I didn’t realise the huge amount of posting there is when it comes to the drive system on here going back to 2016. I’m more confused than ever, and the ideas seem to be bouncing between so many different motors and solutions.
I’m hoping someone could correct me if my guess for my own scanner is wrong or has any flaws please
For my own little 8mm film scanner, I’d like to use cheap geared DC motors with an encoder that would be direct drive.
On the supply side I’d add a dancer arm held by a spring with a high resolution encoder or a pot to give feedback to the motor algorithm to slow or speed up the supply motor only
The spring tension would be adjustable, I don’t know if it’s necessary to have something like a solenoid controlling it.
I’d drive the take up motor to keep the film flowing at a set frame rate (16,18,24,25 etc) by measuring the film speed by a roller(either PTR or printed plastic roller with sprocket teeth) mounted directly on a rotary encoder.
I’d hope to add a magnetic sound head in line which is why speed would be important.
I can’t totally see the need for a capstan if the take up reel is spinning very evenly. Those dc gear motors seem to have a lot of inertia and don’t fluctuate much, but maybe the take up reel might not be spooling very evenly and will cause slight flutters in speed.
Or maybe there could just be a roller with a large flywheel from an old projector?
You’re not wrong! Motors have been a consistently slippery thing to nail down. That’s because there are so many ways to tackle the problem.
In your case, if you are going to be concerned with only 8mm, then you are in luck. You can keep it simple. The easiest setup possible is what the old 8mm projectors used to have. A motor on the take-up reel and some manual friction on the feed reel to give it some tension. In the old days this was a spring loop. Those are hard to find these days but you can accomplish the same thing with other materials, or by putting another motor on the feed side and giving it a very low voltage in the opposite direction of the feed.
In this setup, you do not need a capstan.
Where capstans can be helpful is in maintaining a strictly consistent fps rate. It focuses on that while the feed/take-up reels are only worried about maintaining consistent tension on either side of the capstan. It separates the tension into two “sides” (feed and take-up) and lets each reel manage it independently. This has the effect of consistent tension throughout the scan of a reel.
If you are adding a sound head, the capstan is a good design but not entirely necessary as you have mentioned. It may be harder, however to guarantee 24fps with no speed wobble without one. It all depends on your design.
Personally, with all the variables in the motor design I tend to get lost and confused. Tension, fps, torque, etc. In the end, I just have to try things out for myself and see how they work. Usually, the simpler the better. I’m of the philosophy that if it works for you, then it’s good enough.
Feel free to ask any more specific questions and I’ll try to point you in the right direction to other posts or resources.
I was tossing up the idea of making a friction based supply, I own a Moviestuff mkii scanner that uses one, but in order to reverse the scanner both sides need a little electromagnetic clutch so the supply motor can be disengaged to allow the other side to spin freely and one takes up the film, otherwise the gear motor would hold it rock solid.
Those little electromagnetic clutches are quite expensive which turned me off.
The tension does change as the reels empty and fill, and the tension isn’t constant, if you measured it carefully it would probably slip,hold,slip,hold etc
I thought it would be a nice challenge to make a dancer arm too, and if it works it will be a bit cheaper, and the gearbox would be simpler for reverse without having clutches.
Maybe I’ll try without a capstan first as I’ll be struggling enough as it is.
Is a dc gear motor probably the most suitable for a capstan? Or maybe a stepper?
Yeah friction is the easiest and cheapest. And totally works! It’s a great solution in my book. I’m all about whatever works. And if you need something more complicated, you can always upgrade from there later. The friction could be as simple as a timing. belt flipped inside out like I did here for a test a while back.
Yeah a geared DC motor is a great choice. Go stepper if you plan on doing intermittent motion at some point. Otherwise, DC is fine. It’s a tradeoff of how fast your camera’s shutter speed is, how bright the light is, and how fast you want to be able to scan.
In other words, DC = constant motion, which requires fast camera and bright light. Stepper = constant OR intermittent, which will require less speed in the camera and less light since the film is at rest when it is photographed.
Does that make sense?
The tension hub design (aka dancer arm) I was using is a bit more than a single arm design but if it’s of interest to you, you can find it here on the wiki. You could simplify the design for your needs. For example, you can get rid of the PCB and just connect the potentiometer to your arduino directly.
I too am hoping to implement a single arm design, which I sketched out with another forums member over text recently. It will include a sensor to measure tension and be super cheap (I hope)!
