Which lens to use for macro work?

[Mark Sawyer]

I'd like to do some macro-work on my 8x10, but don't have the luxury of a Luminar. The subjects will come out from almost-actual-size to up to 3x actual size on the film. Could anyone tell me if any of these lenses would be appropriate? (I'm listing some that I know won't cover 8x10 at infinity as the bellows will be out enough to increase coverage.) Also, will there likely be a focus shift or similar problem at these magnifications? Is there a rule-of-thumb for when to turn the lens backwards? (I've read past threads on this, but none quite give me the info I need.) Thanks!

Likely suspects for a macro lens are:

80mm f/2.8 Zeiss Biometar

90mm f/6.8 Angulon

121mm f/8 Super Angulon

135mm f/5.6 Componon enlarging lens

150mm Rodenstock Rodagon enlarging lens

150mm f/9 Eskofot Ultragon

159mm Wollensak EWA

210mm f/6.8 Goerz Dagor

215 f/4.8 Acuton

Note: I tried a 3xlife-size two-day (!) exposure with the 121mm Super Angulon @ f/64 which came out rather fuzzy, although the image had looked sharp on the ground glass @ f/8. The camera was on a big studio majestic tripod with two additional tripods under each end for support, so I doubt motion was a factor. Any thoughts on what went wrong?

[Gem Singer]

Hi Mark,

You didn't include the Nikon AM 210 f5.6 ED Macro on your list. It's a relatively expensive lens, but it's maximized for the type of photography you are attempting .

[Jon Shiu]

Film can slip or warp during long exposures.

[Ernest Purdum]

For 3X magnification, you need a minimum of four focal lengths bellows extension, so you need to be careful that your lens is not too long for your bellows. The 210mm Dagor you mention would need over 33" extension. On the other hand, to get close to 1:1 ratio, you only need close to two focal lengths extension. Assuming you are using 4X5 film, most lenses of 75mm or more would cover satisfactorily. My suggestion would be a 75 or 105mm lens and shutter from the Polaroid MP-4 system. These are still available very cheaply on eBay, although the prices do seem to be increasing rapidly.



Only lenses which are not entirely symmetrical need to be reversed. You turn the lens around when the bellows extension exceeds the distance from the lens to the subject. Another way of putting it is that you reverse when the image on your groundglass is larger than the subject.



If you go back to the home page and look at the index of articles, you'll find more information on this subject.

[Ernest Purdum]

Regarding your fuzzy result, maybe it was subject movement or floor vibration. I once was attempting to make extreme macro photographs, and was startled to find that I could tell on the groundglass every time a truck came by.



The problem might have been optical, though. I have never tried a Super Angulon at f64 in the macro range, but I wouldn't necessarily expect good results.

[Emmanuel BIGLER]

For a 2x to 3x magnification range a reversed 135 or 150 mm enlarger lens should be OK. Those lenses cover 4"x5" used in a an enlarger, so reversed and used as a macro lens they will certainly cover 8"x10" @2x and 12"x15" @3x. At 1:1 a 150 enlarging lens is not at its best but will certainly cover more than 4"x5". This is to be tested.

The total distance beween lens and film will be : ( M+1) x f , where M is the magnification ratio. So @3X with a 150mm lens you'll need 4x150 = 600 mm (2 feet).

But for 1:1 you could consider a lens covering 8"x10" at 1:1 ratio. For this I would suggest a classical apo-repro lens like the apo ronar. An apo ronar covers 48° according to the manufacturer's specs (a bit conservative certainly) i.e. in diameter 90% of its focal length in infinity-focus.
The image circle is roughly doubled @1:1. So very probably a 210 or 240 4-element symmetrical repro lens will suit your needs. @1:1 distance between lens and film with a 240 will be 480 mm.

[Joerg Krusche]

Hi,

take the Eskofot Ultragon, this lens is optimized for 1:3 to 3:1 range, no need to reverse due to symmetrical design.

Best regards

Joerg

[Bob Salomon]

Best reults would come from a macro lens.

The 150mm and the 180mm Apo Macro Sironar both more then cover 810 at 1:1 and deliver spectacular results of 3-dimensional originals at ratios of 1:5 to 5:1.

[Mark Sawyer]

Thanks, guys! I think I'll play with the Eskofot-Ultragon, as Jorge suggested. It sounds like it's optimized for just what I want to do. Thanks, Jorge! I'll eventually post the results, but it may be a week or two...

Vibration and subject or camera movement were *not* a factor with the 121mm SA. (Trust me on this one.) Could have been film movement or warpage, as Jon suggested, or maybe the SA just doesn't function as a macro at small f/stops, as Ernest thought. There was enough depth to the subject that if it was focus shift, something else would have come into focus, but the whole image was out-of-focus.

If you have the patience (and a stable studio floor and tripod), it's a somehow satisfyingly different approach pulling the darkslide on Tuesday afternoon and putting it back in Thursday night...

[Jim Galli]

Mark, my vote would be for the Ultragon also, then the Componon. So if the exposure took 2 days what was the fstop? Beyond about f128 diffraction will begin to make your exposure look like a pinhole pic. Ansel tried to explain that to Eddy Weston but he didn't listen. With the bellows draw for a 3X life size that eskofot will be at f64 wide open. Reciprocity is ok, but diffraction will get you.

[Mark Sawyer]

The 121mm Super Angulon was set at the f/stop marked f/64, but with a bellows draw of around 21", it was pretty much a pinhole.

I've never thought about it this way, but is this a workable way to figure a "real" f/stop, rather than do a bellows extention factor? Let's see. Get out the calculator...

f/64 on 121mm would be 1.89mm.

So I had a 1.89mm f/stop, but on a lens about 540mm away. 1.89 into 540 gives me f/286.

That's just over 4 f/stops darker than f/64, or a bit over 16x less light. (f/256 would be 16x, f/360 would be 32x)

The usual fellows-squared-over-focal-length-squared gives 19.9x less light.

Hmm, it seems to work. Never figured it that way before. Thanks, Jim! Of course with a two day exposure, it's the reciprocity failure that gets you...

BTW, I just made a new lensboard for the Ultragon this afternoon. (Paint's drying as I type this.) One of those $30 purchases-of-opportunity I never knew why I made. Maybe I just figured it out...

[Jim Galli]

Mark, You've got it. At 540mm you could have taken the glass out of the lens and the 1.89mm would have made a fine pinhole. :-0

[Emmanuel BIGLER]

Even if f/286 looks like the numerical aperture of a pinhole, using a lens like Mark did still delivers a much better and much brighter image! The glass is still there !
When placing the film at 540mm from a pinhole, do not forget that the optimum pinhole diameter, the one delivering the best image quality is given according to the excellent book by Leslie Stroebel by

diameter(in mm) = (1/28) x (distance (in mm))^(1/2)
and the corresponding numerical aperture will be : 28 x (distance in mm)^2

for a distance of 540 mm the best aperture accordig to Stroebel's formal is about 0.8 mm and the corresponding f-number is about 650.

So do not worry your image is still MUCH better in quality and 5x (650/286) MUCH brighter than an optimum pinhole image for a distance of 540 mm;-);-)

[Andy Eads]

Mark, I'm a little late to chime in but here goes. The main reason your image is fuzzy has more to do with diffraction at the effective f/stop of 256. No matter how sharp the lens is, once you stop down past a certain point, diffraction limits how sharp the image will be. Edward Weston complained to Ansel Adams that his close up images were not sharp. He had the same problem you do. Adams advised him to open up the lens a few stops to make the image sharper; it worked. As a crude rule of thumb, I stop the lens down to not smaller than f/32 effective aperture. The image you focused at f/8 (aprox. f/16 effective aperture) may be about as sharp as it can get.

[Mark Sawyer]

Never too late to chime in, Andy, unless they change the laws of physics every so often... This is an area of interest for a lot of us, and I really appreciate everyone's input! Here are my areas of curiosity right now:

1.) Yes, you lose resolution to diffraction at much smaller f/stops, but you also lose depth of field as you open them up. Depth of field is at a premium in macro work. Is there a "rule of thumb" for when diffraction causes an image to be unuseable? (I don't mind a little loss, as I'm only contact printing an 8x10, and diffraction seems to be regarded as much less of an issue with b/w film.)

2.) Does the distance to subject or formulation of lens affect the degree to which diffraction degrades an image? I ask this noting that my 480mm Apo Ronar is quite sharp at f/256. (Yeah, I just HAD to shoot my first exposure through it stopped ALL the way down!) Does the apochromatic nature of the lens minimize diffraction?

3.) Would using a polarizing filter or monochromatic filter (say a red 25a or deep red 29a) filter reduce diffraction?

I muchly await any insights the group may have, as this is an area in which I have profound ignorance.

[Jim Galli]

Here's how I remember. LPPM descends in number very roughly equal to reciprocal f stops. This is too easy and the experts will shoot it full of holes but still it's helpful for what it's worth.

f22-64lppm
f32-45lppm
f45-32lppm
f64-22lppm
f90-16lppm
f128-11lppm threshold of human sight
f180-8lppm
f256-5lppm

Though it's imperfect it's easy enough to remember on the fly so I can do the mental gymnastics on wether to use a sheet of film or not. I was playing with some "product shots" this weekend past. Bellows at 360 with a 240 lens. I tested about 15 or 16 lenses over 20 exposures. Only have half the negs developed so far. The clear winner so far is a 240 G-Claron Dagor type. Unprocessed contenders are 2 other Artars and 2 more Dagors though. The 135 Ultragon was a heavy hitter and suffered from my poor focus. Photog error!



http://tonopahpictures.0catch.com/Antiquelenses240G.JPG

Frustratingly bad neg scan. I printed this today in 16X20 and it is among the most lovely photos I've ever printed. Tonality just stunning. Done in 4X5 on 32 asa Panatomic X aerial film.

[Emmanuel BIGLER]

Jim's figures are perfect and perfectly fit the simple formulae where N_eff is the effective f-number :

- diffraction-induced cut-off frequency in lp/mm : 1000/(0.7 x N_eff)

- corresponding cut-off period in microns : 0.7 x N_eff

This is consistent with the theoretical value, cut-off frequency = lambda x N_eff where N_eff is the effective f-number taking into account bellows extension.



The visible spectrum goes from blue @0.4 microns to red @0.8 micron and taking 0.7 as the average value has the advantage of being not too optimistic and to fit well within the conventional series of f-numbers ; 1/0.7 = 1.4 the traditional aperture ratio between one click and another. Moreover if you try to fit the computed MTF values for top-class diffraction-limited lenses (e.g. apo enlarging lenses) with the very simple above mentioned formulae, you find that the values fit with the 0.7 value as well.

Note than N as engraved on the lens, should be replaced in macro by the effective f-mumber N_eff equal to N x (M+1)
where M is the magnification factor. When M+1 at 1:1 ratio, N_eff = 2N. The same (M+1) factor isued in the bellows exposure factor, (M+1)^2

The M+1 factor is also equal to the ratio of the total bellows drag to the focal length.

So now answers to Mark's questions are easy.

Is there a "rule of thumb" for when diffraction causes an image to be unuseable? (I don't mind a little loss, as I'm only contact printing an 8x10...

For far-distant objects manufacturers suggest a best f-stop. Say : f/22 for 4"x5" lenses. It means that at f/22 residual aberrations contribute to the same part as diffraction in the final image qualiaty. From this recommended best f-stop, you may stop two clicks more before things become really unacceptable.

In contact print the eye's resolution limit as seen from the conventional 250 mm (10") distance is somewhere in between 5 lp/mm (for most of us) and 10 lp/mm (for air force pilots ;-)
So let us take 7 lp/mm, this is simple : 1000/(0.7 x N_eff) = 7 ?? => yields N_eff = 200 as the maximum acceptable effective f-stop ; it means f/100 as engraved when shooting at 1:1 magnification. Say : f/64 engraved on the lens to be slighty more stringent and as a reference to the Old Masters ;-);-) and actually compute : f/128 @1:1 ratio.

Does the distance to subject or formulation of lens affect the degree to which diffraction degrades an image?

Formulation of lens : see below, unfortunately there is no way to avoid diffraction. Distance to subject, directly, no, but indirectly the distance to the subject determines the magnification ratio M hence the effective f-number Nx(M+1). In a situation where your magnification ratio is above 2:1, microscopists usually refer to a diffraction limit as measured on the object and not in the image. Simply because the effective numerical aperture as seen from the object goes to a limit equal to N atlarge magnifications. So instead of speaking of a diffraction limit of 0.7 x N (M+1) in image space you can also refer to a diffraction limit in object space, M times smaller, equal to 0.7 x N (M+1)/M ~= 0.7 N when M is large.The classical cut-off period in microscopy, measured inobject space is equal to : lambda / (2. n. sin(alpha)) where n is the refractive index of the entrance medium and alpha the angle between the optical axis and the edge of the aperture stop. This angle can be quite large in high-power microscopes but is susually small in macrophotography. For aficionados of Lord Rayleigh, expressed in terms of cut-off period there is no 1.22 factor ;-);-)

Does the apochromatic nature of the lens minimize diffraction?
A well-corrected lens delivers superior image quality hence blurring by diffraction appears for smaller f-stops than a fuzzy lens for which non-corrected aberrations will hide the diffraction effects until you stop down really "heavily". And I would say as a joke, an excellent apochromatic lens apperas to be plagued by diffraction more than others ;-) I recommend to look at the MTF charts for the apo-ronar 240mm as scanned on Paul Butzi's web site (Thanks to Bob S. for providing the printed material).


http://www.butzi.net/rodenstock/apo-ronar/p12.htm (http://www.butzi.net/rodenstock/apo-ronar/p12.htm)

Usually people say (and they are right of course) that an apo-ronar is optimized for 1:1 ratio. So you'd expect that the MTF curves are much better @1:1 ratio than @1:20. No ! Since the apo-ronar 240 is virtually diffraction-limited @f/22 and @1:20 ratio, keeping the engraved aperture to f/22 and working @1:1 ratio yields f/45 as effective numerical aperture, and consequently MTF curves drop dramatically only by cause of diffraction ! Conversely the useable image circle is doubled, so in fact the total number of usable pixels in the image is exactly the same, the global image quality is exactly the same when counting all available good image elements that you can record in the whole format. In fact the global image quality is probably slightly less @1:20 if you count all possible pixels but not significantly to prevent LF photographers to swear by apo-ronars even at infinity !! Ah why did 'they' discontinue our beloved apo-ronars ;-);-)



Would using a polarizing filter or monochromatic filter (say a red 25a or deep red 29a) filter reduce diffraction?

Polarizing ? no, certainly no effect ; Monochromatic ? yes but you should prefer the blue end of the spectrum to the red one !! blue => lambda = 0.4 mu, red => lambda = 0.8 mu, thhe diffcation spot is twice as big for red than for blue... in theory and for a perfectly aberration-free lens !! But in practical photography I doubt that you'll see anything significant. First because photographic lenses are optimized for a broad band spectrum of visible light and even in B&W you insist on getting all colors rendered in shades of grey as the eye would see them through a green or yellow filter. A monochromatic blue filter would yield contrast effects similar to old non-panchromatic films, so this has been the rule for decades in the early days of photography !

However reducing the wavelength for precision photo-lithography is the only solution to increase the density or electronic circuits on silicon. Current technology based of 365 mercury arc lines delivers the absolute physical limit of 360 nm in terms of periodic images (line width of 180 nm separated by a 180 nm space) and newer equipments with new UV sources of shorter wavelengths will again push down the limits.

[Dan Fromm]

Mark, H. Lou Gibson published a mind-numbing explanation of how to determine the optimal aperture given magnification and subject characteristics in Kodak Publication N-12B "Photomacrography." Optimal means "giving the best balance between rendition of detail in the plane of best focus and rendition of detail in depth."

There were several editions. You want the earliest, which contains the mathematical appendix. I got mine via eBay, also try abebooks.com , addall.com, and amazon.com . Search on title = photomacrography, not on author. For some reason, Kodak pubs. N-12A"Close-up Photography" and N-12B don't give the author's name but Kodak pub N-14 (I think that's it), which is the two bound together, does.

Cheers,

Dan

[Susan]

I learned quite a bit from this forum. I have always wanted to do SUPER closer up. Basically a la Karl Blossfeldt. Does anyone know of a book that gives more info on doing super close up photography? I want to do tabletop, black and white- I am at a loss on how he did 12x-30x magnifation with a pin hole camera!

I lost all camera equipment in a divorce (hey-it was worth it!). I am starting over.

I want to know what I want this time going in, as opposed to what I bought years ago because it was considered the camera to be seen with in class (sinar, hassy, etc).

Suggestions for formats at 4 x 5 are good. Anything larger- not really practical for me.
Even 120- super close up would be nice, but I kind of doubt it could be done.

12x magnification. Wow. Karl Blossfeldt.

[Dan Fromm]

Susan, 12:1 is very hard to do.

If you'll look up this thread, my post immediately above yours gives directions to one book. The other one you want is Lester Lefkowitz' The Manual of Closeup Photograpy. Both are out of print, can be found used on eBay or via amazon.com, abebooks.com, addall.com.

The two books will remove the stars from your eyes, will also give you a good idea of what's possible and how to do it.

Dan