Front Aperture

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

ray_parkhurst
Posts: 3412
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Front Aperture

Post by ray_parkhurst »

I have been experimenting with front apertures, with mixed results. I've found that I can greatly limit the front aperture with seemingly minimal impact to image sharpness and coverage, but I am not sure how to quantify the effects or to calculate the limits. Would appreciate some input on this.

Specifically, I am using a tapered lens hood that extends typically 50mm from the front of the lens, with a ~10mm aperture at the tip. The lens is typically 75mm FL and operated at f5.6, at 0.7:1 magnification. This particular implementation seems to work very well, showing only a small amount of vignetting in the corners, which is fine for the application. Indeed I can tolerate significant darkening when shooting round coins.

I have been deciding on aperture size by stopping-down the lens and observing the vignetting. With smaller lens aperture, vignetting becomes more pronounced. My theory is that if I keep the small-aperture vignetting such that it stays well outside the radius of the coin, then the impact on sharpness will be minimum. Opening the lens to working aperture then shows minimal vignetting and surprisingly little sharpness degradation in the corners, and none visible on the edges of the coin.

So that's fine for the specific application, but I'd like to build a system with variable aperture height, and am not sure of the limits or optimization. I would also like to have zero vignetting at the wide edge of FF so that I can use the technique for 2-shot stitching, and zero vignetting at the corners of FF for detail images. I assume these will have different front aperture requirements.

So my questions are, where is the best placement of the aperture, and minimum size of aperture, when placed in front of the lens to keep from vignetting?

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: Front Aperture

Post by Lou Jost »

Those are good questions, it will be interesting to know the answers. Meanwhile I can report that front apertures (usually step-down rings) are much used in astronomy, and work extremely well for long lenses but not so well on wide lenses (eg. 40mm). Astrophotographers like front apertures because most telescopes don't have diaphragms, and most camera lens diaphragms cause diffraction spikes on stars since they are not perfectly smooth.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Front Aperture

Post by rjlittlefield »

Ray,

Vignetting is caused by a conflict between two apertures: one provided by the lens manufacturer inside the lens and the second one that you have added outside the lens. This conflict causes some rays that would otherwise be passed by the aperture inside the lens, to be blocked by the added front aperture.

To have zero vignetting, you need to keep the external aperture outside the volume of rays that is formed by the rectangular image frame at the subject joined to the entrance pupil of the lens. The general principle is the same as for constructing baffles in extension tubes. There is an extensive discussion of that topic in the thread surrounding viewtopic.php?p=120423#p120423 , which provides a diagram of the construction and the resulting shape.

But I am not sure that I correctly understand what you're doing. Here at PMN I normally understand the word "aperture" to mean something that limits the ray bundles, thus affecting resolution, DOF, and aberrations. In contrast, I think that you're designing a lens hood, and you want the hole in the hood to specifically not be an aperture in the sense of limiting the ray bundles. Is that correct?

Edited to add: Or is it the other way around, and you want to make the front aperture be the only limiting aperture in the system? In this case, you need to keep the lens entrance pupil totally outside the volume of rays defined by the image frame at the subject and the added aperture.

--Rik

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Front Aperture

Post by rjlittlefield »

rjlittlefield wrote:
Tue Nov 23, 2021 11:23 am
you want to make the front aperture be the only limiting aperture in the system? In this case, you need to keep the lens entrance pupil totally outside the volume of rays defined by the image frame at the subject and the added aperture.
On further thought, I realize that this "volume of rays" cannot be computed in exactly the same way as for the other case. The basic concept is the same: every ray cast from every point on the subject through the front aperture must also make it through the rear aperture = entrance pupil of the lens. However, these rays include the ones that join opposite sides of the image frame and the added aperture, not just the ones that join same sides of the frame and aperture. As a result, the required volume spreads out farther behind the added aperture than is suggested by the diagram that I linked earlier.

--Rik

patta
Posts: 86
Joined: Mon Apr 20, 2020 9:51 am
Location: Stavanger Norge
Contact:

Re: Front Aperture

Post by patta »

some qualitative thoughts

if you put the aperture pretty near to the subject, it will work as "field stop", meaning it is not aperture anymore; it may be useful to restrict the view (extreme vignetting) so there is less haze, better contrast in the center

otherwise, middle distances, if you keep the "real" aperture wide open, then your external aperture becomes the only ruling one, no vignetting
if you see vignetting, that's because the two apertures are fighting against each other

the ruling aperture defines the "point of view", from where the image is seen. Putting the aperture outside, nearer to the subject, you get a nearer point of view - instead of the "distant" perspective of normal macro.
You can experiment, make your external aperture very small (like, 1mm) and near to the subject; you should get a "near" perspective.
Here below example, lens is a 50mm, working as "near perspective macro"
The resulting photo looks like if taken with a phone camera from near.

But, anyway guess that the best results (sharpness and aberration-wise) come from putting the aperture where was designed for; and that is usually in the middle of the lenses stack. Anyway it's fun to move things. Now I'm working overtime to push this aperture back to infinity...

More edit:
Eyepieces are lenses designed to work with a front aperture - where our eye stays.
Also simple small objectives - like from old compact cameras - have the aperture in front, but is really basically attached to the lens.

.
prime-aperure-swap-bare.jpg

ray_parkhurst
Posts: 3412
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Front Aperture

Post by ray_parkhurst »

rjlittlefield wrote:
Tue Nov 23, 2021 11:23 am
In contrast, I think that you're designing a lens hood, and you want the hole in the hood to specifically not be an aperture in the sense of limiting the ray bundles. Is that correct?
Yes, this is the goal, ie to find the position and size of the smallest "hole in the hood" such that it is not the limiting aperture.

It seems to be a fairly complex question. I am planning to do some experiments.

patta
Posts: 86
Joined: Mon Apr 20, 2020 9:51 am
Location: Stavanger Norge
Contact:

Re: Front Aperture

Post by patta »

ok I misunderstood.

if you have a DSRL, shine a flashlight in the viewfinder.
the lens will project a cone of light that is the limit for vignetting.
can look at it with a piece of paper

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Front Aperture

Post by rjlittlefield »

patta wrote:
Tue Nov 23, 2021 4:29 pm
if you have a DSRL, shine a flashlight in the viewfinder.
the lens will project a cone of light that is the limit for vignetting.
can look at it with a piece of paper
I agree, this works well and should be a sensitive indicator.

I use this technique for aiming and rough focusing, as a brighter equivalent to laser aiming through the viewfinder.

One caution: be careful to check that the flashlight is illuminating the entire frame. Even some wide-beam flashlights may not reach the corners of the frame. I eventually settled on unscrewing and removing the front lens assembly of https://www.amazon.com/gp/product/B082X2415P , so as to place the bare LED close to the point where my eye pupil would be.

--Rik

ray_parkhurst
Posts: 3412
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Front Aperture

Post by ray_parkhurst »

patta wrote:
Tue Nov 23, 2021 4:29 pm
ok I misunderstood.

if you have a DSRL, shine a flashlight in the viewfinder.
the lens will project a cone of light that is the limit for vignetting.
can look at it with a piece of paper
I'm using an A7Rm4, so no chance for viewfinder optics. So I tried adding a variable iris between the lens and subject. What I found was that when the iris was close to the subject, it acted as a field stop as described earlier, with strong vignetting. The diameter of the vignetting was weakly proportional to the lens aperture setting. Moving the iris closer to the lens increased the diameter of the vignetting. When the iris was 'close' to the lens, the aperture could be made very small (~1mm) before visible corner vignetting. Clearly the 1mm iris was acting as the limiting aperture as there is visible blurring of the subject.

Still more experiments to do, but I think I have experimentally determined that my earlier method is correct. With the lens stopped down, as long as the "hole" is big enough to not vignette the subject, then when the lens is used at normal taking aperture, there does not seem to be any degradation.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Front Aperture

Post by rjlittlefield »

ray_parkhurst wrote:
Tue Nov 23, 2021 5:17 pm
When the iris was 'close' to the lens, the aperture could be made very small (~1mm) before visible corner vignetting. Clearly the 1mm iris was acting as the limiting aperture as there is visible blurring of the subject.
When the external aperture is so small and so close to the lens that no part of the subject field can "see" the lens aperture at all, then there will be no vignetting due to the lens and added aperture fighting each other. So, if your words means that when smaller than 1 mm you do get darkening of the corners, then it's an interesting question what else in the system is causing that.

--Rik

ray_parkhurst
Posts: 3412
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Front Aperture

Post by ray_parkhurst »

rjlittlefield wrote:
Tue Nov 23, 2021 6:00 pm
When the external aperture is so small and so close to the lens that no part of the subject field can "see" the lens aperture at all, then there will be no vignetting due to the lens and added aperture fighting each other. So, if your words means that when smaller than 1 mm you do get darkening of the corners, then it's an interesting question what else in the system is causing that.
I'm not sure about front apertures smaller than 1mm as that's the smallest my iris goes, but for sure at 1mm there was a small amount of visible vignetting at lens f4, and strongly proportional vignetting at smaller lens apertures. Note that by "close" to the lens, I mean as close I can physically get them, which is ~10mm spacing.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Front Aperture

Post by rjlittlefield »

Interesting. Then I'm guessing the entrance pupil is so far back that on FF you simply cannot hide its edge from the corners of the frame. Some careful measurements could clear that up, but it's probably not worth the trouble since the issue is not related to what you're trying to accomplish.

--Rik

patta
Posts: 86
Joined: Mon Apr 20, 2020 9:51 am
Location: Stavanger Norge
Contact:

Re: Front Aperture

Post by patta »

Maybe the concept of "Kohler illumination" from microscopy could be relevant here:

the subject is illuminated by a system of lenses with TWO apertures.
One is the normal one, determines the f/#
the second aperture is the "field stop" and blocks all light outside the subject (like a spotlight, illuminating only a single actor on a scene) its purpose is to block unnecessary light that will only cause haze; and give perfectly uniform illumination across the field.
That's for transmitted illumination though.

Translating this Kohler to reflected light and coins instead, it would mean that this "field stop" should be placed at the focal plane, just around the coin
But instead we need some distance, to get illumination through.
So yes needs some investigation

ray_parkhurst
Posts: 3412
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Front Aperture

Post by ray_parkhurst »

patta wrote:
Tue Nov 23, 2021 11:42 pm
Maybe the concept of "Kohler illumination" from microscopy could be relevant here:

the subject is illuminated by a system of lenses with TWO apertures.
One is the normal one, determines the f/#
the second aperture is the "field stop" and blocks all light outside the subject (like a spotlight, illuminating only a single actor on a scene) its purpose is to block unnecessary light that will only cause haze; and give perfectly uniform illumination across the field.
That's for transmitted illumination though.

Translating this Kohler to reflected light and coins instead, it would mean that this "field stop" should be placed at the focal plane, just around the coin
But instead we need some distance, to get illumination through.
So yes needs some investigation
Good thoughts, but unfortunately you're correct that a field stop at the focal plane would not leave any room for lighting. It also results in the largest required "hole" in the hood. Raising the hole a bit above the coin gives some room for lighting, but the lighting is at too low an angle to give good results. Ultimately, my goal for this type of hood is two-fold:

- Lighting from as close to center of lens axis as possible. This means small diameter hole and/or hole high as possible vs coin
- Masking of stray light coming from around the coin, directly from the light, and other sources

I have built such hoods for many years, but was never sure how far I could push the concept without negative optical effect. I generally believed that the optimum placement was somewhere between the coin and lens, but indeed yesterday's experiments show that closer to the lens is best. Luckily, this is also best for illumination, ie high as possible, with smallest hole possible, gives illumination closest to the optical axis.

A few years ago I made this graphic to explain the concept:
.
Hood2.jpg
.
With the availability of small-diameter ringlights, I was wondering if a ringlight could be used instead of a diffuser. The 20mm Angel Eyes have 4mm center holes that can be expanded to 7mm. Is 4mm large enough to not vignette or act as a limiting aperture? Is 7mm? Experimentally, it seems that 7mm may work, if close to the lens. So instead of a tapered hood, I'd mount the ringlight to a flat plate with a 7mm hole, mounted to the filter threads. I will experiment with this.

Edited to add: I think I may have interpreted "focal plane" incorrectly. Do you mean a vignetting mask at the Image/Sensor plane? I interpreted it to mean a mask at the subject plane.

patta
Posts: 86
Joined: Mon Apr 20, 2020 9:51 am
Location: Stavanger Norge
Contact:

Re: Front Aperture

Post by patta »

Aha, thanks for the illustration, now your aim is clear. Near ring illumination and minimize haze/reflections.
It is definitely a "sophisticate lens hood" and not an aperture; since it shouldn't stop any ray going from the coin to the sensor (otherwise will vignette; but it may vignette a bit the corners since the coin is round and in the center)

My ballpark method to find the "forbidden vignetting zone" where to put your cone will be to:
- have the camera set up, in live view ( & with the iris closed at the aperture you'll work with)
- move the finger (or a piece of paper) around, between lens and coin; when the image show some defocused finger - there it is!
Hmm guess you've done that already...

Focal Plane, that that was my slip, since for the Kohler illumination in microscopy, the field stop is not on the subject plane, but at another place (near the lamp) then "conjugate" / refocused on the subject plane by the illumination optics. Like, the subject plane is conjugate/refocused toward the image on the sensor (which is again a focal plane?). But, anyway, this story of Kohler, was my fantasy because I'm doing mostly microscopy, it isn't actually so relevant to your setup.

You have already a "mask" in the subject plane, it is the stage under the coin, that (hopefully) is black and does not reflect light; so there isn't unnecessary light coming from there, which causes only haze etc.
I'd try to put black velour / flocked paper around or under the coin so to minimize this haze. Like, jewels are usually photographed against a pitch-black background
Sorry this too isn't relevant to the cone/aperture placement... but may give a tiny kick to the contrast too...

Post Reply Previous topicNext topic