Any difference between diaphragm blades crossing in front of and behind aperture?

<p>I noticed that the diaphragm blades of some lenses cross in front of the aperture, meaning you can see the pointy end of the blades when looking through the front element, while others cross behind, so the tail of each blade is covered by the adjacent one as the aperture is closed.</p>

<p>Are there any benefits or drawbacks to either design compared to the other one? Why does Nikon make some lenses this way, others that way, and why are two otherwise so similar lenses like the 105 DC and 135 DC different in this? Any significance?</p>

<p><em> ' . . . diaphragm blades of some lenses cross in front of the aperture,"</em><br>

<em>"In front of the aperture..."?</em> The "diagram blade" is the moving part of the "aperture", and as a whole unit, We call it, as "aperture". ONE unit mechanism.<br>

Or, the question was something els.</p>

<p>Sorry, I may not have been clear enough. By "aperture" I meant the opening itself. Here is a picture I just took to illustrate my question. The lens on the left is the 105 DC lens with the blades going <em>under</em> the adjacent one, and the lens on the right is the 135 DC with the blades going <em>over</em> the adjacent one.</p><div></div>

<p>Well, at least I'm not the only one that cares:-)</p>

<p>http://www.jcolwell.ca/photography/x-arc/BellyButtonSurvey.pdf</p>

Certainly the "crossing under" design looks nicer. I had an old 90mm/4 Elmar with a

crazy number of diaphragm blades - 10, 14 - something like that. Watch it operate

was mesmerizing. While the number of blades (determining the roundness of the

aperture) might be an issue, I can't think how the orientation of the blades would

affect the function of the diaphragm.

<p>OK Oliver, I can see now what is your question.<br>

I wouldn't bother with such a detail, functionally and optically, doesn't meter, it is doing the job they supposed to do. I have lenses like this, it is a way sometime a physical functioning design has ben executed.</p>

<p>I think it's just aesthetic, since most of the lenses we see cross under in front, and whichever side is under, the back is over. When the lens is on, the front is what you see. I don't see any reason why it makes a functional difference which side is which. The only possible reason I can think of would be internal reflection when the blades wear, but you rarely if ever see any wear, and when you do, it's usually on both sides. The one other issue I might think of is if lens design requires the diaphragm to operate very close to an optical element, it might be turned so that the "over" side is a bit further away from glass. </p>

<p>Oliver: I have wondered the same thing re "overs vs unders." I am interested to see whether any of the community members have insight as this design feature. Maybe Bela is right, but I would at least like to know what lens construction considerations are in play. In other words, why do they do it differently?</p>

<p>Overs vs unders: the aperture opening location at the lens axis changes a bit when stopping down. See e.g. the image attached by Oliver. <br>

This might have been taken into account in the lens design. Maybe affects/compensates the focus shift?</p>

<p>Interestingly, all the Nikkor auto lenses I've seen are "innies" with the ends visible on the inside. But I have four preset lenses of various sorts, and three of these are "smoothies" with ends visible on neither side. I am guessing that free ended diaphragms are lighter in action. Of my four presets, the 85/2.8 PC is a nine bladed innie, but the 35/2.8 PC is a smoothie, as are an old Hanimex 28, and an ancient Spiratone 135/2.8 which has sixteen diaphragm blades, way too stiff to work as an auto diaphragm.</p>

Matthew, the "new" 24-120 f/4 VR zoom lens is also an outie, although you really have to set it to 120 mm to see the blades

at all. I remember some of the kit zooms to also be outies.

 

I wonder if there's anything more than the toss of a coin at play during the design process.

<p>Oliver, you're right. I forgot to look at my two DX zooms. The 18-55 and the 55-300 are both outies, also pretty hard to see. </p>

<p>In large format where the things are more "modular" there are shutters in front of the lense, behind the lense or inside.</p>

<p>Hi<br>

Yes is does make a difference, the exact placing of the aperture has an optical effect and the the side that the blades cross over on has to be further away from a lens element than the other side. So depending on where the designer wants to place the aperture it may be close to te surface of the element and thus have to be one way round or the other.<br>

Then once the internal mechanism has been designed it simply gets copied for any other lens that it is suitable for.</p>

 

<p>Simon: So you're claiming that it's because the optimal aperture position is not perpendicular to the optical axis? (That the aperture needs to move either closer to or farther from the focal point as it stops down?) That's a possible consideration, but I'd have thought it was minor - but I'm not a lens designer.<br />

<br />

Mag: The shutter does need to be at the appropriate position in the optical design, or you just get vignetting. Where that point is depends somewhat on the lens design - particularly whether it's a telephoto or reverse-telephoto design. I think that's secondary to the direction of the blade overlap. But not just large format - I have an old Leica lens where the blades are clearly exposed on the camera side.<br />

<br />

One concern I could think of is that I imagine the blade tips are more reflective, or at least visually intrusive, than the sides of the blades. This should be minor, because of course they're painted black, but if you can see them it can't be entirely negligible. I'd kind of expect blades that face towards the sensor might bounce back more light that's reflecting off the sensor (less an issue for film) than if the blades face forward, which might discourage their use for digital sensors, just as Sigma have "DG" coatings on their rear elements to avoid sensor reflections. On the other hand, forward-facing blades are obviously exposed to more light in the first place, and there may be a risk of that light bouncing off the inside of the front lens elements contributes to the image. Therefore I suppose the optical design of the front elements and whether they reflect enough light back to affect contrast might influence a decision about which way to face the blades.<br />

<br />

Or it could all be about mechanical convenience. I'm in no way a lens designer, I just thought the reflection issue might be a consideration and wanted to share.</p>

<p>I've wondered this too, and it seems a good one for a lens designer. One thought is that there are differences between light entering, and light exiting the aperture. Is there a difference the way the light is projected onto a sensor plane, or film plane from a blade face that is flat, and smooth, verses the opposing surface, which is not. I would think there would be more variation in image result with the amount of aperture blades, verses the positioning. Simons point of the aperture assembly position relative to element surfaces could very well be the probable solution.</p>

<p>Since aperture blades are opaque and therefore shade whatever is directly in front or behind them, it makes no difference to their function which way round they're orientated. In effect they behave as if they were a truly 2 dimensional mask with however many sides as blades.</p>

<p>Their position along the optical axis is important, and affects not only the vignetting performance of the lens but quite a few other aberrations as well. Notably curvature of field and lateral CA.</p>

<p>Simon has a point that the optimum aperture position may be close to a lens element and so the protruding "tails" of the iris blades would be better placed facing away from proximity to that element. However I've seen forward facing blade ends in Nikon designs where that definitely wasn't a consideration. It's even possible that such oddities are the result of accidental incorrect assembly. No matter though, because which way round the blades are doesn't really matter as long as they don't scrape any glass.</p>

<blockquote>

<p>Their position along the optical axis is important...</p>

</blockquote>

<p>As we can see in the image - I have seen it when doing aperture blade cleaning (e.g. the famous Ai-s 55/2.8 micro) - the blades may bend a little when the aperture is small. When bending the placement of the aperture opening changes along the opical axis. Generally the opening displacement is towards the aperture blade tips.<br>

This is just a fact that you can see too. However I do not know if this is a optical design decision or an assembly related subject.</p>

<p>Quite - as Kari says, there's no requirement that aperture blades are planar. In fact, they pretty much can't be. That's what I was trying to get at - and it might have an interesting effect on focus shift.<br />

<br />

RJ: Blades are opaque, but not perfectly black. Most of my argument is based around which side of the blades you'd rather get reflections from. I don't think it's a simple decision, but it may be a very minor factor in lens performance. Still, every little helps.</p>

<p>Kari, by using an optical design simulation program, you can see that aperture positioning can be quite critical. However, that positioning becomes slightly less critical as the aperture is made smaller. Therefore any bending of the iris blades on stopping down (which really shouldn't happen) has little practical effect on image quality. AFAIK there's no deliberate design intent to move the position of the iris along the optical axis as it's stopped down. Also, some of the apparent 'bending' may simply be an optical illusion caused by looking through the curvature of the glass elements.</p>

<p>Andrew, I don't think either the character or amount of reflected light is much changed from one side of the iris assembly to another. The area directly exposed to light stays pretty much the same, as does the reflectivity of the surface.</p>

<p>RJ, I fully agree. Actually f 2.8 wide angle lenses have so small lens physical opening and so thin blades that they do not bend , in practise, even at f22. The bending is more probable in faster lenses like 100 f2 or 135 f2 and like.<br>

To the OP's question: it seems that the aperture tip position is not "field relevant" nor is a "design parameter" ;-)</p>