Photos always tilted clockwise, even though crosshairs are aligned when taking photo

[jasonxweb]

Camera Info:

Type: DSLR
Make & Model: Nikon D3000
Lens: AF-S DX Nikkor 18-55mm f/3.5-5.6G VR
Year: 2010 - purchased new


Hi everyone,

I am having a problem where all the photos I take always show a slight clockwise tilt - most noticeably near the top edge of the photo. While this tilting is slight, it is very noticeable, and has a very negative effect on my photos, because I am using my camera to document artworks.

To further explain, my artwork documentation photos usually contain a rectangular piece of artwork that fills most of the photograph - so the edges of the artwork paper are close to the edges of the photograph, and the slight tilt becomes very noticeable and makes the artwork documentation look terrible and misaligned.

Of course the first thing that I checked was to make sure that my camera is aligned properly when I take the photo. I made sure that the axis of the camera is perpendicular to the artwork, and that it is centred on the artwork, and that the angle of the vertical and horziontal crosshairs line up with the paper's angle. However, my photos always still show a clockwise tilt.

Have you seen this kind of problem before? Maybe it is a problem with the alignment of the crosshairs in the viewfinder? Or may an issue with the alignment of the digital sensor? Or could it be a problem with the lens?

To demonstrate the problem, I have taken some test photos of a grid on a piece of paper. The views through the viewfinder are also shown, and were taken with a separate camera. I hope these photos will help you to see the problem.

As you can see in

19151

, when I took a photo of the test paper, even though my crosshairs were aligned to the grid on the test paper, the resulting photo shows a very noticeable clockwise tilt - especially near the top edge of the photo.

Now, to make sure that the problem isn't caused by my camera being off-axis when I took the photo, I have taken some off-axis photos: (When the camera is off the center axis, I still pointed and aligned the angle of the crosshairs to the grid on the test paper)

19152

19153

19154

19155

The results of these test photos seem to rule out the possiblity that the problem is being caused by the camera being off-axis when the photo was taken, and the results seem to confirm that there is always a clockwise tilt, regardless of whether the camera is on-axis or off-axis in taking the photo.

Do you have any ideas about what is causing the clockwise tilt?

Thank you very much!

Jason

[Marko]

Hi Jason,

Welcome to the forum.

Thanks for posting - weird.
First questions though are designed to eliminate possibilities - don't mean to be daft in any way.

1 - are you using a level on your camera (you never mentioned how you are checking alignment) and is your wall perfectly square?
2 - Is your mat cut perfectly even at the top and the bottom
3- Is this a task that you do all the time with flawless results? If so, using this same gear?

[ericmark]

It is an interesting problem. I would normally not bother too much at getting camera square I would correct after taking the photo with software. I am sure one can buy a lens designed for a set distance with spot on optics but in the main to get spot on with a lens as a set distance needs some correction be it for vignetting, distortion, or other errors. When converting from the RAW file RawTherappee has a 85 profiles just for Nikon cameras to correct the lens. Other makes are similar.

Using a plate camera with adjustment screws for the lens mounted on a bellows one could using the opaque viewing screen position the lens to correct many problems but today in the main we use a software fix.

I have not really got into using Nikon but with my Pentax the focus screen which superimposes details onto the viewed image is removable one can buy special versions clearly if this is not fitted correctly or if there is a fault on this it could stop accurate aliment. Again on the Pentax the CCD is fully floating held with magnets as part of the anti-shake system a fault on this could also cause a problem in accurate aliment. However the viewed area is only around 95% of the area of the CCD so framed correct it would always leave a small amount requiring trimming so when one crops off the extra any small error would be corrected.

However I would be looking at the focus screen first 19156 I would guess either faulty or more likely not fitted correctly. Can't think of any other likely cause. One shown is Nikon D800 similar system with my Pentax but there is nothing in the instruction book about removal or refitting this item and on the Pentax it was really a fiddle to remove and refit. With Nikon it shows a special tool used to remove it. I would say this is really a job for the pros and not a DIY thing. I did remove mine on the Pentax dirt got behind the focusing screen when I swapped lens without enough care.

Personally I would not bother trying to remove and would just correct in Photoshop. But out of interest what is it like to manual focus with your camera? If the focus screen was not fitted squarely I would have expected there to be other faults like being hard to manual focus.

I would agree with Marko using a level on camera and item on the wall will show if the CCD or other sensor is miss aliened or viewfinder. I see replacement focus screens are available Nikon D3000 D5000 Focusing Screen - KatzEye Optics (http://www.katzeyeoptics.com/item--Nikon-D3000-D5000-Focusing-Screen--prod_D5000.html) with Architectural grid lines which makes me think the camera should not have this sort of fault.

I think the Nikon anti-shake is in lens not in camera like on my Pentax not a clue how it works in lens but if anything like the Pentax then a fault could cause what you describe so question again does it do the same with all lenses?