[Les Olson]

If diffraction is an important issue at small apertures, how are macro lenses made to minimize that?

They aren't and they can't be.

If you are interested in the science there is a series of excellent tutorials on diffraction (and on many other things relevant to photography) at http://micro.magnet.fsu.edu/primer/lightandcolor/diffractionhome.html  (There is a fair bit of science involved, but anyone with some tertiary physics or good secondary school physics and a willingness to work at it should be able to follow). 

The skinny is that resolution increases as the aperture gets bigger: every decrease in aperture reduces resolution.  This is a result of the nature of light, and it applies to reflecting as well as refracting imaging systems (which is why astronomers always want bigger telescopes).  The catch is that lens aberrations also reduce resolution, and some aberrations improve as the aperture gets smaller (especially spherical aberration - related to the cube of aperture - but also coma, astigmatism and field curvature).  So as the aperture gets smaller it is a race between more diffraction making the image worse and less aberration making the image better.  A "diffraction limited" lens is one with no aberrations, and for practical purposes they do not exist, but folk talk about lenses that are "diffraction limited at f/5.6" and so on, meaning that at f/5.6 the diffraction effect is worse than the aberrations.  The bigger the aperture at which the diffraction effect is worse than the aberrations the better corrected the lens. 

Close focus means shallow depth of field, so you often need a small aperture to get adequate depth of field, but that is always bad for resolution.  Macro photography is a struggle with optical limitations. 
 

[richardHaw]

Macro photography is a struggle with optical limitations. 
 

100%

i never got this until i started shooting greater than 1:1

[Bjørn Rørslett]

The dedicated photomacrographic lenses should always be used "wide" open as they are designed for that use.

The technique of focus stacking has allowed wonderful solutions to the limited depth of field issues troubling us before. However no tree grows into the sky and stacking can only be of so much help, plus introduces separate issues.

[rolubich]

Long answer:   They aren't, but some lenses seem to produce slightly more diffraction induced softness than other similarly spec'd lenses, and used under similar conditions.

That is what I have found out testing my lenses (mostly AIS Nikon lenses) at small apertures: there are some difference between lenses, they are very small (I would say less than 1 stop, I mean that the best lenses at f/22 are not better than the worse ones at f/16) but still visible.

So what can make a lens behave better than another in the diffraction field?

I think that at small apertures (f/11 and smaller) the different optical quality of different lenses is very small compared to the deterioration of IQ due to diffraction. So in my opinion differences of IQ at small apertures is mainly due by different level of diffraction.

As diffraction is produced by an angular deviation, could the position of the iris diaphragm have impact on diffraction effect? Consider two lenses of the same FL at the same aperture, lens A has the iris diaphragm closer to the sensor than lens B, given that the light rays are bended by the same angle, rays of lens A should arrive at the sensor with a smaller displacement.

As far as I know aperure is determined by the diameter of the entrance pupil and not by the diameter of the physical aperture of the iris diaphragm. So at the same f/stop lens A could have a bigger physical aperture of the iris diaphragm than lens B and be less affected by diffraction. Could this be another reason for little differences between lenses in the diffraction territory?

[David H. Hartman]

one more thing that puzzles me is that FX cameras get at least 1 stop more before diffraction starts to be obvious. i tested this on a D7000 and a D800 a long time ago so the pixel pitch is roughly the same.

Might that be because the DX image must be magnified 1.5x to make the same size image be it print or Web display? Then again the DX image has more DoF so for DX you might not stop down as far as you would with FX.

Dave

[Bjørn Rørslett]

f-numbers are relative .... f/N really where f=focal length, N=aperture number. That'll take care of the absolute size.

The physical aperture opening is usually placed at the internal lens position where the ray bundle is near-parallel.

FYI, it is the exit pupil that acts as the film/sensor illuminating source on behalf of the optical system.

[rolubich]

f-numbers are relative .... f/N really where f=focal length, N=aperture number. That'll take care of the absolute size.

The physical aperture opening is usually placed at the internal lens position where the ray bundle is near-parallel.

FYI, it is the exit pupil that acts as the film/sensor illuminating source on behalf of the optical system.

Thank you Bjørn for the replay.

I was referring to two lenses of the same FL at the same aperture, having different position of the iris or/and physical aperture size.

So there is no much room for lens designer to move the iris closer to the sensor? But could it help?

Thanks for the correction about exit/entrance pupil.

[Bjørn Rørslett]

"So there is no much room for lens designer to move the iris closer to the sensor? But could it help?"

Affirmative on the first question, NO on the other.