Focal length (zoom ring) vs focus (focus ring)

[spacediver]

Trying to wrap my head around these concepts and lens controls.

I have a kit EF-S 18-55 mm lens (canon EOS 450D), and the zoom ring is labeled 18-55 mm, which indicates the focal length. Adjusting this changes magnification.

But this is different from the focusing ring. When you adjust the focusing ring, the magnification doesn't change. So what does the focus ring adjust, if not focal plane?


Some other relevant observations: When I adjust the focusing ring, the outer barrel telescopes outwards and inwards independently of the inner barrel. When I adjust the zoom ring, the outer barrel is at its most retracted position at 28 mm, and at its most unretracted position at 18mm and 55 mm. So what seems to be happening is that the focus ring is being adjusted automatically to compensate for different zoom settings. Not sure why the relationship between the focus and zoom ring is nonmonotonic.

Is there a simple explanation for all this, and why is the focusing ring called a focusing ring if it doesn't adjust focal plane?

 

[Overread]

Few thoughts - note I don't know the physics of it all well enough to give a comprehensive answer.

1) The focusing ring changes the position of the plane of focus in the shot itself. The focal length should remain the same, ergo the magnification is the same, but the plane of focus shifts through the scene from near to far (infinity). At least in theory till you get to point 2

2) The focal length is measured when the focusing ring is set to infinity and this is because many lenses often change their focal length through their focusing range. As a result the focal length at the nearest focusing point might be shorter than the stated focal length on the lens. Thus the focusing ring does (in some lenses) have an effect on focal length and thus magnification; however its typically a point most ignore in casual descriptions of lenses and their properties. (esp since it does not apply equally to all lenses)

 

[spacediver]

thanks for the detailed reply. I think the concept of "plane of focus" vs. "focal length" that you introduced will be helpful in helping me grapple further with these ideas

 

[bratkinson]

Perhaps the best explanation I can provide is the zoom ring 'moves' the subject nearer or further to you by 'zooming in' or 'zooming out' respectively. In doing so, as you zoom in (increasing the focal length from 18mm to 55mm), the subject will get larger and larger but the surrounding area will be simultaneously reduced. Holding the camera eyepiece to your eye, twist the zoom ring and watch what happens. You'll go from, say, seeing the entire person in the eyepiece to seeing only their head and shoulders, for example. Without using the zoom ring, the subject (a person in this case) could simply walk closer to you to accomplish the same result.

It doesn't matter 'what' the lens does while you are zooming. Most zoom lenses these days extend an inner barrel outwards as it zooms in. This is akin to extending a handheld telescope longer and longer to see 'further and further'. Some zoom lenses don't 'extend' or 'retract' at all while zooming. The repositioning of various glass elements for zooming are all done internal to the lens without moving the front element.

The focus ring is a small adjustment to make the subject 'sharp' in the eyepiece and ultimately the photograph. Otherwise, the image will be blurred. As your camera can automatically perform all focusing functions, it's probably best to set the AF switch on the side of the lens to AF (Auto Focus) and let the camera do the work. The autofocus functions are so good these days I haven't taken my camera and lenses out of autofocus (AF) in several years! Back in the days before autofocus was commonly available, most cameras had a 'split ring' focusing area projected into the eyepiece that made manual focusing easily accomplished. The split ring feature disappeared with autofocus. These days, manual focusing is difficult to perform accurately without significant technique and in some instances, a tripod and other equipment.

 

[spacediver]

I'm using my lens in reverse (taking macro images of a CRT screen), so I can't use autofocus. But even if I could, I'd use manual focus as I need to preserve the exact same focus and zoom as the configuration used during the vignetting calibration phase of my project (I'm measuring vignetting so I can factor it out of my images as I need very precise luminance measurements across space). So I'll be using a focusing rail and manually focusing.

The interesting thing is that with the lens reversed, the zoom lens now seems to adjust focus, while the focus ring seems to adjust magnification. It's hard to say for sure, need to do some more experimentation.

 

[wfooshee]

Focal length and focus are two different concepts. This post is probably WAY more than you bargained for, but understanding the difference holds the answer to your question.

In a simple single-element lens, just a plain piece of glass like a magnifying glass, the focal length is literally the distance at which light through the lens converges to a point. Not an image, but a point. Basically you can measure the focal length by holding the lens over a piece of paper and under a light source. Move the lens up and down until you find where it makes a point instead of spreading the light out on the paper. That distance to the paper is that lens's focal length. A lens element with a short focal length will be closer to the paper, and a long focal length will be further away.

But a naked lens element doesn't serve much purpose in photography. The glass has to be contained in a light-proof barrel, and has to direct the light to a surface, either film or a digital sensor. And a single-element lens is also not much use in photography, as the light through the edges bends differently than the light through the center, and different colors of light bend differently from each other. You can get a picture, but it won't be sharp, and it will have color fringes around the edges.

Multi-element lenses are there simply to correct these flaws that occur with a single element. This piece fixes that issue, and that piece corrects another issue.

Another thing that multiple elements can do is "fake out" the focal length. With a single-element lens, a 400mm lens has to be literally 400 millimeters from the imaging surface. a 10mm single-element lens would have to be only 10mm from that surface. Obviously we don't mount a lens two feet in front of the camera, or actually inside the camera. Some of the glass elements in those lenses serve no purpose other than artificially changing the focal length of the assembly, so that the back of the assembly is about the right distance from the imaging surface to fit in the lens mount of the camera.

This, by the way, is the reason a 50mm lens is so much cheaper than other focal lengths. The back of that lens actually is about 50 millimeters from the imaging surface. No corrective retro-focus or telephoto elements are needed to change the focal length to something shorter or longer than the mounting surface of the camera.

Going back to a single-element lens, a single piece of glass.... If you have a lens with a short focal length close to your imaging surface, the view through that lens covers a wide field. If you have a long focal length lens sitting way out from your imaging surface, the field of view of that lens is much narrower. The image seems magnified. Think of standing near a window looking out, or well away from it. The further you are from it, the narrower your view. Unlike you, though, the lens fills the imaging surface with its view, so it "squeezes" or "magnifies" what it sees (technically not the correct terminology but the concept is clear,) depending on whether it's a short or long focal length. That concept carries over into the multi-element designs. A lens assembly with a short focal length has a wide field of view, for the same reason that a single-element glass does. The focal length is close to the imaging surface, showing a wide field of view.

All of that so far describes focal length, and how it comes to be measured the way it is.

Any lens needs a slight adjustment to make a sharp image of some object or scene, based on how far away that object or scene is from the lens. The "focal length" is not an indicator that something is "in focus." If you're imaging something fairly close to the lens, then the lens needs to fudge a bit out from the imaging surface to make a sharp image. If the subject is well away, the lens need to be fudged back the other way, closer to the imaging surface, to make a sharp image. That is what we call focusing the lens, and the distance from the lens to the subject is the focusing distance. That has nothing to do with the focal length of the lens, which describes the width of the field of view. Any SLR lens, zoom or not, has to be able to be focused, to adjust for the distance to the subject.

Now the fact that you have to move the lens in or out to focus explains the fact that most lenses actually have a very small change in actual focal length depending on the range to the subject. A "50mm" lens that's focused on something close may actually exhibit a precisely measured focal length of 51.8 millimeters and while focused on something very far away actually be at 49.2 millimeters. (These are numbers I pulled out a dark place behind me, by the way. I have no idea what the actual variance might be.)

So: focal length is a measure of the lens's field of view. Focus distance is the very slight adjustment needed to converge the light rays into a sharp image, depending on how far away the subject is.

 

[spacediver]

thanks for that incredibly lucid tutorial, I had a couple revelatory moments reading that.

A few questions to help consolidate my understanding:

Based on your post, I take it that a single-element lens (focusing a point light source onto a plane) has a focal length, and a focusing distance.

1: Am I correct in assuming that there is exactly one value for each of these parameters for this lens? Or are there an infinite number of combinations of focal length and focusing distance for a given single-element lens?

2: With a single-element lens, if you were to move the lens farther away from the imaging plane than the focal length of that lens, would the field of view (and thus magnification) change? In such a case, would any depth plane of the 'world-being-imaged' (what's the technical name for this?) be in focus?

3: Is the ability to change the distance of the lens from the sensor, independently of the width of the field of view, a marvel that can only be achieved by a lens assembly?

4: Why do you refer to an "SLR" lens. My understanding of SLR is that it refers to the rotating mirror system that allows a direct view through the lens. Why would mirrorless systems use a different type of lens?

 

[wfooshee]

1. One focal length for a single-element lens. Focusing distance still would be depends on distance to subject. A very simple single-element camera, like the disposable film cameras, don't need focusing, everything beyond a certain minimum distance is clear and sharp.

2. No. Moving it away from its focal length destroys whatever image it is resolving. Play with a magnifying glass; look at your fingertip. There is a specific distance the lens has to be both from your fingertip and from your eye before you get a useful image from it. You can't just say, "Well, I'll look at something across the room with it."

3. I used the term "assembly" to refer to a multi-element lens in its housing. We simply call that assembly a "lens." I neede to differentiate that from the single-element lenses I was describing, that's all. Many of the elements serve only to modify the focal length of the entire group. The purpose of those is to move the resolved image some distance from the actual focal length behind the lens. A 14mm lens can't be 14mm from the sensor, it would interfere with the camera's mechanicals. Some of the glass is used to move that sharpness point further from the back of the lens, without changing the field of view. The term is retrofocus. The focus is actually behind the point specified by the focal length. For a high focal length, say 200mm, we want the focus point to be much closer to the back of the lens, so we don't have to hang the lens out half a foot from the camera. That's where the term telephoto comes from; the focal length is longer (tele) than the physical distance tot he imaging surface. Telephoto has come to mean any long lens, but technically it means a lens whose physical focus is shorter than its focal length.

4. I probably shouldn't have phrased this the way I did. My meaning was any multi-element lens.

 

[Dave442]

It seems like your project has a number of variables to overcome and using a reversed zoom lens makes it a bit more difficult than using a dedicated macro lens that goes to 1:1. You probably want to stop down the aperture to reduce lens vignette, probably in the f/8 to f/16 range is where I would start. If your not at the 55mm end then might consider taping down the zoom ring until all tests are done, especially if you have to remove and remount the lens anytime during the tests.

 

[spacediver]

thanks for the clarification wfooshee

It seems like your project has a number of variables to overcome and using a reversed zoom lens makes it a bit more difficult than using a dedicated macro lens that goes to 1:1. You probably want to stop down the aperture to reduce lens vignette, probably in the f/8 to f/16 range is where I would start. If your not at the 55mm end then might consider taping down the zoom ring until all tests are done, especially if you have to remove and remount the lens anytime during the tests.

Yes I might experiment with a smaller aperture to reduce vignetting, though this will increase diffraction and I'm not yet sure if this will compromise my measurements. Either way, I am characterizing the vignetting by taking a large number of images at random locations and orientations on the screen and then averaging them. That way I can factor out the vignetting in all future measurements (I'm shooting RAW and using Matlab to process all the data).

 

[Dave442]

Good point on the diffraction, f/11 and f/16 will probably start to show some limit to sharpness.

 

[spacediver]

Here's an image I just took. Colors are off (once I finish the code, I'll be subsampling the raw image and will be able to derive objective luminance and chromaticity measurements - see my post here if anyone is interested in how to do this), and this is just a quick jpeg with auto white balance. Based on my calculations, the magnification is about 4x!

I'm also going to experiment with a smaller aperture like you suggested Dave - this will make it much easier to obtain an in focus image, and I have a feeling that the phosphor coating itself may not be on a perfect plane, so having a wider depth of field will enable me to have a clearer image without focus stacking.

Anyway, this is probably the most detailed image of an aperture grille phosphor mask that's online right now! No tripod or focusing rail - just a camera resting on a stack of books on a flimsy desk. The width of each RGB triad is about 0.23 mm.

 

[Dave442]

Is that 1:4 or 4:1 magnification, or were you referring to a 4x magnification?
Interesting test, is this only for CRT displays or can it be used for other display types?

 

[spacediver]

I (mistakenly?) wrote 1:4 originally as I thought that was the correct way to describe magnification. Now I'm not sure what the proper convention is. What I meant to say was that a mm of the phosphor screen took up about 4 mm of my sensor (using rough calculations). So it was magnified 4x, and technically I guess this is super macro!

This can be used for any display type.

 

[Dave442]

Nice, I guess you were at the 18mm end of that reversed zoom lens.