Film Flatness Test

[ic-racer]

I just got a camera with 1/2 roll of expired film. I removed the film from the camera and put it in a Rolleiflex SL35 to test frame-by-frame flim flatness.

This Rolleiflex SL35 was set to focus 0.15mm in front of the pressure plate.

With in mind, I used an Autocollimator to measure the front surface of the film on each frame from 1 to 24 and recorded the deviation from 0.15mm.

The film was Kodak Supra 100.

 

[koraks]

At least it's fairly consistent.
Also interesting to know would be what the measurements are if the film is left in position for 1 hour and/or 1 day.
And also what the numbers are in the center of the frame vs. the top and bottom edge. I expect the latter to show more deviation.

 

[Mr Bill]

At least it's fairly consistent.
Also interesting to know would be what the measurements are if the film is left in position for 1 hour and/or 1 day.
And also what the numbers are in the center of the frame vs. the top and bottom edge. I expect the latter to show more deviation.

Yeah, I would agree. I wet expect that relative humidity would have the most significant effect, if you had a way to manipulate it.

I don't think there's any way to measure anything other than near the center of the lens axis with a normal autocollimator.

Something I did a bunch of years back came from Norman Goldberg's book, Camera Technology - the Dark Side of the Lens (I think). The idea is to put a beamsplitter right in front of the camera lens, at a 45 degree angle. A light box, with a grid drawn over the front, is set to be reflected into the camera lens. With the shutter open the grid is imaged onto the film (focus the camera on the grid). And... a person can look directly into camera lens (through the beamsplitter) to see the grid. If the film is bowed to any significant degree the grid will appear distorted.

I did this just for fun with one of our (the company's) studio cameras. And to my surprise there was a distortion near the center on one of the long sides. These custom built cameras, in use for years, and this had never been noticed. We actually went back and slightly repositioned a roller on every film magazine (during routine periodic service over several years); this removed the slight ripple.

 

[ic-racer]

Great comments.

The Nikon N90S I just acquired was shown in the ad with the back open and half a roll of film exposed. Since the camera had been sitting some time with the film that way, I thought that roll would be good for this test. So, I did remove the film from the N90S and re-loaded it into the SL35 test camera. The film did have an obvious curl to it. It was sitting at about the 10th exposure in the N90S.

After loading into my SL35 test camera with 50mm test lens, all the measurements were done from the center of the film, though the NatCam c6400 Autocollimator can observe all 4 corners of the film.

T

 

[BrianShaw]

Two potentially confounding or causal factors to consider are film tension consistency and film base thickness uniformity.

Very nice measures. I haven't seen your hypothesis or conculsion yet, though.

 

[ic-racer]

The data are a spinoff of understanding the answers to these questions: 'Where should the focal point at infinity reside' and 'irrespective of theory, what is a good practical target'

Where should the focal point at infinity reside? On the film surface.

What is a good practical target? I found the film surface is NOT a good practical target, due to its irregularity (as shown in the first post). The pressureplate minus 0.13 (HP5 thickness) to 0.17 (shop manual spec) is reasonable. However, every test lens will have a different reading at 0.17 on the vernier scale of the collimator. Of course the expected readings on the vernier scale for every lens can be calculated. What I wound up doing is using the focus screen as a practical target. The focus screen on the test camera is exactly 0.17mm away from the pressureplate.

 

[Harry Callahan]

If you`re measuring in the range of 0.01mm dust should be a factor to consider. The digital picture of the film sitting in the camera isn`t that clear, but i think i see some dust on the film. If the inside of the camera hasn`t been cleaned thorough, there also can be dust on the film outside the cartridge.
At 0.01mm even a fingerprint on the film base may be measurable.

Also an acetate-base does shrink by age. I don`t know how much it will shrink in thickness and how old the film was you tested - but as acetate also does age depending on humidity you may have gotten these results because some of the film was sitting inside the cartridge (for years?) and some outside.

I`d use new film for such a test - and watch for fewest dust possible.

"Where should the focal point at infinity reside? On the film surface."

Not quite sure about this. If a film has many layers, like color film, the focal point of infinity may reside in the emulsion - not on the surface. I`m not sure about this, maybe this was from a time when emulsions of films were thicker than today... but i think i read something about this, somewhere, sometime...

 

[ic-racer]

The 'old' film in post #4 was transferred to a Rollei SL35 for the test. Rather than wait a week between each frame, I used the 'old' film that had been sitting in a camera for a few years.

 

[ic-racer]

Again, interested in anyone else's results. Prior to getting the Autocollimator, I simply took a pencil and poked the emulsion. It was easy to tell if it was sitting flat against the pressure plate or not.

I had a Rollei SL2000F that would not hold the 1980s version of Tmax film flat at all. It bulged every frame; at least it was consistent and the focus screen and lens were adjusted to match.

 

[BrianShaw]

The data are a spinoff of understanding the answers to these questions: 'Where should the focal point at infinity reside' and 'irrespective of theory, what is a good practical target'

Where should the focal point at infinity reside? On the film surface.

What is a good practical target? I found the film surface is NOT a good practical target, due to its irregularity (as shown in the first post). The pressureplate minus 0.13 (HP5 thickness) to 0.17 (shop manual spec) is reasonable. However, every test lens will have a different reading at 0.17 on the vernier scale of the collimator. Of course the expected readings on the vernier scale for every lens can be calculated. What I wound up doing is using the focus screen as a practical target. The focus screen on the test camera is exactly 0.17mm away from the pressureplate.

Thank you!

 

[naaldvoerder]

A case for theContax RTS III…

 

[Harry Callahan]

Apparently i am talking incomprehensible.