Lens distortion?

[atlcruiser]

thought of that but it is straight

[Frank Petronio]

Maybe your lens is assembled wrong or there is a spacer missing, the one that precisely separates the elements at an ideal fixed distance? Sometimes there are shims in lenses that go missing (!) when you buy them from sloppy or ignorant sellers.

But I don't know if that is the case for certain. What do your other lenses do? If you can repeat it with another lens, the you are a step closer to figuring out the problem.

The other thing I would do is sacrifice the negative and scratch it up by putting a square down and seeing if it is just as distorted on the film - then you know it was the scanning doing this.

Just an off chance but could one of your editing programs be forgottenly set by default to do a lens correction?

[Doremus Scudder]

You may indeed have a lens with slight (I wouldn't call 6% massive) barrel distortion.

However, before you blame the lens, there's a lot a testing/experimenting to do to determine if, indeed, that lens is the problem. With the quality of lens you are using, I would tend to think that the problem lies elsewhere... :-) There are a lot of weird combinations of "not square" that will get you things that look a lot like barrel distortion.

FWIW, I do a lot of walls and architecturals here in Vienna, and find getting the image perfectly square to be one of my biggest challenges. Here are a few tips from my experience and what I would do to test that may help.

First, although you admirably level your camera back using levels, you really need to be using the grid on your ground glass for the fine leveling adjustments. Levels just aren't accurate enough and wooden cameras like your 'Dorf probably do not have many plane surfaces that are straight and parallel to each other.

Using the grid is the best way if: (I'm assuming you do have a gridded ground glass, if not, you need to get a good one!) 1. your grid is square on the ground glass (check it with a good square!), 2. it is aligned properly with the way your filmholders fit in the camera back, i.e., not skewed rotationally and, 3.) all four corners and the center are at the same place as the film when the filmholder is inserted the back (this is harder to check, so just check the first two and save it till after the following).

If, after an inspection, you think your gg is mounted square and parallel to the film, then try this:

1: find your self a square test subject (wall) that you can photograph with several (at least two) lenses and still have a square subject for each.

2. Set up with the camera in "0" position, leveling with the levels as you do. This is my preliminary leveling step. If you need front rise to compose, now is the time to apply it. Do not tilt or swing. This is the last time I use the levels. getting the tripod head as level as possible at this point will help save time later, but is not indispensable.

3. Find a vertical line at the center of the image and use side-to-side tilt to align this with the grid and lock down the control. Next, find vertical lines at the sides of the image and use forward/backward tilt to make sure they align with the grid lines at the sides the gg. Sometimes I flip the back to vertical position to aid with this, even though I want a horizontal composition. This second step is necessary because a center vertical line will be vertical if side-to-side is correct, but the back will still not be parallel top-to-bottom. Once the vertical alignment is correct, check framing and adjust front rise/fall. Check again. Lock down these tripod controls.

4. Using the pan on your tripod head, pan the camera horizontally to get horizontal lines at the top and bottom of your image parallel. When they are exactly parallel, lock down the pan control. Don't use a line crossing the center of the image since if the camera is leveled side-to-side, this line will be horizontal regardless (you've already done this, above). If panning destroys your composition, use camera shift to re-frame.

5. If you were lucky enough to have your tripod head exactly leveled in relation to the subject, your verticals will still be vertical. However, most likely the verticals are now off again, especially if you panned a lot to achieve horizontal parallelism. So go back and repeat steps 3 and 4 till everything is lined up.

6. When you are 101% happy with the image alignment, lock everything down, and take the photographs with several lenses without moving the camera. Try to use lenses close to the lens you suspect of distortion. I you had another 240mm, I'd use that too. Develop and inspect the negatives for proper image alignment. If, indeed, your 240 shows distortion, but the other lenses don't, then you've found your culprit.

If not, start in on getting your gridded gg better aligned with the actual film position and/or refining your alignment technique. Sometimes the former can be maddening. Keep in mind though, that if you align well to the grid, even if the gg is skewed from the film rotationally, the image lines will still be parallel to each other, just not to the film edges. If there are other gg/film plane alignment issues, these will usually show up as focus problems in a corner or center., etc. The image will not be square in direct relation to the out-of-focus areas (of course, you need to shoot with a very shallow depth of field to check this, stopping down a lot will hide the focus error).

Finally, don't discount film warp (especially if you are shooting 8x10) causing a problem like you have. Check how a scrap sheet sits in the filmholders in all orientations.

Also, realize that many structures are no longer square. Brick walls on old building are especially prone to sagging. And, as hard as I try and as careful as I am in leveling and aligning images, I still have a few that I miss on and have to correct at the printing stage (which can be maddeningly fiddley as well). I print analog, so use easel/lens stage controls. Many times, if the original structure was not square, one simply has to arrive at the best visual compromise. (BTW, you can correct barrel distortion by using a curved paper surface on the easel as well as digitally. If I were printing your sample negative, I might use a pencil or thin rod under the center of the paper on the easel to achieve parallelism.)

Hope this helps,

Doremus Scudder

[Ken Lee]

"Film was dead flat on the scanner."

Yes, when you laid it down, but did you tape it to the glass? A sheet of film will generally curl, after you close the lid on the scanner.

If your ground glass has lines, it will be very easy to see any distortion without having to expose film. Lines also helps position the camera when shooting architecture.

Also, one shot does not a test make. Have you repeated the test at another location?

What size film? You didn't mention that.

[atlcruiser]

Frank:

Not impossible at all. lens came from KEH. In defense of the lens I have not seen this before on other images, not to say it did not happen and I did not notice it. I normally am either closer or farther away from walls with this lens...(I know that might not make sense but it is true)



Ken:

I did not tape it. I shot 10 frames of wall stuff similar to this and 9 were with the 360. None of them show any issue. A simple check will be to tape it down and I will try that. The negative on a light table shows the distortion as well so I don't think it is the scanner.



This was not a test and I had no idea I had an issue. I have a very hard time getting it all lined up on shots like this so my first thought is that it is me not the lens.



I have a gridded GG and I burned my retinas working the GG lines against the image I agree I could have seen it but it is so slight I did not notice it.



This 810 film



Doremus:



Thanks for the detailed suggestions!



I have read some of your suggestions before and they have helped. I will start at the top and run through all of your points. I have a feeling your point #3 is the core issue with this problem.



I have found some wonky walls in my neighborhood but this building is newer and I actually checked the wall.



The deardorff has no shift and that can be an issue in these situations.



I was out with Michael A Smith working as an indentured servant for him a while back. I asked him some of these same questions on how he gets walls aligned....doors, bricks windows etc... Michael will not use a gridded GG nor does he use a level. His answer was simple........."I don't take pictures of those things, too hard to get it level"



Thanks for all of the detailed replies. I will work through this a bit more but I am almost convinced the issue lies in me and not the lens.

[GPS]

the middle is pushed out. the wall is flat but the middle is concave.
...

It's an optical illusion due to the different colors and forms of the windows on the sides. Cover them and you'll see. Nothing to care about.

[Jim Noel]

I agree that your film plane is not vertical. I have used Deardorfs for more than 50 years. Getting the back vertical is the most easily overlooked move. Use a good accurate level directly on the ground glass.
A good and accurate level is one in which the edges of the bubble exactly touch the engraved lines when level.They are difficult to find.
Jim

[E. von Hoegh]

I agree that your film plane is not vertical. I have used Deardorfs for more than 50 years. Getting the back vertical is the most easily overlooked move. Use a good accurate level directly on the ground glass.
A good and accurate level is one in which the edges of the bubble exactly touch the engraved lines when level.They are difficult to find.
Jim

I find it far better to use a grid on the GG. Your subject may not be level/plumb. And as you pointed out, accurate levels are hard to find. You won't find them at bestbuy or wallyworld.

Edit. After looking at the photo, I can't see an issue of parralellism. It really does look bulged - convex. Vey odd.

[Nathan Potter]

Your second image shows a small degree of classic "pincushion" distortion when examining the four edges of the film. Note the indented frame at the center of the four edges. It is less than 1% indentation at the center. This may not be unusual for the lens you used.

Petronio has a good point. Is the lens up to factory spec or could it have possibly been modified slightly in the past? Check very carefully that the pincushion is on the film; which seems likely, rather than a function of the scanner.

Nate Potter, Austin TX.

[E. von Hoegh]

Your second image shows a small degree of classic "pincushion" distortion when examining the four edges of the film. Note the indented frame at the center of the four edges. It is less than 1% indentation at the center. This may not be unusual for the lens you used.

Petronio has a good point. Is the lens up to factory spec or could it have possibly been modified slightly in the past? Check very carefully that the pincushion is on the film; which seems likely, rather than a function of the scanner.

Nate Potter, Austin TX.