I am opening this thread because I have had a long story with my Ensign Selfix 820 folder camera, with the Ross Xpres 105/3.8 lens. And i have had many other folder cameras in the past and there's always the task of aligning the focus scale to the actual focus achieved on the film plane.
* a note on the film plane: Finding out the real film plane is a task in itself in many folder cameras, because, for example in the Agfa Record series, there are no "film rails" a la 35mm cameras, where the film contacts the rail. On the Agfa Record series, and also other folders like the Zeiss Nettar, the film rollers and the pressure plate set the film plane position, so for those cameras, IMO, it's best to use some Scotch "Magic" tape rolled onto two 120 spools, and then fitting it very taut into the camera, so the Scotch "magic" tape simulates film under tension. And then focus on that 'film plane'. Now, the Ensign Selfix, fortunately, is a different story, because it does have film guide rails, and you can fortunately put a ground glass over them.
The 105/3.8 is a four-element tessar-like lens, and at f3.8 it is a fast lens for a 6x9 machine. It is also a front-cell-focusing lens, like in the vast majority of MF folder cameras. Thus, the lens design is compromised in performance, due to the fast speed and due to the front cell focusing.
What happens with this? The problem is that the performance at close distances and at infinity will differ strongly. And there will be strong curvature of field at one or the other distances. I stumbled with this problem frontally with the Ross Xpres. But it's not the only lens that shows this problem and what I'll mention here will be applicable to other folder cameras.
A camera lens never focuses to a 100% flat image plane. Thus, if one calibrates the lens for maximum central sharpness, the edges will be a bit soft, and if one calibrates for best edge sharpness, the center will suffer in resolution. Of course, some lenses are better corrected for "flatness" than others (sometimes to detriment of other parameters). But the effect is that if you calibrate for center sharpness, edges will suffer, and viceversa. This can be readily seen in this link:
http://www.hevanet.com/cperez/MF_testing.html
See the resolution figures for the miscalibrated Rolleiflex with the 75/3.5 Xenar. Or the figures for the Kodak Special Six-20 at different focusing distances.
Now, what happened on the Ensign Selfix 820 is that when i calibrated the infinity focus for maximum central sharpness at wide open aperture, infinity-focused (landscape) pictures were sharp centrally, but the edges were terrible until stopped down to about f16... AND, normal distance (3m-8m) pictures were not having correct focus.
So I researched the subject. I've read on the 'net that a front-cell-focusing lens is usually optimized for best performance at "40x the focal length", so in the 105mm Xpres that would be about 4m. Thus, i reasoned that, in distances closest to 4m, the lens should perform good at both center and edge definition and the focus scale ring would be calibrated using such distance as a starting point. The lens has no mark for 4 meters, but it does have one for 5 meters.
What i did, then, was to calibrate the focus ring so wide-open central focus at 5m is correct for a subject located exactly 5m away. Good, because images at 3m and 2m also were being correctly focused on the ground glass. But what about the infinity focus mark?
Now here comes the SURPRISE: Afterwards, when setting such lens to the infinity position, the lens was NOT correctly focused for best wide-open sharpness in the center!! In fact it was focused "beyond" infinity by some margin. But, at such position, the "middle" or "near edge" zone of the image was correctly focused. When stopping down to about f8, the center became sharp again and the performance was uniform in the ground glass. Except for the extreme edges, which only became pin-sharp at about f16, but I can live with that.
So, what does this means? That doing the procedure of "aligning the focus ring's infinity mark with a sharp center image on the ground glass" is NOT always a good idea!!
Of course, i need to test the camera with film, since the film image plane position will not always be exactly where the groundglass was sitting, but it gives me a starting point for giving such machines a better calibration.
* a note on the film plane: Finding out the real film plane is a task in itself in many folder cameras, because, for example in the Agfa Record series, there are no "film rails" a la 35mm cameras, where the film contacts the rail. On the Agfa Record series, and also other folders like the Zeiss Nettar, the film rollers and the pressure plate set the film plane position, so for those cameras, IMO, it's best to use some Scotch "Magic" tape rolled onto two 120 spools, and then fitting it very taut into the camera, so the Scotch "magic" tape simulates film under tension. And then focus on that 'film plane'. Now, the Ensign Selfix, fortunately, is a different story, because it does have film guide rails, and you can fortunately put a ground glass over them.
The 105/3.8 is a four-element tessar-like lens, and at f3.8 it is a fast lens for a 6x9 machine. It is also a front-cell-focusing lens, like in the vast majority of MF folder cameras. Thus, the lens design is compromised in performance, due to the fast speed and due to the front cell focusing.
What happens with this? The problem is that the performance at close distances and at infinity will differ strongly. And there will be strong curvature of field at one or the other distances. I stumbled with this problem frontally with the Ross Xpres. But it's not the only lens that shows this problem and what I'll mention here will be applicable to other folder cameras.
A camera lens never focuses to a 100% flat image plane. Thus, if one calibrates the lens for maximum central sharpness, the edges will be a bit soft, and if one calibrates for best edge sharpness, the center will suffer in resolution. Of course, some lenses are better corrected for "flatness" than others (sometimes to detriment of other parameters). But the effect is that if you calibrate for center sharpness, edges will suffer, and viceversa. This can be readily seen in this link:
http://www.hevanet.com/cperez/MF_testing.html
See the resolution figures for the miscalibrated Rolleiflex with the 75/3.5 Xenar. Or the figures for the Kodak Special Six-20 at different focusing distances.
Now, what happened on the Ensign Selfix 820 is that when i calibrated the infinity focus for maximum central sharpness at wide open aperture, infinity-focused (landscape) pictures were sharp centrally, but the edges were terrible until stopped down to about f16... AND, normal distance (3m-8m) pictures were not having correct focus.
So I researched the subject. I've read on the 'net that a front-cell-focusing lens is usually optimized for best performance at "40x the focal length", so in the 105mm Xpres that would be about 4m. Thus, i reasoned that, in distances closest to 4m, the lens should perform good at both center and edge definition and the focus scale ring would be calibrated using such distance as a starting point. The lens has no mark for 4 meters, but it does have one for 5 meters.
What i did, then, was to calibrate the focus ring so wide-open central focus at 5m is correct for a subject located exactly 5m away. Good, because images at 3m and 2m also were being correctly focused on the ground glass. But what about the infinity focus mark?
Now here comes the SURPRISE: Afterwards, when setting such lens to the infinity position, the lens was NOT correctly focused for best wide-open sharpness in the center!! In fact it was focused "beyond" infinity by some margin. But, at such position, the "middle" or "near edge" zone of the image was correctly focused. When stopping down to about f8, the center became sharp again and the performance was uniform in the ground glass. Except for the extreme edges, which only became pin-sharp at about f16, but I can live with that.
So, what does this means? That doing the procedure of "aligning the focus ring's infinity mark with a sharp center image on the ground glass" is NOT always a good idea!!
Of course, i need to test the camera with film, since the film image plane position will not always be exactly where the groundglass was sitting, but it gives me a starting point for giving such machines a better calibration.
