Number of aperture blades

Lens specifications are not complete without detailing the number of aperture blades. But what does it mean in practice?

Of course, the shape of the aperture blades is important for the out of focus highlights, the bokeh. I have previously seen that the Lumix 45mm f/2.8 1:1 macro lens has somewhat more rounded aperture when stopped down than the Olympus 45mm f/1.8. When not stopped down, i.e., at the largest aperture, the diaphragm blades move out of the way, and the opening is usually perfectly round.

Adding a higher number of diaphragm blades can make the aperture opening more rounded when stopping down. However, lenses seldom go beyond nine blades, probably for the reasons of cost, complexity, and the risk of having one of them break down.

It turns out that the number of blades is also related to the rendering of flare, strong light sources inside the image frame. I'll look at that later in this article.

Samyang and Lumix fisheye lenses

There are currently two fisheye lenses available for the Micro Four Thirds system, the Samyang 7.5mm f/3.5 and the Lumix G 8mm f/3.5:



Of course, the major difference between these lenses is the price (the Samyang is the cheaper), the image quality (the Samyang is better, in my opinion), and the focus mechanism (only the Lumix lens has autofocus). But in addition: The Samyang has six aperture blades, and the Lumix has seven. These close up pictures show the rear exit pupil at f/5.6 for both lenses. The Lumix G 8mm lens has more rounded aperture blades, but you can still make out seven segments:

Samyang @ f/5.6	Lumix @ f/5.6

Here is a video showing how the aperture blades open up on the Samyang lens:



Affecting the flare

It turns out that the number of aperture blades affect the flare in a fundamental way. It is common to get a star shaped flare when using small apertures. And the number of spikes in the star is the number of diaphragm blades (when that number is even), or twice the number of diaphragm blades (when that number is odd). Hence, using the Samyang should give us stars with six spikes, and using the Lumix should give fourteen. Let's check, by taking pictures at f/22:

Samyang @ f/22	Lumix @ f/22

To see this more clearly, here are enlargements at f/22 and f/11:



It's not so easy to count the number of spikes on the Lumix image to the right, but I think it is still quite clear that it is fourteen.

Conclusion

To my experience, most lenses have an odd number of aperture diaphragm blades, which produces the most spikes on the star shaped flare around bright objects, e.g., the sun. I guess this is because a larger number makes the flare look more blurred, whereas a small number makes the flare more distracting.

A bit of trivia is that Canon has chosen an even number for most of their lenses, while Nikon is going for an odd number. Hence, it is often easy to guess what brand a photographer uses, based on an image where the sun is inside the frame. However, the aperture also needs to be fairly small to see this effect, e.g., f/16 or f/22.  So you might not be able to make this out based on any picture.

Bad aperture diaphragm in Lumix G 14-42mm f/3.5-5.6

When investigating the bokeh of some Panasonic Lumix lenses, it came to my attention that out of focus highlights using the Lumix G 14-42mm f/3.5-5.6 kit lens was irregularly shaped when stopped down. I decided to take a closer look, to see what the problem is.

First, I took a picture in which I set the lens to tele (42mm), focused as close as possible, and placed a flashlight in the background. The flashlight renders out of focus. Here is the full image at f/5.6, the maximum aperture. The flashlight is placed in the centre of the image:

To see how the roundness changes when stopping down, I have made 100% crops from the centre at various apertures:

This verifies the problems I saw when studying the bokeh. The out of focus highlights are definitively not circular. It looks like the aperture diaphragm blades are misaligned.

Micro Four Thirds lenses are always wide open when powering the camera down. This means that normally, you cannot look at the diaphragm blades from the inside, since the aperture is wide open.

However, a trick is to stop down the lens, and then remove the camera battery. That way, you can remove the lens while stopped down. This procedure is not exactly recommended by the manual, so use with caution.

Using this trick, I could photograph the back side of the lens when stopped down:

Here it is clear that some of the blades are misaligned. Thus, the resulting image has non-round out of focus rendering.

I've made a video showing the stopping down of the aperture blades. The apertures goes from f/3.5 down to a full close in 1/3 stops.



I filmed it using the Panasonic Lumix GH2 and the Leica Lumix DG 45mm f/2.8 macro lens. To get the needed magnification, I used the new Extra Tele Converter (ETC) mode.

Here's a photo of the setup for capturing the video:

Conclusion

My lens most certainly has a bad aperture mechanism. Whether this is a one-off bad copy, or a systematic problem with the lens line is hard to say. I would guess it's an example of poor quality checking, and that most lenses are ok.

This problem annoys me a bit. I've previously found the basic kit lens to be a good one, despite the mixed reception it generally gets online.

Now, this is not really a big issue. Generally, you don't get much bokeh with kit zoom lenses anyway. So the problem is not very likely to show in images. If using the camera at full auto, it generally chooses the maximum aperture anyway, in which case the aperture opening is round.

This problem might affect the exposure correctness. The defective aperture blades could cause slight exposure irregularities. But again, this is not likely to be a big problem

Epilogue

If found the aperture to be so bad, that I took the lens back to the shop where I bought it in the first place.

The store keeper has some problem verifying that his off the shelf lens did not exhibit the same non-round aperture. I helped him by taking a photo with the lens mounted to the Panasonic GF2 camera at f/9, 2 seconds, and removed the lens during the exposure. Looking towards the light through the lens showed that his copy had a round aperture.

So he accepted my lens as defective, and sent it for repair.

After one month, I started enquiring about the lens. I always got the same answer: "The lens is just around the corner, should be in our store the beginning of the next week."

It was not until after three months that the lens finally did arrive in the store. When I went to pick it up, I was told that they simply replaced the lens with a new one. Why let me wait for three months when they would just give me a new copy?

Coming home, I once again checked if the aperture was rounded. I found that my new lens had exactly the same problem, the aperture blades are misaligned. I did check that the new lens has a different serial number, so it is not the same lens that I returned.

Letting me wait three months for a new lens is bad. But giving me a new lens which has the same problem as the one I returned is simply appalling.