Hi all,
I purchased this toy projector recently and was amazed by its solution for intermittent motion. I’m putting this video here for you so you can help me evaluate it’s good/bad implications. Please post your thoughts in this thread!
It looks like it would put a lot of tension on the topside of the sprocket during that process, the topside is usually in good shape as it is the bottom-side that has been hit by the pull-down usually during projection on a normal projector.
When designing scanners that are sprocket driven, we usually make them so that the pull-down or sprocket feed engages the top of the sprocket instead of the bottom for that reason.
I personally wouldn’t want to put my film through something like the mechanism in the video, especially acetate film, but it certainly is cheap and simple.
I have been thinking on this more, and perhaps to make it less harsh on the film, but utilise the same principle, instead of having the lever engage and disengage the sprocket-holes directly, you could have a sprocket wheel, that had ‘bumps’ on it, and a similar lever mechanism would allow the wheel to turn ‘one frame’ and then lock. A bit like a ratchet.
The sprocket-wheel would then pull the film down, (or more to the the point, it would stop the film after it had moved one frame through the gate) and the lever mechanism would only allow the wheel to advance exactly one frame. If 3D printing the sprocket-wheel it would be an easy design, and barely more complex than the design shown, but a lot less stressful on the film. If it was a larger diameter sprocket wheel, then it would handle worn or even missing sprockets.
Maybe I’m misunderstanding, but the graphic above wouldn’t actually stop the wheel from moving forward (counter-clockwise), but it would prevent backlash or reverse (clockwise) motion. Now if the sprocket wheel was turning clockwise, it would most certainly only allow movement through one notch each time the lever is released, But to release the lever, you’d need some other mechanism to lift it up far enough.
I’d think that for a “best case” scenario, a large drive wheel that engages several sprocket holes from the top would be the easiest/cheapest to implement. That single wheel would spread any force over several sprocket holes, and if engaged from the top of the hole similar to the toy projector, it would avoid some issues with worn film. Additionally, that wheel could be used as the reference for when to take each image similar to the original Kinograph.
Maybe via having a large drive wheel as I’ve said, we could devise an approach that pulls the lever up based on the movement of the main drive wheel rather than from film tension?
The graphic is just something I grabbed off the net quickly to demonstrate the ratchet concept, not what would be used.
I’ll draw up a design and print one and test it.
Just a more basic question - If the drive wheel in the toy projector engages the sprocket holes of the film, then why is it even necessary to engage them at or near the gate?
Edit - I watched again and I get it now. It’s using the tension of the film against the lever arm to move the sprocket-catching nubbin away from the film. That allows it to advance, which creates enough slack that the arm re-engages the sprocket holes. This creates tension again, but the film is stopped until enough tension builds up to pull the lever arm again.
I guess I don’t like that idea all that much since you’re effectively putting the film in tension in multiple spots. It would also seem difficult to control the tension within the gate, since the are of tension (between the engaged sprocket hole and the lever arm) all sits below the gate. This leaves the film within the gate under very little tension and would rely on squeezing the film to keep it flat, right?
Just another thought…how do the optical sensors in a computer mouse detect motion? Many gaming mice specify dpi measurements in the 1000’s. Would it be possible to use one of those to detect how far/fast the film is moving?
The sprocket wheel on the toy projector is just pulling the film, providing motion, the lever at the gate engages the film to hold it stationary in the gate for projection.
Optical mice work by looking at the difference between what its sensor sees in a small area under the mouse. You might be able to get the speed of the sprocket hole as it moves by… But most of the time there might be no signal when there is no sprocket in the field of view of the mouse.
You could change the lens perhaps, to see a larger area, but then you would probably lose the resolution you need.
This sensor has 900 pixels, which I would guess are in a 30x30 array.
at the highest resolution, 2000dpi ~80px/mm, you would see <0.5mm, how big is a sprocket hole?
I think you might be able to get the average film speed from a mouse sensor, but not any absolute position information about the sprockets
When I was A very little kid, (half century ago now) I had a Brumberger toy projector that worked on more of less that same principle, but just had a broad flat spring resting on the film, and once the spring came up enough, the sproket would escape the pin, and the spring would pull the film quickly to the next position where the sproket hole would again be caught. (same principle but even fewer parts.) there was only regular 8 of course in those days.
The tops of the sprocket holes did get worn on the three films that came with the unit. plus there was some scratches from all the sliding.
I guess as a Kid I ignored the lack of a shutter.
Looking arround here is one of the ones like I had being sold at a nostalia Price. http://www.ioffer.com/i/brumberger-battery-operated-8mm-movie-projector-549653844
I have not gone through the complete forum, and may be missing if a decision was made on a continuous motion, sprocket, etc. I came across this video that has an interesting mechanism to move the film (see second 28).
I was thinking if it would be possible to build a system that instead of inserting the pin in the sprocket hole would be integrated by two arms opposite to each other, which would press the film in opposite direction and move it forward.
1 Like
I had not seen the blog corresponding to the video, which includes additional information (http://www.sabulo.com/). The mechanics come from a Bolex 8 driven by a continous servo… see detail in this video:
Arduino driving a Bolex 8mm camera film gate - YouTube
1 Like