I have been testing the 127mm Ektar of my 'Crown and I have noted a strange phenomenon.
When testing for definition in the corners and centre I shot a builiding.
In the centre all is normal. The edge is soft, but then I am talking an equivalent of a 8X magnification of the neg.
The front door of the building had verticals aligned thus: / . All horizontal lines were undistorted.
Looking along the edges of the image I noted that verticals half way along the edge were normal.
What is this?
The effect is still there even stopped down to f22.
Its not as if the negative was improperly loaded in the holder because I check when loading the holders.

Does the effect become more noticeable as you move away from the center of the image? If I were you, I'd square the camera up and take a photo of a relatively new brick wall from a distance of about 15 feet. It's hard to tell, with architectural features -- I have seen a lens accused of producing "barrel distortion" by someone focusing it on a row of support columns which were, on closer inspection, decidedly bowed to the left!

Are you sure the camera (film plane) was aligned square and level with the building?

How flat was the film in the holder?

I have used these holders a lot and take care when loading so that I don't get the edge of the film in the groove for the slide.

It just seemed very odd the way the door appeared skewed, because I assure you it is perfectly vertical in reality!

The effect is more pronounced towards the corners of the image. It was only when I aligned the subject along the bottom of the negative that the effect became apparent.

The camera front was not absolutely vertical because I was tilting the camera up on the tripod.

Any guesses fellow Graphic users?

It would help greatly if we could see the photo. Very difficult to diagnose by description only...

Do the verticals slant to the right all the way across the frame, as in your illustrative ASCII character, or is there a transition from a rightward to a leftward slant from one side of the negative to the other? You'd expect them to do that if you were tilting the camera up, of course. How does what you're looking at differ from perspective convergence?

Quote:

On 2004-03-24 12:05, primus96 wrote: The camera front was not absolutely vertical because I was tilting the camera up on the tripod.

So does this mean the back was tilted to? Or does your camera have rear movements?

Well if you were using your crown as stated then your back was tilted and probably cause the problem your describing as the crown has a solid non-moving back and if you were tilting it up then your top was out of the focal plane..

Dave

The / character represents the effect on verticals towards the bot LH corner.
At the bot RH corner the effect seems to be \.
Exactly half way between corners there is no effect.

How does the effect differ from the convergence that you'd expect to occur from tilting the camera upward?

When tilting the camera upwards you get converging verticals : / \.
This effect is // in the lower LH corner & || in the RH corner

Check for lens board flatness, mounting square and flat to standard and standard being square to bed/body/film plane.
Remove the front and rear cells of the lens and look for uneveness between any elements, correct as needed.
It is possible someone had it apart to clean and did not get the elements back in correctly. Remounting the lens on the board to a different orientation will cause the error to shift accordingly if camera/lens board are correct.

Typing error in my last post!
I should have said the effect on verticals was / / in the bot LH corner and \ \ in the bot RH corner.
If necessary i,ll go shoot another test.

I think you'd do best to choose some sort of rectilinear grid as a test subject -- a brick or concrete-block wall, for example -- and get the camera back as parallel to it as you can, using a level and a tape measure as necessary. It also might be instructive to make a horizontal exposure, then a vertical, and see whether the effect rotates with the camera or is independent of orientation.