[Building a camera] I would like help developing a function.

[Zippie]

Hello.

So, I am building a camera and have ran into a problem that I feel could be solved with a semi-simple function.

There is a reel-to-reel system for winding the film from a spool to a dowel of 3/4 inch diameter. I need to know how many turns of this dowel would result in advancing the film a certain distance. I could find the circumference and multiply that by number of turns, but as the circumference increases each turn, this would prove inaccurate after the dozens of turns it would take to reach the end of the reel.

I am asking for a function that takes the number of times turned (or rather which turn it is, to avoid a zero) and outputs the number of turns next needed to advance the film a constant amount.

Thank you.

 

[berkeman]

Hello.

So, I am building a camera and have ran into a problem that I feel could be solved with a semi-simple function.

There is a reel-to-reel system for winding the film from a spool to a dowel of 3/4 inch diameter. I need to know how many turns of this dowel would result in advancing the film a certain distance. I could find the circumference and multiply that by number of turns, but as the circumference increases each turn, this would prove inaccurate after the dozens of turns it would take to reach the end of the reel.

I am asking for a function that takes the number of times turned (or rather which turn it is, to avoid a zero) and outputs the number of turns next needed to advance the film a constant amount.

Thank you.

Welcome to the PF.

This would be a simple problem involving the thickness of the film (right?), but it is complicated in the real world by the tension of the film affecting how tightly it packs on the pickup dowel.

You could try to experimentally find out what the linear film rate is versus time versus the tension, or a better approach by far would be to use film that has perforations along the edges like regular 35mm film. It will be challenging to make an accurate system without some sort of fiducial markings or perforations on the film.

Can you say more about the application and design constraints?

 

[Zippie]

Thank you.

It is indeed a matter of how tightly it is about the pickup dowel, but it will be kept quite taunt so the variance from an ideal situation could be left out.

The film has marking on the back that could be used as reference, but no suitable method for reading them could be devised for this particular design (it would be far too awkward to add a peep-hole as the plane of focus is actually rounded).

I had thought that a function could be devised that took into account the number of turns, n, and the starting radius(.75) plus the thickness of the film(w), and used the circumference(advance) desired to solve for N... to give the desired number of turns. Or, something of the like. I can't manage to create a suitable method of calculating this.

 

[Stephen Tashi]

Is there any reason why this camera does not use the typical sort of film that has evenly spaced holes in the margin where a sprocket gear can engage the film?

 

[Zippie]

Yes, the camera's lens system allows for a very wide view angle of 152° which at the focal length would develop a circle of film [upon said curved 'plane'] of 3,1 in or 78mm. Typical 35mm film would not be able to take advantage of this. 120mm film also has a paper backing which makes it easier to avoid light leaks.

 

[berkeman]

Then I would suggest using pinch rollers along one or both edges of the moving film, to measure the movement of the film. It's not going to be accurate enough to use the turning motion of the pick-up dowel alone, IMO.

 

[Zippie]

That's a good idea. Thank you, Sirs.