PDA

View Full Version : Mysterious focus issue



Julian Boulter
28-Jun-2005, 08:34
Hi all,

just got my chromes back from a recent trip with very disapointing results. I have a nice shot of a dried up river bed, rocks converging into the distance with a centred tree in middle distance and distant mountains. I took 2 shots, one with standards parallel (focussed about 1/3 in which was approx just in front of the tree), and one with rear tilt both at f32. Both shots exhibit an obviously out of focus area of the right side of the river bank (viewed with a 4 x loupe) which is to the right hand side of the tree which stretches to the edge of the frame and upwards to the top of the frame. The thing that confuses me is that the equivalent area on the other side (the left river bank) which is at the same distance as the out of focus area, is perfectly sharp. I have been wracking my brains but I just can not see how the plane of focus especially when parallel (and even when tilted) can allow this to happen! My only thoughts were the following:

1. Everything was not centered and somehow swing was introduced (but I'm almost certain this was not the case)

2. Temperature changes (I was camping - cold and wet at night and hot during the day) had caused the camera to warp somehow and produce the effect

3. The film was unevenly placed in the holder (but the effect was very pronounced even under a low powered 4x loupe)

Equipment used -

Wista DX II with Nikon 75mm SW 4.5 lens, Fuji QL back - everything purchased new and used throughout the last year without problem.

I am at a loss and it's really bugging me!!!

Steve Hamley
28-Jun-2005, 08:42
Julian,

If you're using conventional film holders and manage to get one side of the film on top of the rail rather than under it - well, that side will be out of focus. Just a thought.

Steve

Brian Sims
28-Jun-2005, 09:57
Steve's guess could explain it if the orientation was vertical. However, you said the problem was on both shots...so you would have had to mis-load both sheets of film. The way I make sure I've loaded the sheet correctly is to pull up slightly on the end of the film. There should be equal resistance on both sides to indicate both sides are under the rails. Don't pull to hard...just takes a little touch. If you were shooting horizontal a mis-loaded sheet would not explain the problem. Then I would look to the possibility that you knocked the film plane out of alignment while loading the holder...although with a 75mm at f32 it would have to be a significant knock. Even after many years of shooting with a view camera, I still worry that I've screwed some adjustment up after loading the film holder. On most shots, I will recheck the gound glass after exposing the film. It only takes 30 seconds...a small price considering the the long hike in and the time to set up the shot.

Julian Boulter
29-Jun-2005, 02:59
Thanks for the response guys only its a Fuji Quick Load back and film so not a possibility that the film was loaded incorrectly - I know that quite a few have reported that these are not as sharp as conventional holders but I would not expect such a pronounced out of focus area on just one area of the film. Having said that I have just located an article on this site indicating that under test conditions areas of the QL sheet were found to be out of focus while other areas were sharp but this was only obvious under 20x magnification.

Leonard Evens
29-Jun-2005, 07:42
The others are probably right, but there is also the possibility that the standards are not parallel. Sometimes there is play in the lock down knobs or whatever you have, and that could result in lack of alignment. That could produce an unwanted swing, which could produce an effect like that you describe. A simple way to check if the standards are parallel when in the default detent positions is to turn the camera so it points vertically down. Put a level on the rear standard and adjust the tripod so that it is level in all directions. Then move the level to the lens and press up on the lens housing with it. For the latter a small torpedo level should work. The two standards may not be completely parallel, but they should be close to parallel.

I've found with my camera that the shorter the focal length of the lens, the more obvious the effect on focus from a small departure from parallelism.

N Dhananjay
29-Jun-2005, 07:53
It is a little puzzling. Some possibilities to chew on...

1) Your camera is out of whack (either or both standards). Don't know if it has detents, but if it has taken a knock, now in the detent or zero position maybe it has a small amount of swing. If this was the case, you would see this on the ground glass.

2) Crud under the back. Sand particles etc can result in your film holder not sitting absolutely flat in the camera back, same result as introducing a small amount of swing. Although again, with a 75 mm lens, it seems unlikely such minor shifts would affect focus so much.

3) Maybe your QL is problematic. Maybe it took a blow sometime and the pressure plate (or whatever mechanism is inside a QL) is bent/damaged. Is this problem on every sheet of film you shoot?

4) I'm sure you worked carefully but is it possible you were on the edge of DOF limits? If you check focus on the GG with a 4X loupe and stop down just enough to get the extremes into focus, there are enough small slops in the system that you might not have enough DOF on the film. And while the left bank is in focus, maybe there was enough difference in the distances that it just snuck in under the DOF limit while the right bank was out. Sounds a little implausible but if it is the only sheet on which you had a problem....

Shooting some test shots (a flat plane, something like a brick wall or sheet of newspaper) should allow you to rule out most of these sources, at least those related to film plane and parallel standards etc.

Cheers, DJ

Leonard Evens
29-Jun-2005, 13:49
It seems paradoxical that a short focal length lens should show a small swing more than a longer lens. But I believe that to be the case. As I noted above, I never noticed a problem with my lenses down to 90 mm, but when I purchased a 75 mm lens, I found there was clearly an assymetric loss of focus. I thought initially there was something wrong with the lens, but further testing showed it was fine. When I checked the standards I found that the small amount of play in the detent positions resulted in a small swing. I took greater care in making sure the standards were parallel, and I also was much more careful to be sure the image was symmetically in focus across the field, and that resolved the problem. My 75 mm lens is a f/4.5 Rodenstock-M, and I have a Maxwell precision Optics focusing screen. So image dimness is not really an issue. But it was still easy to miss the fact that the image was going out of focus on one edge while remaining in focus in the middle and the other edge. I later did some theoretical analysis of the situation and convinced myself that in fact a swing has a more dramatic effect for shorter focal length lenses. I'm not sure of my analysis, so I would be interested in any comments anyone else has about the matter.

Julian Boulter
30-Jun-2005, 05:51
DJ,

I am fanatical when it comes to looking after and cleaning my gear (too much so actually!) so that rules out 2) and 3) and I cannot see that 4) is possible given that other areas at the same distance as the off-focus area are perfectly sharp however 1) is possible I guess which makes Leonards post very interesting.

The Wista DX has 2 levers (one on each side) to lock the front standards at the required infinity stops. I purchased the camera new and there was no play in the attachment of the lever mechanism to the base however it has become somewhat loose (mabye 3 or 4 mm of swing is possible) so I have been very careful to try and keep the standards parallel prior to shooting but I would never have expected such a dramatic result with perhaps just a tiny swing error due to this looseness. I'm suprised to hear that Leonard that the wider lens enhances the effect! I will obviously have to perform some tests myself to really pinpoint the problem but I woould be interested to hear if any other Wista DX users have experienced this issue with play in the front standard and how to address it.

Thanks

N Dhananjay
30-Jun-2005, 20:00
I think Leonard is right that the effects of swing/tilt will be greater with shorter focal lengths. I haven't had the time to do the more formal analysis to confirm to myself. But it sounds correct. If we draw the Scheimpflug line and the hinge line, a given angular displacement results in more dramatic shifts in the plane of focus with a shorter focal length than a longer focal length. I haven't thought about the DOF issues here though - I would imagine the shorter focal length would have greater DOF but maybe that is insufficient to compensate for the greater shift in the plane of focus.

The easiest way to check if your standards are out of true is probably to use something like the zig align mirrors. If you don't have those, you could still set up the camera level and perpendicular to a brick wall with all controls zeroed and see if there is difference in sharpness across the frame.

Cheers, DJ

N Dhananjay
30-Jun-2005, 22:18
OK, I was curious enough to try thinking this through and I think this is correct but I'm a little stuck.

DOF at the hyperfocal distance with a tilted lens is given by J = f/sin(a) where f is the focal length and a is the angle of lens plane relative to film plane. The location of the Scheimpflug line can be calculated as Ds = BE/tan (a) where Ds is the distance below the centre of the film plane (assuming no tilts and shifts) to the Scheimpflug line and BE is the bellows extension (which would be f when the subject is at infinity). The location of the hinge line can be similarly calculated as Dh = f/tan (a) where Dh is the distance below the lens nodal point to the hinge line. When a = 90 (i.e., all controls are zeroed), We can see that (for any gievn a) Ds and Dh are proportional to f (i.e., a longer focal length will result in Scheimpflug and hinge lines being far below the center of the film and the nodal point respectively i.e., smaller change in the plane of focus whereas shorter focal lengths will result in more dramatic changes to the anglular displacement of the plane of focus). So far, so good. But I'm stuck in terms of trying to draw the wedges of DOF to figure if there ends up being a difference or not. It sort of makes sense intuitively but I'd be interested in other thoughts.

Cheers, DJ

N Dhananjay
30-Jun-2005, 22:37
Alright, some more addition. Imagine a line emanating from the Scheimpflug line to a point in subject space that is the point focussed on. This distance can be calculated as sqrt of (O^2 + Ds^2). We already know that Ds is a function of f and can substitute that. So, the ratio of the lengths of these lines as a function of the focal length can be calculated.

Now imagine the wedges of DOF emanating from the Scheimpflug line. Obviously the longer these project into object space, the greater the DOF. And they project out longer for the longer focal length. If we have the angle of the DOF wedges for the two focal lengths (or the ratio of the same), and the ratio of the length of the two lines referred to in the previous paragraph, we should be able to prove this.

Sorry for the clumsy writing - this was much more obvious and easy with the figures I drew out.

Cheers, DJ

Julian Boulter
1-Jul-2005, 06:30
actually DJ you have raised a point which I have been pondering recently (apologise it is off your main subject though).

I had always assumed that a DOF area projected out indefinitley- so for example, if the standards were parallel, the 2 limits of DOF in front and behind the plane of focus project upwards and sideways way outside the picture area. Are you saying the limits of the projection depend on focal length?

Julian

N Dhananjay
1-Jul-2005, 06:49
In the normal situation when standards are parallel to one another, you can focus on one plane in image space parallel to the film plane. This plane extends infinitely above and below but that is the plane of focus. The DOF area extends in front of and behind this plane. When you tilt the back or front to establish the Scheimpflug relationship, the plane of focus has also got tilted, so it is no longer parallel to the film back, although this plane extnds infinitely out into space (except that the plane no longer runs vertically up and down but at an angle). Now the DOF limits extend on either side of this plane of focus. However, since different parts of the plane are now at different distances from the lens, DOF varies with how far from the lens each point on the plane is. In effect, the DOF limits look like a wedge emanating from the Scheimpflug line. This is independent of focal length. Cheers, DJ

Julian Boulter
1-Jul-2005, 11:20
Hi DJ,

I understand the principals of Scheimpflug - can we summarise by saying that:

1) when standards are parallel both the POF and DOF planes extend infinitley upwards and sideways

2) With tilt both POF and DOF planes extend infinitley outwards (in a wedge)

?

Thanks

Julian