Aperture looks off by a couple stops

[Tim Carroll]

Just picked up a couple of lenses (a 32mm and a 50mm) and the glass is very nice on them, and they are collimated pretty well, but the apertures seem to be off.

 

From what I can tell the apertures are off by about two stops on each lens. And I want to see if I understand this aperture thing correctly. On the 32mm lens, when the aperture is set at T8, by my reckoning the opening in the aperture iris should be 4mm. Am I understanding this correctly? And by extension, then the aperture opening on the 50mm lens, at say T22, should be about 2mm, correct? Or am I missing something here?

 

And when I say aperture opening, I mean measuring with a calipers the actual opening in the iris blades, (so with the back elements of the lens removed). Is that the way you measure aperture, or am I totally bassackwards on this?

 

Any and all info would be appreciated.

 

Thanks,

-Tim

[Chris Keth]

Something tricky creeps in, actually. Your math is correct but the method of measurement is off. What you need to know is the diameter of the entrance pupil, or the diameter of the front of the lens that is actually working. This is basically a measurment of the iris with the effect of the glass taken into account.

 

To do this is slightly tricky on anything but a lens made for large format. This is how I do it on large format equipment, with the directions altered to work with a film camera or an SLR. I think it should work fine on that type of lens. I see no reason why it wouldn't.

 

Focus the lens to infinity and put it on the camera. Poke a small pinhole in a piece of paper or film and place it in the gate of the camera. In a dimly lit room, shine a bright light through that pinhole so it projects a beam out through the lens. The diamater of that beam is the diameter of the entrance pupil. Measure it as close to the front element of the lens as possible for best accuracy and for the brightest light.

[Tim Carroll]

Chris,

 

Thanks for the response.

 

I did what you described above, wasn't too tough. What I came up with for the 32mm lens is approximately 18mm. And with the 50mm lens it was approximately 26mm. Now I did this with both lenses focused at infinity and with the apertures wide open, (with both lenses that was T2.3).

 

What does this tell me? I assume from your description it tells me the diameter of the entrance pupil. How do I use that information to check or measure if the aperture scales on the lenses are accurate?

 

If I am supposed to substitute the diameter of the entrance pupil for the size of the lens, i.e. instead of using 32mm as I did above, use the diameter of the entrance pupil (18mm), (which would give me an iris opening of 2.25mm at T8), then the aperture scales on both lenses are WAY OFF.

 

Thanks for any clarification anyone can give,

-Tim

[Mike Rizos]

Hi Tim,

I don't know how to figure out the correct actual measurement of the iris, but the numbers you give seem to work out. You should be using f/stops instead of T stops for this by the way.

Relative aperture or f/number is found by dividing the focal length by the effective aperture. You found the effective aperture the way Chris described.

The 32mm lens with entrance pupil of 18mm (32/18) gives a maximum f/stop of f1.8. And the 50mm with entrance pupil of 26mm (50/32) gives a maximum f/stop of f1.9. These f/stops seem to agree with the T2.3 of these lenses.

 

From what I can tell the apertures are off by about two stops on each lens. And I want to see if I understand this aperture thing correctly. On the 32mm lens, when the aperture is set at T8, by my reckoning the opening in the aperture iris should be 4mm. Am I understanding this correctly? And by extension, then the aperture opening on the 50mm lens, at say T22, should be about 2mm, correct? Or am I missing something here?

 

The effective aperture (entrance pupil) is not the same size as the physical aperture (actual size of the iris opening), because the lens compresses the beam of light entering into a cone, therefore trasmitting more light through a smaller opening.

 

My guess is your lenses are fine.

[Tim Carroll]

Hi Tim,

I don't know how to figure out the correct actual measurement of the iris, but the numbers you give seem to work out. You should be using f/stops instead of T stops for this by the way.

Relative aperture or f/number is found by dividing the focal length by the effective aperture. You found the effective aperture the way Chris described.

The 32mm lens with entrance pupil of 18mm (32/18) gives a maximum f/stop of f1.8. And the 50mm with entrance pupil of 26mm (50/32) gives a maximum f/stop of f1.9. These f/stops seem to agree with the T2.3 of these lenses.

 

The effective aperture (entrance pupil) is not the same size as the physical aperture (actual size of the iris opening), because the lens compresses the beam of light entering into a cone, therefore trasmitting more light through a smaller opening.

 

My guess is your lenses are fine.

 

Mike,

 

Thanks for the info. The issue I have with the two lenses is not that the maximum aperture looks off, when the lenses is wide open the iris blades are completely out of the light path of the lens. Where they appear to be off, to me, is when you start stopping down the lens. When I look at the lens stopped down to f22 and it still looks pretty open, that tells me something isn't right. When I need to turn the aperture scale past 2.8, almost to 4 before I see the iris blades coming into the light path, that tells me something isn't right.

 

Using the description/calculations above, when I stop the 32mm lens down to f8, and do the method Chris described, I get a spot of light that is 9mm in diameter. Which, if I understand what you wrote above, means that the actual f/stop at that point is f3.5. And that is pretty far off from f8.

 

What I'm trying to determine is a testing method so I can set the aperture scale accurately on these two lenses. Is what Chris described the way to test the aperture throughout its range? Can this method be used to check the f/stops from wide open to f22 (which is as far as these lenses close down)? Is there some kind of f/stop machine that lens houses put these lenses up on to test to make sure the aperture scales are accurate?

 

Thanks,

-Tim

[Dominic Case]

THe physical measurement process that's been described should work at any aperture setting. So that's one method.

 

However, to confirm wheter your results make sense, why don't you try exposure tests? Surely, if you compare the lenses you are testing with another one that you rely on, youwill have the answer.

 

Shoot an 18% grey card, making sure that it is uniformly lit (I find skylight (not direct sunlight) is the best for this.) Expose a bracket of a couple of stops each side of your meter reading, and find which setting for each lens matches your known lens.

 

If you can get your lab to read the density of the grey card in each test, you will have the most accurate results.

[Stephen Williams]

Hi Tim,

 

If you have a SLR with a built in meter you should be able to get an idea what is going on with those lenses, obviously they won't mount properly. Some cardboard & gaffer tape should help!

 

Stephen

[Tim Carroll]

Thanks everyone.

 

Figured out how to set it up and it turned out that only the 32mm is off, the 50mm turned out to be okay.

 

I set the lens up with a 52mm standard, black paper with a pin hole, a diffusion glass, a Super White LED flashlight, and a piece of tracing paper. Shot the flashlight through the diffusion glass, through the pin hole, through the 52mm standard into the back of the lens set at infinity and projected the image through the tracing paper which was sitting against the face of the lens.

 

Slowly opened the aperture until the disk of light shining through the tracing paper no longer got bigger. Set that on the aperture scale to 2.3 (widest aperture on the lens). So when I start closing the aperture from 2.3, the disk of light shining through the tracing paper immediately starts to get smaller. Turns out the scale was off by almost exactly two stops, so either I had a very lucky guess, or my eyes are calibrated. :rolleyes:

 

Thanks again for everyone's help,

-Tim

[Chris Keth]

Thanks everyone.

 

Figured out how to set it up and it turned out that only the 32mm is off, the 50mm turned out to be okay.

 

I set the lens up with a 52mm standard, black paper with a pin hole, a diffusion glass, a Super White LED flashlight, and a piece of tracing paper. Shot the flashlight through the diffusion glass, through the pin hole, through the 52mm standard into the back of the lens set at infinity and projected the image through the tracing paper which was sitting against the face of the lens.

 

Slowly opened the aperture until the disk of light shining through the tracing paper no longer got bigger. Set that on the aperture scale to 2.3 (widest aperture on the lens). So when I start closing the aperture from 2.3, the disk of light shining through the tracing paper immediately starts to get smaller. Turns out the scale was off by almost exactly two stops, so either I had a very lucky guess, or my eyes are calibrated. :rolleyes:

 

Thanks again for everyone's help,

-Tim

 

That certainly sounds like a logical adjustment method to me. Be sure you shoot a good controlled exposure test before you do anything important with it.